We investigated whether ischemic preconditioning (PC) produced a second window of protection by delayed synthesis of cardioprotective proteins. Anesthetized open-chest rabbits were subjected to 30 min of coronary occlusion and 3 h of reperfusion. PC was elicited by 5 min of ischemia and was separated from sustained ischemia by 5 min, 2 h, or 24 h of reperfusion. Infarct size (% area at risk) was markedly limited by PC with 5 min of reperfusion when compared with controls (13.3 +/- 2.5 vs. 46.8 +/- 7.0%; P < 0.05). This protective effect was lost when the interval between PC and sustained ischemia was extended to 2 h (47.8 +/- 4.8%; P = NS vs. control) and did not reoccur even when it was extended to 24 h (44.2 +/- 6.5%; P = NS vs. sham-operated control). To potentiate induction of heat shock proteins (HSPs), a PC protocol involving four 5-min episodes of ischemia and reperfusion was also used and was separated from sustained ischemia by 24 or 48 h of reperfusion. However, neither of these protocols was protective, and limitation of infarct size was not observed (55.5 +/- 5.9 and 53.4 +/- 6.5% in 24 and 48 h of reperfusion, respectively; P = NS vs. corresponding sham-operated control). Myocardial expression of HSPs was examined using a monoclonal antibody against 72- to 73-kDa HSP in additional rabbits. Immunoreactivity was observed in the myocardium at 24 and 48 h after PC, but not immediately after PC.(ABSTRACT TRUNCATED AT 250 WORDS)