How tumour-induced vascular changes alter angiogenesis: Insights from a computational model.

A computational model was developed to describe experimentally observed vascular changes induced by the introduction of a tumour on a mouse equipped with a dorsal skinfold chamber. The vascular structure of the host tissue was segmented from in vivo images and transposed into the computational framework. Simulations of tumour-induced vascular changes were performed and include the destabilizing effects of the growth factor VEGF on the integrity of the vessels walls. The integration of those effects, that include alteration of the vessel wall elasticity and wall breaching, were required to realistically reproduce the experimental observations. The model was then used to investigate the importance of the vascular changes for oxygen delivery and tumour development. To that end, we compared simulations obtained with a dynamic vasculature with those obtained with a static one. The results showed that the tumour growth was strongly impeded by the constant vascular changes. More precisely, it is the angiogenic process itself that was affected by vascular changes occurring in bigger upstream vessels and resulting in a less efficient angiogenic network for oxygen delivery. As a consequence, tumour cells are mostly kept in a non-proliferative hypoxic state. Tumour dormancy thus appears as one potential consequence of the intense vascular changes in the host tissue.

A mathematical model of HiF-1α-mediated response to hypoxia on the G1/S transition.

Hypoxia is known to influence the cell cycle by increasing the G1 phase duration or by inducing a quiescent state (arrest of cell proliferation). This entry into quiescence is a mean for the cell to escape from hypoxia-induced apoptosis. It is suggested that some cancer cells have gain the advantage over normal cells to easily enter into quiescence when environmental conditions, such as oxygen pressure, are unfavorable [43,1]. This ability contributes in the appearance of highly resistant and aggressive tumor phenotypes [2]. The HiF-1α factor is the key actor of the intracellular hypoxia pathway. As tumor cells undergo chronic hypoxic conditions, HiF-1α is present in higher level in cancer than in normal cells. Besides, it was shown that genetic mutations promoting overstabilization of HiF-1α are a feature of various types of cancers [7]. Finally, it is suggested that the intracellular level of HiF-1α can be related to the aggressiveness of the tumors [53,24,4,10]. However, up to now, mathematical models describing the G1/S transition under hypoxia, did not take into account the HiF-1α factor in the hypoxia pathway. Therefore, we propose a mathematical model of the G1/S transition under hypoxia, which explicitly integrates the HiF-1α pathway. The model reproduces the slowing down of G1 phase under moderate hypoxia, and the entry into quiescence of proliferating cells under severe hypoxia. We show how the inhibition of cyclin D by HiF-1α can induce quiescence; this result provides a theoretical explanation to the experimental observations of Wen et al. (2010) [50]. Thus, our model confirms that hypoxia-induced chemoresistance can be linked, for a part, to the negative regulation of cyclin D by HiF-1α.

STAT1 deficiency in the heart protects against myocardial infarction by enhancing autophagy.

Previous studies have shown that the transcription factor signal transducer and activator of transcription 1 (STAT1) activation is increased in primary cardiac myocytes exposed to simulated ischaemia/reperfusion injury. This promotes apoptotic cell death by enhancing the expression of pro-apoptotic proteins. Autophagy has been demonstrated to play a cardioprotective role in the heart following myocardial infarction (MI). We therefore investigated the role of STAT1 in the intact heart subjected to MI and examined the contribution of autophagy in modulating the protective effect of STAT1 after MI injury. STAT1-deficient hearts had significantly smaller infarcts than wild-type hearts and this correlated with increased levels of autophagy shown by light chain 3 (LC3)-I/LC3-II conversion, and up-regulation of Atg12 and Beclin 1. Moreover, pre-treatment with the autophagy inhibitor 3-methyladenine reversed the cardioprotection observed in the STAT1-deficient hearts. These results reveal a new function of STAT1 in the control of autophagy and indicate a cross-talk between the cardioprotective versus the damaging effects of STAT1 in the intact heart exposed to MI injury.

The rigidity in fibrin gels as a contributing factor to the dynamics of in vitro vascular cord formation.

