Several lines of evidence indicate that the megakaryocyte/platelet lineage is crucial in myelofibrosis induction. The demonstration that NOD/SCID mice with functionally deficient monocytes do not develop fibrotic changes when exposed to thrombopoietin (TPO) also suggests an important role for monocyte/macrophages. However, in this animal model, the development of myelofibrosis is dependent on the level of TPO. This study was conducted to investigate whether NOD/SCID mice exposed to high TPO levels mediated by a retroviral vector would be refractory to the development of bone marrow fibrosis. We show that TPO and TGF-β1 in plasma from NOD/SCID and SCID mice engrafted with TPO-overexpressing hemopoietic cells reach levels similar to the ones reached in immunocompetent mice, and all animals develop a myeloproliferative disease associated with a dense myelofibrosis at 8 wk posttransplantation. Monocytes in NOD/SCID mice are functionally deficient to secrete cytokines such as IL-1α in response to stimuli, even under TPO expression. Surprisingly, the plasma of these mice displays high levels of IL-α, which was demonstrated to originate from platelets. Together, these data suggest that completely functional monocytes are not required to develop myelofibrosis and that platelets are able, under TPO stimulation, to synthesize inflammatory cytokines, which may be involved in the pathogenesis of myelofibrosis and osteosclerosis.

Fibrosis is a prominent clinical complication in several disorders and occurs particularly in the lung, kidney, heart, or liver. In human hemopoietic tissues, myelofibrosis is less frequent but is typically associated with two distinct clinical entities: chronic idiopathic myelofibrosis (IM)4 with myeloid metaplasia and acute panmyelosis with myelofibrosis (1). Bone marrow fibrosis occurs as a cytokine-mediated secondary response to a clonal malignant event originating in a multipotent hemopoietic stem cell (2, 3) and is characterized by excessive deposits of extracellular matrix proteins (4). In vivo and in vitro studies have involved several cytokines in the aberrant stromal reaction, with a strong emphasis on the pleiotropic cytokine TGF-β1. Through fibroblast stimulation, this cytokine is able to produce extracellular matrix, induce cell adhesion proteins, and enhance expression of proteases that inhibit enzymes involved in the degradation of the extracellular matrix. TGF-β1 is secreted by numerous cell lineages, but mainly by monocytes/macrophages and megakaryocytes (MKs)/platelets. Several lines of evidence obtained both from studies of patients with IM (5, 6) or murine models ending with myelofibrosis (7, 8, 9) are in favor of a crucial role of MK in myelofibrosis induction. However, other studies suggested a central role for the monocyte/macrophage lineage. This theory is based on two different observations: 1) Rameshwar et al. (10) reported that monocytes from patients with IM are spontaneously activated and secreted abnormally TGF-β1; and 2) Frey et al. (11) reported that thrombopoietin (TPO) overexpression in SCID mice led to myelofibrosis, but not in NOD/SCID mice. Because NOD/SCID mice differ from SCID mice by impaired mononuclear phagocyte functions (12, 13), the authors postulated that monocytes and macrophages must be required for promotion of myelofibrosis in TPO-overexpressing mice. However, these experiments were performed with a human TPO cDNA vectorized in an adenovirus construct that leads to a relatively low elevation of plasma TPO. In the present study, to ensure a high and sustained TPO level, we used a retroviral vector encoding a mouse TPO cDNA to overexpress TPO in NOD/SCID, SCID, and immunocompetent wild-type (WT) mice. All TPOhigh mice invariably developed a myeloproliferative syndrome associated with a myelofibrosis and osteosclerosis. At any time during the follow-up, TPO and TGF-β1 plasma levels were substantially and similarly elevated in the three groups of mice. Our data provide evidence that the monocyte/macrophage lineage does not play a crucial role in the development of the myelofibrosis induced by TPO. In addition, platelets and MKs when stimulated are able to synthesize IL-1α and thus to replace monocytes in the synthesis of inflammatory cytokines, which may be involved in the pathogenesis of myelofibrosis and osteosclerosis.

All procedures were approved by the local Institut Gustave Roussy ethics committee. WT C57BL/6J mice (Janvier) were maintained at the Institute Gustave Roussy animal facility under specific pathogen-free conditions. SCID mutant mice (C57BL/6J-scid/scid) (14) (Charles River Laboratories) and double NOD/SCID mutant mice (15) (NOD/LtSz-scid/scid obtained from J. Dick (University Health Network, Toronto, Ontario, Canada)) were bred under sterile conditions and maintained in isolators. All mice used in this study have the same C57BL/6J genetic background.

Six- to 8-wk-old WT, SCID, or NOD/SCID male mice were used as bone marrow donors. Seven- to 10-wk-old WT, SCID, or NOD/SCID female mice were recipients. The infection procedure was performed as described previously (7, 16). Briefly, 4 days after 5-fluorouracil treatment (one injection of 150 mg/kg administered i.p.), total bone marrow cells were cocultured with 1 × 106 MPZenTPO virus-producing GP+E-86 cells in DMEM supplemented with antibiotics, 20% FBS and murine IL-3 (100 U/ml), murine IL-6 (20 ng/ml), and murine stem cell factor (20 ng/ml). All cytokines were purchased from R&D Systems. After 4 days, nonadherent cells were harvested. An aliquot was used immediately in clonogenic progenitor assays to determine the percentage of infected colony-forming cells (CFC) as described previously (7, 16). The remaining cells were inoculated i.v. via the retro-orbital sinus into irradiated hosts (9.5 Gy for WT mice or 2.5 Gy for SCID and NOD/SCID mice, x-ray apparatus, single dose) in a ratio of two donors per one recipient. Fifty-one recipients were studied.

Two approaches were used in parallel. First, a PCR analysis was performed on myeloid colonies derived from the bone marrow of sacrificed animals at week 8 after transplantation, with primer sets corresponding to specific sequences on the Y chromosome and with actin primers as control (16). The second approach was a fluorescent in situ hybridization (FISH) analysis performed on bone marrow cells from mice sacrificed at week 8 after transplantation. Briefly, 1 × 105 bone marrow cells were cytospun and fixed with a 4% formaldehyde solution. Slides were successively incubated for 5 min in 0.1 M HCl and three times in a permeabilization solution containing 0.5% saponin and 0.5% Triton X-100 (Sigma-Aldrich), and then washed in a solution containing 20% glycerol. After three freezing/thawing, slides were incubated in 20% glycerol, deshydrated with ethanol, incubated in a 70% formamide solution at 75°C, and deshydrated again. A denatured rhodamine-labeled Y probe (purchased by QBiogene) was hybridized overnight at 37°C in a humidified atmosphere. The slides were incubated in a 60% formamide solution at 43°C, washed in SSC 0.1% Tween 20, and counterstained with Vectashield and 4′,6′-diamidino-2-phenylindole (Vector Laboratories). Numeration of Y-labeled cells was performed on 500 cells for each slide using a fluorescence microscope (Nikon Eclipse 600).

Blood from the orbital plexus was collected in citrated tubes at indicated times. Nucleated blood cells, hematocrit level, and platelet counts were determined using an automated blood counter calibrated for mouse blood (MS9; Schloesing Melet). Differential cell counts were performed after May-Grünwald-Giemsa staining. Platelet-poor plasma was used for determination of TPO, IL-1α, IL-1β, and TGF-β1 levels. Eight weeks after transplantation, mice in each group were killed. Bones were excised and cleaned of soft tissue. One femur and one tibia were fixed in formaldehyde, decalcified, and paraffin embedded. Spleen, liver, and pulmonary samples were fixed and embedded in the same manner. Sections (4.5 μm) were stained with H&E, periodic acid-Schiff, and Giemsa for overall cytology analysis. Reticulin fibers were revealed by silver staining according to Gordon Sweet method. In parallel, 105 bone marrow, spleen, and blood cells were grown in semisolid medium for CFC analysis as described previously (16, 17).

