Journal of Vascular Surgery
Volume 43, Issue 3 , Pages 601-608, March 2006

Monocyte recruitment in venous thrombus resolution

This study was presented at the Seventeenth Annual American Venous Forum, San Diego, CA, Feb 9-13, 2005.

  • Tahir Ali, MRCS

      Affiliations

    • Academic Department of Surgery, Cardiovascular Division, GKT, King’s College, St Thomas’s Hospital, London, United Kingdom
    • ,
  • Julia Humphries, PhD

      Affiliations

    • Academic Department of Surgery, Cardiovascular Division, GKT, King’s College, St Thomas’s Hospital, London, United Kingdom
    • ,
  • Kevin Burnand (FRCS)

      Affiliations

    • Academic Department of Surgery, Cardiovascular Division, GKT, King’s College, St Thomas’s Hospital, London, United Kingdom
    • ,
  • Barbara Sawyer

      Affiliations

    • Academic Department of Surgery, Cardiovascular Division, GKT, King’s College, St Thomas’s Hospital, London, United Kingdom
    • ,
  • Christina Bursill, PhD

      Affiliations

    • Department of Cardiovascular Medicine, University of Oxford, Oxford, United Kingdom
    • ,
  • Keith Channon (FRCP)

      Affiliations

    • Department of Cardiovascular Medicine, University of Oxford, Oxford, United Kingdom
    • ,
  • David Greaves, PhD

      Affiliations

    • Department of Cardiovascular Medicine, University of Oxford, Oxford, United Kingdom
    • ,
  • Barrett Rollins, MD

      Affiliations

    • Department of Adult Oncology, Dana-Faber Cancer Institute, Harvard Medical School, Boston
    • ,
  • Israel F. Charo, MD, PhD

      Affiliations

    • University of California, Gladstone Institute of Cardiovascular Disease, San Francisco, Calif
    • ,
  • Alberto Smith, PhD

      Affiliations

    • Academic Department of Surgery, Cardiovascular Division, GKT, King’s College, St Thomas’s Hospital, London, United Kingdom
    • Corresponding Author InformationCorrespondence: Dr Alberto Smith, Academic Department of Surgery, Cardiovascular Division, GKT, King’s College, St. Thomas’s Hospital, London SE1 7EH, UK.

Received 28 June 2005; accepted 18 October 2005.

Article Outline

Objective

To investigate the importance of monocyte recruitment in thrombus resolution and the role of cysteine-cysteine (CC) chemokines and the CC chemokine receptor, CCR2, in this process.

Methods

Peritoneal macrophages, monocyte chemotactic protein 1 (MCP1), or carrier solutions were injected into thrombi induced in the vena cava of rats. Caval thrombi were also formed in CCR2−/− and MCP1−/− mice and in wild-type mice transfected with an adenoviral construct expressing a broad-spectrum CC receptor antagonist.

Results

Direct administration of peritoneal macrophages decreased thrombus size by more than fivefold and increased recanalization by more than fourfold compared with controls (P < .001). A 100-ng MCP1dose reduced thrombus size by more than sixfold (P < .01) and increased recanalization by more than sevenfold (P < .01), without affecting macrophage recruitment. Deletion of CCR2 or blockade of all CC chemokines inhibited both monocyte recruitment (P < .05) and thrombus resolution (P < .01), but knocking out MCP-1 had no effect.

Conclusion

Increasing macrophage numbers in the thrombus enhances its resolution. MCP1 treatment enhances resolution by stimulating recanalization, independent of an effect on monocyte recruitment. CCR2 deficiency has the same effect as blockade of all CC chemokines. CCR2 receptor activation may therefore be an important mechanism in monocyte recruitment into venous thrombi and could be targeted to promote their resolution.

Clinical Relevance

Deep vein thrombosis may lead to residual venous obstruction or reflux and result in post-thrombotic complications, which are debilitating and have a substantial socioeconomic impact. Enhancing the resolution of venous thrombi may reduce post-thrombotic complications.

 

The rapid and complete resolution of deep venous thrombosis may conserve valve integrity and, as a consequence, reduce the incidence of post-thrombotic syndrome.1 Incomplete resolution of the thrombus encourages rethrombosis, further propagation, and valve destruction. Valvular incompetence and obstruction is ultimately responsible for ambulatory venous hypertension and cutaneous ulceration.2

Anticoagulation diminishes thrombus propagation, the risk of rethrombosis, and life-threatening thromboembolism. Anticoagulants do not, however, promote thrombus resolution.3 The use of systemic and local thrombolytic treatment is effective in reducing thrombus burden but is complicated by a risk of hemorrhage.4 Thrombolytic agents remove thrombus rapidly in patients with proximal ileofemoral thrombi who are treated early. They can restore luminal patency and maintain distal valve function, but their use also carries a risk of major bleeding and has not been widely adopted in the United Kingdom.