While the formation of vascular cords in in vitro angiogenesis assay is commonly used to test the angiogenic properties of many molecular or cellular components, an extensive characterisation of the dynamics of this process is still lacking. Up to now, quantitative studies only focused on the resulting capillary structures characterised through static morphometric approaches. We therefore propose in this paper a rather extensive characterisation aiming to identify different stages in the dynamics of this process, through the investigation of the influence of the rigidity of the fibrin extracellular matrix on the growth of the vascular cords. Using time lapse videomicroscopy, the time evolution of relevant morphodynamical parameters has been considered both at the cell level and at the cell population level. At the cell level, a trajectography analysis of individual cells observed in different locations of the growing network has been conducted and analysed using a random walk model. From image sequence analysis and segmentation i.e. extraction of the boundaries of the lacunae formed through matrix degradation and cell tractions, the evolution of the lacunae surface has been precisely quantified, revealing different phases and transitions in the growth patterns. Our results indicate that the rigidity of the extracellular fibrin matrix strongly influences the different stages, i.e. the dynamics of the angiogenic process. Consequently, optimal rigidity conditions for the formation of stable vascular cord networks could be identified in the context of our experiments.

Opposing actions of STAT-1 and STAT-3.

The signal transducers and activators of transcription (STATs) are a family of transcription factors, which were originally identified on the basis of their ability to transduce a signal from a cellular receptor into the nucleus and modulate the transcription of specific genes. Interestingly, recent studies have demonstrated that STAT-1 plays a key role in promoting apoptosis in a variety of cell types, whereas STAT-3 has an anti-apoptotic effect. Moreover, whilst STAT-3 promotes cellular proliferation and is activated in a variety of tumour cells, STAT-1 appears to have an anti-proliferative effect. Although the initially characterised signal transduction events mediated by STAT-1 and STAT-3 involve the DNA binding and transcriptional activation domains of the factor, some of their other effects appear not to require DNA binding. Therefore, STAT-1 and STAT-3 can mediate the regulation of gene transcription both by direct DNA binding and via a co-activator mechanism and despite their very similar structures, have antagonistic effects on cellular proliferation and apoptosis.

Cardioprotection mediated by urocortin is dependent on PKCepsilon activation.

Urocortin (Ucn) is an endogenous cardioprotective agent that protects against the damaging effects of ischemia and reperfusion injury in vitro and in vivo. We have found that the mechanism of action of Ucn involves both acute activation of specific target molecules, and using Affymetrix (Santa Clara, CA) gene chip technology, altered gene expression of different end effector molecules. Here, from our gene chip data, we show that after a 24 h exposure to Ucn, there was a specific increase in mRNA and protein levels of the protein kinase C epsilon (PKCepsilon) isozyme in primary rat cardiomyocytes compared with untreated cells and in the Langendorff perfused ex vivo heart. Furthermore, a short 10 min exposure of these cells to Ucn caused a specific translocation/activation of PKCepsilon in vitro and in the Langendorff perfused ex vivo heart. The importance of the PKCepsilon isozyme in cardioprotection and its relationship to cardioprotection produced by Ucn was assessed using PKCepsilon-specific inhibitor peptides. The inhibitor peptide, when introduced into cardiomyocytes, caused an increase in apoptotic cell death compared with control peptide after ischemia and reperfusion. When the inhibitor peptide was present with Ucn, the cardioprotective effect of Ucn was lost. This loss of cardioprotection by Ucn was also seen in whole hearts from PKCepsilon knockout mice. These findings indicate that the cardioprotective effect of Ucn is dependent upon PKCepsilon.

Role of STAT-1 and STAT-3 in ischaemia/reperfusion injury.

Ischaemia/reperfusion (I/R) injury results in the death of irreplaceable cardiac myocytes by a programme cell death or apoptosis. The signal transducers and activators of transcription (STAT) factors function as modulators of cytokine signaling and sensors responding to cellular stress. Interestingly, many studies have demonstrated that although they have a similar structural organization, STAT-1 and STAT-3 have apposing effects on processes such as differentiation or apoptosis. For example, STAT-1 has been shown to induced apoptosis, whilst STAT3 is able protect cardiac myocytes following ischaemia/reperfusion (I/R) injury. Many of the effects of STAT-1 and STAT-3 involve the direct binding to DNA and transcriptional activation of target genes. However, recent studies have shown that for STAT-1 some of its effects appear not to require DNA binding. For example, induction of apoptosis by STAT-1 can be produced by the C-terminal activation domain in the absence of the DNA binding domain. This therefore, appears to involve a co-activator effect in which STAT-1 is recruited to DNA via a DNA-bound transcription factor. In this regard, it is of interest that STAT-1 but not STAT-3 has been shown to interact with p53 and enhance its growth arrest and apoptosis- inducing properties. Hence, the finding that STAT-1 and STAT-3 can modulate the apoptotic programme both by direct DNA binding or via a co-activator mechanism and despite their very similar structures, suggests that these related factors may be therapeutic targets against the damage myocardium following I/R injury. Recently, we reported that the polyphenolic agent epigallocatechin-3-gallate (EGCG), a major constituent of green tea and a potent inhibitor of STAT-1 activation, protects the myocardium against I/R injury.

