Low Molecular Weight Fucoidan Prevents Neointimal Hyperplasia After Aortic Allografting

Abstract

Background. 

Fucoidan, a new low molecular weight sulfated polysaccharide (LMWF), has previously been shown to mobilize bone marrow-derived progenitors cells via stimulation of stromal derived factor (SDF)-1 release. Mobilized progenitor cells have been suggested to repair intimal lesions after immune-mediated endothelial injury and thus prevent intimal proliferation. The aim of this study was to evaluate the effect of LMWF treatment in a rat aortic allograft model of transplant arteriosclerosis (TA).

Methods. 

Aortic grafts were performed in Brown Norway (BN, donor) and Lewis (Lew, recipient) rats. The recipient rats were treated with LMWF (5 mg/kg/day) and sacrificed at 30 days. To determine the role of SDF-1 in mediating the effects of LMWF, a specific inhibitor of the SDF-1 receptor CXCR4, AMD 3100 (20 μg/kg/day), was used. The grafted segments were evaluated by morphometric (histochemical) analyses.

Results. 

Untreated aortic allografts exhibited severe intimal proliferation, indicative of TA. In contrast, LMWF treatment significantly prevented allograft intimal proliferation as compared with controls (5.7±3 vs. 66.2±6 μm, P<0.01) and permitted a normalization of the intima/media ratio (0.1±0.1 vs. 1.7±0.3, P<0.01). Further, LMWF treatment stimulated allograft reendothelialization, as evidenced by strong intimal endothelial nitric oxide synthase antibody and CD31 signals. Unexpectedly, AMD treatment failed to prevent the protective effect of LMWF on intimal thickening and AMD treatment alone was found to reduced intimal proliferation in allografts.

Conclusions. 

We found that LMWF treatment reduced intimal thickness and induced the presence of an endothelial cell lining in the vascular graft at 30 days. Our findings may suggest a novel therapeutic strategy in the prevention of TA.

Chronic vascular rejection-induced transplant arteriosclerosis (TA) is considered to constitute the main cause of organ loss 1 year after allografting. TA is characterized by progressive intimal thickening and perivascular inflammation, resulting in the development of an arterial occlusive lesion leading to loss of organ function due to reduced blood perfusion. The TA process is initiated by an immune and inflammatory attack on the different cellular components of the arterial wall. In response to endothelial injury, occlusive neointimal lesions develop as an arterial repair mechanism. Smooth muscle cells (SMC) migrate from the media and proliferate in the intima in response to injury (¹). Therefore, the intimal proliferation in TA was originally considered to be of grafted tissue origin (i.e., donor-derived). However, there is now increasing evidence from experimental and clinical studies which show that neointimal cells are primarily of recipient origin (^2–6).

Bone marrow progenitor cells may play a pivotal role during TA as they could contribute to vascular remodeling and repair of damaged tissue (^7–10), and promote both endothelial and vascular smooth muscle cell recolonization of the arterial wall (¹¹). Bone marrow-derived vascular progenitor cell mobilization has recently been reported to inhibit intimal proliferation in a mechanical model of arterial wall injury, such as balloon-induced deendothelialization (^12–16).

We hypothesized that after initial immune-induced endothelial damage early intimal repair by autologous cells may prevent intimal proliferation. In a previous study, we have shown that in situ seeding of recipient syngeneic SMC limited intimal proliferation by accelerating the intimal healing process (¹⁷). A low molecular weight sulfated polysaccharide, Fucoidan P240Red (LMWF), has been shown to be able to mobilize bone marrow-derived progenitor cells (^18–21). The molecular mechanisms of LMWF may involve stromal derived factor-1 (SDF-1), as treatment with LMWF was found to increase the plasma concentration of SDF-1, in a dose and time-dependent fashion that correlated with the mobilization of bone marrow-derived progenitors cells (^18–21). Additionally, bone marrow-derived progenitors cells homing to sites of vascular injury has also been proposed to involve the specific molecular interaction of SDF-1 with its receptor CXCR4 (²²). In the present study, we investigated the effects of LMWF treatment on prevention of neointima formation in a rat aortic allograft model, characterized by arterial wall immune aggression.

MATERIALS AND METHODS

All experiments performed in this study conformed to the Guiding Principles in the Care and Use of Animals approved by the American Physiological Society. Rats from the two inbred strains Brown Norway (BN) and Lewis (Lew) were purchased from Iffa-Credo Laboratory (Labresle, France). Male rats weighing between 225 g and 250 g were used. The animals were allowed access to food and water ad libitum before and after surgery.