Venous thrombus resolves by a process of organization and recanalization that is similar to the formation of granulation tissue in healing wounds. The recruitment of inflammatory cells is an important component of both processes. An initial neutrophil infiltrate is replaced by monocyte-derived macrophages5, 6, 7 that have the capacity to express a host of chemotactic agents, proteases, and growth factors that orchestrate tissue remodelling and revascularization.8, 9 Our previous studies have shown that monocytes are recruited in large numbers into maturing human and experimental venous thrombi thrombi,6 and thrombus resolution does not occur if their recruitment is restricted.10 Monocyte chemoattractants such as the cysteine-cysteine (CC) chemokine, monocyte chemotactic protein-1 (MCP1), are also expressed in the thrombus as it organizes, and treatment with exogenous MCP1 enhances resolution,11, 12 although evaluation of monocyte recruitment was not assessed.11 These data suggest that increasing monocyte recruitment into the thrombus might improve its resolution.

The aim of this study was to determine whether enhanced thrombus resolution could be achieved by increasing thrombus macrophage numbers through either the direct injection of isolated macrophages or by stimulating their ingress with the addition of MCP1. The importance of the endogenous expression of MCP1, its cognate receptor, CCR2, and the CC chemokine system was also investigated by using gene knockout mice and a selective CC chemokine inhibitor.

Back to Article Outline

Materials and methods 

Rat and mouse models of venous thrombosis 

This project was carried out in accordance with the United Kingdom Animal (Scientific Procedures) Act 1986. Laminated thrombus was produced in the inferior vena cava knockout mice and their wild types13, 14 by using previously described flow models of venous thrombosis.6, 15

Analysis of the thrombus 

Measurement of thrombus size 

Digital images of hematoxylin and eosin stained sections of thrombi were taken at ×50 magnification (Nikon Coolpix 990). Acquisition, spatial calibration, and analysis of captured images were achieved, in a blinded manner, using Image-Pro Plus analysis software (Media Cybernetics UK, Berkshire, UK). Thrombus size was estimated by measuring the cross-sectional area of thrombus in sections taken at defined intervals (500 μm in the rat and 300 μm in the mouse) and adding the areas obtained along the entire length of the specimen. Thrombus volume was estimated by multiplying the mean cross sectional area by the length of the thrombus.

Measurement of recanalization area of thrombus 

Recanalizing channels were identified as spaces, lined by flat endothelial-like cells, found between resolving thrombus and the vein wall and within the main body of the thrombus. Cross-sectional images were taken and analyzed as described above. The area of recanalization was measured in each section and expressed as a percentage of the luminal area of the vein. This allowed estimation of the total percentage recanalization for each thrombus. We have previously examined the interoperator variability of this method and found this to be <10%.12

Measurement of macrophage content 

Rat thrombus. The endogenous peroxidase activity in sections was quenched by immersing dewaxed paraffin sections for 10 minutes in a solution of 1% hydrogen peroxide in methanol. The rat macrophage CD68 antigen was retrieved by pressure-cooking sections in 0.01 M citrate buffer (pH 6.0) for 2 minutes. Sections were blocked with 10% normal horse serum in Dulbecco phosphate-buffered saline (DPBS) containing 0.01% bovine serum albumen. CD68 expression was located using 5 μg mouse antirat CD68 antibody (ED1, Serotec, Oxford, UK). A secondary rat adsorbed biotinylated horse antimouse immunoglobulin G (IgG) (Vector Laboratories, Peterborough, UK), avidin-biotinylated peroxidase complex (Vectastain Elite ABC, Vector), and SG chromogen (Vector) were then used to detect primary antibody binding. Sections were counterstained with Nuclear Fast Red (Vector).

Mouse thrombus. Antigen retrieval was done on paraffin sections of mouse thrombus as described above. Sections were blocked with serum-free Protein Block (Dako, Ely, Cambridgeshire, UK) and the mouse macrophage marker, Mac3, located using a rat antimouse Mac3 antibody (BD PharMingen, Oxford, UK). A rabbit antirat IgG and the Vectastain ABC reagent (Vector) were used to detect primary antibody binding as described above. The macrophage content of the thrombus was measured as the percentage area of the thrombus stained by the ED1 or Mac3 antibodies. Digitized images were taken and analyzed as described above.

