Wragg et al., in this issue, report on a novel model of angiogenesis [1]. The authors sampled rat bone marrow and cultured bone marrow-derived cells under conditions designed to produce so-called multipotent adult progenitor cells (MAPC). They termed their cells, MAPC-derived progenitor cells (MDPC). MDPC expressed appropriate surface markers, such as MHC class I, CD44, CD90, but not MHC class II, CD45 or CD73. The cells expressed vascular endothelial growth factor receptor (VEGFR) 1 and the chemokine receptor CXCR4. The authors measured in vitro cytokine levels produced in culture over 24 h and found MDCP produced monocyte chemoattractant protein (MCP)-1, growth-related oncogene (Gro/KC), stromalderived factor (SDF)-1, and macrophage inflammatory protein (MIP)-2 in relative abundance. Wragg et al. then transplanted MDPC into the neurovascular bundle of ischemic rat hindlimbs. In this model, a segment of femoral artery is removed and substantial distal ischemia develops. MDPC were labeled with a β-Galexpressing retrovirus. Four weeks later, the cells were readily identified as LacZ positive. Significant engraftment of MDPC was detected around the neurovascular bundles. The cells seemed to be concentrated around arterioles. In contrast, sham-operated rats did not accumulate MDCP. Interestingly, the authors did not find that MDCP had assumed a vascular cell phenotype. MDCP did not stain positively for von Willebrand’s factor and did not resemble vascular smooth muscle cells. Nor did they express CD34. Nonetheless, MDCP resided in close proximity to CD34 positive cells. The authors were successful in measuring cytokine levels in vivo. MCP-1, SDF-1, MIP-2, and Gro/ KC again were present in increased amounts compared to animals that received vehicle. The capillary to muscle fiber ratio was increased in rats receiving MDCP into their ischemic hindlimbs; however, the effect was not dramatic. MDCP also increased muscle perfusion, as measured with microspheres. Since MDCP did not themselves differentiate into vascular structures, the question of how they may work remained. The authors were aware that SDF-1 facilitates the retention of CD11b-positive myelo-monocytic cells at the site of wound repair. This state of affairs caused the authors to focus on the spatial relationship between MDPC, infiltrating CD11bpositive cells, and proliferating endothelial cells. Indeed, about a three-fold increase in CD11b-positive cells were found in preparations that had received MDPC, compared to vehicle. Blocking the receptor for SDF-1, namely CXCR4, virtually eliminated arrival of CD11b-positive cells. The authors next performed monocyte-depletion experiments with clodronate-laden liposomes. This maneuver lowered local MIP-2 levels in tissue, but had no effect on SDF-1 or MCP-1 expression that presumably predominantly came from MDCD. Wragg et al. end their story at this point [1]. However, the questions have just started. What is the connection between monocyte-macrophage cells and angiogenesis? The authors point out that several studies have focused on monocyte recruitment to ischemic tissues and subsequent angiogenesis. For instance, Van Weel et al. studied amputated ischemic human extremities and presented evidence that bone marrow-derived cells expressing CXCR4 are retained in close proximity to angiogenic vessels by SDF-1 [2]. Arras et al. [3] investigated the association of monocyte accumulation along with cytokine production and vessel proliferJ Mol Med (2008) 86:1193–1195 DOI 10.1007/s00109-008-0401-8