Aortic Transplantation

Rats were anesthetized by intraperitoneal injections of xylazine (5 mg/kg, ROMPUM 2% Bayer, France) and ketamine (100 mg/kg, IMALGENE 1000, Merial, France). Recipient rats were transplanted orthotopically with a 1-cm long segment of allogeneic or isogeneic abdominal aorta using end-to-end sutures as previously described (⁶). No immunosuppressive or anticoagulant treatment was used. All groups of rats underwent the surgical procedure. Allografts were performed in BN donor to Lew recipient and isografts in BN donor to BN recipient.

Pharmacological Products

LMWF was obtained by radical processing of high molecular weight extracts from brown seaweed (Iffremer, France) (¹⁸). The characteristics of LMWF are as follows: weight-average molecular mass 7±2 kDa; fucose content 35% (w/w); uronic acid content 3% (w/w); and sulfate content 34% (w/w). LMWF treatment was initiated immediately after aortic transplantation and maintained for 30 days at a dose of 5 mg/kg/day by daily, intramuscular injections.

AMD 3100 (AMD, Sigma, France), a specific CXCR4 antagonist which blocks the binding of SDF-1 to its receptor (²²), was administered at a dose of 20 μg/kg/day by daily, subcutaneous injections for 30 days.

Experimental Design

The following groups were used: an allografted group (BN-Lew) treated with LMWF (n=10), an isografted group (BN-BN) treated with LMWF (n=10), an untreated allografted group (BN-Lew) (n=10), and an untreated isografted (BN-BN) group (n=10). Two additional allografted groups (BN-Lew) (n=10) were treated with AMD 3100, alone or concomitantly with LMWF-treatment during 30 days. All rats were sacrificed at 30 days with an overdose of pentobarbital (50 mg/kg) by intraperitoneal injection. The infrarenal aorta was rapidly dissected and removed for histological evaluations.

SDF-1 Levels in Plasma

Plasma concentrations of SDF-1 were measured at day 3, day 15, and day 30 throughout the entire period of LMWF treatment in the four groups: untreated isografts (n=5), LMWF treated isografts (n=5), untreated allografts (n=5) and LMWF-treated allografts (n=5). One hour after the last injection of LMWF, blood was extracted by ophthalmic vein puncture and immediately centrifuged to collect plasma. Plasma concentrations of SDF-1 were determined using an enzyme-linked immunoabsorbent assay kit (Quantikine, Oxon, UK).

Histology

Histological examination was performed in all grafted groups. For this purpose, one half of the grafted aorta was filled with Tissue-TEK and snap-frozen immediately in isopentane precooled by liquid nitrogen and stored at –80°C until sectioning. The other half of the transplant was postfixed in 4% formalin for at least 24 hr. The tissue was embedded in paraffin and transversal sections (5 μm) were obtained to allow light microscopic examination. Slides were stained with hematoxylin and eosin. Each slide was examined for qualitative and quantitative evaluation.

Morphometric Analyses

Semiautomatic image analysis was used to quantify histological features in formalin-fixed sections. The image system: included a light microscope (Nikon E600 microscope), a color video camera (Spot Software v2.21 Camera, Diagnostic Instruments), an image analysis processor (Software image pro-plus 4.1 for Windows 95), and a personal computer connected to the processor that permitted the automatic analysis of all morphological transformations and sequential storage of the results. For each stain, measurements were performed on the intimal, medial and adventitial tunicae. Two approaches were used in this study. The first approach was to analyze the elastin structure of the aortic wall. The elastin content permitted the determination of the intimal and medial thickness (μm) and the intima/media ratio. Thus, intimal thickness was defined and measured between the luminal limit of the intima and the internal elastic limiting laminae. The second approach was to analyze the nuclear content of the graft vessel wall. The mean cell density (number of nuclei per millimeter square of the aortic wall; C/mm²) was measured in the three components of the aorta: in intima, media, and adventitia.

Immunohistochemistry

The cryopreserved half of the transplant was stored at −80°C prior to cryo-sectioning (5 μm). Immunohistological staining was performed using streptavidine peroxidase technique (Amersham, France) in order to reveal the three different monoclonal antibodies used in this study: an anti endothelial nitric oxide synthase antibody (eNOS/NOS type III, BD Transduction Laboratories, San Diego, CA) reactive against endothelial cells, an anti-CD31 antibody (CD31, BD Pharmingen, France) reactive against endothelial cells and an anti-alpha smooth muscle actin (αSMA) antibody (monoclonal mouse anti-human αSMA, Dakocytomation, France) reactive against SMCs.

