In a murine heart perfusion system, we were able to "turn off" the transport of derivatized albumin [dinitrophenylated albumin (DNP-albumin)] from the perfusate to the tissue, by preperfusing the system with 1 mM N-ethylmaleimide (NEM) for 5 min at 37 degrees C, followed by a 5-min perfusion of DNP-albumin in the presence of NEM. Using a postembedding immunocytochemical procedure, we showed that (i) a 30-sec to 1-min treatment of heart vasculature with 1 mM NEM reduces the transendothelial transport of DNP-albumin and nearly stops it after 5 min, and (ii) DNP-albumin is detected exclusively in plasmalemmal vesicles (PVs) while in transit across endothelial cells. Perfusion with 10 mM dithiothreitol for 1 min before NEM prevents the inhibition of vesicular transport. To quantitate the NEM effect on vesicular transport inhibition, we developed an ELISA and a dot-blot assay for measuring DNP-albumin in supernatants of perfused whole-heart homogenates. The results obtained indicate that the treatment of the heart vasculature with 1 mM NEM decreases the vesicular transport of DNP-albumin by 78-80%. Since NEM is known to inhibit the fusion of different types of vesicular carriers with their target membranes in other cell types and in in vitro reconstituted cellular systems, by alkylating a NEM-sensitive factor, we assume that the same mechanism applies in our in situ system. The decrease of vesicular transport can be explained by NEM preventing the fusion of recycling vesicles with their targets--i.e., the abluminal and luminal domains of the plasmalemma. The results open to question previous interpretations from other laboratories according to which plasmalemmal vesicles are sessile, immobile structures.