The nitric oxide hypothesis of late preconditioning

Abstract

Ischemic preconditioning (PC) occurs in two phases: an early phase, which lasts 2-3 h, and a late phase, which begins 12-24 h later and lasts 3-4 days. The mechanism for the late phase of PC has been the focus of intense investigation. We have recently proposed the "NO hypothesis of late PC", which postulates that NO plays a prominent role both in initiating and in mediating this cardioprotective response. The purpose of this essay is to review the evidence supporting the NO hypothesis of late PC and to discuss its implications. We propose that, on day 1, a brief ischemic stress causes increased production of NO (probably via eNOS) and .O2-, which then react to form ONOO-, ONOO-, in turn, activates the epsilon isoform of protein kinase C (PKC), either directly or via its reactive byproducts such as .OH. Both NO and secondary species derived from .O2- could also stimulate PKC epsilon independently. PKC epsilon activation triggers a complex signaling cascade that involves tyrosine kinases (among which Src and Lck appear to be involved) and probably other kinases, the transcription factor NF-kappa B, and most likely other as yet unknown components, resulting in increased transcription of the iNOS gene and increased iNOS activity on day 2, which is responsible for the protection during the second ischemic challenge. Tyrosine kinases also appear to be involved on day 2, possibly by modulating iNOS activity. According to this paradigm, NO plays two completely different roles in late PC: on day 1, it initiates the development of this response, whereas on day 2, it protects against myocardial ischemia. We propose that two different NOS isoforms are sequentially involved in late PC, with eNOS generating the NO that initiates the development of the PC response on day 1 and iNOS then generating the NO that protects against recurrent ischemia on day 2. The NO hypothesis of late PC puts forth a comprehensive paradigm that can explain both the initiation and the mediation of this complex phenomenon. Besides its pathophysiological implications, this hypothesis has potential clinical reverberations, since NO donors (i.e., nitrates) are widely used clinically and could be used to protect the ischemic myocardium in patients.

Fig. 1

Schematic representation of the pathways involved in the genesis of late preconditioning against myocardial stunning. A brief episode of myocardial ischemia/reperfusion causes increased production of NO and reactive oxygen species (ROS), which serve as triggers for the development of late PC. NO and ROS activate a complex signal transduction cascade (which involves PKC (most likely PKC ε), tyrosine kinases, and possibly MAPKs), which leads to activation of transcription factors, upregulation of cardioprotective genes, and increased activity of NOS (specifically, iNOS) 24–72 h later. This increased NOS activity confers protection during the second ischemic stress.

Fig. 2

Systolic thickening fraction in the ischemic-reperfused region in L-NA treated rabbits before administration of L-NA (baseline), 9 min after the end of the infusion of L-NA (immediately before the first occlusion) (preocclusion (Pre-O)), 3 min into each coronary occlusion (O), 3 min into each reperfusion (R), and at selected times during the 5 h reper-fusion interval following the sixth occlusion. Conscious rabbits underwent a sequence of six 4-min coronary occlusion/4-min reperfusion cycles for three consecutive days (day 1, 2, and 3). L-NA was given on day 1 prior to the first occlusion/reperfusion cycle. ○ indicates measurements taken on day 1; ●, measurements taken on day 2; and ▲, measurements taken on day 3. To facilitate comparisons, the data pertaining to day 1 of the control group are also shown (thick interrupted line without symbols). Thickening fraction is expressed as a percentage of preocclusion values. Data are mean ± SEM. In contrast to control rabbits, in which the recovery of wall thickening was markedly enhanced on day 2 compared with day 1 (data not shown in this figure), in L-NA-treated rabbits the recovery of wall thickening was similar on days 1 and 2, indicating that administration of L-NA on day 1 abrogated the development of late PC against stunning on day 2. The expected late PC effect became apparent on day 3, as a result of the ischemic PC stimulus applied on day 2 (when rabbits were not treated with L-NA). Inset. Total deficit of wall thickening during the 5 h reperfusion period following the 6th reperfusion. The total deficit of wall thickening is an integrated measure of the overall severity of myocardial stunning. Notice that the total deficit was similar on day 1 and day 2, indicating that L-NA prevented late PC against stunning. The deficit of wall thickening decreased on day 3, indicating that the ischemic stimulus on day 2 induced a late PC effect on day 3. (Reproduced with permission of the American Heart Association from Bolli et al. (1997) Circ Res 81: 42–52.)