Statistical Analyses

The results of the morphometric study are expressed as means±SEM. To analyze differences in morphometric measurement, Student's two-tailed t test was performed to compare two groups. The difference was considered significant for P<0.05 (Statview Software).

RESULTS

LMWF Treatment Reduces Neointimal Formation in Allografts

To determine the structural aspects of the vascular grafts, we performed a qualitative histochemical approach and quantitative morphometrical measurements.

No sign of rejection was observed in the isografted aorta: no intimal proliferation, no inflammatory adventitial infiltration, and the medial thickness was preserved (Fig. 1A, B).

FIGURE 1.: Qualitative analysis: LMWF treatment prevents intimal hyperplasia in allografted aortas. (A) No sign of rejection was observed in untreated isografted aortas. No intimal proliferation or inflammatory adventitial infiltration was observed, and the medial thickness was respected (hematoxylin & eosin, 10×). (B) LMWF did not alter the histological aspects of isografted aortas (hematoxylin and eosin, 10×). (C) In the absence of LMWF treatment, the three layers of aortic allografts displayed significant alterations at 30 days, including increased adventitial cell infiltration, medial smooth muscle cells disappearance, reduced medial thickness, and intimal thickening (hematoxylin and eosin, 10×). (D) LMWF-treated allografts were devoided of intimal thickening at 30 days. Qualitative histological examination did not permit analyses of medial thickness modifications (hematoxylin and eosin, 10×).

In contrast, all untreated aortic allografts exhibited histological signs of severe arterial graft rejection at 30 days with a concentric intimal proliferation (Fig. 1C). The intimal thickness was significantly increased in allografts as compared with isografts (Table 1). Medial allograft thickness was dramatically reduced as compared to isografts, due to the disappearance of SMCs in the allograft (Table 1, Figure 1C). Thus, the intima/media ratio was substantially elevated in untreated allografts as compared with untreated isografts (Table 1). Further, adventitial cell density was higher in allografts than in isografts due to inflammatory cell infiltration (Table 1, Figure 1C).

TABLE 1: Quantitative analyses: LMWF prevents intimal hyperplasia and partially rescues medial layer thickness

As compared to the untreated allograft group, LMWF treatment attenuated the rejection reaction in the allografted aorta (Fig. 1D). LMWF treatment dramatically reduced the intimal thickening of the allografts as compared to the untreated allograft control group (Table 1). Indeed LMWF treatment restored the intimal layer in allografts to levels similar to that of isograft controls (5.7±3 μm in treated allografts vs. 4.2±3 μm in untreated isografts, P=0.7). However, intimal cell density was not modified by LMWF (Table 1). Furthermore, LMWF significantly prevented destruction of the media (Table 1, Figure 1D). Nevertheless, medial thickness in allografts treated with LMWF remained lower than in isografts (59.8±2 μm in treated allografts vs. 89.8±6 μm in untreated isografts, P<0.01). LMWF did not significantly increase medial cell density in allografts (Table 1). Thus, LMWF treatment partially restored intima/media ratio to isograft levels (Table 1). Additionally, LMWF partially but significantly decreased adventitial cell density (Table 1) but failed to normalize the adventitial characteristics (8.2×10³± 0.4×10³ C/mm² in treated allografts vs. 2.5×10³± 0.4×10³ C/mm² in untreated isografts, P<0.01). In contrast, LMWF treatment of isografted aortas did not affect any histological aspects, including degree of intimal thickness, medial thickness, or cell density in the media or the adventitia (Fig. 1B, Table 1).

LMWF Treatment Induces a Reendothelialization in Aortic Allografts

To determine the presence of an endothelial cell lining in the vascular graft, we performed immunohistochemical analyses of endothelial cell and SMC markers. First, eNOS immunostaining revealed the presence of a single layer of endothelial cells in untreated isografts (Fig. 2A), indicative of an intact intima. Conversely, eNOS staining was absent in the untreated allograft group (Fig. 2C). Promisingly, strong eNOS signal was found on endothelial cells in the intima in the LMWF-treated allografts group (Fig. 2D) displaying a level of endothelial staining comparable to both treated and untreated isografts (Fig. 2A, B). The endothelial phenotype was confirmed by CD31 staining present in the intima and vasavasorum of the adventitia of treated and untreated isografts (Fig. 2E, F). We found that untreated allografts exhibited markedly intimal thickening covered with poor CD31 staining (Fig. 2G), whereas LMWF-treated allografts displayed strong CD31 signals (Fig. 2H), confirming the intact endothelial cell lining.

