Biomolecular and biochemical response of myocardial cell to ischemia and reperfusion in the course of heart surgery.

BACKGROUND: Previous studies have shown that biomolecular and biochemical adaptive changes antagonize oxidative damage due to hypoxia and ischemia in myocardial cells. The aim of our study was to verify in human ischemic and reperfused cardiac tissue the relationship between mitochondrial enzyme activities and the activation of HSP70 and c-fos syntheses in the context of a cytoprotective mechanism. Nitric oxide (NO) modulating effects on mitochondrial respiratory chain enzyme activities in ischemic and reperfused tissue were investigated (preliminary report). METHODS: During elective coronary artery bypass grafting, in 30 consecutive patients ventricle samples were taken one before aortic clamping the second after 55+/-8 min ischemic period and the third 34+/-5 after final reperfusion. Coronary sinus blood samples were taken in parallel to assess free radical release measured by malonaldehyde (MDA) levels. In a small number of patients (N=5) nitric oxide tissue levels were analyzed. RESULTS: When compared with normoxic tissue, a significant decrease in cytochrome Coxidase (COX) and succinate Cyt-c reductase (SCR) activities in ischemic and reperfused samples were observed. The activation of HSP70-72 and c-fos transcription factor was evident in courses of ischemia and reperfusion. Blood MDA levels underline the concept that oxyradical generation characterize the peroxidative damage in reoxygenated myocardial tissue while adaptive changes which occur in ischemic cells seem to antagonize the oxyradical injury. CONCLUSIONS: In the course of heart surgery the myocardial cell seems to prevent ischemic damage by activating some peculiar biomolecular and biochemical adaptive changes which permit the reversibility of the oxidative injury. In contrast it appears evident that massive and rapid reoxygenation of the cardiac tissue leads to peroxidative damage due to oxyradical generation. Nitric oxide seems to play a crucial role in cellular adaptation to ischemia even if further studies will be needed to elucidate these findings. From the data obtained in this work we cannot draw certain conclusions in terms of human cardiac cell adaptation to ischemia whereas it seems convincible that reoxygenation, as actually employed in clinical practice, compromises the integrity of the cells.

Influence of proteasome and redox state on heat shock-induced activation of stress kinases, AP-1 and HSF.

We studied the pattern of activation of stress kinases and of transcription factors activator protein-1 (AP-1) and heat shock factor (HSF) in FAO cells by combining two treatments, i.e. heating (42 degrees C for 1 h) and proteasome inhibition, each known to cause cellular heat shock response. The co-treatment heat shock (HS) and proteasome inhibitor (a peptidyl aldehyde or lactacystin) showed cumulative effects on the intensity and duration of activation of c-Jun N-terminal kinase (JNK) and p38 mitogen-activated protein kinase (MAPK) at the end of the HS period and during recovery. Similarly, the thiol-reducing agents N-(2-mercaptoethyl)-1,3-diaminopropane and dithiothreitol strongly activated both JNK and p38 MAPK in cells undergoing HS. AP-1 DNA binding activity in response to proteasome inhibitors was so strong that it shadowed the stimulatory effect of HS in the combined treatment, but lactacystin, which is the most potent and specific proteasome inhibitor, decreased the binding late during recovery from HS. Thiol-reducing agents prevented AP-1 DNA binding induced by HS. The combined HS/proteasome inhibitors or HS/thiol-reducing agents treatments cooperatively activated HSF DNA binding. Expression of collagenase I and hsp 70 mRNAs reflects the different behavior of AP-1 and HSF transcription factors in cells exposed to HS and proteasome inhibition. The data seem to indicate that JNK and p38 MAPK activations are not necessarily coupled to DNA binding of AP-1, which can be either increased or inhibited when these kinases are activated. AP-1 and HSF show opposite patterns of response to HS in the presence of proteasome inhibitors or reducing agents.

Transferrin receptor induction by hypoxia. HIF-1-mediated transcriptional activation and cell-specific post-transcriptional regulation.