To detect and quantify MK apoptosis, we performed in situ TUNEL analysis on histological sections of bone marrow and spleen from control mice (WT, SCID, and NOD/SCID) or TPOhigh mice (WT, SCID, and NOD/SCID). DNA strand breaks generated during apoptosis were identified using In Situ Cell Death Detection Kit (POD; Roche) according to the manufacturer’s instructions. Samples were analyzed under a light microscope, and quantification of apoptotic MKs was performed. Results were expressed as the percentage of apoptotic MKs among the total MK population.

Bone marrow cells were harvested from femurs and tibia of age- and sex-matched control or TPOhigh mice (WT and NOD/SCID). Cells were cultured (1.6 × 106 cells/ml), as described previously (18), in DMEM supplemented with antibiotics, 10% FBS, and 10% CSF-1-conditioned medium (provided by M. Lebastard, Unité Immunophysiologie et Parasitisme Intracellulaire, Institut Pasteur, Paris, France) with and without 10 U/ml IFN-γ. Bone marrow cells were cultured at 37°C in a humidified incubator (5% CO2) with an addition of fresh medium at day 3. After 5 days in culture, nonadherent cells were removed and centrifuged for 10 min at 400 × g. Pellets were suspended in DMEM with antibiotics and 10% FBS by flushing through, successively, 25-, 27-, and 30-gauge needles, and single-cell suspensions were produced. Cells (2 × 105 cells/ml) were seeded in fresh medium alone (DMEM supplemented with antibiotics, 10% FBS, and 10% CSF-1-conditioned medium) or in fresh medium containing Escherichia coli LPS at 10 μg/ml. After a 24-h incubation period, culture supernatants were harvested and used for determination of IL-1α level.

Mice were anesthetized and bled by cardiac puncture. Blood from age- and sex-matched control or TPOhigh mice (WT and NOD/SCID) was collected onto 3.8% Na citrate (9 vol blood/1 vol citrate). Whole blood (500 μl) was mixed with 500 μl of buffer saline glucose citrate (BSGC) (129 mM NaCl, 13.6 mM Na3 citrate, 11.1 mM D+ glucose, 1.6 mM K2PO4, and 8.6 mM NaH2PO4) and spun at 400 × g. Platelet-rich plasma was layered over a 10-ml Sepharose B column (Pharmacia), and pellets were used for leukocyte RNA analysis after red cell lysis. To minimize leukocyte contamination, only the upper (9 of 10) platelet-rich plasma was used for gel filtration. Platelets were collected, adjusted at a concentration of 50 × 106/ml, and pelleted by centrifugation. Pellets were used for protein or RNA analysis. For protein analysis, platelets were lysed by three successive freezing/thawings, suspended in 1 ml of BSGC, and centrifuged to remove platelet debris. Supernatants were used to determine IL-1α level.

Plasma TPO levels and IL-1α and IL-1β levels in plasma or supernatants were determined with the appropriate murine Quantikine Kits from R&D Systems, according to manufacturer’s instructions. Sensitivity limits of the assays were 62.5, 4.69, and 7.8 pg/ml, respectively, for TPO, IL-1α, and IL-1β. Human TGF-β1 immunoassay, which detects only active forms of TGF-β1, was used for circulating TGF-β1 level determination (Quantikine Kit; R&D Systems). Samples were assayed after acidification (active and latent forms) according to manufacturer’s instructions. The sensitivity of the assay was 31.2 pg/ml active TGF-β1.

Total RNA extraction from circulating platelets and leukocytes of age- and sex-matched control or TPOhigh WT mice was performed using either TRIzol reagent (Invitrogen Life Technologies) or the RNeasy Mini kit (Qiagen), following the manufacturer’s recommendations. Total RNA was reverse transcribed in first-strand cDNA using random hexamers (Roche Diagnostics) and Moloney Murine Leukemia Virus Reverse Transcriptase (Invitrogen Life Technologies).

PCR amplifications were conducted on the LightCycler (Roche Diagnostics) in duplicate or triplicate capillaries in a final volume of 11 μl. Two microliters of cDNA along with primers (500 nM; Roche Diagnostics) and SYBG Green containing mix (QuantiTect SYBR Green Kit; Qiagen) were used per reaction. CD45-, CD11b-, IL-1α-, IL-1β-, and TGF-β1-specific primers were obtained from published literature or designed by the LightCycler Probe Design software (version 1.0; Idaho Technology). Identical thermal cycling conditions were used for all targets, allowing efficient synthesis of all amplicons (Table I). Crossing Points were determined with the FitPoint method (LightCycler software, version 3.3) at a constant fluorescent level in the early stage of the exponential phase of the reaction. Using the tool Gene Expression Macro from Bio-Rad, relative expression values were expressed for each sample compared with a sample calibrator (a pool of platelets from control WT mice). The underlying principles and calculations are described in studies by VandeSompele et al. (19). Briefly, the Bio-Rad tool calculates a gene expression normalization factor for each sample based on the geometric mean of a user-defined number of housekeeping genes. To allow a better normalization between samples within this study, three housekeeping genes were used, namely β-actin, GAPDH (21), and integrin αIIb/CD41. Then, taking into account the PCR efficiencies for each target mRNA as described in the mathematical model developed by M. W. Pfaffl (22), the algorithm provided values for each sample relative to the sample calibrator.

Table I.

Primers and amplicons for real-time PCR analysisa

Gene NameAccession no.SourcePrimer Sequences (5′-3′)Amplicons
TmLength (bp)
GAPDH NM_001001303 Lean et al. (21FW CggATTTggCCgTATTgg 83.4 218 
   RV ggTCTCgCTCCTggAAgATg   
β-actin NM_007393 Colle et al. (20FW ggACTCCTATgTgggTgACgAgg 84.8 366 
   RV gggAgAgCATAgCCCTCgTAgAT   
αIIb/CD41 NM_010575 Probe Design FW AgTTCTCCgTCTATgCAg 82.5 123 
   RV CTTgACTTgCgTTTAggg   
IL-1α NM_010554 Probe Design FW gTATgCCTACTCgTCgg 80.7 235 
   RV ggTggCAATAAACAgCTC   
IL-1β NM_008361 Probe Design FW AggCAggCAgTATCAC 81.4 167 
   RV CACACCAgCAggTTATC   
TGF-β1 NM_011577 Probe Design FW gCggACTACTATgCTAAAgA 82 243 
   RV gTAACgCCAggAATTgT   
CD11b NM_008401 Probe Design FW TTTAggAgCACCTCgg 82.8 223 
   RV ACACACTgACACCTgg   
CD45 NM_011210 Probe Design FW TATgACATCATCgCCAgC 79.2 127 
   RV CTggACggACACAgTT   
Gene NameAccession no.SourcePrimer Sequences (5′-3′)Amplicons
TmLength (bp)
GAPDH NM_001001303 Lean et al. (21FW CggATTTggCCgTATTgg 83.4 218 
   RV ggTCTCgCTCCTggAAgATg   
β-actin NM_007393 Colle et al. (20FW ggACTCCTATgTgggTgACgAgg 84.8 366 
   RV gggAgAgCATAgCCCTCgTAgAT   
αIIb/CD41 NM_010575 Probe Design FW AgTTCTCCgTCTATgCAg 82.5 123 
   RV CTTgACTTgCgTTTAggg   
IL-1α NM_010554 Probe Design FW gTATgCCTACTCgTCgg 80.7 235 
   RV ggTggCAATAAACAgCTC   
IL-1β NM_008361 Probe Design FW AggCAggCAgTATCAC 81.4 167 
   RV CACACCAgCAggTTATC   
TGF-β1 NM_011577 Probe Design FW gCggACTACTATgCTAAAgA 82 243 
   RV gTAACgCCAggAATTgT   
CD11b NM_008401 Probe Design FW TTTAggAgCACCTCgg 82.8 223 
   RV ACACACTgACACCTgg   
CD45 NM_011210 Probe Design FW TATgACATCATCgCCAgC 79.2 127 
   RV CTggACggACACAgTT   
a