Generation and characterization of peritoneal mononuclear cells 

Three donor rats were given an intraperitoneal injection (20 mL/kg) of >2-month-old, 4% sterile thioglycollate broth (Biotec Laboratories Ltd, Suffolk, UK). This induced a macrophage-rich exudate that was recovered by peritoneal lavage with 40 mL sterile DPBS 5 to 7 days after thioglycollate administration.16 The lavage fluid was pooled, centrifuged at 400g for 5 minutes at room temperature, and the recovered peritoneal cells were washed twice with sterile DPBS containing 1% bovine serum albumin. The cell pellet was then prepared for differential cell count, flow-cytometric, and immunohistochemical analyses (ED1) (Serotec).

Flow-cytometric analysis of peritoneal lavage fluid cell populations 

The monocytic cell population in the pooled lavage fluid was quantified by differential expression of CD3 and CD4, because conventional antirat monocyte antibodies such as ED1 are not informative in flow cytometric analysis. Cells were first blocked with mouse antirat Fcγii receptor (BD PharMingen). CD3 and CD4 antigens were visualized using mouse antirat CD3 (R-Phycoerythrin-conjugate) and CY-Chrome-conjugated antirat CD4 antibodies (BD PharMingen). Flow-cytometric analysis was performed on a FACScan flow cytometer (Becton Dickinson UK Ltd, Oxford, UK). Gating was based on forward and side scatter characteristics.

Differential and viability counts of lavage cells 

Lavage smears were stained by using the conventional Wright-Giemsa method. A standard 200-cell differential count was carried out by two independent observers, and the percentage number of macrophages was determined. The percentage number of viable cells was measured in a Neubauer hemocytometer by using the trypan-blue dye exclusion technique.

Immunohistochemical analysis peritoneal lavage macrophage content 

Lavage fluid containing 1 × 104 cells/100 μL was spun onto Vectabond (Vector) coated slides (cytospin) at 900g for 10 minutes and allowed to air-dry. Rat Fc blocking antibody (2 μg anti-rat CD32/Fcγii receptor) (BD PharMingen) was applied for 5 minutes at room temperature, and the cells were fixed and permeabilized using IntraPrep reagent (Immunotech, Marseille, France) for 15-minutes. The slides were washed in PBS, and endogenous peroxidase was quenched in a solution of 1% hydrogen peroxide in methanol. The slides were washed in distilled water, and Protein Block (Dako) was applied for 5 minutes at room temperature. The rat macrophage marker (ED1) was detected by using 5 μg mouse antirat ED1 (Serotec). Primary antibody binding was located with the Envision detection kit (Dako) and SG chromogen (Vector).

Fibrinolytic activity of lavage cells 

The fibrinolytic activity of the lavaged cells was measured by incubating the cells (105) with a solution containing a plasmin substrate (Chromagenix S-2403) and lys-plasminogen (a gift from Prof P Gaffney, NIBSC, UK). The optical density of this, and a series of uPA standards or blank controls, was measured at 405 nm.

Peritoneal-derived mononuclear cell injection into thrombus 

Caval thrombi were induced in a cohort of eight rats. DPBS (50 μL) containing about 2 × 107 peritoneal lavage mononuclear cells was injected directly into the thrombus 48 hours later. Thrombus formed in a second group of eight animals was injected with the same volume of DPBS carrier alone. The rats were killed after 7 days, and the vena cavae containing the thrombus were harvested. Specimens were fixed in 4% formalin and embedded in paraffin. Thrombus size, recanalization, and macrophage content were measured in 5-μm sections taken at 500-μm intervals throughout the length of the specimen, as described above.

The fate of MCP1 protein injected into formed thrombus 

A thrombus was formed in the vena cava of four groups of three rats each. Twenty-four hours after thrombus induction, 1 μCi of 125I-MCP1 (Amersham, Little Challfont, Oxon, UK) was directly injected into the thrombus of each animal by using a gas tight Hamilton syringe and 36-guage needle. Rats were killed after 3 hours, 1 day, 3 days, and 7 days, and samples of blood, caval wall, thrombus, aorta, heart, lung, spleen, liver, and skin were taken. All tissues were rinsed in DPBS. The amount of radioactivity they contained counted on a gamma counter (LKB, Turku, Finland). The counts were expressed as a percentage of the total count used and normalized for the wet weight of the sample.

The effect of MCP1 on venous thrombosis resolution 

Caval thrombus was induced in four groups of eight rats. Doses (10 μL) of 1000 ng, 100 ng, or 33 ng of recombinant rat MCP1 (Serotec, UK) were injected directly into the thrombus formed in three groups of animals 48 hours after induction of thrombus. An equal volume of carrier (DPBS) was injected in the fourth cohort to act as a control. The rats were killed 7 days later, and the length of cava from the left renal vein to the iliac bifurcation was harvested. Each specimen was rinsed in DPBS solution, fixed in 4% formalin, and embedded in paraffin. Thrombus size, recanalization, and macrophage content were measured in 5-μm sections taken at 500-μm intervals throughout the length of the specimen, as described above.