The tight relationship between oxygen and iron prompted us to investigate whether the expression of transferrin receptor (TfR), which mediates cellular iron uptake, is regulated by hypoxia. In Hep3B human hepatoma cells incubated in 1% O(2) or treated with CoCl(2), which mimics hypoxia, we detected a 3-fold increase of TfR mRNA despite a decrease of iron regulatory proteins activity. Increased expression resulted from a 4-fold stimulation of the nuclear transcription rate of the TfR gene by both hypoxia and CoCl(2). A role for hypoxia-inducible factor (HIF-1), which activates transcription by binding to hypoxia-responsive elements in the activation of TfR, stems from the following observations. (a) Hypoxia and CoCl(2)-dependent expression of luciferase reporter gene in transiently transfected Hep3B cells was mediated by a fragment of the human TfR promoter containing a putative hypoxia-responsive element sequence, (b) mutation of this sequence prevented hypoxic stimulation of luciferase activity, (c) binding to this sequence of HIF-1alpha, identified by competition experiments and supershift assays, was induced in Hep3B cells by hypoxia and CoCl(2). In erythroid K562 cells, the same treatments did not affect iron regulatory proteins activity, thus resulting in a stimulation of TfR gene expression higher than in hepatoma cells.

Influence of polyamines on DNA binding of heat shock and activator protein 1 transcription factors induced by heat shock.

Polyamine depletion, obtained in FAO cells with specific inhibitors of biosynthetic enzymes, prevents or decreases the accumulation of hsp 70 mRNA following heat shock [Desiderio et al., Hepatology 24 (1996) 150-156]. The present study shows that under conditions of spermidine depletion caused by alpha-difluoromethylornithine, the DNA binding capacity of the transcription factor HSF induced by heat shock undergoes a severe and prompt deactivation. Replenishment of the spermidine pool before heat shock re-establishes the DNA binding activity of HSF and the inducibility of hsp 70 mRNA. Similar to HSF, but with a different time-course, the DNA binding of the transcription factor AP-1 activated by heat shock is also impaired in spermidine-depleted cells and reversed by exogenous spermidine. STAT3 provides an example of a transcription factor slightly activated by heat shock but insensitive to polyamine decrease.

Differential activation of some transcription factors during rat liver ischemia, reperfusion, and heat shock.

Cells respond to external stimuli by changes in gene expression that are largely dependent on transcription factors (TFs). We studied the behavior of some TFs in rat liver during ischemia, postischemic reperfusion, and heat shock. Knowledge of the conditions at the end of ischemia is essential to understand changes occurring at reperfusion. The TFs investigated are known to be typically responsive to heat shock (HSF), hypoxia (HIF-1), pro- and antioxidant conditions (AP-1), or to various environmental changes (HNF-1 and ATF/CREB family). The most relevant new information includes the following: 1) Liver ischemia activates extremely rapidly the DNA binding capacity of HSF, soon followed by analogous activation of HIF-1 and AP-1. 2) After a certain lag time from the activation of HIF-1, mRNAs accumulate for two glycolytic enzymes, in particular Aldolase A and Heme Oxygenase 1, which contain HIF-1 sequences in their promoters. 3) Reperfusion, which is known to further increase the binding of HSF and to induce NFkappaB binding, abrogates or decreases the binding of HIF-1 and AP-1, stimulated by ischemia, and activates the binding of ATF/CREB. Later on, a second peak of AP-1 binding is induced. 4) Heat shock activates both ischemia-responsive and reperfusion-responsive TFs. 5) Preliminary experiments of supergelshift reveal that the activation of AP-1 at reperfusion or upon heat shock may result from the different involvement of the component subunits.

Effect of reactive oxygen species on iron regulatory protein activity.

Iron may be important in catalyzing excessive production of reactive oxygen species (ROS). Cellular iron homeostasis is regulated by iron regulatory proteins (IRPs), which bind to iron-responsive elements (IRE) of mRNAs for ferritin and transferrin receptor (TfR) modulating iron uptake and sequestration, respectively. Although iron is the main regulator of IRP activity, IRP is also influenced by other factors, including the redox state. Therefore, IRP might be sensitive to pathophysiological alterations of redox state caused by ROS. However, previous studies have produced diverging evidence on the effect of oxidative injury on IRP. Results obtained in an animal model close to a pathophysiological condition, such as ischemia reperfusion of the liver as well as in a cell-free system involving an enzymatic source of O2 and H2O2, indicate that IRP is downregulated by oxidative stress. In fact, IRP activity is inhibited at early times of post-ischemic reperfusion. Moreover, the concerted action of O2 and H2O2 produced by xanthine oxidase in a cell-free system caused a remarkable inhibition of IRP activity. IRP seems a direct target of ROS; in fact, in vivo inhibition can be prevented by the antioxidant N-acetylcysteine and by interleukin-1 receptor antagonist. In addition, modulation of iron levels of the cell-free assay did not affect the downregulation imposed by xanthine oxidase. Conceivably, downregulation of IRP activity by O2 and H2O2 may facilitate iron sequestration into ferritin, thus limiting the pro-oxidant challenge of iron.