FIGURE 2.: LMWF therapy induces reendothelialization and prevents intimal SMC proliferation but only partly inhibits medial SMC destruction. eNOS immunostaining revealed the presence of a single layer of endothelial cells in untreated isografts (A, 25×) and in LMWF-treated isografts (B, 25×). In contrast, no eNOS staining was observed in untreated allografts (C, 25×). Strongly eNOS positive cells (D) were however found in LMWF-treated allografts (10×). Similarly, CD31 positive staining was observed in the intima and vasa vasorum of the adventitia of untreated isografts (E, 25×) and in treated isografts (F, 25×). Untreated allografts were CD31 negative (G, 25×), whereas CD31 immunostaining confirmed the endothelial phenotype in LMWF-treated allografts (H, 25×, and H', 40×; hematoxylin counterstained section). (I and J) αSMA positive immunostaining was prominent in the media, but absent in the intima, of untreated or LMWF-treated isografts (10×). Untreated allografts exhibited high levels of αSMA staining (K) in the intima while αSMA positive cells disappeared in the media (10×). LMWF-treated allografts (L) displayed only a few αSMA stained cells in the intima whereas the media was weakly αSMA positive (10×).

To determine the localization of SMCs in the aortic segment αSMA immunostaining was performed. As expected, SMC were prominent in the media, but absent in the intima, of untreated isografts (Fig. 2I). In contrast, untreated allografts exhibited high levels of αSMA staining in the intima, indicative of the intimal proliferation whereas αSMA positive cells disappeared from the media (Fig. 2K). However, LMWF-treated allografts showed only a few αSMA positive cells in the intima, a finding in line with the previous observation of reduced intimal proliferation by LMWF treatment. Nevertheless, the media was only weakly αSMA positive, again indicating that LMWF did not significantly protect against immune-mediated medial cell destruction (Fig. 2L).

LMWF Increases SDF-1 Plasma Levels

In order to analyze the effects of LMWF treatment on SDF-1 in our model, we measured SDF-1 levels in plasma by enzyme-linked immunosorbent assay (Fig. 3).

FIGURE 3.: Increase in plasma concentration of SDF-1 by LMWF treatment. The plasma concentrations of SDF-1 were measured at days 3, 15, and 30 after transplantation. The measurements were performed in four groups of rats (n=5/group): untreated isografts, LMWF-treated isografts, untreated allografts and LMWF-treated allografts. Dissimilar to in untreated isografts, SDF-1 levels gradually increased in untreated allografts after transplantation. LMWF further significantly increased SDF-1 plasma levels in the allografts groups as compared with untreated allografts, at day 3 but not at day 15 and day 30.

SDF-1 levels gradually increased in untreated allografts after transplantation (1553 pg/ml±39 at day 3; 1793 pg/ml±173 at day 15; and 2069 pg/ml±200 at day 30). LMWF injection markedly increased SDF-1 plasma levels in the isograft group as compared to the untreated isograft group throughout the entire period of the study (2208 pg/ml±109 vs. 1305 pg/ml±205 at day 3, P<0.05; 1806 pg/ml±96 vs. 1467 pg/ml±209 at day 15, P<0.05; 1809 pg/ml±156 vs. 1409 pg/ml±156 at day 30, P<0.05, respectively). Similarly, SDF-1 levels were significantly increased by LMWF treatment in allografts as compared to untreated allografts at day 3 (2268 pg/ml±535 in treated allografts vs. 1553 pg/ml±39 in untreated allografts at day 3, P<0.05). However, the effects of LMWF on SDF-1 levels in allografts were not significant at day 15 and day 30 (1827 pg/ml±300 in treated allograft vs. 1793 pg/ml±173 in untreated allograft at day 15, P=0.2; 2115 pg/ml±228 in treated allograft vs. 2069 pg/ml±200 in untreated allograft at day 30, P=0.15).