Accession nos. of housekeeping and target genes along with characteristics of primers and amplification products are depicted. The LightCycler run protocol used in this study included a denaturation and Taq polymerase activation program (95°C for 15 min), an amplification program repeated 40 times with a temperature transition rate of 20°C/s (denaturation step, 95°C for 10 s; annealing step, 54°C for 30 s; elongation step, 72°C for 25 s with a single fluorescent measurement), a melting curve program (72–95°C with a heating rate of 0.1°C/s and a continuous fluorescent measurement), and finally a cooling step to 35°C. FW, Forward primer; RV, reverse primer; Tm, melting temperature is the temperature, where 50% of the DNA is single stranded.

Results are presented as mean ± SEM, and data were analyzed with the two-tailed Student t test.

The different x-ray irradiation protocols used for SCID and NOD/SCID mice compared with WT may lead to different levels of chimerism and subsequently different pathological processes in hemopoietic tissues. We analyzed the host chimerism in sex-mismatched transplantations using two approaches: PCR on Y chromosome (donors were males, recipients were females) on bone marrow-derived myeloid colonies, and FISH on Y chromosome analysis on whole nucleated bone marrow cells. Chimerism levels were similar in TPOhigh WT, SCID, and NOD/SCID mice when studied on myeloid colonies 8 wk posttransplantation (69 ± 18% (n = 3 mice), 62 ± 15% (n = 4 mice), and 67 ± 16% (n = 5 mice), respectively). Moreover, studies by FISH on marrow cells gave identical results as those from PCR.

To obtain a highly elevated and long-lasting overexpression of a mouse TPO molecule in immunodeficient SCID and NOD/SCID mice, we used a previously described retroviral gene transfer protocol in which lethally irradiated hosts are hematologically repopulated with transduced stem cells (7). Because SCID and NOD/SCID mice could not be heavily irradiated before bone marrow transplantation (23), and to overcome a possible risk of a low chimerism, 10 times more cells (6–8 × 106 cells) were engrafted into 2.5 Gy-irradiated SCID and NOD/SCID mice as compared with 6 × 105 cells injected in 9.5 Gy-irradiated WT mice. Transduction efficiencies were determined at the end of the infection protocol before transplantation in myeloid progenitor-derived colonies. For each mouse genotype, two independent infection experiments were performed with a total of 51 hosts engrafted. As shown from PCR analysis, the percentage of colonies demonstrating the integrated TPO cDNA was comparable in the three groups of mice (Table II). Plasma levels of TPO were determined in the three groups of engrafted mice over time using an ELISA. Six weeks postengraftment, TPO concentration in plasma was 1,000- to 10,000-fold higher than normal. Thereafter, the magnitude of the increase was comparable in the three groups (Fig. 1).

Table II.

Analysis of transduction efficiency in progenitor cells injected to WT, SCID, and NOD/SCID micea

Bone Marrow CellsTPO/ActinTransduced (%)
WT 20/23 87 
SCID 65/73 89 
NOD-SCID 66/74 89 
Bone Marrow CellsTPO/ActinTransduced (%)
WT 20/23 87 
SCID 65/73 89 
NOD-SCID 66/74 89 
a

WT, SCID, and NOD/SCID pooled bone marrow cells were transduced and used to reconstitute 10 WT, 19 SCID, and 22 NOD/SCID. Each pool was plated at the end of the infection protocol, and CFC-derived colonies were picked from methylcellulose. Samples were analysed by PCR with specific primers for the viral TPO gene and with actin primers to ascertain the presence of material. No significant differences were observed between the three groups of mice with more than 80% transduced progenitor cells.

FIGURE 1.

TPO quantification in plasma. Results are presented as the mean ± SEM of four to 10 animals per experimental group. Values in control WT, SCID, and NOD/SCID were 0.78 ± 0.04, 0.56 ± 0.07, and 1.05 ± 0.08 ng/ml, respectively. Results of statistical analysis with the two-tailed Student t test are as follows: SCID and NOD/SCID mice vs WT mice; ∗, p < 0.05 and ∗∗, p < 0.001.

FIGURE 1.

TPO quantification in plasma. Results are presented as the mean ± SEM of four to 10 animals per experimental group. Values in control WT, SCID, and NOD/SCID were 0.78 ± 0.04, 0.56 ± 0.07, and 1.05 ± 0.08 ng/ml, respectively. Results of statistical analysis with the two-tailed Student t test are as follows: SCID and NOD/SCID mice vs WT mice; ∗, p < 0.05 and ∗∗, p < 0.001.

Close modal

Platelet numbers in TPOhigh WT mice increased over 8 wk, achieving values 4-fold higher than control WT mice. TPOhigh SCID mice present the same features. In TPOhigh NOD/SCID mice, platelet numbers increased less significantly with barely a 2-fold increment over controls (Fig. 2,A). Macroplatelets were observed in all mice for which blood smears were performed. As previously reported, no correlation between TPO levels and platelet numbers was observed (17). Nucleated blood cells were increased in all groups of mice (Fig. 2,B) due to a striking increment in mature polymorphonuclear neutrophils in association with immature myeloid precursor cells (data not shown). Leukocyte numbers were more elevated in TPOhigh WT and SCID mice than in TPOhigh NOD/SCID mice, without significant difference. TPOhigh WT and NOD/SCID mice became progressively anemic; however, TPOhigh SCID mice spontaneously corrected this anemia over time (Fig. 2 C).

FIGURE 2.

Blood parameters and TGF-β1 level of TPOhigh mice. Evolution in platelet number (A), leukocyte number (B), hematocrit (C), and TGF-β1 level (D) in TPOhigh mice. Results are presented as the mean ± SEM of four to 10 animals per experimental group. Some WT mice displayed major leukocytosis (233 × 109/l), explaining large SEM. For the TGF-β1 level (D), bars indicate TGF-β1 levels after acidification of the samples. Values in control WT, SCID, and NOD/SCID were 7.9 ± 2.0, 9.4 ± 1.4, and 8.7 ± 2.1 ng/ml, respectively. No spontaneously active TGF-β1 was detected before acidification of the samples. Results of statistical analysis with the two-tailed Student t test are as follows: SCID and NOD/SCID mice vs WT mice; ∗, p < 0.05 and ∗∗, p < 0.001.

FIGURE 2.

Blood parameters and TGF-β1 level of TPOhigh mice. Evolution in platelet number (A), leukocyte number (B), hematocrit (C), and TGF-β1 level (D) in TPOhigh mice. Results are presented as the mean ± SEM of four to 10 animals per experimental group. Some WT mice displayed major leukocytosis (233 × 109/l), explaining large SEM. For the TGF-β1 level (D), bars indicate TGF-β1 levels after acidification of the samples. Values in control WT, SCID, and NOD/SCID were 7.9 ± 2.0, 9.4 ± 1.4, and 8.7 ± 2.1 ng/ml, respectively. No spontaneously active TGF-β1 was detected before acidification of the samples. Results of statistical analysis with the two-tailed Student t test are as follows: SCID and NOD/SCID mice vs WT mice; ∗, p < 0.05 and ∗∗, p < 0.001.