Natural thrombus resolution in MCP1−/− and CCR2−/− mice 

Thrombosis was induced in the vena cavae of four groups of control and knockout-MCP1−/− and CCR2−/− mice (n = 48/group). Mice were killed at 3, 7, 14, and 21 days, and the vena cavae containing the thrombus were harvested, washed in sterile PBS, fixed in 4% formalin solution, and embedded in paraffin. Thrombus size, recanalization, and macrophage content were measured in 5-μm sections taken at 300-μm intervals throughout the length of the specimen, as described above.

Generation of adenovirus construct expressing 35Kd protein (ad.35K) 

An adenoviral construct containing the 750 bp fragment of 35K of the vaccinia virus (Lister strain), incorporating a carboxy-terminal HA epitope tag, was produced as described previously.17 A recombinant adenovirus, ad.35K, was generated after transfection of 293 cells using the AdEasy system (Stragene, La Jolla, CA).18 A control recombinant adenovirus, containing the gene encoding enhanced green fluorescence protein (ad.EGFP, Clonetech, Mountain View, Calif) was used as a control for viral infection and was also prepared as described above. Viruses were isolated and purified as previously described.19

Assessment of cell migration 

A modified Boyden chamber assay with transwell membranes (6.0-mm diameter, 8-μm pore size) (Receptor Technologies, Adderbury, Oxon, UK) was used to assess specific CCR5 receptor-directed cell migration, as previously described.17 Briefly, 293 cells were grown to 50% confluence in Dulbecco modified Eagles media (DMEM) and 10% (v/v) fetal calf serum. The cells were then co-transfected (Fugene6, Roche, Lewes, East Sussex, UK) with plasmids encoding CCR5 and enhanced green fluorescence protein (EGFP) to facilitate localization. Transfected cells were harvested and allowed to migrate overnight towards serum samples placed in the lower chamber. Cells that migrated to the underside of membranes were fixed and quantified by computer analysis of EGFP fluorescence in confocal microscope images. Each experimental sample was analyzed in duplicate, and three separate images were quantified for each membrane.

Confirmation of adenoviral construct transfection 

Indirect confirmation of construct expression was carried out by fluorescence microscopy on sections of liver taken from animals treated with the control ad.EGFP. Direct confirmation of 35K expression in the plasma of ad35K treated animals, was carried out by Western blotting. Plasma, taken 2 weeks after injection, was incubated for 2 hours with monoclonal anti-HA agarose conjugated beads (Sigma, Gillingham, Dorset, UK). The beads were washed, diluted 1:1 in 2 × sodium dodecyl sulfate (SDS) sample buffer and the proteins denatured by heating at 95°C for 3 minutes. Beads were pelleted by centrifugation and the supernatant separated on 14% SDS-polyacrylamide gel electrophoresis gels. After transfer to polyvinylidene fluoride membranes, 35Kd protein was detected using rat monoclonal anti-HA high affinity antibody (Roche) followed by an anti-rat secondary antibody conjugated to horseradish peroxidase.

CC chemokine blockade, monocyte recruitment, and thrombus resolution 

Two cohorts of ten 8- to 12-week-old wild-type mice were injected with ad.35K or ad.EGFP (1 × 109 plaque-forming units in 300 μL sterile PBS) through the tail vein. The mice were anesthetized 48 hours after the injection, and thrombosis was induced in their vena cavae as described above. The mice were killed 2 weeks after thrombus induction, and thrombus and blood (ethylenediamine tetraacetic acid anticoagulant) were collected. The thrombus was washed in sterile PBS, fixed, and embedded in paraffin. Thrombus size, recanalization, and macrophage content were measured in 5-μm sections taken at 500-μm intervals throughout the length of the specimen, as described above.

Statistical analysis 

Data are presented as means ± standard error of the mean (SEM). Intergroup differences were determined by the Student’s t test or the Mann-Whitney U test, as appropriate. Differences in the MCP1 dosing experiments were analyzed with one-way analysis of variance (ANOVA) with post-test analysis (Tukey multiple comparison test). Differences between control and knockout mice resolution curves were analyzed by using two-way ANOVA. All data were analyzed with PRISM (version 3) statistical software (GraphPad Software, San Diego, Calif). A P < .05 was accepted as significant.

Back to Article Outline

Results 

Characteristics of thioglycollate-induced peritoneal macrophages 

Cell viability was >80%. At least 75% of the recovered cells were of a mononuclear cell phenotype and the remaining cells were polymorphonuclear in morphology as determined by Wright-Giemsa staining. Flow-cytometric and immunohistochemical analyses showed that 65% of the gated events were CD4+/CD3− cells (Fig 1, A), and >70% of the cells were ED1+ (Fig 1, B). The cells generated fibrinolytic activity that was equivalent to just over 1ng/mL of urokinase activity.