A mathematical model for the dynamics of large membrane deformations of isolated fibroblasts.

In this paper we develop and extend a previous model of cell deformations, initially proposed to describe the dynamical behaviour of round-shaped cells such as keratinocytes or leukocytes, in order to take into account cell pseudopodial dynamics with large amplitude membrane deformations such as those observed in fibroblasts. Beyond the simulation (from a quantitative, parametrized model) of the experimentally observed oscillatory cell deformations, a final goal of this work is to underline that a set of common assumptions regarding intracellular actin dynamics and associated cell membrane local motion allows us to describe a wide variety of cell morphologies and protrusive activity. The model proposed describes cell membrane deformations as a consequence of the endogenous cortical actin dynamics where the driving force for large-amplitude cell protrusion is provided by the coupling between F-actin polymerization and contractility of the cortical actomyosin network. Cell membrane movements then result of two competing forces acting on the membrane, namely an intracellular hydrostatic protrusive force counterbalanced by a retraction force exerted by the actin filaments of the cell cortex. Protrusion and retraction forces are moreover modulated by an additional membrane curvature stress. As a first approximation, we start by considering a heterogeneous but stationary distribution of actin along the cell periphery in order to evaluate the possible morphologies that an individual cell might adopt. Then non-stationary actin distributions are considered. The simulated dynamic behaviour of this cytomechanical model not only reproduces the small amplitude rotating waves of deformations of round-shaped cells such as keratinocytes [as proposed in the original model of Alt and Tranquillo (1995, J. Biol. Syst. 3, 905-916)] but is furthermore in very good agreement with the protrusive activity of cells such as fibroblasts, where large amplitude contracting/retracting pseudopods are more or less periodically extended in opposite directions. In addition, the biophysical and biochemical processes taken into account by the cytomechanical model are characterized by well-defined parameters which (for the majority) can be discussed with regard to experimental data obtained in various experimental situations.

MR image-guided investigation of regional signal transducers and activators of transcription-1 activation in a rat model of focal cerebral ischemia.

BACKGROUND AND PURPOSE: STAT-1 is a member of a family of proteins called signal transducers and activators of transcription (STATs), and recent studies have shown its involvement in the induction of apoptosis. There is limited information on the role of STAT-1 following stroke. In this study we use MRI measurements of cerebral perfusion and bioenergetic status to target measurements of regional STAT-1 activity. METHODS: Rats were subjected to 60 or 90 min of middle cerebral artery occlusion with and without reperfusion. MRI maps of the apparent diffusion coefficient of water and cerebral blood flow were acquired throughout the study. After the ischemia or reperfusion period, the brain was excised and samples were analyzed by Western blots using anti-phospho-STAT1 and anti-Fas antibodies. Regions were selected for analysis according to their MRI characteristics. RESULTS: Transcriptional factor STAT-1 was enhanced in the lesion core and, to a lesser extent, in the lesion periphery, following ischemia and reperfusion. This level of activity was greater than for ischemia alone. Western blots demonstrated STAT-1 phosphorylation on tyrosine 701 and not serine 727 after ischemia and 3 h of reperfusion. Enhanced expression of the apoptotic death receptor Fas was confirmed after ischemia followed by reperfusion. CONCLUSIONS: This study demonstrates that focal ischemia of the rat brain can induce STAT-1 activation, particularly following a period of reperfusion. The activation occurs not only in the lesion core, but also in the lesion periphery, as identified using MRI. STAT-1 may play an important role in the induction of cell death following stroke.

The protective effect of moderate hypothermia during intestinal ischemia-reperfusion is associated with modification of hepatic transcription factor activation.