Induction of ferritin synthesis in ischemic-reperfused rat liver: analysis of the molecular mechanisms.

BACKGROUND & AIMS: Iron may catalyze the production of reactive oxygen species (ROS) during postischemic reoxygenation. Ferritin, a cellular iron storage protein, can either represent a source of iron or perform a cytoprotective action against ROS. The aim of this study was to address the role of ferritin in postischemic reperfusion. METHODS: Transcriptional and posttranscriptional mechanisms controlling ferritin gene expression were studied in reperfused rat livers. RESULTS: Proteolysis reduced ferritin levels 2 hours after reperfusion, but a concomitant increase of synthesis, accompanied by enhanced transcription and accumulation of H and L ferritin subunit messenger RNAs (mRNAs), almost re-established normal ferritin content at 4 hours. Pretreatment with interleukin 1 receptor antagonist (IL-1RA) did not prevent the rise of ferritin mRNAs. RNA bandshift assays showed that the activity of the iron regulatory proteins (IRPs), which control ferritin mRNA translation, declined early after reperfusion and recovered progressively thereafter. Pretreatment with either the antioxidant N-acetyl cysteine or IL-1RA was sufficient to prevent almost completely down-regulation of IRP activity. CONCLUSIONS: Postischemic reperfusion causes degradation of ferritin, possibly increasing iron levels. However, induction of ferritin gene transcription, possibly mediated by ferritin-derived iron and ROS-mediated inactivation of IRP, which allows translation of ferritin mRNAs, counteracts this effect and concurs to reestablish the amount of ferritin, which may thus act to limit reperfusion damage.

Differential activation of heat shock and nuclear factor kappaB transcription factors in postischemic reperfused rat liver.

The aim of this study was to investigate the behavior of the transcription factors, heat-shock factor (HSF) and nuclear factor kappaB (NF-kappaB), in postischemic reperfused liver, with particular attention paid to possible differences in the time-course and mechanism of activation, which may help in defining their role in the response of the liver to reperfusion. Ischemia was induced by clamping the hilar pedicle of the left lateral and median liver lobes; the clamp was removed after 1 hour. Some rats were treated intraperitoneally with IL-1 receptor antagonist (IL-1RA) 30 minutes before ischemia and at the time of reperfusion. Binding of NF-kappaB to the corresponding consensus sequence is activated after 30 minutes of reperfusion, and is still increased 1 hour after reperfusion. Activation is suppressed in rats treated with IL-1RA; NF-kappaB persists in the cytosol associated with the inhibitor, IkappaB, and can be artifactually activated in vitro. Super-gel shift experiments revealed that the two subunits, p50 and p65, are involved in the activation of binding. In contrast, binding of HSF to the corresponding consensus sequence, heat shock element (HSE), is already activated at the end of ischemia, shows a further increase after 30 minutes of reperfusion, but declines 1 hour after reperfusion; more importantly, it is not inhibited by pretreatment of the rat with IL-1RA. In conclusion, although both HSF and NF-kappaB are activated by ischemia-reperfusion, there are clear differences in time-course and mechanism of activation of the two transcription factors. Activation of HSF depends directly on some events occurring during ischemia; NF-kappaB is activated only after reperfusion and the concurrent oxidative stress, by an indirect mechanism that can be suppressed by IL-1RA. The possibility of dissociating the activation of these two transcription factors in postischemic reperfusion can have a prospective clinical relevance.

The MAP kinase cascades are activated during post-ischemic liver reperfusion.

We have investigated the involvement of MAP kinase cascades in the response of the liver to post-ischemic reperfusion. Both JNKs and ERKs are activated but the duration and magnitude of the increase in their activities appear to be different. JNK activation is more marked but shorter than that of ERKs. The increase observed in the phosphotyrosine content of the 52 kDa Shc protein, accompanied by an increased amount of co-immunoprecipitated Grb2, and the activation of Raf-1 kinase provide evidence of the involvement of a Ras-Raf-dependent pathway, with a time course that is similar to that of ERK activation. The treatment of rats with IL-1 receptor antagonist modified all of the described effects, suggesting that IL-1 plays a role in the response of the liver to reperfusion.

