Nitric oxide (NO) is an important regulator of blood flow, but its role in permeability is still challenged. We tested in vivo the hypotheses that: (a) endothelial nitric oxide synthase (eNOS) is not essential for regulation of baseline permeability; (b) eNOS is essential for hyperpermeability responses in inflammation; and (c) molecular inhibition of eNOS with caveolin-1 scaffolding domain (AP-Cav) reduces eNOS-regulated hyperpermeability. We used eNOS-deficient (eNOS-/-) mice and their wild-type control as experimental animals, platelet-activating factor (PAF) at 10(-7) m as the test pro-inflammatory agent, and integrated optical intensity (IOI) as an index of microvascular permeability. PAF increased permeability in wild-type cremaster muscle from a baseline of 2.4 +/- 2.2 to a peak net value of 84.4 +/- 2.7 units, while the corresponding values in cremaster muscle of eNOS-/- mice were 1.0 +/- 0.3 and 15.6 +/- 7.7 units (P < 0.05). Similarly, PAF increased IOI in the mesentery of wild-type mice but much less in the mesentery of eNOS-/- mice. PAF increased IOI to comparable values in the mesenteries of wild-type mice and those lacking the gene for inducible NOS (iNOS). Administration of AP-Cav blocked the microvascular hyperpermeability responses to 10(-7) m PAF. We conclude that: (1) baseline permeability does not depend on eNOS; (2) eNOS and NO are integral elements of the signalling pathway for the hyperpermeability response to PAF; (3) iNOS does not affect either baseline permeability or hyperpermeability responses to PAF; and (4) caveolin-1 inhibits eNOS regulation of microvascular permeability in vivo. Our results establish eNOS as an important regulator of microvascular permeability in inflammation.