Close modal

The number of progenitor cells in bone marrow, spleen, and blood was studied at week 8 posttransplantation in the three groups of mice. Bone marrow progenitor numbers were 3- to 5-fold decreased in all TPOhigh mice in comparison to controls, parallel to decrease of total bone marrow cell number. All TPOhigh mice developed a splenomegaly. Probably due to the myeloproliferative disorder, spleen CFC increased achieving values 22-fold higher than control in TPOhigh WT mice, 17-fold higher than control in TPOhigh SCID mice, but only 2-fold higher than control in TPOhigh NOD/SCID mice. Finally, blood-circulating CFC were 30- to 1,500-fold increased at 8 wk posttransplantation (Table III).

Table III.

CFC analysis of TPOhigh mice 8 wk postengraftment with MPZen-TPO-bone marrow cellsa

MiceTissue
Bone marrowSpleenBlood
Cells × 106/femurCFC × 103/femurWeight, mgCFC/106 cellsNucleated cells × 106/mlCFC × 102/ml
Control WT 14.8 ± 2.4 21.7 ± 3.4 91 ± 9 90 ± 31 6.6 ± 0.7 <1 
Control SCID 12.2 ± 2.0 14.9 ± 0.8 52 ± 3 150 ± 39 3.5 ± 0.4 1 ± 0.1 
Control NOD/SCID 10.3 ± 1.0 17.9 ± 2.7 54 ± 5 753 ± 195 2.1 ± 0.4 4.5 ± 1.6 
TPOhigh WT 3.1 ± 1.0 4.4 ± 0.8 894 ± 141 1950 ± 579 84.9 ± 23 1146 ± 627 
TPOhigh SCID 1.1 ± 0.1 3.5 ± 1.2 384 ± 63 3033 ± 501 100 ± 31 2335 ± 1450 
TPOhigh NOD/SCID 5.1 ± 0.9 5.2 ± 0.3 366 ± 52 1126 ± 147 32.3 ± 4.3 140 ± 25 
MiceTissue
Bone marrowSpleenBlood
Cells × 106/femurCFC × 103/femurWeight, mgCFC/106 cellsNucleated cells × 106/mlCFC × 102/ml
Control WT 14.8 ± 2.4 21.7 ± 3.4 91 ± 9 90 ± 31 6.6 ± 0.7 <1 
Control SCID 12.2 ± 2.0 14.9 ± 0.8 52 ± 3 150 ± 39 3.5 ± 0.4 1 ± 0.1 
Control NOD/SCID 10.3 ± 1.0 17.9 ± 2.7 54 ± 5 753 ± 195 2.1 ± 0.4 4.5 ± 1.6 
TPOhigh WT 3.1 ± 1.0 4.4 ± 0.8 894 ± 141 1950 ± 579 84.9 ± 23 1146 ± 627 
TPOhigh SCID 1.1 ± 0.1 3.5 ± 1.2 384 ± 63 3033 ± 501 100 ± 31 2335 ± 1450 
TPOhigh NOD/SCID 5.1 ± 0.9 5.2 ± 0.3 366 ± 52 1126 ± 147 32.3 ± 4.3 140 ± 25 
a

Values are means ± SEM of data from three controls and three TPOhigh mice in each group. Progenitor cells numbers (CFC) were calculated from the number of colonies obtained from 105 cells grown in semisolid medium.

These data indicate that overproduction of TPO in hosts repopulated with WT, SCID, and NOD/SCID hemopoietic cells resulted in a similar myeloproliferative syndrome. A similar feature was observed at the histological examination. However, some hematological differences were observed. A lower increase in all cell types was observed in TPOhigh NOD/SCID mice compared with TPOhigh WT and SCID mice.

Histological sections revealed a massive hyperplasia of maturing granulocytic cells and dysmorphic MKs found in large clusters. Erythroblasts were rare. Increased reticulin fibers were noted in bone marrow samples at 8 wk postengraftment. No differences were observed between the TPOhigh WT, SCID, or NOD/SCID mice (Fig. 3). Osseous cortex was a bit thicker in TPOhigh NOD/SCID than in TPOhigh SCID or WT, but some bone trabeculae were seen in the marrow cavity in the three lineages. In spleen samples, hyperplasia of the red pulp was observed without white pulp in TPOhigh SCID and NOD/SCID. MKs were extremely abundant as well as neutrophils. Numerous MKs were observed in large sheets in TPOhigh WT in the red pulp and in TPOhigh NOD/SCID (more frequently than in TPOhigh SCID mice). Apoptotic figures of MKs were often seen in TPOhigh WT and SCID mice. Erythroid cells were observed. Similar reticulin deposits were seen 8 wk posttransplantation in all TPOhigh mice analyzed.

FIGURE 3.

Studies of myelofibrosis and apoptosis on histological sections of femur and spleen. Three mice of each group were killed 8 wk after transplantation. Representative sections are shown. Longitudinal femur sections stained by H & E from TPOhigh mice WT 2 (A), SCID 2 (C), and NOD/SCID 2 (E) show the hyperplasia of MK and granulocytes. Silver staining of longitudinal femur sections from mice WT 2 (B), SCID 2 (D), and NOD/SCID 2 (F) reveals similar myelofibrosis with reticulin fibers surrounded MKs. Original magnification, ×250. Sections of spleen from WT control (G) and TPOhigh WT (H) after TUNEL technology. Apoptotic MKs are stained (black arrow). Original magnification, ×600.

FIGURE 3.

Studies of myelofibrosis and apoptosis on histological sections of femur and spleen. Three mice of each group were killed 8 wk after transplantation. Representative sections are shown. Longitudinal femur sections stained by H & E from TPOhigh mice WT 2 (A), SCID 2 (C), and NOD/SCID 2 (E) show the hyperplasia of MK and granulocytes. Silver staining of longitudinal femur sections from mice WT 2 (B), SCID 2 (D), and NOD/SCID 2 (F) reveals similar myelofibrosis with reticulin fibers surrounded MKs. Original magnification, ×250. Sections of spleen from WT control (G) and TPOhigh WT (H) after TUNEL technology. Apoptotic MKs are stained (black arrow). Original magnification, ×600.

Close modal

To study whether myelofibrosis could be the consequence of a dysmegakaryopoiesis, we studied whether apoptotic MKs were increased in TPOhigh mice. Apoptotic MKs were detected using TUNEL analysis on histological sections of both bone marrow and spleen from control and TPOhigh mice (WT, SCID, and NOD/SCID). Percentages of apoptotic MKs are illustrated in Table IV. Apoptosis of bone marrow and spleen MKs was increased in all TPOhigh mice in comparison to controls. Representative sections of spleen from WT control (Fig. 3,G) and TPOhigh WT (Fig. 3,H) mice are shown in Fig. 3.

Table IV.

TUNEL analysis of apoptotic megakaryocytes in bone marrow and spleen of TPOhigh mice 8 wk postengraftment with MPZen-TPO-bone marrow cellsa

MiceBone Marrow (%)Spleen (%)
Control WT 3/164 (1.8) 0/27 (0) 
Control SCID 6/226 (2.7) 6/315 (1.9) 
Control NOD/SCID 7/486 (1.4) 3/85 (3.5) 
TPOhigh WT 32/992 (3.2) 52/1478 (3.5) 
TPOhigh SCID 42/716 (5.9) 67/783 (8.6) 
TPOhigh NOD/SCID 63/981 (6.4) 57/862 (6.6) 
MiceBone Marrow (%)Spleen (%)
Control WT 3/164 (1.8) 0/27 (0) 
Control SCID 6/226 (2.7) 6/315 (1.9) 
Control NOD/SCID 7/486 (1.4) 3/85 (3.5) 
TPOhigh WT 32/992 (3.2) 52/1478 (3.5) 
TPOhigh SCID 42/716 (5.9) 67/783 (8.6) 
TPOhigh NOD/SCID 63/981 (6.4) 57/862 (6.6) 
a

Apoptosis of MK was detected using TUNEL analysis on histological sections of both bone marrow and spleen from control and TPOhigh mice. Results are presented as number of apoptotic stained MKs per number of total MKs (percentage of apoptotic MK).

