Bone marrow-Derived cells have been proposed to form new vessels or at least incorporate into growing vessels in adult organisms under certain physiological and pathological conditions. We investigated whether bone marrow-Derived cells incorporate into vessels using mouse models of hindlimb ischemia (arteriogenesis and angiogenesis) and tumor growth. C57BL/6 wild-type mice were lethally irradiated and transplanted with bone marrow cells from littermates expressing enhanced green fluorescent protein (GFP). At least 6 weeks after bone marrow transplantation, the animals underwent unilateral femoral artery occlusions with or without pretreatment with vascular endothelial growth factor or were subcutaneously implanted with methylcholanthrene-induced fibrosarcoma (BFS-1) cells. Seven and 21 days after surgery, proximal hindlimb muscles with growing collateral arteries and ischemic gastrocnemius muscles as well as grown tumors and various organs were excised for histological analysis. We failed to colocalize GFP signals with endothelial or smooth muscle cell markers. Occasionally, the use of high-power laser scanning confocal microscopy uncovered false-positive results because of overlap of different fluorescent signals from adjacent cells. Nevertheless, we observed accumulations of GFP-positive cells around growing collateral arteries (3-fold increase versus nonoccluded side, P <0.001) and in ischemic distal hindlimbs. These cells were identified as fibroblasts, pericytes, and primarily leukocytes that stained positive for several growth factors and chemokines. Our findings suggest that in the adult organism, bone marrow-Derived cells do not promote vascular growth by incorporating into vessel walls but may function as supporting cells.

Two signaling receptors for vascular endothelial growth factor (VEGF) in the vasculature are known with not yet well-understood roles in collateral vessel growth (arteriogenesis). In this study, we examined the involvement of the two VEGF receptors in arteriogenesis. Therefore, we used the VEGF homologue placenta growth factor (PlGF), which only binds to VEGFR-1 and VEGF-E, which only recognizes VEGFR-2. These peptides were locally infused over 7 days after ligation of the femoral artery in the rabbit. Evaluation of collateral growth by determining collateral conductance and angiographic scores demonstrated that the VEGFR-1–specific PlGF contributed significantly more to arteriogenesis than the VEGFR-2 specific VEGF-E. The combination of VEGF-E and PlGF did not exceed the effect of PlGF alone, indicating that cooperation of the two VEGF receptors in endothelial cell signaling is not required for arteriogenesis. In an in vitro model of angiogenesis, VEGF and VEGF-E were comparably active, whereas PlGF displayed no activity when given alone and did not further increase the effects of VEGF or VEGF-E. However, PlGF was as potent as VEGF when monocyte activation was assessed by monitoring integrin surface expression. In addition, accumulation of activated monocytes/macrophages in the periphery of collateral vessels in PlGF-treated animals was observed. Furthermore, in monocyte-depleted animals, the ability of PlGF to enhance collateral growth in the rabbit model and to rescue impaired arteriogenesis in PlGF gene–deficient mice was abrogated. Together, these data indicate that the arteriogenic activity observed with the VEGFR-1–specific PlGF is caused by its monocyte-activating properties.

Arteriogenesis has been associated with the presence of monocytes/macrophages within the collateral vessel wall. Induced macrophage migration in vivo is driven by the binding of monocyte chemoattractant protein-1 (MCP-1, or CCL2 in the new nomenclature) to the CCR2-chemokine receptor on macrophages. To determine whether the CCL2-CCR2 signaling pathway is involved in the accumulation of macrophages in growing collateral vessels, we used mice that are deficient in CCR2 in a model of experimental arterial occlusion and collateral vessel growth. In an in vitro CCL2-driven chemotaxis assay, mononuclear cells isolated from wild-type BALB/c mice exhibited CCL2 concentration–dependent migration, whereas this migration was abolished in cells from CCR2 ^−/− mice on a BALB/c genetic background. In vivo, blood flow recovery as measured by laser Doppler (LDI) and MRI (MRI) was impaired in CCR2 ^−/− mice on either the BALB/c or C57BL/6 genetic backgrounds. Three weeks after femoral artery ligation, LDI perfusion ratio of operated versus nonoperated distal hindlimb in BALB/c wild-type mice increased to 0.45±0.06 and in CCR2 ^−/− animals only to 0.21±0.03 ( P <0.01). In C57BL/6 mice, ratio increased to 0.96±0.09 and 0.85±0.08 ( P <0.05), respectively. MRI at 3 weeks (0.76±0.06 versus 0.62±0.01; P <0.05) and hemoglobin oxygen saturation measurements confirmed these findings. Active foot movement score significantly decreased and gastrocnemius muscle atrophy was significantly greater in CCR2 ^−/− mice. Morphometric analysis showed a lesser increase in collateral vessel diameters in CCR2 ^−/− mice. Importantly, the number of invaded monocytes/macrophages in the perivascular space of collateral arteries of CCR2 ^−/− animals was dramatically reduced in comparison to wild-type mice. In conclusion, our results demonstrate that the CCR2 signaling pathway is essential for efficient collateral artery growth.

Fibroblast growth factors (FGFs) have been applied in a variety of therapeutic and experimental studies to improve collateral blood flow. However, the pathophysiological role and the temporospatial expression of the FGFs and their receptors during arteriogenesis have never been elucidated in vivo. Here, we report that collateral artery growth in its early phase is associated with an increased expression of FGF receptor-1 (FGFR-1) and syndecan-4 on mRNA and protein levels as well as with an increased kinase activity of FGFR-1 in a rabbit model of arteriogenesis. However, the mRNA levels of FGF-1 and -2 remained constant. Our data suggest that these growth factors are supplied by endothelial attracted monocytes that, in turn, produce and deliver the FGFs to growing collateral arteries. Monocyte chemoattractant protein-1-stimulated arteriogenesis was strongly reduced in rabbits by application of the FGF inhibitor polyanetholesulfonic acid, indicating that the monocyte-related arteriogenesis (as well as the unstimulated adaptation proper) is promoted by FGFs. In summary, this study shows that arteriogenesis is associated with an increased expression of the FGFRs at the site of the vessel, whereas the growth-promoting ligands are supplied by monocytes in a paracrine way.

There are few detailed descriptions of the coronary arterial patterns in the mouse. Some recent reports on coronary anomalies in mutant mouse models have uncovered the importance of several genes (i.e. iv and connexin43) in coronary morphogenesis. These mutations spontaneously appeared (iv) or were generated (connexin43) in a C57BL/6 background, which is widely used for the development of mutant mice. We have studied the origin and course of the main coronary arteries of two C57BL/6 mouse strains. Unusual anatomical coronary arterial patterns were found, including: solitary ostium in aorta, accessory ostium, high take‐off, aortic intramural course, slit‐like ostium, sinus‐like ostium and origin of a septal artery from the left coronary artery. In humans, some of these conditions are clinically relevant. Most of these patterns, which differ from those observed in wild mice and Swiss albino mice, coincide with those previously found in iv/iv and connexin43 knockout mice. The results indicate that there is variability in the coronary arterial arrangement of the laboratory mouse. Care should be taken when analysing coronary phenotypes of mutant mouse models.