Expression of CYP4A1 in U251 human glioma cell induces hyperproliferative phenotype in vitro and rapidly growing tumors in vivo.

Exogenous 20-hydroxyeicosatetraenoic acid (20-HETE) increases the growth of human glioma cells in vitro. However, glioma cells in culture show negligible 20-HETE synthesis. We examined whether inducing the expression of a 20-HETE synthase in a human glioma U251 cell line would increase proliferation. U251 cells transfected with CYP4A1 cDNA (termed U251 O) increased the formation of 20-HETE from less than 1 to over 60 pmol/min/mg proteins and increased their proliferation rate by 2-fold (p < 0.01). Compared with control U251, U251 O cells were rounded, smaller, showed a disorganized cytoskeleton, exhibited reduced vinculin staining, and were easily detached from the growing surface. They showed a marked increase in dihydroethidium staining, suggesting increased oxidative stress. The expression of phosphorylated extracellular signal-regulated kinase 1/2, cyclin D1/2, and vascular endothelial growth factor was markedly elevated in U251 O. The hyperproliferative and signaling effects seen in U251 O cells are abolished by selective CYP4A inhibition of 20-HETE formation with HET0016 [N-hydroxy-N'-(4-butyl-2-methylphenyl)-formamidine], by small interfering RNA against the enzyme, and by the putative 20-HETE antagonist, 20-hydroxyeicosa-5(Z),14(Z)-dienoic acid. In vivo, implantation of U251O cells in the brain of nude rats resulted in a approximately 10-fold larger tumor volume (10 days postimplantation) compared with animals receiving mock-transfected U251 cells. These data show that elevations in 20-HETE synthesis in U251 cells lead to an increased growth both in vitro and in vivo. This suggests that 20-HETE may have proto-oncogenic properties in U251 human gliomas. Further studies are needed to determine whether 20-HETE plays a role promoting growth of some human gliomas.

Role of angiotensin II in retinal leukostasis in the diabetic rat.

To study if the endogenous renin-angiotensin system affects diabetic retinal leukostasis, rats with streptozotocin-induced diabetes were treated with an ACE inhibitor (ramipril), an angiotensin II AT(1) receptor antagonist (losartan) and the Ca channel blocker, (nifedipine). In the diabetic rats, these drug treatments reduced systolic blood pressure by approximately 16 mmHg but did not change blood glucose. After 2 weeks, the rats were examined for retinal leukostasis in vivo with a scanning laser ophthalmoscope (SLO). Retinal leukostasis, which was defined as no movement of arrested leukocytes over 2 min, was markedly higher in diabetic rats than normal controls (P<0.01). Leukostasis was significantly decreased by ramipril and losartan (P<0.01 vs. untreated diabetic rats) but was still higher than normal. Retinal leukostasis after nifedipine treatment was not significantly different than in untreated diabetic rats. The same trend was observed when leukostasis was analyzed on retinal flat mounts with concanavalin A and CD45 immunofluorescence; ramipril and losartan treatment, however, decreased leukostasis to values no different than controls. Retinal leukostasis was lowered by nifedipine (P<0.05, untreated diabetes vs. nifedipine-treated) but was still higher than in normal, ramipril-, or losartan-treated rats. Assays of gene expression of retinal intercellular adhesion molecule (ICAM-1) by semi-quantitative RT-PCR indicated that ICAM-1 mRNA was increased in diabetic rats but was decreased markedly by treatment with losartan or ramipril, and modestly by nifedipine. In summary, suppressing the activity of the endogenous renin-angiotensin system markedly decreases, perhaps even normalizes, the retinal leukostasis that accompanies type I diabetes in rats. These effects seem to be partly independent of blood pressure and to be associated with a decrease in ICAM-1 gene expression. Angiotensin II may, thus, mediate retinal leukostasis in early diabetes.

Inhibition of vascular endothelial growth factor-associated tyrosine kinase activity with SU5416 blocks sprouting in the microvascular endothelial cell spheroid model of angiogenesis.

The angiogenic vascular endothelial growth factor (VEGF) is believed to play a critical role in endothelial cell proliferation, differentiation, and sprouting. Small molecules that selectively inhibit the VEGF receptor-associated tyrosine kinase activities of Flk-1 (KDR) and Flt-1 have been developed. These agents, a prototype being SU5416, have effects on the proliferation of cultured endothelial cells, constrain angiogenesis in vivo, and have been proposed as antitumor drugs. Although SU5416 inhibits in vivo angiogenesis, it is not clear which of the complex processes leading to angiogenesis are impacted by VEGF receptor-associated tyrosine kinase inhibition. We utilized SU5416 and a microvascular endothelial cell line derived from mouse heart (SMHEC4) to specifically examine the role of VEGF receptor-associated tyrosine kinase activity on in vitro models of angiogenesis. We characterized spheroid formation and sprouting, a new model of angiogenesis, in this stable cell line. SU5416 inhibits (approximately 50%) VEGF (50 ng/ml) stimulated and basal DNA synthesis of SMHEC4 cultured in monolayer. SU5416 does not prevent the aggregation and organization of SMHEC4 into tri-dimensional spheroids. CD31, a marker of differentiated endothelial cells, is negligibly expressed in monolayer cultures but highly expressed in SMHEC4 spheroids. The content and biochemical characteristics of spheroidal CD31 are unaltered by SU5416. SU5416 also does not prevent the spontaneous and rapid (approximately 3-h) alignment into cords by SMHEC4 on Matrigel. These two models suggest that the organization and differentiation of endothelial cells is independent of VEGF receptor-associated tyrosine kinase signaling. SMHEC4 spheroids embedded in collagen gels spontaneously and rapidly (approximately 6 h) sprout capillary-like projections and subsequently (1-2 days) form complex self-anastomosing networks. In addition, VEGF (50 ng/ml) markedly stimulates sprouting of capillary-like projections from SMHEC4 spheroids. Both the spontaneous and the VEGF-stimulated sprouting are nearly eliminated by SU5416. This demonstrates that VEGF receptor-associated tyrosine kinase activity is essential to the formation of capillary-like structures from SMHEC4 spheroids. Overall, these observations demonstrate that (a) the spheroid sprouting model is appropriate for the study of angiogenesis since it appears to recapitulate many of its steps and (b) SU5416 can inhibit endothelial cell proliferation and sprouting without impacting the organization and differentiation of endothelial cells.