Superoxide and hydrogen peroxide-dependent inhibition of iron regulatory protein activity: a protective stratagem against oxidative injury.

Cellular iron homeostasis is regulated by the cytoplasmic iron regulatory protein (IRP), which binds to iron-responsive elements (IRE) of mRNAs, modulating iron uptake and sequestration, respectively. When iron is scarce, IRP binds to IRE and coordinately increases the synthesis of transferrin receptor and decreases that of ferritin, thus providing the cell with readily available free iron. When iron is in excess, IRP does not bind and iron sequestration prevails over iron uptake. We have found that incubation of rat liver lysates with xanthine oxidase (XO), which generates superoxide (O2-.) and hydrogen peroxide (H2O2), caused a remarkable but reversible inhibition of IRP activity, as the formation of IRE-IRP decreased by 70-80% but returned to baseline values upon exposure to a reducing agent like 2-mercaptoethanol. IRP inhibition was prevented by separate or simultaneous addition of superoxide dismutase and catalase, showing that both O2-. and H2O2 were involved. By contrast, iron chelators and hydroxyl radical scavengers did not impede the inhibition of IRP, suggesting that O2-. and H2O2 acted independently of free iron sources. Ferritin enhanced IRP inhibition, but this process involved tightly bound iron centers that shunted reducing equivalents from XO and returned them to oxygen, thus increasing the formation of O2-. In agreement with the exclusive role of O2-. and H2O2, XO also inhibited recombinant human IRP in the absence of iron. These results demonstrate that O2-. and H2O2 can directly but reversibly down-regulate the RNA-binding activity of IRP, causing transient decrease of free iron that otherwise would convert them into more potent oxidants such as hydroxyl radicals or equally aggressive iron-peroxo complexes. This establishes a novel protective stratagem against oxidative injury under pathophysiologic conditions characterized by the excessive generation of O2-. and H2O2.

Effects of polyamine imbalance on the induction of stress genes in hepatocarcinoma cells exposed to heat shock.

The expression of hsp70-the inducible member of the corresponding heat shock gene family-of the oxidative stress marker gene heme oxygenase (HOx), and of the immediate early response genes c-fos and c-jun has been studied in FAO hepatocarcinoma cells depleted of polyamines and exposed to heat shock. Depletion of polyamines was obtained in short-term experiments (24-48 hours) by the use of alpha difluoromethylornithine (DFMO), a classical inhibitor of ornithine decarboxylase (ODC), or of the combination of the newly available inhibitors of ODC and S-adenosylmethionine decarboxylase, i.e., (2R,5R)-hept-6-yne-2,5-diamine (MAP) and 5'{[(Z)-4-aminobut-2-enyl]methylanino}-5-deoxyadeno-si ne (AbeAdo). Under our experimental conditions polyamine imbalance was realized without appreciable growth-related genes. Decreases of putrescine and spermidine 48 hours after DFMO prevented the induction of hsp70 messenger RNA (mRNA), whereas depletion spermidine and spermine obtained with MAP/AbeAdo decreased intensity and duration of post-heat shock accumulation of hsp70 mRNA. Inductions of HOx, c-jun and c-fos were also inhibited. Because MAP/AbeAdo caused also an intracelluar accumulation of putrescine, we tested the effect of exogenous putrescine, which was found to stabilize the mRNAs for hsp70 and c-jun. Hsp70 and HOx are thought to play a protective role, and the proteins of c-jun and c-fos constitute the transcription factor activator protein-1, which is involved in the transcription of many defensive products. Therefore, the integrity of polyamine pool seems to be a necessary permissive condition for an effective response of the cells to adverse environmental changes.

Post-transcriptional control of increased hepatic catalase gene expression in response to oxidative stress.

Catalase is an important member of the antioxidant network that protects the cell against reactive oxygen species (ROS). We studied catalase gene expression in the liver of rats exposed to oxidative stress induced by the ROS-generating drug nitrofurantoin (NF). Catalase enzymatic activity and content are enhanced by NF treatment. The corresponding increase in the steady state level of the messenger ribonucleic acid (mRNA) occurs without significant changes in transcription and seems therefore controlled post-transcriptionally. Indeed, RNA band-shift assays demonstrated a reduced binding of redox-sensitive cytoplasmic protein(s) to the 3' region of catalase mRNA in NF-treated rats, thus suggesting that the redox state of protein that binds to an antioxidant enzyme mRNA may play a role in the hepatic response to oxidative stress.

