Adenosine can be formed extracellularly and is part of the adenine nucleotide cascade involving CD39 and CD73 (ecto-5'-nucleotidase). It acts via specific adenosine receptors and plays an important role in the regulation of vascular tone and inflammatory processes. Furthermore, adenosine has been reported to be a potent stimulator of angiogenesis. However, its role in arteriogenesis is not known. To address this issue we induced hindlimb ischemia in a mouse mutant lacking CD73. The phenotype of these mice was previously shown to be associated with enhanced monocyte adhesion to the endothelium, considered to be an important initial trigger for arteriogenesis. Collateral vessel growth was assessed by 3D MR angiography (MRA) and metabolic restoration by in vivo ³¹P MR spectroscopy (MRS).

Experiments were performed at a Bruker 9.4 T NMR spectrometer equipped with microimaging unit and an actively shielded 40-mm gradient set. MR angiograms were acquired with a 30-mm ¹H saw resonator, while ³¹P MR spectra were recorded with a dual-tuned ¹H/³¹P 10-mm tilt resonator. Hindlimb ischemia was induced by ligation of the femoral artery. For direct visualization of newly developed collateral vessels we established a high resolution 3D MRA protocol. A 3D gradient echo sequence was used (FOV 2.56·2.56·1.28 cm³, matrix 256·256·128) and the FOV of 128 mm in z-direction was covered by three overlapping slabs of 64 mm. Datasets were subsequently 'sticked' together, segmented, and quantified with the self-developed image processing software ECCET for 3D MIP reconstruction.

With this protocol the femoral vessel system of the mouse could be imaged in great detail with an isotropic voxel size of 1 nl. The typical corkscrew-like structure of the collateral vessels could be clearly recognized. To account for differences in animal size and vessel geometry, newly developed vessels within the ischemic side were normalized by quantification of an anatomically defined artery segment on the contralateral side. For this segment - between the proximal end of the femoral artery and the branching into the popliteal artery - we determined an average length of 8.10.2 mm and a volume of 653134 nL (n=6). Collateral flow was detected as early as 3 d after ligation reaching a transient maximum 7 d after occlusion. At this time the extent of collateral formation in the ischemic area was substantially higher in CD73^-/- mice (+70% vs. wild-type (WT), p<0.05, n=7). The improved blood supply was also reflected by the faster recovery of hindlimb muscle energetics in the mutant compared with WT controls as assessed by in vivo ³¹P MRS. Histology confirmed the enhanced development of collaterals in CD73^-/- mice.

In conclusion, our protocol allows the quantitative assessment of growing collateral vessels in mice. We found that lack of CD73-derived adenosine promotes arteriogenesis.