BACKGROUND/PURPOSE: Moderate hypothermia throughout intestinal ischemia-reperfusion (IIR) injury reduces multiple organ dysfunction. Heat shock proteins (HSPs) have been shown to be protective against ischemia-reperfusion injury, and STAT (Signal Transducers and Activators of Transcription) proteins are pivotal determinants of the cellular response to reperfusion injury. The aim of this study is to investigate the mechanism of hypothermic protection during IIR. METHODS: Adult rats underwent intestinal ischemia-reperfusion (IIR), 60-minute ischemia and 60-minute reperfusion, or sham (120 minutes) at either normothermia or moderate hypothermia. Four groups of animals were studied: (1) normothermic sham (NS), (2) normothermic IIR (NIIR), (3) hypothermic sham (HS), and (4) hypothermic IIR (HIIR). Western blotting measured heat shock protein expression, phosphorylated (p-) and total (T-) hepatic STAT-1 and STAT-3. RESULTS: There were no differences in expression of HSPs 27, 47, 60, i70, c70, or 90 between any of the experimental groups. NIIR caused a significant increase in p-STAT-1 compared with normothermic sham (P <.05) and a highly significant increase in p-STAT-3 (P <.001), both these increases were completely abolished by moderate hypothermia (P <.01 v NIIR.) CONCLUSIONS: The protective effect of moderate hypothermia on liver is not mediated by HSP expression at this time-point. Hypothermia may act by decreasing hepatic STAT activation, supporting the potential therapeutic role of moderate hypothermia. Modulation of STAT activation may also provide novel therapeutic targets.

Protective effects of the urocortin homologues stresscopin (SCP) and stresscopin-related peptide (SRP) against hypoxia/reoxygenation injury in rat neonatal cardiomyocytes.

Urocortin (UCN), a member of the Corticotropin-Releasing Factor (CRF) family of peptides is a well described cardioprotective agent. UCN is able to bind to two types of G-protein coupled receptors: CRF receptor type 1 (CRFR1) and CRF receptor type 2 (CRFR2), whereas, two homologues of UCN, stresscopin (SCP) or also known as urocortin III (UCNIII) and stresscopin related peptide (SRP), or urocortin II (UCNII), bind exclusively and with high affinity to CRFR2, we hypothesised that they will exhibit more pronounced cardioprotective effects than UCN. We show for the first time that SCP is expressed in rat cardiomyocytes and that the levels of SRP and SCP are increased by hypoxic stress. All three peptides have potent cardioprotective effects in cells exposed to hypoxia/reoxygenation. When used at 10(-8) M they increased the amount of live cells by 25% when added prior to hypoxia, and by 20% when UCN and SCP were added at the onset of reoxygenation. In addition, the peptides are equally are more potent antiapoptotic factors than UCN. The antiapoptotic effects of SCP were more pronounced than SRP and UCN at a concentration of 10(-10) M. Furthermore, SCP and SRP protect cardiomyocytes better than UCN at concentrations up to and including 10(-10) M and reduced the amount of TUNEL positive cells almost by half at concentrations of 10(-12) to 10(-10) M. More importantly, we demonstrate that SCP and SRP are able to protect cardiomyocytes even if they are administered after the hypoxic insult and prior to reoxygenation. In this case SCP was more potent than UCN and SRP at 10(-12) M and both SCP and SRP exhibited higher protection at 10(-8) M compared to UCN. Cardioprotection of cardiomyocytes by 10(-8) M of peptides was abolished when treated with 50 microM LY294002 or 100 microM PD98059, but not by 10 microM SB203580 prior to the hypoxic insult. Transfection of dominant negative Akt and MEK1 also blocked protection by the peptides, whereas dominant negative MEKK6 had no effects, demonstrating that SCP and SRP, like UCN, require activation of p42/44 Mitogen activated protein kinase and Akt/Protein Kinase B in order to produce their cardioprotective effects. In addition, we showed that SCP and UCN are potent activators of the p42/44 MAPK pathway, with SRP able to induce phosphorylation of p42/44 MAPK as well, albeit not as pronounced.

Urocortin protects cardiac myocytes from ischemia/reperfusion injury by attenuating calcium-insensitive phospholipase A2 gene expression.