Inhibition of CXCR4 Reduces Intimal Thickening in Allografts

To determine the role of SDF-1 in mediating the effects of LMWF in protecting the allograft, we examined the effect of a specific inhibitor of CXCR4, AMD. We surprisingly found that the ratio of intima/media was decreased in AMD alone treated allografts as compared to untreated allografts (0.4±0.1 in AMD-treated allografts vs. 1.7±0.3 in untreated allografts, P<0.01; Fig. 4). However the intima/media ratio remained higher in AMD alone treated allografts than in LMWF alone treated allografts indicating that LMWF treatment better protects the allograft than AMD treatment (Fig. 4). When AMD treatment was added to LMWF treatment, the intima/media ratio was less favorable in allografts as compared to LMWF treatment alone (0.4±0.2 in LMWF and AMD-treated allografts vs. 0.1±0.1 in LMWF-treated allografts, P<0.01). Thus the result was similar to AMD-alone treated allografts, but remained dramatically improved as compared to untreated allografts (Fig. 4).

FIGURE 4.: LMWF treatment normalizes the intima/media ratio but the effect is not altered by AMD-mediated CXCR4 blockage. Two allografted groups (BN-Lew, n=10) were daily treated subcutaneously with AMD, alone or in combination with LMWF, during 30 days. Qualitative (hematoxylin & eosin, 10×) and quantitative (E) morphological evaluations revealed that LMWF treatment alone (A) or AMD treatment alone (C), compared to untreated allografts (B), dramatically reduced the intima/media ratio (P<0.01). The two combined treatments were effective to reduce the intima/media ratio in allografts (D), showing that AMD failed to inhibit LMWF-mediated effects on intima/media ratio (P=0.95).

DISCUSSION

In the present study we demonstrated that LMWF treatment protected, for up to 30 days, against severe aspect of arterial allograft lesions, such as intimal thickening in response to chronic vascular rejection. Graft vascular disease remains a major limitation to long-term organ graft survival. Conventional immunosuppressive agents generally fail to be effective for the treatment or the prevention of TA. Promisingly, in our rat aortic allograft model, 5 mg/kg/day LMWF significantly reduced the gravity of the immune-mediated intimal lesion.

Intimal proliferation occurs after vascular wall immune aggression due to endothelial destruction. Many clinical and experimental studies suggest that the neointimal cells are of recipient origin (^2–6). We hypothesized that early autologous endothelial repair could limit the intimal proliferation. Previously, we have evaluated the effect of direct cell seeding in mechanically deendothelialized arterial segments in rats prior to allogeneic aortic transplantation. Indeed, we showed that syngeneic recipient SMC seeding reduced the intimal proliferation 2 months after arterial transplantation (¹⁷). Other studies have obtained similar results using recipient syngeneic bone marrow-derived mononuclear cell seeding of deendothelialized aortic segments followed by allograft transplantation (¹⁶). On the other hand side, the ability of progenitor cells to repair an endothelial injury in response to cell mobilizing therapy has been previously shown in several studies (^{13, 14}). For example, Werner et al., using a carotid artery balloon injury model, demonstrated that bone marrow-derived progenitor cell mobilization was able to reduce neointima formation and stimulate vascular re-endothelialization (¹³). We expected early intimal healing by LMWF-induced mobilization of autologous bone marrow-derived cells, to prevent intimal thickening during vascular wall immune-rejection. Similarly, LMWF has previously been shown to prevent intimal proliferation in a mechanically deendothelialized arterial model (²³). Interestingly, the ability of LMWF to prevent intimal proliferation may also be dependant on other biological effects. LMWF was found to inhibit intimal hyperplasia in a rat balloon-carotid injury model via a direct effect on SMC proliferation and migration, in vitro and in vivo (^{24, 25}). LMWF was also shown to promote a pro-angiogenic stimulation during ischemia/reperfusion and may further be implicated in endothelial cell differentiation or survival (^{26, 27}). Although the main protective effect of LMWF demonstrated in our study was inhibition of intimal thickening, we also observed that the allograft medial thickness was markedly preserved by the treatment. Similarly, we recently demonstrated a significant parenchymal protective effect of LMWF in a rat cardiac allograft model, including cardiomyocyte preservation and reduced inflammatory cell infiltration and fibrosis (²⁸). Reduced adventitial cell infiltration and medial preservation in the allograft may be explained by the general anti-inflammatory effects of LMWF. For instance, LMWF has been found to interfere with the inflammatory response, in particular the release of cytokines by activated monocytes (^{29, 30}). Indeed, direct inhibition by LMWF of selectin-mediated leukocyte recruitment and neutrophil infiltration may be implicated in the protective effects of LMWF on the arterial wall (³¹). Nevertheless, in our study, progenitor cell mobilization is considered to be the main effect of LMWF in prevention of intimal proliferation.