  • View full-size image.
  • Fig 1. 

    A, Fluorescence activated cell sorting analysis of peritoneal lavage cells demonstrates the CD4+/CD3− macrophage population in the upper left-hand quadrant. B, Peritoneal lavage fluid cytospin shows an abundance of mononuclear cells (arrows) stained by the rat macrophage marker, ED1 (brown stain, ×100 original magnification).

Intrathrombus injection of peritoneal macrophages 

Significantly greater numbers of macrophages (ED-1+ cells) were found in treated animals compared with controls (37.5% ± 2.4% vs 27.5% ± 3.1% at 7 days, P < .05). Thrombus size was reduced by over fivefold compared with controls (12.1 ± 1.7 mm3 vs 63.3 ± 7.3 mm3, P < .001), and thrombus recanalization was more than fourfold greater in animals treated with peritoneal-derived macrophages compared with controls (14.1% ± 1.4% vs 3.1% ± 0.7%, P < .001).

Intrathrombus injection of MCP1 

125-I-MCP1 was predominantly found in the thrombus at all the time intervals studied (Fig 2). Levels of 125-I MCP1 were greatest 3 hours after injection, but significant amounts were still present after 7 days. Radioactivity was also found in the vena cavae of treated animals, but was not detectable in other tissues (blood, aorta, lung, spleen, liver, or skin).

Effect on thrombus resolution 

Treatment with MCP1 resulted in a reduction in thrombus volume compared with controls (P < .001, ANOVA) (Fig 3, A). All three doses significantly reduced thrombus size compared with control (P < .05). The dose effect was greatest after injection with 100 ng MCP1 (9.8 ± 3.0 mm3 vs 63.3 ± 7.3 mm3 in controls, P < .001). Thrombus macrophage content did not differ from controls in any of the MCP1-treated groups (P = 0.1, ANOVA).

Treatment with MCP1 increased thrombus recanalization (P = .002, ANOVA) (Fig 3, B). This difference was significant at the 30- and 100-ng doses (P < .01), with the greatest response achieved at 100 ng of MCP1 (22.4% ± 4.0% vs 3.1% ± 0.7%, P < .01). Treatment with 1000 ng MCP1 also increased the recanalization of thrombi compared with control, but this failed to reach statistical significance (6.8% ± 2.0% vs 3.1% ± 0.7%; P > .05).

Monocyte recruitment and thrombus resolution in MCP-1−/− and CCR2−/− mice 

A similar temporal pattern of resolution was observed in MCP1−/− and wild-type controls (Fig 4, A). The size of thrombus produced in CCR2−/− mice was not significantly different from their wild-type controls (10.7 ± 1.9 mm3 vs 11.1 ± 1.9 mm3; P = .89), but subsequent resolution of the thrombus was significantly delayed in CCR2−/− mice compared with wild-type controls (P < .001, ANOVA, Fig 4, B). Significant differences in thrombus size were evident at 7, 14, and 21 days (P < .05), although by 21 days this difference was small.

  • View full-size image.
  • Fig 4. 

    A, Longitudinal study of thrombus resolution in MCP−/− mice (grey circles) and wild-type controls (black circles) and in (B) CCR2−/− mice (grey circles) and wild-type controls (black circles). *Denotes intergroup significance (P < .05). Values are ± SEM. ANOVA, Analysis of variance.

The thrombus macrophage content was significantly lower in the CCR2−/− mice than in their wild-type controls at day 7 (0.5% ± 0.02% vs 0.63% ± 0.02%; P = .006) and day 14 (0.5% ± 0.02% vs 0.61% ± 0.02%; P = .009).

CC chemokine system blockade 

Widespread expression of EGFP was detected in the liver parenchyma of the ad.EGFP-transfected mice (Fig 5, A). Detection of the 35Kd protein by Western blotting of plasma from animals injected with ad.35K mice indicated successful viral transfection. No protein could be detected in plasma taken from ad.EGFP-treated controls (Fig 5, B). Plasma obtained from ad.35K transfected mice almost halved the in vitro migration of CCR5-transfected 293 cells compared with plasma from ad.EGFP-treated mice (1193 ± 68 cells vs 1994 ± 385 cells, P = .037).

  • View full-size image.
  • Fig 5. 

    A, Section of liver from ad.EGFP-treated mouse showing successful gene adenoviral transfection and EGFP expression (green fluorescence arrows, ×100 original magnification). B, Western blot of from five ad.35K transfected mice. The 35K protein and plasma from ad.EGFP treated animal were used as positive and negative controls respectively.