We have used Affymetrix gene chip technology to look for changes in gene expression caused by a 24 h exposure of rat primary neonatal cardiac myocytes to the cardioprotective agent urocortin. We observed a 2.5-fold down-regulation at both the mRNA and protein levels of a specific calcium-insensitive phospholipase A2 enzyme. Levels of lysophosphatidylcholine, a toxic metabolite of phospholipase A2, were lowered by 30% in myocytes treated with urocortin for 24 h and by 50% with the irreversible iPLA2 inhibitor bromoenol lactone compared with controls. Both 4 h ischemia and ischemia followed by 24 h reperfusion caused a significant increase in lysophosphatidylcholine concentration compared with controls. When these myocytes were pretreated with urocortin, the ischemia-induced increase in lysophosphatidylcholine concentration was significantly lowered. Moreover, co-incubation of cardiac myocytes with urocortin, or the specific phospholipase A2 inhibitor bromoenol lactone, reduces the cytotoxicity produced by lysophosphatidylcholine or ischemia/reperfusion. Similarly, in the intact heart ex vivo we found that cardiac damage measured by infarct size was significantly increased when lysophoshatidylcholine was applied during ischemia, compared with ischemia alone, and that pre-treatment with both urocortin and bromoenol lactone reversed the increase in infarct size. This, to our knowledge, is the first study linking the cardioprotective effect of urocortin to a decrease in a specific enzyme protein and a subsequent decrease in the concentration of its cardiotoxic metabolite.

SAG attenuates apoptotic cell death caused by simulated ischaemia/reoxygenation in rat cardiomyocytes.

Sensitive to apoptosis gene (SAG) is a novel RING finger protein that has been shown to be involved in protection against apoptotic cell death induced by oxidative stress in various cell types. As SAG has been previously shown to be expressed in the heart, we assessed its role in cardiac myocytes exposed to ischaemic stress. SAG expression was enhanced by hypoxia in neonatal cardiomyocytes as well as in the intact heart exposed to ischaemia/reperfusion. SAG levels remain elevated during the first 4 h of reoxygenation and return to control levels after 16 h of reoxygenation. We also show that overexpression of SAG in cardiac myocytes is able to protect against simulated ischaemia/reperfusion-induced apoptotic cell death. However, abrogation of the RING finger of the protein eliminates the anti-apoptotic properties of SAG. Furthermore, an antisense SAG construct enhances cell death, both in normoxic and hypoxic conditions. Hence, we conclude that SAG is a cardioprotective agent in cardiac cells exposed to ischaemic stress and an important protein involved in cardiomyocyte survival.

The C-terminal activation domain of the STAT-1 transcription factor is necessary and sufficient for stress-induced apoptosis.

It has previously been demonstrated that the STAT-1 transcription factor plays a key role in apoptosis induced by the cellular regulatory factors interferon gamma and TNF-alpha. Here we demonstrate that cells lacking STAT-1 show reduced cell death/apoptosis in response to stressful stimuli such as heat or ischaemia. Expression of STAT-1 in these cells does not enhance basal cell death but restores sensitivity to stress-induced death whereas this effect is not observed upon over-expression of STAT-3. Enhanced sensitivity to stress-induced cell death requires the C-terminal activation domain of STAT-1 and the phosphorylation sites at tyrosine 701 and serine 727. Moreover, we show for the first time in any system that the isolated C-terminal domain of STAT-1 is able to enhance stress-induced cell death in the absence of the DNA binding domain or any other region of STAT-1. Hence, STAT-1 plays a key role in stress-induced cell death, potentially acting via a novel co-activator-type mechanism and represents a possible therapeutic target for strategies aimed at minimising cell death, for example, following ischaemic injury.

K(ATP) channel gene expression is induced by urocortin and mediates its cardioprotective effect.

BACKGROUND: Urocortin is a novel cardioprotective agent that can protect cardiac myocytes from the damaging effects of ischemia/reperfusion both in culture and in the intact heart and is effective when given at reperfusion. METHODS AND RESULTS: We have analyzed global changes in gene expression in cardiac myocytes after urocortin treatment using gene chip technology. We report that urocortin specifically induces enhanced expression of the Kir 6.1 cardiac potassium channel subunit. On the basis of this finding, we showed that the cardioprotective effect of urocortin both in isolated cardiac cells and in the intact heart is specifically blocked by both generalized and mitochondrial-specific K(ATP) channel blockers, whereas the cardioprotective effect of cardiotrophin-1 is unaffected. Conversely, inhibiting the Kir 6.1 channel subunit greatly enhances cardiac cell death after ischemia. CONCLUSIONS: This is, to our knowledge, the first report of the altered expression of a K(ATP) channel subunit induced by a cardioprotective agent and demonstrates that K(ATP) channel opening is essential for the effect of this novel cardioprotective agent.