In this TPOhigh mouse model, fibrosis development has been reported to be a direct consequence of high TGF-β1 level in blood and bone marrow fluids (17). TGF-β1 level was thus measured in the three groups of TPOhigh mice at different times posttransplantation. As reported before in an immunocompetent background (16), TGF-β1 levels were invariably increased in plasma as soon as 3 wk after transplantation (data not shown). These levels increased with time, leading to a 2- to 4-fold higher level than in control mice after 6 wk (Fig. 2 D). Such an increase in TGF-β1 level was also noticed in TPOhigh SCID mice but was detected later in the TPOhigh NOD/SCID mice background (2-fold higher level than normal at 8 wk only).

Because it has been reported that TGF-β1 is regulated by IL-1 in human IM (24), and that NOD/SCID monocytes have a major defect in the synthesis of IL-1 (12), we investigated the IL-1α and IL-1β levels in these mice. IL-1α and IL-1β were both quantified in plasma with specific ELISA. Surprisingly, we found a dramatic increase (up to 8-fold the normal value) in IL-1α levels in all TPOhigh mice analyzed at 6 and 8 wk posttransplantation (Fig. 4 A). However, IL-1β was detected only 8 wk posttransplantation (data not shown).

FIGURE 4.

IL-1α quantifications. A, IL-1α quantification in plasma. Results are presented as the mean ± SEM of four to 10 animals per experimental group. Values in control WT, SCID, and NOD/SCID were 73 ± 8, 68 ± 9, and 29 ± 4 pg/ml, respectively. Results of statistical analysis with the two-tailed Student t test are as follows: SCID and NOD/SCID mice vs WT mice; ∗, p < 0.05 and ∗∗, p < 0.001. B, IL-1α quantification in platelets. Platelets from control WT and NOD/SCID mice or from TPOhigh WT and NOD/SCID mice were collected and lysed by three freezing/thawing. The lysates (50 × 106 platelets/1 ml) were harvested and assayed for IL-1α levels. Results are presented as the mean ± SEM of three animals per experimental group. IL-1α platelet levels were low for WT mice and under detection threshold for NOD/SCID mice. However, TPOhigh mice, particularly NOD/SCID, showed marked elevation in IL-1α platelet levels. Results of statistical analysis with the two-tailed Student t test are as follows: TPOhigh mice vs control mice.

FIGURE 4.

IL-1α quantifications. A, IL-1α quantification in plasma. Results are presented as the mean ± SEM of four to 10 animals per experimental group. Values in control WT, SCID, and NOD/SCID were 73 ± 8, 68 ± 9, and 29 ± 4 pg/ml, respectively. Results of statistical analysis with the two-tailed Student t test are as follows: SCID and NOD/SCID mice vs WT mice; ∗, p < 0.05 and ∗∗, p < 0.001. B, IL-1α quantification in platelets. Platelets from control WT and NOD/SCID mice or from TPOhigh WT and NOD/SCID mice were collected and lysed by three freezing/thawing. The lysates (50 × 106 platelets/1 ml) were harvested and assayed for IL-1α levels. Results are presented as the mean ± SEM of three animals per experimental group. IL-1α platelet levels were low for WT mice and under detection threshold for NOD/SCID mice. However, TPOhigh mice, particularly NOD/SCID, showed marked elevation in IL-1α platelet levels. Results of statistical analysis with the two-tailed Student t test are as follows: TPOhigh mice vs control mice.

Close modal

To test whether TPO overexpression could correct the IL-1 synthesis defect of NOD/SCID monocytes, monocytes were cultured from control WT, control NOD/SCID, TPOhigh WT, and TPOhigh NOD/SCID bone marrow isolated cells in the presence of CSF-1 with or without IFN-γ. After 5 days, LPS was added (10 μg/ml), and 24 h later, the culture supernatants were harvested and assayed for IL-1α level. Monocytes from NOD/SCID mice secreted less IL-1α in response to IFN-γ stimulation compared with WT monocytes as described previously (12). TPO overexpression did not improve IL-1α secretion, indicating that monocyte dysfunction in NOD/SCID mice was not corrected by TPO overexpression (Fig. 5).

FIGURE 5.

IL-1α secretion by bone marrow-derived monocytes in response to IFN-γ and LPS. Monocytes from control WT and NOD/SCID mice or from TPOhigh WT and NOD/SCID mice were cultivated in CSF-1 with or without IFN-γ. After 5 days, the culture medium was replaced with fresh medium containing LPS (10 μg/ml). After an additional 24-h incubation period, the cultured supernatants were harvested and assayed for IL-1α levels. Monocytes from NOD/SCID mice secreted less IL-1α in response to IFN-γ stimulation compared with WT monocytes as described previously (12 ). TPO overexpression does not improve IL-1α secretion. Results are presented as the mean ± SEM of three animals per experimental group. Bone marrow cells from each mouse were assayed separately in duplicate culture. Results of statistical analysis with the two-tailed Student t test are as follows: CSF-1 with IFN-γ vs CSF-1 alone.

FIGURE 5.

IL-1α secretion by bone marrow-derived monocytes in response to IFN-γ and LPS. Monocytes from control WT and NOD/SCID mice or from TPOhigh WT and NOD/SCID mice were cultivated in CSF-1 with or without IFN-γ. After 5 days, the culture medium was replaced with fresh medium containing LPS (10 μg/ml). After an additional 24-h incubation period, the cultured supernatants were harvested and assayed for IL-1α levels. Monocytes from NOD/SCID mice secreted less IL-1α in response to IFN-γ stimulation compared with WT monocytes as described previously (12 ). TPO overexpression does not improve IL-1α secretion. Results are presented as the mean ± SEM of three animals per experimental group. Bone marrow cells from each mouse were assayed separately in duplicate culture. Results of statistical analysis with the two-tailed Student t test are as follows: CSF-1 with IFN-γ vs CSF-1 alone.

Close modal

Lindeman et al. (25) have reported that human platelets synthesize and secrete IL-1β in response to activation stimuli. Because we observed an increase in IL-1α level in TPOhigh NOD/SCID mice that have deficient monocytes, we hypothesized that platelets from these mice could also synthesize high levels of IL-1α in response to TPO.

Platelets from control WT and NOD/SCID mice or from TPOhigh WT and NOD/SCID mice were collected and lysed. Lysates (50 × 106 platelets/ml) were harvested and assayed for IL-1α concentration. Levels of IL-1α in platelets from control WT mice were low and below the threshold of detection in control NOD/SCID mice (Fig. 4 B). However, when platelets were collected from TPOhigh WT and NOD/SCID mice, a marked elevation in IL-1α content in platelets was detected, indicating that TPO could regulate IL-1α. Using this approach, we failed to detect IL-1β in these platelet lysates.

Quantitative RT-PCR on Sepharose column-purified platelets from control WT and TPOhigh WT mice was performed to determine whether TPO overexpression could be responsible for IL-1α and TGF-β1 protein overexpression. IL-1α, IL-1β, and TGF-β1 quantification was performed on two independent WT platelet pools used as “calibrator” and “control” (pools were performed to lower interindividual variations, and because RNA from nonthrombocythemic mice was obtained in low quantity) and in five individual TPOhigh WT mice, TPOhigh1–TPOhigh5. Fold changes are expressed relative to calibrator. The normalization factor used for each sample was based on the geometric mean of three housekeeping genes (β-actin, GAPGH, and αIIb).