The liver response to in vivo heat shock involves the activation of MAP kinases and RAF and the tyrosine phosphorylation of Shc proteins.

We have investigated the mechanisms of signal transduction in the response of liver to heat shock in vivo. By immunoblot experiments we have shown that heat shock decreases the electrophoretic mobility of the 40 and 43 kDa mitogen activated protein kinases (MAPKs) and we have found a significant increase of MAPK activity measured as phosphotransferase capacity of both cytosolic extracts and MAPK immunoprecipitates. To elucidate the signalling pathway which accounts for MAPK activation, we focused our attention on its upstream factors, Raf and Ras. We have shown that, heat shock activates Raf-1 kinase and causes an increase in phosphotyrosine content of the 52 kDa Shc protein accompanied by an increment in the amount of coimmunoprecipitated Grb2. These findings provide the first evidence that the Ras-Raf-MAPK pathway is activated in liver during heat shock in vivo.

Heat shock activation of NFkB in rat liver is mediated by interleukin-1.

Exposure to high temperature (heat shock) activates the transcription factor NFkB in the liver of the living rat, but is not effective in hepatoblastoma cells in culture: on the contrary, activation of the heat shock transcription factor (HSF) occurs under both conditions. Pre-treatment of the rat with IL-1 receptor antagonist suppresses the activation of NFkB, which seems to be mediated by the release of this cytokine, but does not hamper the activation of HSF and the concurrent induction of hsp 70 mRNA. IL-1 activity actually shows a strong, albeit transient, increase in the blood of heat shocked rats.

Differential activation of heat-shock and oxidation-specific stress genes in chemically induced oxidative stress.

Post-ischaemic reperfusion increases the level of the major heat-shock (stress) protein hsp 70 and of its mRNA by transcriptional mechanisms, and activates the binding of the heat-shock factor HSF to the consensus sequence HSE. In common with CoCl2 treatment, post-ischaemic reperfusion increases the level of haem oxygenase mRNA, an indicator of oxidative stress, but CoCl2 does not seem to induce the expression of the hsp 70 gene [Tacchini, Schiaffonati, Pappalardo, Gatti and Bernelli-Zazzera (1993) Lab. Invest. 68, 465-471]. Starting from these observations, we have now studied the expression of two genes of the hsp 70 family and of other possibly related genes under conditions of oxidative stress. Three different chemicals, which cause oxidative stress by various mechanisms and induce haem oxygenase, enhance the expression of the cognate hsc 73 gene, but do not activate the inducible hsp 70 gene. Expression of the other genes that have been studied seems to vary in intensity and/or time course, in relation to the particular mechanism of action of any single agent. The pattern of induction of the early-immediate response genes c-fos and c-jun observed during oxidative stress differs from that found in post-ischaemic reperfused livers. Oxidative-stress-inducing agents do not promote the binding of HSF to its consensus sequence HSE, such as occurs in heat-shock and post-ischaemic reperfusion, and fail to activate AP-1 (activator protein 1). With the possible exception of Phorone, the oxidative stress chemically induced in rat liver activates NFkB (nuclear factor kB) and AP-2 (activator protein 2) transcription factors.

Induction of ferritin synthesis by oxidative stress. Transcriptional and post-transcriptional regulation by expansion of the "free" iron pool.

Ferritin, by regulating the "free" intracellular iron pool, controls iron-catalyzed generation of reactive oxygen species, but its role in oxidative damage is still unclear. We show that ferritin synthesis is significantly stimulated in the liver of rats subjected to oxidative stress by treatment with phorone, a glutathione-depleting drug. RNA-bandshift assays document reduced activity of iron regulatory factor, in particular of IRFB, the cytoplasmic protein that post-transcriptionally controls ferritin mRNA translation. Furthermore, Northern blot analysis shows increased accumulation of H and L subunit mRNAs, and nuclear run-on experiments provide evidence of transcriptional activation. Direct measurements of intracellular free iron levels by EPR indicate that the increased ferritin synthesis can be mediated by an expansion of the free iron pool. An early drop of ferritin content after phorone treatment indicates that part of the iron that fuels the free pool might derive from ferritin degradation. Present data seem to suggest that, under conditions of oxidative stress, liver ferritin can represent either a pro- or an anti-oxidant in a time-dependent manner. In fact, its early degradation contributes to expand the intracellular free iron pool that, later on, activates multiple molecular mechanisms to reconstitute ferritin content, thus limiting the pro-oxidant challenge of iron.

