Ca2+ changes induced by nitric oxide (NO.) were investigated in cultured human endothelial cells. Sodium nitroprusside (SNP) (1-100 mumol/L) and S-Nitroso-N-acetylpenicillamine (SNAP) (100 mumol/L) were used as NO. donors. The cytoplasmatic Ca2+ concentration was calculated using ratiometric FURA2 fluorescence measurements. Both NO. donors caused transient oscillatory Ca2+ changes, which were not detectable in the presence of oxyhemoglobin (50 mumol/L). Digital ratio imaging revealed initiation sites within cells where Ca2+ increases started spreading, which indicates that nonuniformly distributed targets might be involved in these reactions. Calcium was released from intracellular stores as indicated by experiments performed in Ca(2+)-free buffer. L-type Ca(2+)-channel blocker diltiazem (100 mumol/L) was not able to block these responses. NO.-induced Ca2+ release from intracellular stores caused capacitative Ca2+ entry. Both thapsigargin (1 mumol/L) and cyclopiazonic acid (10 mumol/L) inhibited the SNP response completely, whereas neither ryanodine (up to 100 mumol/L) nor dantrolene (100 mumol/L) was able to inhibit Ca2+ changes induced by SNP, indicating that primarily inositol 1,4,5-triphosphate (IP3)-dependent stores are released upon stimulation with NO.. A small inhibitory effect of ATP- and SNP-induced peak [Ca2+]i increase was measured in the presence of both caffeine (20 mmol/L) and procaine (1 mmol/L). Evidence is presented that cGMP is not involved in NO.-induced Ca2+ signals, as neither inhibitors of guanylate cyclase (methylene blue and LY 83583) nor cell permeant analogues of cGMP altered or simulated [Ca2+] changes. An inhibitor of cGMP-dependent protein kinase was also ineffective. We therefore propose that endothelial cells have specific targets proximal or at IP3 receptors to induce Ca2+ changes in endothelial cells stimulated with NO..