Angiopoietin-2 (Ang2) exhibits broad expression in the remodeling vasculature of human tumors but very limited expression in normal tissues, making it an attractive candidate target for antiangiogenic cancer therapy. To investigate the functional consequences of blocking Ang2 activity, we generated antibodies and peptide-Fc fusion proteins that potently and selectively neutralize the interaction between Ang2 and its receptor, Tie2. Systemic treatment of tumor-bearing mice with these Ang2-blocking agents resulted in tumor stasis, followed by elimination of all measurable tumor in a subset of animals. These effects were accompanied by reduced endothelial cell proliferation, consistent with an antiangiogenic therapeutic mechanism. Anti-Ang2 therapy also prevented VEGF-stimulated neovascularization in a rat corneal model of angiogenesis. These results imply that specific Ang2 inhibition may represent an effective antiangiogenic strategy for treating patients with solid tumors.

Background-Genetic mutations of dystrophin and associated glycoproteins underlie cell degeneration in several inherited cardiomyopathies, although the precise physiological role of these proteins remains under discussion. We studied the distribution of dystrophin in relation to the force-transducing vinculin-rich costameres in left ventricular cardiomyocytes from normal and failing human hearts to further elucidate the function of this protein complex. Methods and Results-Single-and double-label immunoconfocal microscopy and parallel high-resolution immunogold fracture-label electron microscopy were used to localize dystrophin and vinculin in human left ventricular myocytes from normal (nϭ6) and failing hearts (idiopathic dilated cardiomyopathy, nϭ7, or ischemic heart disease, nϭ5). In control cardiomyocytes, dystrophin had a continuous distribution at the peripheral sarcolemma, with concentrated bands corresponding to the vinculin-rich costameres. Intracellular labeling extended along transverse (T) tubule membranes. Fracture-label confirmed this distribution, showing significantly greater label on plasma membrane fractures overlying I-bands (I-band 4.1Ϯ0.3 gold particles/m; A-band 3.3Ϯ0.2 gold particles/m: meanϮSE, Pϭ0.02). Hypertrophied myocytes from failing hearts showed maintenance of this surface distribution except in degenerating cells; there was a clear increase in intracellular dystrophin label reflecting T-tubule hypertrophy. Conclusions-Dystrophin partially colocalizes with costameric vinculin in normal and hypertrophied myocytes, a distribution lost in degenerating cells. This suggests a primarily mechanical role for dystrophin in maintenance of cell membrane integrity in normal and hypertrophied myocytes. The presence of dystrophin in the cardiac T-tubule membrane, in contrast to its known absence in skeletal muscle T-tubules, implies additional roles for dystrophin in membrane domain organization.

Dystrophin is a key component of the subsarcolemmal skeleton of muscle cells, and lack of dystrophin is the direct cause of Duchenne muscular dystrophy. In skeletal muscle, dystrophin is reported to be localized specifically at costameres, transversely oriented riblike subsarcolemmal plaques that mechanically couple the contractile apparatus to the extracellular matrix. Costameres are characteristically rich in vinculin and are prominent in cardiac as well as skeletal muscle. To define the precise spatial relationship between dystrophin in relation to the costamere in cardiac muscle, we applied high-resolution single- and double-immunolabeling techniques, under a range of preparative conditions, with visualization of vinculin (as a costamere marker) and dystrophin by confocal microscopy and by the freeze-fracture cytochemical technique, fracture label. Immunoconfocal visualization revealed dystrophin as a continuous uniform layer at the cytoplasmic surface of the peripheral plasma membrane of the rat cardiac myocyte at both costameric and noncostameric regions. The pattern of labeling was reproducible with three different antibodies and was independent of time and antibody concentration. Platinum/carbon replicas and thin sections of fracture-label specimens permitted high-resolution visualization of the distribution of dystrophin in plane views of the freeze-fractured plasma membrane and in relation to the sarcomeric banding patterns of the underlying myofibrils. These results confirmed no preferential association of dystrophin with costameres or with any region of the sarcomeres of underlying myofibrils in rat cardiac tissue. We conclude that in contrast to skeletal muscle, dystrophin in cardiac muscle is not exclusively a component of the costamere.

The synchronous contractions of the uterus in labour depend on electrical coupling of myometrial smooth muscle cells by gap junctions. In the human myometrium, gap junctions are scarce in the non-pregnant uterus, but become abundant at term in preparation for labour. We have previously demonstrated that in the human myometrium at term, three different gap-junctional proteins are expressed, connexins 43, 45, and 40. These connexins are known to have distinctive functional capacities in in vitro expression systems but whether, in the human myometrium in vivo, they are co-assembled into the same gap junction or form different types of gap junction has previously been unclear. By applying triple immunogold labelling to sections of Lowicryl-embedded tissue for electron microscopy, together with complementary immunoconfocal microscopy, we demonstrate here that connexins 43, 45, and 40 are commonly present as mixtures within the same gap-junctional plaque. While all gap junctions contain connexin43, the relative signal for each connexin type varies between individual junctions. The presence within single gap-junctional plaques of three different connexins, each with the potential for conferring distinctive channel properties, suggests an inherent versatility for modulation of smooth muscle cell intercellular communication properties during human parturition.