Fig. 3

Effect of the NOS inhibitor L-NA on late PC against infarction in conscious rabbits. All rabbits underwent a 30-min coronary occlusion followed by 3 d of reperfusion. Rabbits in group I received neither ischemic PC nor drug treatment. Rabbits in group II were preconditioned 24 h earlier with a sequence of six 4-min coronary occlusion/4-min reper-fusion cycles. Rabbits in group III underwent the same protocol as group II, except that they received L-NA prior to the six 4-min occlusion/reper-fusion cycles. Rabbits in group IV received L-NA without ischemic PC 24 h before the 30-min occlusion. In group III, administration of L-NA prior to the PC ischemia abrogated the infarct-sparing effect of late PC, indicating that the development of late PC against infarction is triggered by NO. In group IV (administration of L-NA without PC), infarct size did not differ from that observed in controls, demonstrating that administration of L-NA in itself did not exert a delayed deleterious effect on myocar-dial infarction. Infarct size is expressed as a percentage of the region at risk of infarction. Open circles represent individual rabbits, whereas solid circles represent means ± SEM. * P < 0.05 vs. group I (controls); ^§P < 0.05 vs. group II. (Reproduced with permission of the American Physiological Society from Qiu et al. (1997) Am J Physiol 273: H2931–36.)

Fig. 4

Total deficit of wall thickening (WTh) after the 6th reperfusion on days 1, 2, and 3 in the control (n = 6), DETA/NO (n = 5), SNAP (n = 5), and DETA/NO+MPG groups (groups I, II, III, and I V, respectively). Conscious rabbits underwent a sequence of six 4-min coronary occlusion/4-min reperfusion cycles for three consecutive days (day 1, 2, and 3). Twenty-four hours before the first sequence of occlusion/reperfusion, rabbits received either no treatment (control), the NO donor DETA/NO, the NO donor SNAP, or DETA/NO in conjunction with the antioxidant MPG. Pretreatment with either DETA/NO or SNAP resulted in an attenuation of the total deficit of WTh on day 1 equivalent to that induced by ischemic PC on day 2 in control rabbits. This effect was abrogated by MPG, indicating that DETA/NO-induced late PC against stunning is mediated by MPG-sensitive oxidants (e.g., ONOO⁻ and/or ^•OH). The values of total deficit of WTh in individual rabbits are illustrated in the left panel; the mean ± SEM values of total deficit of WTh are depicted in the right panel. The total deficit of WTh was measured in arbitrary units, as described in the text. (Reproduced with permission of the American Heart Association from Takano et al. (1998) Circ Res 83: 73–84.)

Fig. 5

Myocardial infarct size in groups V (n = 10, control group), VI (n = 10, ischemic PC group), VII (n = 8, DETA/NO group), VIII (n = 5, DETA/NO high dose group), IX (n = 7, SNAP group), and X (n = 7, DETA/NO+MPG group). Conscious rabbits underwent a 30-min coronary occlusion and three days of reperfusion. Twenty-four hours before the 30-min occlusion, they received either ischemic PC (six 4-min occlusion/reperfusion cycles), DETA/NO (at two doses), SNAP, or DETA/NO in conjunction with MPG. Pretreatment with either DETA/NO or SNAP resulted in a reduction of infarct size equivalent to that induced by ischemic PC in group VI. This effect was abrogated by MPG, indicating that NO-induced late PC against infarction is mediated by MPG-sensitive oxidants. Infarct size is expressed as a percentage of the region at risk of infarction. Open circles represent individual rabbits, whereas solid circles represent means ± SEM. *P < 0.05 vs. group V (control group); ^§P < 0.05 vs. group VII (DETA/NO group). (Reproduced with permission of the American Heart Association from Takano et al. (1998) Circ Res 83: 73–84.)

Fig. 6

Total deficit of systolic wall thickening after the sixth reperfusion on days 1, 2 and 3 in nine groups of conscious rabbits. All groups underwent a sequence of six 4-min occlusion/4-min reperfusion cycles for three consecutive days (days 1, 2, and 3). Rabbits in group I received no treatment. Rabbits in groups II (L-NA), IV (AG), and VI (SMT) received L-NA, AG, and SMT, respectively, before the first coronary occlusion on day 2. Rabbits in groups III (L-NA-pre), V (AG-pre), and VII (SMT-pre) received the same doses of L-NA, AG, and SMT, respectively, before the first coronary occlusion on day 1. Rabbits in group VIII received the same dose of SMT in conjunction with L-arginine before the first coronary occlusion on day 2. Rabbits in group IX received L-arginine alone before the first coronary occlusion on day 2. Data are mean SEM. Pre indicates treatment on day 1. Administration of either a nonselective (L-NA) or an iNOS-selective (AG and SMT) NOS inhibitor on day 2 completely abrogated late PC against myocardial stunning (groups II, I V, and VI, respectively), indicating that NOS (specifically, iNOS) is the mediator of this cardioprotective phenomenon. The abrogation of late PC by SMT was completely reversed by L-arginine (group VIII), indicating that SMT acted specifically by inhibiting NOS activity. (Reproduced with permission of the American Heart Association from Bolli et al. (1997) Circ Res 81: 1094–1107.)

