Angiogenesis is an ordered process requiring the inter-play of numerous cellular and humoral factors. Studies over the past 20 years have identified several growth factors, cytokines, and enzymes that promote blood vessel formation. Most have revealed how individual factors promote an angiogenic phenotype in endothelial cells in vitro or contribute to blood vessel formation in vivo. However, the fundamental question that remains unanswered is how the cellular microenvironment contributes to angiogenesis. Fibrocytes are a recently characterized mesenchymal cell type isolated from peripheral blood that rapidly enter subcutaneously implanted wound chambers and sites of tissue injury. Here we describe the induction of an angiogenic phenotype in microvascular endothelial cells in vitro and promotion of angiogenesis in vivo by cultured fibrocytes. Fibrocytes constitutively secrete extracellular matrix-degrading enzymes, primarily matrix metalloproteinase 9, which promotes endothelial cell invasion. In addition, fibrocytes secrete several proangiogenic factors including VEGF, bFGF, IL-8, PDGF, and hematopoietic growth factors that promote endothelial cell migration, proliferation, and/or tube formation. By contrast, they do not produce representative antiangiogenic factors. Finally, both autologous fibrocytes and fibrocyte-conditioned media were found to induce blood vessel formation in vivo using the Matrigel angiogenesis model.

No abstract

Endowing tumor cells with costimulatory signals for T cell activation has emerged as a promising strategy for tumor immunotherapy. Costimulatory molecules were either transfected into tumor cells to generate vaccines or were fused, e.g. to antibodies against tumorassociated antigens, to achieve targeted T cell costimulation in vivo. Here we report the production and purification of rM28, a recombinant bispecific single-chain antibody directed to a melanoma-associated proteoglycan and to the costimulatory CD28 molecule on human T cells. We found that a dimer of the recombinant molecule, bound to tumor target cells, induced pronounced T cell activation in peripheral blood mononuclear cell preparations without additional TCR/CD3 stimulation being required. The lytic activity generated after 3 days of stimulation effectively prevented tumor cell growth. However, it was unspecific and predominantly mediated by non T cells. Our findings demonstrate that presentation of a CD28 antibody within a suitable recombinant, bispecific format may result in a "targeted supra-agonistic stimulation" of the CD28 molecule, which leads to effective tumor cell killing after induction of unspecifically lytic cells.

No abstract