Rapid stimulation of mitogen-activated protein kinase of rat liver by prolactin.

Intraperitoneal prolactin administration to female rats caused a rapid and transient stimulation of hepatic mitogen-activated kinase (MAP kinase) activity measured in vitro as cytosolic phosphotransferase capacity towards two specific substrates. Myelin basic protein kinase activity of MAP kinase immunoprecipitates confirmed the specificity and magnified the prolactin effect. Immunoblot experiments with anti-(MAP kinase) and anti-phosphotyrosine antibodies showed changes in both electrophoretic mobility and phosphotyrosine content of 40 and 44 kDa isoenzymes suggesting that prolactin affects these isoforms. Concomitant with the increase in MAP kinase activity, prolactin induced tyrosine phosphorylation in a number of liver proteins, suggesting a rapid involvement of tyrosine kinases which might be correlated in some way with MAP kinase activation. Protein kinase C activity, which has been implicated in the regulation of MAP kinase and in mediating the prolactin effect, does not seem to participate in MAP kinase activation.

Expression of a growth arrest specific gene (gas-6) during liver regeneration: molecular mechanisms and signalling pathways.

A set of growth arrest-specific (gas) genes negatively regulated by serum has been identified. To define the role of gas genes in a model of cell proliferation in vivo we analyzed the expression of one of these genes (gas-6) during liver regeneration after partial hepatectomy (PH). We found that gas-6 mRNA was down-regulated 4 hours after PH, within the G0 to G1 transition. Later on, gas-6 mRNA increased over the level found in normal liver with a peak at 16 hours, before the onset of DNA synthesis. This surge was probably triggered by an inflammatory response caused by the surgical trauma, because an increase of similar extent occurring with the same time course was present in livers of sham-operated and turpentine-treated rats. Comparison of mRNA steady state levels with nuclear transcription rates indicated that gas-6 expression is post-transcriptionally regulated. As we found that down-regulation of gas-6 expression was prevented by treatment with Actinomycin D, a labile protein might be involved in the determination of gas-6 mRNA stability. To investigate the mitogenic signals controlling gas-6 expression during liver regeneration we treated hepatectomized rats with a specific alpha-1-adrenoceptor blocker (prazosin) as well as with drugs which modify intracellular calcium levels. The decrease of gas-6 mRNA 4 hours after PH was prevented by prazosin and by neomycin, an inhibitor of calcium release from endogenous stores. These findings suggest that down-regulation of gas-6 expression during hepatic regeneration is triggered by catecholamines interaction with alpha-1-adrenergic receptors and by subsequent calcium release. In addition we found that the rise of gas-6 gene expression occurring at 16 hours after PH was not affected by prazosin but was inhibited by trifluoperazine. Therefore, we suggest that up-regulation of gas-6 gene expression is mediated by the interaction of calcium with calmodulin, independently of catecholamines.

Protein synthesis and gene expression in transplanted and postischemic livers.

The expression of some genes has been comparatively studied in transplanted rat liver and in liver reperfused after ischemia in situ. Experiments on protein synthesis by tissue slices from cold-stored or transplanted livers show that rat livers that retain a good capacity for protein synthesis during storage undergo a profound impairment in the capacity for protein synthesis during the first hours after transplantation. This recovers in the following hours. There is never any indication of synthesis of stress proteins, and of hsp 70 in particular. The steady-state level of mRNAs for albumin, transferrin, and beta-actin, which are well expressed in reperfused postischemic livers in vivo, are reduced early after transplantation and recover only many hours later. Run-on analysis shows that an early defect in transcription and a partial recovery of this process later on are responsible for these changes. The steady-state levels of the same mRNAs are well maintained in donor livers preserved in University of Wisconsin solution for at least 12 hr, and less satisfactorily in Euro-Collins solution. Results of run-on analysis parallel the data on mRNA levels. The behavior of these mRNAs is, therefore, clearly different in reperfused and transplanted liver. The early stages of liver transplantation seem to be characterized by a depressed capacity of gene expression, without the reactive phenomenon of activation of stress protein genes that occurs in reperfused postischemic livers.