Effect And Mechanism Of Bone Marrow Cell Mobilization On The Resolution And Recanalization Of Venous Thrombi

Background:Venous thromboembolism manly includes deep vein thrombusis and pulmonary embolism. The annual incidence of venous thromboembolism is estimated to be between 100 and 200 per 100,000 of the general population based on recent estimates made in Europe and North America. There is no accurate data in China. Venous thromboembolism is a known serious condition which morbility increased in recent years. Venous thrombi often induced postthrombotic syndrome, cost treatment and unsatisfactory results. The classic treatment of this disease is non-operative therapy that mainly includes anticoagulation and thrombolysis. Anticoagulant therapy, the standard therapy of venous thromboembolism, and thrombolysis both have limited effects to the existing thrombi and has a small but significant risk of severe hemorrhage. Thrombolytic therapy does not consistently reduce the severity of postthrombotic syndrome and is also associated with an increased risk of hemorrhage. The resolution of venous thrombi, progress mainly by body itself, is naturally by a slow process of organization and recanalization. As thrombus organizing, inflammatory cells, such as neutrophils, monocytes, and endothelial cells, enter into the thrombus, the new vessels appear within the body of the thrombus and coalesce and enlarge, and then blood flow is established through the thrombus. Complete resolution of venous thrombi would restore the recanalization, otherwise stenosis or obstruction of vein, and lead to postthrombotic syndrome. Accelerating organization and recanalization of thrombus may prevent postthrombotic complications by reducing valvular damage and residual obstruction and would also be valuable when thrombolysis is contraindicated.Macrophage plays an important role in the resolution of venous thrombi. The resolution and recanalization of thrombi was increased by the injection of macrophage into thrombi. The neovascularization in thrombi mostly exist at the area of macrophage accumulation in human and animal studies. Bone marrow–derived cells were recruited into the thrombus during resolution. Many of these cells expressed a macrophage phenotype and may represent a population of plastic stem cells that orchestrate thrombus recanalization. Macrophages derived from bone marrow, which mostly expressing endothelial cell marker (VEGFR), were present mainly at the periphery of the thrombus at 7th day and throughout the thrombus at 14th day. Bone marrow–derived cells, however, did not line the new recanalizing channels that formed either between the thrombus and the venous wall or within the thrombus itself. Bone marrow derived cells may not differenciate into endothelial cell of new vessel, otherwise stimulate the neovascularization in thrombi.Cysteine-cysteine chemokine receptor 2 (CCR2) and MCP-3/MCP-1 were critical for monocyte mobilization. Deletion of CCR2 or blockade of all CC chemokines inhibited both monocyte recruitment and thrombus resolution. Genetic deletion of CCR2 was associated with larger thrombi, increased thrombus collagen, fewer thrombus monocytes, reduction in both matrix metalloproteinase (MMP)-2 and MMP-9 activity, and significantly impaired neovascularization. The chemokines, such as MCP-1, in thrombus had no effect on monocytes recruitment.We hypothesized that increasing the accumulation and activity of macrophage in thrombi would promote the resolution and neovascularization of thrombus, realize the early recanalization of thrombi and reduce the postthrombotic syndrome. The purpose of the present study was to explore the effect and mechanism of bone marrow mobilization with the recombinant human granulocyte colony-stimulating factor (rhG-CSF) or recombinant human granulocyte-macrophage colony-stimulating factor (rhGM-CSF) on accelerating the venous thrombus resolution and macrophage accumulation. Methods:1 Animal experimentThe animals were assigned to normal, sham-operated, thrombus and mobilization group. Normal group experienced untreatment. Sham-operated group underwent the same procedure, in which a silk ligature was passed around the left infrarenal vena cava but left untied. Thrombus group were received only venous thrombi induced. Mobilization group were received venous thrombi induced and bone marrow mobilization with a daily subcutaneous injection of rhG-CSF or rhGM-CSF for 7 days from operative day (the final dose was 25ug/kg). Other groups received a daily saline at same time.Thrombus was produced in the left infrarenal vena cava of