To check potential leukocyte contamination, we used CD45- and CD11b-specific primers (two highly expressed genes in purified leukocytes from the same animals). No transcript was detected, illustrating a very low presence or absence of leukocyte. A 4- to 12-fold overexpression of IL-1α was found in the five TPOhigh mice (Fig. 6). This observation was strictly correlated to the protein overexpression determined in platelet extracts from the same animals. A slight increase in TGF-β1 was also found in these mice. According to the lack of IL-1β protein, no increase in IL-1β transcript was found.

FIGURE 6.

Real-time PCR analysis of IL-1α, IL-1β, and TGF-β1 transcripts in platelets from control or TPOhigh WT mice. Fold changes of IL-1α, IL-1β, and TGF-β1 transcripts in various individual platelet samples from TPOhigh WT mice (TPOhigh1–TPOhigh5) are expressed relative to a pool of platelets from control WT mice (calibrator). Control is a second pool sample from control WT mice, which is different from the calibrator. The normalization factor used for each sample was based on the geometric mean of three housekeeping genes (β-actin, GAPGH, and αIIb). Data processing and presentation were realized using the Gene Expression Macro (Microsoft Excel Applet) and SigmaPlot 2004 software (Systat), respectively.

FIGURE 6.

Real-time PCR analysis of IL-1α, IL-1β, and TGF-β1 transcripts in platelets from control or TPOhigh WT mice. Fold changes of IL-1α, IL-1β, and TGF-β1 transcripts in various individual platelet samples from TPOhigh WT mice (TPOhigh1–TPOhigh5) are expressed relative to a pool of platelets from control WT mice (calibrator). Control is a second pool sample from control WT mice, which is different from the calibrator. The normalization factor used for each sample was based on the geometric mean of three housekeeping genes (β-actin, GAPGH, and αIIb). Data processing and presentation were realized using the Gene Expression Macro (Microsoft Excel Applet) and SigmaPlot 2004 software (Systat), respectively.

Close modal

IM is a myeloproliferative disease known as a stem cell disorder with involvement of both the myeloid and the lymphoid lineage (3), whereas the associated stromal reaction in the bone marrow environment is considered reactive and cytokine mediated (2). The current pathogenetic hypothesis is that clonal proliferation of MKs or monocytes or both leads to collagen fibrosis and new bone formation called osteosclerosis. Development of mouse models has provided important insights into the pathophysiological mechanisms of myelofibrosis with special regard to the role of MK. The two most important mouse models are characterized by either systemic overexpression of TPO (7, 11, 26, 27, 28) (TPOhigh mice) or MK lineage restricted under expression of the transcriptional factor GATA-1 (8) (GATA-1low mice). These models display hyperplasia of MK, a result of TPO-driven proliferation in the TPOhigh mice (29), and impaired MK proliferation and maturation in the GATA-1low mice (30). However, the observation that TPO overexpression in SCID mice (T and B cell deficient) (14) led to myelofibrosis, but not in NOD/SCID mice (T and B cell as well as multiple other defects in innate immunity including NK and monocyte/macrophage function (12, 13)), suggested an important role for the monocyte/macrophage lineage (11). We hypothesized that insufficiently high TPO levels, and not lack of a specific cell population, was the reason why NOD/SCID mice failed to develop myelofibrosis.

In this study, we demonstrated, using a retroviral vector, that high TPO levels induce myelofibrosis in the NOD/SCID mice similar to the one developed by WT or SCID mice. All of these mice developed a very similar disease, but, strikingly, SCID mice had only a mild anemia. This demonstrates that T and B cells (deficient in SCID and NOD/SCID mice) as well as monocyte/macrophage (deficient in NOD/SCID mice) are probably not directly involved in the development of myelofibrosis in this murine model. In fact, both level and site of TPO overexpression appear to influence myelofibrosis development (31). For example, first-described transgenic mice overexpressing TPO (32) using a liver-specific apolipoprotein E promoter did not exhibit myelofibrosis or osteosclerosis, whereas recently described TPO transgenic mice driven by the IgH promoter (33) developed myelofibrosis and osteosclerosis with the same plasma TPO level as the former model.

The TPOhigh and GATA-1low mice develop a myeloproliferative disorder featuring numerous aspects of the human disease including dysmegakaryopoiesis (5). Histological analyses at the time of myelofibrosis occurrence showed very large-sized platelets with empty granules and dysmorphic MK with pycnotic nuclei and increased emperipolesis (34). Recently, it has been demonstrated that a unique and acquired mutation of JAK2 (JAK2V617F) is associated with BCR/ABL-negative myeloproliferative disorders including IM (35, 36, 37, 38). When the JAK2V617F is expressed in mice, it leads to the development of a polycythemia vera, which progresses to myelofibrosis (C. Lacout, D. F. Pisani, M. Tulliez, F. Moreau-Gachelin, W. Vainchenker, and J. L. Villeval, submitted for publication). The fact that a constitutively active JAK2 may be associated with IM in human suggests that alteration in the TPO/c-mpl signaling may be at the origin of this dysmegakaryopoiesis. It is possible that in the TPOhigh mice, a long-lasting stimulation leads to a Mpl traffic defect and an altered signaling (S. Constantinescu, unpublished observation). This may secondarily imply transcription factors such as GATA-1, which are implicated in MK survival, proliferation, and differentiation. In favor of this hypothesis, it has been shown by Vannucchi et al. (39) that in TPOhigh and GATA-1low mice, TPO, GATA-1, and TGF-β1 are linked in an upstream-downstream relationship.

However, how a high level of TPO can induce myelofibrosis development remains unknown. TPO could regulate synthesis and release of cytokines or induce a dysmegakaryopoiesis. In this study, we show that TPO overexpression directly or indirectly induces apoptosis of MKs in bone marrow and spleen from TPOhigh mice using TUNEL technology, regardless of genetic background. This result suggests that the main mechanism of TPO-induced myelofibrosis is related to a dysmegakaryopoiesis.

In this study, we were able to demonstrate that TPO induced an increased synthesis of IL-1α by platelets. Indeed, numerous cytokines are involved in myelofibrosis and osteosclerosis, including TGF (TGF-β1), platelet-derived growth factor, basic fibroblastic growth factor, and IL-1. Experiments in TPOhigh mice have established the pivotal role of the hemopoietic cell derived-TGF-β1 in the promotion of myelofibrosis (17) and the role of the stromal osteoprotegerin produced by the microenvironment in the promotion of osteosclerosis (16). Accordingly, we found elevated TGF-β1 plasma levels in all TPOhigh mice analyzed and a slight increase of the TGF-β1 transcript in platelets. Thus, TGF-β1 appears to originate from platelets. The slight increase of its transcript could be explained by the 2- to 4-fold higher thrombocytosis displayed by the TPOhigh mice. Moreover, we found increased IL-1α plasma levels in all mice analyzed, including in the monocyte-deficient NOD/SCID. This means that IL-1α may have another cellular origin than monocytes in this model. There is strong evidence that platelets have transcripts for both IL-1α and IL1-β (40) in humans and are innate, inflammatory cells able to release inflammatory cytokines after activation (25). To clearly determine whether platelets may be involved in this excess level of IL-1α in TPOhigh mice, we studied the presence of the IL-1α protein and transcript in platelet extracts from these mice. We found high levels of IL-1α protein and transcript in TPOhigh mice, indicating that platelets (and probably MKs) can replace the monocyte/macrophage function to produce IL-1α when TPO levels become sufficiently high. However, we cannot exclude that nonhemopoietic cells could be also involved in this synthesis of IL-1α. The role of IL-1 in human IM is not completely solved. It has been suggested that IL-1 and probably the NF-κB pathway play a role in the pathogenesis of IM (10). IL-1 may indirectly regulate the TGF-β1 synthesis and, consequently, the development of myelofibrosis as suggested by Rameshwar et al. (24), and we recently have shown that the NF-κB pathway is also activated in MK from IM patients (41). In addition, there is strong evidence that IL-1 is implicated in the regulation of the bone mass. Notably, IL-1 is able to induce osteoprotegerin synthesis (42) by stromal cells, inhibiting the development of osteoclasts and consequently inducing the development of osteosclerosis. Thus, in the future, TPO overexpression in double IL-1α/IL-1β knockout mice (43) may be important in understanding the precise role of IL-1 in the promotion of both myelofibrosis and osteosclerosis. In addition, platelets are able to synthesize several other inflammatory molecules such as CD40-L (44), which are involved in the development of fibrosis (45). It remains to determine whether these molecules could also play a role in the pathogenesis of myelofibrosis. If this hypothesis is validated, it could open a new avenue in the therapeutic approach of IM by targeting the NF-κB pathway.