LMWF has been shown to release sequestered chemokines, such as SDF-1 (^{18, 21}), from their heparan sulfate proteoglycan anchors in vitro. Recently Luyt et al. showed that LMWF treatment in rats increased serum levels of SDF-1 in a dose-dependent fashion (³²). These results were confirmed in our models where serum levels of SDF-1 were found to be increased after 3 days of LMWF treatment in both isografted and allografted rats. SDF-1 is known to stimulate the recruitment and migration of leukocytes and progenitor cells, through interaction with its receptor CXCR4 (¹⁸). However, the beneficial effects of SDF-1 in vascular pathologies remain controversial. Sakihama et al. demonstrated that SDF-1/CXCR4 interaction contributes to the development of neointima formation in aortic allografts in mice (³³). In contrast, in several other studies, SDF-1 in close interaction with CXCR4 appears to play a central role in the recruitment of progenitor cells thus potentially enhancing the intimal repair process in different endothelial injury models (^34–37). Indeed, SDF-1 seems to play a key role for hematopoietic stem cell recruitment and vascular remodeling after tissue/organ injury (³⁶). Interestingly, in our study, in the absence of any LMWF treatment, SDF-1 serum levels appeared to be significantly higher in allografted than in isografted rats at the three time points after surgery. In response to an arterial allograft, SDF-1 overexpression may be attributed to vascular wall remodeling (³³). Our hypothesis was that after arterial wall immune- aggression, early treatment with LMWF should increase SDF-1 leading to enhanced mobilization of vascular progenitor cells which participate in the endothelial healing process, and thus prevents intimal proliferation. In contrast, as developed by Sakihama et al., a late increase of SDF-1 has been proposed to delay endothelial repair and thus to interfere with intimal remodeling (³³).

Additional molecular mechanisms of the progenitor cell-mobilizing effects of LMWF therapy include stimulation of endothelial progenitor cell proliferation in vitro (²⁶) and induction of the release of monocyte chemoattractant protein-1 (MCP-1), another progenitor cell regulatory cytokine, leading to improved progenitor cell recruitment during endothelial repair (²⁰). Fujiyama et al. showed that intra-arterial delivery of bone marrow-derived cells depended on MCP-1-mediated mechanism for cellular adhesion to the injured endothelium and for reduction of neointima formation (³⁸).

To test the role of the SDF-1/CXCR4 axis during progenitor cell mobilization and local recruitment after LMWF treatment, the inhibitor AMD was used in combination with LMWF in the aortic allograft model. AMD, a selective antagonist of CXCR4, has been demonstrated to block the SDF-1/CXCR4 interaction, and to inhibit in vitro chemotactic responses of human monocytes to SDF-1 (^{34, 39}). First, we surprisingly found that AMD used alone was able to reduce intimal thickness in the aortic allograft model. This result is to be considered in relation to recent studies that unexpectedly revealed that AMD was able to mobilize stem cells in a mouse model of hematopoietic cells transplantation (⁴⁰). Other effects attributed to AMD include proangiogenic properties in a murine hind limb ischemic model, which was related to stem cell mobilization and endothelial progenitor cell recruitment to areas of vascular injury (⁴¹). Secondly, we observed that AMD used in combination with LMWF did not prevent the reduction in intimal thickening in the aortic allograft model. Thus, despite our initial objective to block the SDF-1/CXCR4 axis, the use of AMD failed to inhibit the intimal protective effect of LMWF. In fact, the effects of LMWF and AMD may be independent. Increases in SDF-1 plasma levels induced by LMWF treatment appeared as plausible evidence for LMWF-induced bone marrow progenitor cells mobilization via the establishment of a SDF-1 gradient (²¹). Conversely, AMD may release bone marrow progenitor cells via CXCR4/SDF1 anchorage in bone marrow (⁴²). Therefore, this pharmacological approach did not allow us to explore the role of the SDF-1/CXCR4 axis in the mechanism of LMWF during vascular repair and remodeling.

In conclusion, we found that the natural polysaccharide compound LMWF reduced intimal hyperplasia and stimulated the formation of an endothelial cell lining in the vascular allograft after 1 month of treatment. Our findings suggest novel therapeutic strategies in the prevention of human TA.

ACKNOWLEDGMENT

The authors thank Dr. Brakenheielm, Inserm U644, Rouen, France, for valuable advice in editing the manuscript.