animals as previously described and modified. The infrarenal portion of the vena cava was exposed. The segment of the vena cava from just below the left renal vein to the confluence of the common iliac veins was dissected and the tribulatory were ligated. A neurosurgical vascular clip was applied to this portion of the vein for 60 seconds on 2 occasions, 30 seconds apart, to induce endothelial damage. A needle was then placed alongside the vena cava. A stenosis was produced in the vein by tying a 4-0 silk suture around the vena cava to include the needle. The needle was then pulled out to allow some flow within it. The abdominal wall was sutured. The animals were then allowed to recover from the anesthesia and access to water and chow.At 7th and 14th day, cohorts of rats were reanesthetized, and a laparotomy was performed to excise the infrarenal vena cava, which was prepared for histology and image analysis, from just above the ligature to the confluence of the common iliac veins.The peripheral blood was harvested at 30min of pre-operation and 1, 3, 5, 7 and 10d of post-operation. White blood cell count was performed by using an automated hematology analyzer, and the percentage of mononuclear cells in white blood cells was counted under microscope.The infrarenal vena cava containing the thrombus was fixed in 4% paraformaldehyde overnight and embedded in paraffin wax. Transverse 3 sections with 5μm thick were cut at 50μm intervals below the ligature and stained with hematoxylin and eosin (HE). Sections of tissues were viewed and digitized using a microscope-mounted camera. The area of thrombus in each section was measured with image analysis software. The mean percentage of orgonized part to whole area of thrombus in 3 sections was the orgorization rate of each thrombus. The sections were processed for immunohistochemistry to localize macrophages using the monoclonal antibody against the CD68 antigen. The percentage area of the thrombus containing the stained CD68 antigen (macrophage density) was measured with the image analysis software described above. The ultramicroscopic observation of thrombus was also conducted. The levels of MCP-1 and MIP-1αin the peripheral blood were analyzed via sandwich-ELISA assays. Used ELISA Kit was as follows: rat MCP-1 ELISA Kit (BioSource) and rat MIP-1αELISA Kit (Antigenix). The assay was conducted in accordance with the manufacturer's instructions. The expression of CCR2 mRNA in peripheral blood was analyzed via RT-PCR. The total RNA extract was conducted in accordance with the instructions of Catrimox-14TM RNA Isolation Kit Ver.2.11(TaKaRa). The RT-PCR was conducted in accordance with the instructions of TaKaRa RNA PCR Kit (AMV) Ver.3.0. PCR products were subjected to electrophoresis on 2% agarose gel, stained with ethidium bromide, and photographed. The results were semiquantified by densiometric scanning.2 Cell experimentMouse peritoneal marcrophage line RAW264.7 were purchased from Institute of Biochemisry and Cell Biology, SIBS, CAS and stored by Surgical Laboratory of Second Affiliated Hospital, Chongqing University of Medical Science. RAW264.7 cells were grown in RPMI 1640 with 10% calf serum, 100U/ml penicillin and 100μg/ml streptomycin. RAW264.7 cells were seeded at a density of 1×105 cells/well into 24-well plates or 6×105 cells/ flask into 100ml culture flask。Culture media contained the rhG-CSF or rhGM-CSF with the concentration of 0,20,40,80,160,320ng/ml. The influence of treatment on the secretion of MCP-1 and MIP-1αby macrophage was determined by ELISA. Used ELISA Kit was mouse MCP-1 ELISA Kit (Pierce) and mouse MIP-1αELISA Kit (Biosource). The assay was conducted in accordance with the manufacturer's instructions.The expression of CCR2 mRNA in culture medium was analyzed via RT-PCR. The total RNA extract was conducted in accordance with the instructions of Catrimox-14TM RNA Isolation Kit Ver. 2.11(TaKaRa). The RT-PCR was conducted in accordance with the instructions of TaKaRa RNA PCR Kit (AMV) Ver.3.0. PCR products were subjected to electrophoresis on 2% agarose gel, stained with ethidium bromide, and photographed. The results were semiquantified by densiometric scanning.Protein from RAW264.7 cells was extracted according to manufacture's instructions with a tatol RNA extraction kit. To prevent phosphatase activity, all procedures were performed on ice and phosphatase inhibitors were added. Protein content was quantified with a BCA protein quantity Kit with bovine serum albumin (BSA) standard. Equal amounts of protein were resolved on 12% Tris-HCl polyacrylamide gels under 60 mA and transferred to PVDF blotting membranes. The membrane was blocked for 1 h in TBS-Tween supplemented with 5% bovine serum albumin. The membrane was incubated with anti-CCR2 (final dilution 1:1,000; Abcam) followed by washes and incubation with HRP-conjugated rabbit anti-goat IgG antibody (final dilution 1:5,000). The membrane was blocked again as previously described. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody (final dilution 1:1,000, Abcam) was then incubated with the membrane followed by washes and incubation with HRP-conjugated goat anti-rabbit IgG antibody (final dilution 1:5,000). The proteins were detected with DAB kit. The immunoblots were semiquantified by densiometric scanning.3 Statistical analysisSPSS10.0 was used to analyze the data. F test, Dunnett test, tukey's test and t test were used according to the difference of data. Statistical significance was assumed for P < 0.05.Results:1 Mononuclear cells count in peripheral blood Mononuclear cells significantly increased in mobilization group at post-operative 3rd, 5th and 7th day (rhG-CSF: 2.1±0.3, 3.1±0.2, 4.4±0.3; rhGM-CSF: 1.8±0.3, 2.4±0.6, 3.0±0.7) versus to that of thrombus group (1.7±0.2, 1.5±0.3, 1.3±0.4) (P<0.05). Comparing with rhGM-CSF mobilization group, the mononuclear cells significantly increased in rhG-CSF mobilization group at post-operative 5th(t=2.95, P<0.05) and 7th day (t=3.53, P<0.01).2 Organization rate of thrombus and ultramicroscopic observation Comparing with thrombus group (15.6±2.3%, 63.2±8.6%), the organization rate in mobilization group (30.2±3.5%, 89.5±5.6%) at 7th or 14th day have significant difference, respectively(P﹤0.01). No significant difference was observed in the organization rate between rhG-CSF and rhGM-CSF mobilization group. Ultramicroscopic observation suggested that increasing macrophages and its activity with amounts of neovessels were observed in mobilization group with rhG-CSF or hGM-CSF treatment versus to that in thrombus group.3 Macrophage content of thrombusCompared with thrombus group (1.47±0.8%) at postoperative 7th day, the macrophage (CD68 staining) content of thrombus in mobilization group rhG-CSF (8.46±1.2%, P=0.0026) and rhG-CSF (7.84±1.1%, P=0.0041) was significant increased. There was also a significant difference in the macrophage content of the thrombus at 14th day in mobilization group of rhG-CSF(13.52±1.3%, P=0.0087) or rhGM-CSF (12.13±1.4%, P=0.0092) versus to thrombus group (5.52±0.7%). No significant difference observed between rhG-CSF and rhGM-CSF group at 7th and 14th day, respectively.4 Levels of MCP-1 and MIP-1αThere were no statistical difference observed in MCP-1 level of peripheral blood between groups at postoperative 7th and 14th day (P>0.05). The levels of MIP-1αin the peripheral blood of rats at postoperative 7th and 14th day were similar to those seen in MCP-1 level. The secretion of MCP-1 and MIP-1αby RAW264.7 after rhG-CSF or rhGM-CSF stimulation has no significant change, respectively. 5 Expression of CCR2 mRNAAs compared to normal group, levels of CCR2 mRNA increased at postoperative 7th and 14th day in sham-operated, thrombus and mobilization group (P﹤0.05), respectively. Levels of CCR2 mRNA increased in mobilization group versus to that in thrombus group(P﹤0.05). There were no differences in CCR2 mRNA levels between the rhGM-CSF and rhG-CSF groups.The secretion of CCR2 mRNA by RAW264.7 cells after 24h of 80ng/ml rhGM-CSF or rhG-CSF stimulation significantly increased in comparison with other concentrations, respectively. Levels of CCR2 mRNA stably high expressed after 36 and 48h of 80ng/ml rhGM-CSF or rhG-CSF stimulation. Comparing to rhG-CSF stimulation, after 12h of culture, there was significant difference in expression of RAW264.7 CCR2 mRNA with rhGM-CSF stimulation.6 Expression of CCR2 proteinThe secretion of CCR2 protein by RAW264.7 cells after 48h of 80ng/ml rhGM-CSF stimulation significantly increased in comparison with other concentrations. Levels of CCR2 protein increased after 36, 48 and 60h of 80ng/ml rhGM-CSF stimulation. There were no differences in the ratio of CCR2 to GAPDH between the rhG-CSF stimulation and not. Conclusion:1. Venous wall damage combined with stenosis is effective to induce rat venous thrombosis.2. Mobilization of bone marrow accelerates the resolution and recanalization of thrombus, and increases macrophages accumulation in thrombus.3. rhGM-CSF or rhG-CSF increase chemokine CCR2 expression of macrophages.4. The increase of macrophages in thrombus, after mobilization by rhGM-CSF or rhG-CSF, may correlate to the increase of macrophages chemokine CCR2 expression.