Macrophage recruitment in 2-week-old venous thrombi was significantly reduced in ad.35K-infected animals (0.55% ± 0.03%) compared with controls (0.47% ± 0.01%; P = .036). The residual thrombus volume was more than fivefold greater in ad.35K-infected animals (2.59 ± 0.62 mm3) compared with controls (0.51 ± 0.22 mm3, P = .005) (Fig 6).

  • View full-size image.
  • Fig 6. 

    The effect of vaccinia virus-derived 35Kd protein blockade of C-C chemokines on (A) thrombus size and (B) thrombus recanalization harvested 14 days after tail vein gene transfer.

Back to Article Outline

Discussion 

Increasing the numbers of macrophages by direct injection of fibrinolytically active peritoneal cells into the thrombus resulted in a large reduction in thrombus size and an increase in recanalization. Although the peritoneal lavage fluid contained predominantly macrophages, it did contain significant numbers of other leukocytes, mainly polymorphonuclear cells, which may have influenced resolution. Neutrophils, for example, express a variety of enzymes (urokinase, cathepsin G, and elastase) that can degrade fibrin,20 and spontaneous lysis has been observed in polymorphonuclear-enriched thrombi produced in vitro.21 These cells also secrete matrix metalloproteinases such as MMP9 and angiogenic cytokines such as interleukin 8 that influence angiogenesis and tissue remodelling.22, 23

Treatment of experimental venous thrombi with MCP1 at doses between 30 ng and 1000 ng significantly reduced thrombus size, which is in agreement with our original pilot data in which a 1000-ng dose was used.11 A maximal effect (more than sixfold reduction in thrombus size) was achieved at a 100-ng dose of MCP1 in this study, with a lesser effect at the higher dose. The reason for this nonlinear response is not clear, but a similar effect has been reported for cells treated with vascular endothelial cell growth factor (VEGF) in vitro24 and thrombus treated with VEGF in vivo.12

The effect of MCP1 was, however, independent of monocyte numbers, as increasing MCP1 levels in the thrombus did not lead to enhanced monocyte recruitment. This dissociation between increased local or circulating MCP1 levels and monocyte chemotaxis has previously been demonstrated in MCP1 over-expressing mice.25 Our previous work has also shown that the thrombus contains comparatively lower levels of MCP1 during natural resolution.26 Treatment with a high MCP1 dose may have overwhelmed the subtle chemotactic gradient that exists in vivo.

Although the amount of MCP1 remaining within the thrombus after injection was not quantified, tracking studies demonstrated measurable amounts of labelled MCP1 remaining in the thrombus for at least 7 days. Only very small quantities of labelled MCP1 were found in distant tissues and the viscera. These data are consistent with the distribution of other injected radiolabeled cytokines, such as VEGF, that we have previously used in this thrombus model.26

It is probable that MCP1 enhanced thrombus resolution by stimulating recanalization, as treatment with this cytokine caused a more than a sevenfold increase in the recanalization area; but the mechanisms by which MCP1 regulates this process in the thrombus remain to be defined. MCP1 can promote angiogenesis through a direct action on endothelial cells27 and can stimulate macrophages to form vascular channels or express potent angiogenic growth factors such as VEGF.28, 29 Increased thrombus recanalization may, however, be stimulated by angiogenic cytokines without any effect on thrombus size.30 Improved thrombus contraction has been reported after treatment of the thrombus with angiogenic cytokines such as basic fibroblast growth factor.30 It is possible, therefore, that MCP1 may have stimulated thrombus recanalization in a similar manner in this study.

Deletion of the MCP1 gene had no effect on monocyte recruitment and thrombus resolution. This was not surprising, as a high degree of redundancy and promiscuity is known to exist between CC chemokines and their receptors.31, 32 Deletion of MCP1 may have simply resulted in the upregulation of other MCPs (MCP2, 3, and 4) and CC chemokines, which replaced the activity of MCP1. The design of this study centred, however, around histologic and image analytic end-points that were not amenable to the measurement of changes in the expression of other MCPs and chemokines. Blockade of all members of the CC chemokine family with a vaccinia-derived 35Kd protein, which has been shown to decrease monocyte recruitment into aortic plaques in mice,17, 33 also inhibited monocyte recruitment into thrombus and reduced thrombus resolution and recanalization in this study. The effects of blockade by the 35K protein were, however, no greater than those seen in the CCR2−/− mice, in whom there was also a reduced monocyte ingress and a marked delay in thrombus resolution. These data suggest that CCR2 activation is an important mechanism in the regulation of monocyte migration into thrombus.