Urocortin increases the expression of heat shock protein 90 in rat cardiac myocytes in a MEK1/2-dependent manner.

We have previously demonstrated that urocortin protects cultured cardiac myocytes from ischaemic and reoxygenation injury and decreases the infarct size in the rat heart exposed to regional ischaemia and reperfusion. Urocortin-mediated cardioprotection is via activation of the mitogen-activated protein kinase (MAP kinase, MEK1/2) pathway. In addition, it is well documented that heat shock protein (hsp) 70 and hsp90 are cardioprotective against lethal stress. In this study we show, for the first time, that urocortin induces the expression of hsp90 but not hsp70 in primary cultures of rat neonatal cardiac myocytes. Levels of hsp90 protein increase by 1.5-fold over untreated cells within 10 min of urocortin treatment and are sustained for 24 h with a maximal increase of 2.5-fold at 60 min (P<0.05 at all time points). The increase in hsp90 expression by urocortin was not inhibited by actinomycin D, and urocortin failed to increase hsp90 promoter activity. Urocortin induction of hsp90 was inhibited by the MEK1/2 inhibitor PD98059 (P<0.001) and by cycloheximide, and both inhibitors abrogate urocortin-mediated cardioprotection (P<0.05 for cycloheximide, P<0.001 for PD98059). Hence, MEK1/2 and protein synthesis are involved in the cardioprotective effect of urocortin against hypoxic-mediated cell death, possibly due to an increase in expression of hsp90 protein. This is the first report of heat shock protein induction by urocortin or any other member of the corticotrophin-releasing hormone family.

CT-1 mediated cardioprotection against ischaemic re-oxygenation injury is mediated by PI3 kinase, Akt and MEK1/2 pathways.

Cardiotrophin-1 protects cardiac myocytes from ischaemic re-oxygenation (IR) injury. CT-1 activates MEK1/2,p42/44MAPK as well as the phosphatidylinositol (PI) 3-OH kinase (PI3) protein kinase B (PKB/Akt) pathway. In this study we investigate the signalling pathways that mediate the anti-apoptotic cell survival effect of CT-1 in IR. Dominant negative gene based inhibitors of MEK1/2, PI3-kinase and Akt inhibited CT-1 mediated cardioprotection in re-oxygenation as did chemical inhibitors of the PI3-kinase pathway. Hence the PI3-kinase/Akt pathway is required in addition to MEK1/2 to mediate CT-1 cardioprotection in IR.

Apoptosis of endothelial cells precedes myocyte cell apoptosis in ischemia/reperfusion injury.

BACKGROUND: Apoptosis contributes to cell loss after ischemia/reperfusion injury in the heart. This study describes the time course and level of apoptosis in different cell types in the intact heart during ischemia/reperfusion injury. METHODS AND RESULTS: Isolated Langendorff-perfused rat hearts were subjected to perfusion alone (control) or to 35 minutes of regional ischemia, either alone or followed by 5, 60, or 120 minutes of reperfusion. Sections were stained by terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) and propidium iodide and with anti-von Willebrand factor, anti-desmin, or anti-active caspase 3 antibodies; they were then visualized by confocal microscopy. Sections were also examined by electron microscopy. No TUNEL-positive cells were seen in control hearts or hearts exposed to ischemia alone. Early in reperfusion, TUNEL staining was colocalized with endothelial cells from small coronary vessels. Endothelial apoptosis peaked at 1 hour of reperfusion and, at this time, there was clear perivascular localization of apoptotic cardiac myocytes, whose number was inversely proportional to their distance from a positive vessel. After 2 hours of reperfusion, apoptotic cardiac myocytes assumed a more homogeneous distribution. Active caspase 3 labeling was seen independent of DNA fragmentation during ischemia alone, but it colocalized with TUNEL staining over the 3 time points of reperfusion. Immunocytochemical findings were confirmed by electron microscopy and Western blotting. CONCLUSIONS: In the very early stages of reperfusion, apoptosis is first seen in the endothelial cells from small coronary vessels. The radial spread of apoptosis to surrounding cardiac myocytes suggests that reperfusion induces the release of soluble pro-apoptotic mediators from endothelial cells that promote myocyte apoptosis.