Fig. 7

Effect of the nonselective NOS inhibitor L-NA and the relatively-selective iNOS inhibitor aminoguanidine (AG) on late PC against infarction in conscious rabbits. All groups underwent a 30-min coronary occlusion followed by 3 d of reperfusion. Group II was preconditioned 24 h earlier with six 4-min occlusion/reperfusion cycles. Group III underwent the same protocol, except that L-NA was given prior to the 30-min occlusion. Group IV received L-NA as group III, but was not preconditioned 24 h earlier. Group V underwent the same protocol as group II, except that AG was given before the 30-min occlusion. Group VI received AG without undergoing ischemic PC 24 h earlier. Administration of either L-NA or AG completely abrogated the infarct-sparing effect of late PC, indicating that NOS (specifically, iNOS) is the mediator of late PC against infarction. Infarct size in groups IV and VI was not different from controls, indicating that L-NA and AG did not exert a delayed deleterious effect on infarct size independent of PC. Infarct size is expressed as a percentage of the region at risk of infarction. Open circles represent individual rabbits, whereas solid circles represent means ± SEM. *P < 0.05 vs. group I (controls); ^§P < 0.05 vs. group II (PC group). (Reproduced with permission of the American Physiological Society from Takano et al. (1998) Circulation 98: 441–449.)

Fig. 8

Myocardial infarct size in wild-type (WT) and iNOS knockout (KO) mice. Mice were subjected to 30 min of coronary occlusion and 24 h of reperfusion, in the absence of prior ischemic PC (groups I and II), 24 h after ischemic PC (six 4-min occlusion/4-min reperfusion cycles) (groups III and IV (late PC groups)), or 10 min after the same ischemic PC protocol (groups V and VI (early PC groups)). Infarct size is expressed as a percentage of the region at risk of infarction. Note that late PC produced a significant reduction in infarct size in wild-type mice but not in iNOS knockout mice, whereas early PC produced marked infarct size reduction in both. *P < 0.05 vs. Group I; ^#P < 0.05 vs. Group II.

Fig. 9

Measurements of NF-κB DNA binding activity (electrophoretic mobility shift assay) in the ischemic/reperfused left ventricular region of conscious rabbits undergoing ischemic PC. All values are expressed as a percentage of the values in the control group (group I), which did not undergo ischemia. A marked rise in the DNA binding activity was noted 30 min after the ischemic PC protocol (six 4-min coronary occlusion/4-min reperfusion cycles) (group II). This increase in DNA binding activity was completely abrogated when DDTC (an inhibitor of NF-κB, group III), MPG (a scavenger of ^•OH and ONOO, group IV), L-NA (a NOS inhibitor, group V), chelerythrine (an inhibitor of PKC, group VI), or LD-A (an inhibitor of tyrosine kinases, group VII) were administered before the ischemic PC protocol. Interestingly, when DETA/NO was administered in the absence of PC ischemia (group VIII), there was a marked increase in NF- κ B DNA binding activity similar to that in preconditioned rabbit hearts (group II). *P< 0.05 vs. group I; ^§P< 0.05 vs. group II (n = 4 in all groups).

Fig. 10

Schematic representation of the cellular mechanisms involved in triggering late PC against myocardial stunning. A brief episode of myocardial ischemia/reperfusion causes increased production of NO and ^•O₂⁻, which then react to form ONOO⁻. ONOO⁻, in turn, activates the novel subgroup of PKCs (most likely PKC ε) either directly or via its reactive byproducts, such as ·OH. ^•O₂⁻ could also generate secondary ROS capable of activating PKC, and it is possible that NO could directly activate PKC. PKC activation triggers a complex signaling cascade that involves tyrosine kinases and probably other kinases, the transcription factor NF-κB, and probably other as yet unknown components, resulting in increased synthesis of new proteins and increased NOS activity on day 2, which is responsible for the protection. Among the numerous tyrosine kinases, two elements of the Src family (Src and Lck) are likely to be involved on day 1. Tyrosine kinases also appear to be involved on day 2, possibly by modulating iNOS activity. The increased activity of iNOS on day 2 requires synthesis of tetrahydrobiopterin (BH₄) to support NO production. According to this paradigm, late PC against stunning can be abrogated by several interventions targeted at different components: by blocking increased NO synthesis following the initial stress with L-NA on day 1, by scavenging reactive species derived from NO (MPG on day 1), by inhibiting PKC, tyrosine kinases, NF-κB or protein synthesis (with chelerythrine, lavendustin A, DDTC and cycloheximide, respectively, on day 1), by blocking the enhanced NOS activity on day 2 (L-NA, AG, or SMT on day 2), by blocking tyrosine kinases on day 2 (lavendustin A on day 2), or by blocking sepiapterin reductase (an enzyme involved in BH₄ synthesis) on day 2 (NAS on day 2). Conversely, administration of PMA on day 1 can bypass the triggering events and activate the development of late PC in the absence of ischemia.