Deletion of the gene encoding the CXCR2 receptor, which regulates neutrophil chemotaxis, has been shown to reduce both neutrophil and monocyte recruitment into the thrombus, which causes an initial delay in thrombus resolution that is lost after 8-days.22 It is not clear, however, whether this effect was the result of a direct action on monocyte chemotaxis or whether decreased neutrophil signalling caused a subsequent reduction in monocyte recruitment.

Back to Article Outline

Conclusion 

The findings of this study support the concept that the monocyte/macrophage is a key mediator of venous thrombus resolution6, 10 and suggests that the CC chemokine family and their receptors have an important role in the complex process that regulates their recruitment. Targeting of this receptor family with novel agonists may form the basis of a treatment that accelerates the resolution of venous thrombi and minimizes the effects of the post-thrombotic syndrome.

Back to Article Outline

Author Contributions 

Conception and design: T.A, J.H, K.G.B, C.B, K.C, D.G, B.R, I.F.C, A.S

Analysis and interpretation: T.A, J.H, K.G.B, A.S

Data collection: T.A, J.H, B.S

Writing the article: T.A, K.G.B, A.S

Critical revision of the article: T.A, J.H, K.G.B, B.S, C.B, K.C, D.G, B.R, I.F.C, A.S

Final approval of the article: T.A, J.H, K.G.B, B.S, C.B, K.C, D.G, B.R, I.F.C, A.S

