We used a modification of the isolated perfused rat heart, in which coronary effluent and interstitial transudate were separately collected, to investigate the localization and production of angiotensin II (Ang II) in the heart. During combined renin (0.7 to 1.5 pmol Ang I/mL per minute) and angiotensinogen (6 to 12 pmol/mL) perfusion (4 to 8 mL/min) for 60 minutes (n=3), the steady-state levels of Ang II in interstitial transudate in two consecutive 10-minute periods were 4.3+/-1.5 and 3.6+/-1.5 fmol/mL compared with 1.1+/-0.4 and 1.1+/-0.6 fmol/mL in coronary effluent (mean+/-half range). During perfusion with Ang II (n=5), steady-state Ang II in interstitial transudate was 32+/-19% of arterial Ang II compared with 65+/-16% in coronary effluent (mean+/-SD, P<.02). During perfusion with Ang I (n=5), Ang II in interstitial transudate was 5.1+/-0.6% of arterial Ang I compared with 2.2+/-0.3% in coronary effluent (P<.05). The tissue concentration of Ang II in the combined renin/angiotensinogen perfusions (per gram) was as high as the concentration in interstitial transudate (per milliliter). Addition of losartan (10(-6) mol/L) to the renin/angiotensinogen perfusion (n=3) had no significant effect on the tissue level of Ang II, whereas losartan in the perfusions with Ang I (n=5) or Ang II (n=5) decreased tissue Ang II to undetectably low levels. The results indicate that the heart is capable of producing Ang II and that this can lead to higher levels in tissue than in blood plasma. Cardiac Ang II does not appear to be restricted to the extracellular fluid. This is in part due to AT1-receptor-mediated cellular uptake of extracellular Ang II, but our results also raise the possibility of intracellular Ang II production.