In summary, our results have shown that completely functional monocytes/macrophages are not required for the development of myelofibrosis induced by TPO. Together with the GATA-1low model, the TPOhigh model of myelofibrosis in the mouse underscores the role of the MK/platelet lineage in the development of myelofibrosis.

We acknowledge Francoise Wendling (U362) and Caroline Lefebvre for helpful discussions and improving the English manuscript, and we thank Annie Rouchès and Patrice Ardouin for managing the animals.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by grants from INSERM and La Ligue Nationale contre le Cancer Équipe Labellisée 2000, fellowships from Association pour la Recherche sur la Cancer and INSERM (Poste d’Accueil) (to O.W.-B.), and the Research Ministry (to H.C.).

4

Abbreviations used in this paper: IM, idiopathic myelofibrosis; MK, megakaryocyte; TPO, thrombopoietin; WT, wild type; CFC, colony-forming cells; FISH, fluorescent in situ hybridization.

1
Tefferi, A..
2000
. Myelofibrosis with myeloid metaplasia. N.
Engl. J. Med.
342
:
1255
-1265.
2
Jacobson, R. J., A. Salo, P. J. Fialkow.
1978
. Agnogenic myeloid metaplasia: a clonal proliferation of hematopoietic stem cells with secondary myelofibrosis.
Blood
51
:
189
-194.
3
Reeder, T. L., R. J. Bailey, G. W. Dewald, A. Tefferi.
2003
. Both B and T lymphocytes may be clonally involved in myelofibrosis with myeloid metaplasia.
Blood
101
:
1981
-193.
4
Reilly, J. T..
1997
. Idiopathic myelofibrosis: pathogenesis, natural history and management.
Blood Rev.
11
:
233
-242.
5
Thiele, J., B. Hoeppner, R. Zankovich, R. Fischer.
1989
. Histomorphometry of bone marrow biopsies in primary osteomyelofibrosis/-sclerosis (agnogenic myeloid metaplasia): correlations between clinical and morphological features.
Virchows Arch A Pathol. Anat. Histopathol.
415
:
191
-202.
6
Martyre, M. C., N. Romquin, M. C. Le Bousse-Kerdiles, S. Chevillard, B. Benyahia, B. Dupriez, J. L. Demory, F. Bauters.
1994
. Transforming growth factor-β and megakaryocytes in the pathogenesis of idiopathic myelofibrosis.
Br. J. Haematol.
88
:
9
-16.
7
Villeval, J. L., K. Cohen-Solal, M. Tulliez, S. Giraudier, J. Guichard, S. A. Burstein, E. M. Cramer, W. Vainchenker, F. Wendling.
1997
. High thrombopoietin production by hematopoietic cells induces a fatal myeloproliferative syndrome in mice.
Blood
90
:
4369
-4383.
8
Vannucchi, A. M., L. Bianchi, C. Cellai, F. Paoletti, R. A. Rana, R. Lorenzini, G. Migliaccio, A. R. Migliaccio.
2002
. Development of myelofibrosis in mice genetically impaired for GATA-1 expression (GATA-1low mice).
Blood
100
:
1123
-1132.
9
Toki, T., F. Katsuoka, R. Kanezaki, G. Xu, H. Kurotaki, J. Sun, T. Kamio, S. Watanabe, S. Tandai, K. Terui, et al
2005
. Transgenic expression of BACH1 transcription factor results in megakaryocytic impairment.
Blood
105
:
3100
-318.
10
Rameshwar, P., R. Narayanan, J. Qian, T. N. Denny, C. Colon, P. Gascon.
2000
. NF-κB as a central mediator in the induction of TGF-β in monocytes from patients with idiopathic myelofibrosis: an inflammatory response beyond the realm of homeostasis.
J. Immunol.
165
:
2271
-227.
11
Frey, B. M., S. Rafii, M. Teterson, D. Eaton, R. G. Crystal, M. A. Moore.
1998
. Adenovector-mediated expression of human thrombopoietin cDNA in immune- compromised mice: insights into the pathophysiology of osteomyelofibrosis.
J. Immunol.
160
:
691
-699.
12
Shultz, L. D., P. A. Schweitzer, S. W. Christianson, B. Gott, I. B. Schweitzer, B. Tennent, S. McKenna, L. Mobraaten, T. V. Rajan, D. L. Greiner, et al
1995
. Multiple defects in innate and adaptive immunologic function in NOD/LtSz-scid mice.
J. Immunol.
154
:
180
-191.
13
Serreze, D. V., J. W. Gaedeke, E. H. Leiter.
1993
. Hematopoietic stem-cell defects underlying abnormal macrophage development and maturation in NOD/Lt mice: defective regulation of cytokine receptors and protein kinase C.
Proc. Natl. Acad. Sci. USA
90
:
9625
-9629.
14
Bosma, G. C., R. P. Custer, M. J. Bosma.
1983
. A severe combined immunodeficiency mutation in the mouse.
Nature
301
:
527
-530.
15
Prochazka, M., H. R. Gaskins, L. D. Shultz, E. H. Leiter.
1992
. The nonobese diabetic scid mouse: model for spontaneous thymomagenesis associated with immunodeficiency.
Proc. Natl. Acad. Sci. USA
89
:
3290
-324.
16
Chagraoui, H., M. Tulliez, T. Smayra, E. Komura, S. Giraudier, T. Yun, N. Lassau, W. Vainchenker, F. Wendling.
2003
. Stimulation of osteoprotegerin production is responsible for osteosclerosis in mice overexpressing TPO.
Blood
101
:
2983
-299.
17
Chagraoui, H., E. Komura, M. Tulliez, S. Giraudier, W. Vainchenker, F. Wendling.
2002
. Prominent role of TGF-β1 in thrombopoietin-induced myelofibrosis in mice.
Blood
100
:
3495
-3503.
18
Flamant, S., M. Lebastard, P. Pescher, C. Besmond, G. Milon, G. Marchal.
2003
. Enhanced cloning efficiency of mouse bone marrow macrophage progenitors correlates with increased content of CSF-1 receptor of their progeny at low oxygen tension.
Microbes Infect.
5
:
1064
-109.
19
Vandesompele, J., K. De Preter, F. Pattyn, B. Poppe, N. Van Roy, A. De Paepe, F. Speleman.
2002
. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes.
Genome Biol.
3
:
RESEARCH0034
20
Colle, J. H., P. B. Falanga, M. Singer, B. Hevin, G. Milon.
1997
. Quantitation of messenger RNA by competitive RT-PCR: a simplified read out assay.
J. Immunol. Methods
210
:
175
-184.
21
Lean, J. M., C. Murphy, K. Fuller, T. J. Chambers.
2002
. CCL9/MIP-1γ and its receptor CCR1 are the major chemokine ligand/receptor species expressed by osteoclasts.
J. Cell. Biochem.
87
:
386
-393.
22
Pfaffl, M. W..
2001
. A new mathematical model for relative quantification in real-time RT-PCR.
Nucleic Acids Res.
29
:
e45
23
Fulop, G. M., R. A. Phillips.
1986
. Full reconstitution of the immune deficiency in scid mice with normal stem cells requires low-dose irradiation of the recipients.
J. Immunol.
136
:
4438
-4443.
24
Rameshwar, P., T. N. Denny, D. Stein, P. Gascon.
1994
. Monocyte adhesion in patients with bone marrow fibrosis is required for the production of fibrogenic cytokines: potential role for interleukin-1 and TGF-β.
J. Immunol.
153
:
2819
-2830.
25
Lindemann, S., N. D. Tolley, D. A. Dixon, T. M. McIntyre, S. M. Prescott, G. A. Zimmerman, A. S. Weyrich.
2001
. Activated platelets mediate inflammatory signaling by regulated interleukin 1β synthesis.
J. Cell Biol.
154
:
485
-490.
26
Ulich, T. R., J. del Castillo, G. Senaldi, O. Kinstler, S. Yin, S. Kaufman, J. Tarpley, E. Choi, T. Kirley, P. Hunt, W. P. Sheridan.
1996
. Systemic hematologic effects of PEG-rHuMGDF-induced megakaryocyte hyperplasia in mice.
Blood
87
:
5006
-5015.
27
Abina, M. A., M. Tulliez, C. Lacout, N. Debili, J. L. Villeval, F. Pflumio, F. Wendling, W. Vainchenker, H. Haddada.
1998
. Major effects of TPO delivered by a single injection of a recombinant adenovirus on prevention of septicemia and anemia associated with myelosuppression in mice: risk of sustained expression inducing myelofibrosis due to immunosuppression.
Gene Ther.
5
:
497
-506.
28
Yan, X. Q., D. Lacey, D. Hill, Y. Chen, F. Fletcher, R. G. Hawley, I. K. McNiece.
1996
. A model of myelofibrosis and osteosclerosis in mice induced by overexpressing thrombopoietin (mpl ligand): reversal of disease by bone marrow transplantation.
Blood
88
:
402
-409.
29
Kaushansky, K..
1995
. Thrombopoietin: the primary regulator of platelet production.
Blood
86
:
419
-431.
30
Vyas, P., K. Ault, C. W. Jackson, S. H. Orkin, R. A. Shivdasani.
1999
. Consequences of GATA-1 deficiency in megakaryocytes and platelets.
Blood
93
:
2867
-2875.
31
Tefferi, A..
2005
. Experimental myelofibrosis in mice and the implications to human disease.
Leuk. Res.
29
:
723
-726.
32
Zhou, W., C. F. Toombs, T. Zou, J. Guo, M. O. Robinson.
1997
. Transgenic mice overexpressing human c-mpl ligand exhibit chronic thrombocytosis and display enhanced recovery from 5-fluorouracil or antiplatelet serum treatment.
Blood
89
:
1551
-159.
33
Kakumitsu, H., K. Kamezaki, K. Shimoda, K. Karube, T. Haro, A. Numata, K. Shide, T. Matsuda, K. Oshima, M. Harada.
2005
. Transgenic mice overexpressing murine thrombopoietin develop myelofibrosis and osteosclerosis.
Leuk. Res.
29
:
761
-769.
34
Schmitt, A., H. Jouault, J. Guichard, F. Wendling, A. Drouin, E. M. Cramer.
2000
. Pathologic interaction between megakaryocytes and polymorphonuclear leukocytes in myelofibrosis.
Blood
96
:
1342
-137.
35
Baxter, E. J., L. M. Scott, P. J. Campbell, C. East, N. Fourouclas, S. Swanton, G. S. Vassiliou, A. J. Bench, E. M. Boyd, N. Curtin, et al
2005
. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders.
Lancet
365
:
1054
-1061.
36
James, C., V. Ugo, J. P. Le Couedic, J. Staerk, F. Delhommeau, C. Lacout, L. Garcon, H. Raslova, R. Berger, A. Bennaceur-Griscelli, et al
2005
. A unique clonal JAK2 mutation leading to constitutive signalling causes polycythaemia vera.
Nature
434
:
1144
-118.
37
Kralovics, R., F. Passamonti, A. S. Buser, S. S. Teo, R. Tiedt, J. R. Passweg, A. Tichelli, M. Cazzola, R. C. Skoda.
2005
. A gain-of-function mutation of JAK2 in myeloproliferative disorders.
N. Engl. J. Med.
352
:
1779
-1790.
38
Levine, R. L., M. Wadleigh, J. Cools, B. L. Ebert, G. Wernig, B. J. Huntly, T. J. Boggon, I. Wlodarska, J. J. Clark, S. Moore, et al
2005
. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis.
Cancer Cell.
7
:
387
-397.
39
Vannucchi, A. M., L. Bianchi, F. Paoletti, A. Pancrazzi, E. Torre, M. Nishikawa, M. Zingariello, A. Di Baldassarre, R. A. Rana, R. Lorenzini, et al
2005
. A pathobiologic pathway linking thrombopoietin, GATA-1, and TGF-β1 in the development of myelofibrosis.
Blood
105
:
3493
-3501.
40
Bugert, P., A. Dugrillon, A. Gunaydin, H. Eichler, H. Kluter.
2003
. Messenger RNA profiling of human platelets by microarray hybridization.
Thromb. Haemost.
90
:
738
-748.
41
Komura, E., C. Tonetti, V. Penard-Lacronique, H. Chagraoui, C. Lacout, J. P. Lecouedic, P. Rameau, N. Debili, W. Vainchenker, S. Giraudier.
2005
. Role for the nuclear factor κB pathway in transforming growth factor-β1 production in idiopathic myelofibrosis: possible relationship with FK506 binding protein 51 overexpression.
Cancer Res.
65
:
3281
-329.
42
Penno, H., C. J. Silfversward, A. Frost, H. Brandstrom, O. Nilsson, O. Ljunggren.
2002
. Osteoprotegerin secretion from prostate cancer is stimulated by cytokines, in vitro.
Biochem. Biophys. Res. Commun.
293
:
451
-455.
43
Horai, R., M. Asano, K. Sudo, H. Kanuka, M. Suzuki, M. Nishihara, M. Takahashi, Y. Iwakura.
1998
. Production of mice deficient in genes for interleukin (IL)-1α, IL-1β, IL-1αβ, and IL-1 receptor antagonist shows that IL-1β is crucial in turpentine-induced fever development and glucocorticoid secretion.
J. Exp. Med.
187
:
1463
-1475.
44
Henn, V., S. Steinbach, K. Buchner, P. Presek, R. A. Kroczek.
2001
. The inflammatory action of CD40 ligand (CD154) expressed on activated human platelets is temporally limited by coexpressed CD40.
Blood
98
:
1047
-1054.
45
Zhang-Hoover, J., A. Sutton, J. Stein-Streilein.
2001
. CD40/CD40 ligand interactions are critical for elicitation of autoimmune-mediated fibrosis in the lung.
J. Immunol.
166
:
3556
-3563.