Statistical analysis: T.A, J.H, A.S

Obtained funding: A.S

Overall responsibility: A.S

Back to Article Outline

References 

  1. Meissner MH . Thrombolytic therapy for acute deep vein thrombosis and the venous registry . Rev Cardiovasc Med . 2002;3(Suppl 2):S52–S59
  2. Johnson BF , Manzo RA , Bergelin RO , Strandness DE . Relationship between changes in the deep venous system and the development of the postthrombotic syndrome after an acute episode of lower limb deep vein thrombosis (a one- to six-year follow-up) . J Vasc Surg . 1995;21:307–312
  3. Tanaka K , Hirst AE , Smith LL . Rate of endothelialization in venous thrombi (an experimental study) . Arch Surg . 1982;117:1045–1048
  4. Schweizer J, Kirch W, Koch R, Elix H, Hellner G, Forkmann L, et al. Short- and long-term results after thrombolytic treatment of deep venous thrombosis . J Am Coll Cardiol . 2000;36:1336–1343
  5. Astrup T , Henrichsen J , Kwaan HC . Protease content and fibrinolytic activity of human leukocytes . Blood . 1967;29:134–138
  6. McGuinness CL , Humphries J , Waltham M , Burnand KG , Collins M , Smith A . Recruitment of labelled monocytes by experimental venous thrombi . Thromb Haemost . 2001;85:1018–1024
  7. Wakefield TW, Linn MJ, Henke PK, Kadell AM, Wilke CA, Wrobleski SK, et al  Neovascularization during venous thrombosis organization (a preliminary study) . J Vasc Surg . 1999;30:885–892
  8. Knighton DR , Fiegel VD . Macrophage-derived growth factors in wound healing (regulation of growth factor production by the oxygen microenvironment) . Am Rev Respir Dis . 1989;140:1108–1111
  9. Knighton DR , Fiegel VD . The macrophages (effector cell wound repair) . Prog Clin Biol Res . 1989;299:217–226
  10. Singh I , Burnand KG , Collins M , Luttun A , Collen D , Boelhouwer B , et al.   Failure of thrombus to resolve in urokinase-type plasminogen activator gene-knockout mice (rescue by normal bone marrow-derived cells) . Circulation . 2003;107:869–875
  11. Humphries J , McGuinness CL , Smith A , Waltham M , Poston R , Burnand KG . Monocyte chemotactic protein-1 (MCP-1) accelerates the organization and resolution of venous thrombi . J Vasc Surg . 1999;30:894–899
  12. Waltham M , Burnand KG , Collins M , Smith A . Vascular endothelial growth factor and basic fibroblast growth factor are found in resolving venous thrombi . J Vasc Surg . 2000;32:988–996
  13. Lu B , Rutledge BJ , Gu L , Fiorillo J , Lukacs NW , Kunkel SL , et al.   Abnormalities in monocyte recruitment and cytokine expression in monocyte chemoattractant protein 1-deficient mice . J Exp Med . 1998;187:601–608
  14. Rollins BJ , Morrison ED , Stiles CD . Cloning and expression of JE, a gene inducible by platelet-derived growth factor and whose product has cytokine-like properties . Proc Natl Acad Sci U S A . 1988;85:3738–3742
  15. Singh I , Smith A , Vanzieleghem B , Collen D , Burnand K , Saint-Remy JM , et al.   Antithrombotic effects of controlled inhibition of factor VIII with a partially inhibitory human monoclonal antibody in a murine vena cava thrombosis model . Blood . 2002;99:3235–3240
  16. Fisker S , Kudahl K , Sonne O . In vivo inflammatory stimulation induces a transient change in the binding of thrombin to rat peritoneal macrophages . Exp Cell Res . 1992;201:145–153
  17. Bursill CA , Cai S , Channon KM , Greaves DR . Adenoviral-mediated delivery of a viral chemokine binding protein blocks CC-chemokine activity in vitro and in vivo . Immunobiology . 2003;207:187–196
  18. He TC , Zhou S , da Costa LT , Yu J , Kinzler KW , Vogelstein B . A simplified system for generating recombinant adenoviruses . Proc Natl Acad Sci U S A . 1998;95:2509–2514
  19. Channon KM , Blazing MA , Shetty GA , Potts KE , George SE . Adenoviral gene transfer of nitric oxide synthase (high level expression in human vascular cells) . Cardiovasc Res . 1996;32:962–972
  20. Moir E , Booth NA , Bennett B , Robbie LA . Polymorphonuclear leucocytes mediate endogenous thrombus lysis via a u-PA-dependent mechanism . Br J Haematol . 2001;113:72–80
  21. Mutch NJ , Moir E , Robbie LA , Berry SH , Bennett B , Booth NA . Localization and identification of thrombin and plasminogen activator activities in model human thrombi by in situ zymography . Thromb Haemost . 2002;88:996–1002
  22. Henke PK , Varga A , De S , Deatrick CB , Eliason J , Arenberg DA , et al.   Deep vein thrombosis resolution is modulated by monocyte CXCR2-mediated activity in a mouse model . Arterioscler Thromb Vasc Biol . 2004;24:1130–1137
  23. Henke PK , Wakefield TW , Kadell AM , Linn MJ , Varma MR , Sarkar M , et al.   Interleukin-8 administration enhances venous thrombosis resolution in a rat model . J Surg Res . 2001;99:84–91
  24. Pepper MS , Ferrara N , Orci L , Montesano R . Potent synergism between vascular endothelial growth factor and basic fibroblast growth factor in the induction of angiogenesis in vitro . Biochem Biophys Res Commun . 1992;189:824–831
  25. Gu L , Rutledge B , Fiorillo J , Ernst C , Grewal I , Flavell R , et al.   In vivo properties of monocyte chemoattractant protein-1 . J Leukoc Biol . 1997;62:577–580
  26. Waltham M , Burnand KG , Collins M , McGuinness CL , Singh I , Smith A . Vascular endothelial growth factor enhances venous thrombus recanalization and organisation . Thromb Haemost . 2003;89:169–176
  27. Salcedo R , Ponce ML , Young HA , Wasserman K , Ward JM , Kleinman HK , et al.   Human endothelial cells express CCR2 and respond to MCP-1 (direct role of MCP-1 in angiogenesis and tumor progression) . Blood . 2000;96:34–40
  28. Moldovan NI , Goldschmidt-Clermont PJ , Parker-Thornburg J , Shapiro SD , Kolattukudy PE . Contribution of monocytes/macrophages to compensatory neovascularization (the drilling of metalloelastase-positive tunnels in ischemic myocardium) . Circ Res . 2000;87:378–384
  29. Moldovan NI , Asahara T . Role of blood mononuclear cells in recanalization and vascularization of thrombi (past, present, and future) . Trends Cardiovasc Med . 2003;13:265–269
  30. Varma MR , Moaveni DM , Dewyer NA , Varga AJ , Deatrick KB , Kunkel SL , et al.   Deep vein thrombosis resolution is not accelerated with increased neovascularization . J Vasc Surg . 2004;40:536–542
  31. Rossi D , Zlotnik A . The biology of chemokines and their receptors . Annu Rev Immunol . 2000;18:217–242
  32. Van Coillie E , Van Damme J , Opdenakker G . The MCP/eotaxin subfamily of CC chemokines . Cytokine Growth Factor Rev . 1999;10:61–86
  33. Bursill CA , Choudhury RP , Ali Z , Greaves DR , Channon KM . Broad-spectrum CC-chemokine blockade by gene transfer inhibits macrophage recruitment and atherosclerotic plaque formation in apolipoprotein E-knockout mice . Circulation . 2004;110:2460–2466

 T. A. was funded by a Clinical Research Fellowship from the Guy’s & St Thomas’ Charitable Foundation. J. H. was funded by the British Heart Foundation. I. F. C. is supported by National Institutes of Health grants HL52773 and HL063894.Competition of interest: none.

PII: S0741-5214(05)01924-5

doi:10.1016/j.jvs.2005.10.073

Journal of Vascular Surgery
Volume 43, Issue 3 , Pages 601-608, March 2006

Article Tools

* Image Usage & Resolution