Altered Hepatic Gene Expression Profiles Associated With Myocardial IschemiaCLINICAL PERSPECTIVE
Background— Acute coronary syndrome is sometimes accompanied by accelerated coagulability, lipid metabolism, and inflammatory responses, which are not attributable to the cardiac events alone. We hypothesized that the liver plays a pivotal role in the pathophysiology of acute coronary syndrome. We simultaneously analyzed the gene expression profiles of the liver and heart during acute myocardial ischemia in mice.
Methods and Results— –Mice were divided into 3 treatment groups: sham operation, ischemia/reperfusion, and myocardial infarction. Mice with liver ischemia/reperfusion were included as additional controls. Marked changes in hepatic gene expression were observed after 24 hours, despite the lack of histological changes in the liver. Genes related to tissue remodeling, adhesion molecules, and morphogenesis were significantly upregulated in the livers of mice with myocardial ischemia/reperfusion or infarction but not in those with liver ischemia/reperfusion. Myocardial ischemia, but not changes in the hemodynamic state, was postulated to significantly alter hepatic gene expression. Moreover, detailed analysis of the signaling pathway suggested the presence of humoral factors that intervened between the heart and liver. To address these points, we used isolated primary hepatocytes and showed that osteopontin released from the heart actually altered the signaling pathways of primary hepatocytes to those observed in the livers of mice under myocardial ischemia. Moreover, osteopontin stimulated primary hepatocytes to secrete vascular endothelial growth factor-A, which is important for tissue remodeling.
Conclusions— Hepatic gene expression is potentially regulated by cardiac humoral factors under myocardial ischemia. These results provide new insights into the pathophysiology of acute coronary syndrome.
Received June 1, 2008; accepted November 10, 2009.
In addition to chest pain, acute coronary syndrome (ACS) is sometimes accompanied by systemic manifestations, such as proinflammatory responses, activation of the coagulation-fibrinolytic system, and lipid metabolism.1–3 These are considered to be systemic reactions involving multiple organs, which exacerbate the cardiac events.
Clinical Perspective on p 68
C-reactive protein, coagulation factors, and protein C, the levels of which fluctuate in ACS, are liver-specific factors. Although these reports were based on a limited number of factors, the observations suggest a close relation between the liver and myocardial ischemia and imply that the liver plays a pivotal role in the pathophysiology of ACS.
cDNA microarray technology allows simultaneous analysis of the expression levels of thousands of genes. Genome-based expression profiling provides useful information on the molecular pathogenesis of various diseases as well as disease progression and prognosis.4–7 Previous microarray studies have examined the molecular dynamics of the myocardium induced by myocardial ischemia.8,9 However, global gene expression analyses applied to the liver affected by myocardial ischemia have not been reported.
In this study, we examined the responses of hepatic gene expression to myocardial ischemia. Given the systemic inflammation that characterizes ACSs, we postulated that regulation of hepatic genes occurs by inflammatory mediators and not by alterations in hemodynamics or hepatic perfusion. Therefore, we used whole-genome transcriptional profiling to identify hepatic genes selectively regulated in myocardial ischemia.
This study was approved by institutional and governmental animal research committees and was conducted in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication No. 85-23, revised 1996). C57BL/6J mice (n=46; body weight, 24.1±1.4 g; 8 to 10 weeks of age; Charles River Laboratories, Yokohama, Japan) were divided into the following treatment groups: sham operation (n=11), ischemia/reperfusion (I/R; n=10), myocardial infarction (MI; n=10), liver I/R (n=10), and sham operation plus hydralazine (n=5). Hepatic gene expression was evaluated among these groups, and the results were further investigated in primary mouse hepatocytes.
An expanded Methods section containing details of animal surgery, hydralazine group, liver I/R group, blood sampling and analysis, histopathological analysis, blood pressure and heart rate measurements, microarray experiments, processing of cDNA microarray data, extraction of significantly upregulated cardiac and hepatic genes, pathway analysis, ELISA for secreted osteopontin and vascular endothelial growth factor (VEGF), primary hepatocyte experiments, and quantitative real-time detection polymerase chain reaction (RTD-PCR) is available in the online-only Data Supplement.
All microarray data have been deposited in the National Center for Biotechnology Information Gene Expression Omnibus database with the series accession number GSE14843.
The data are presented as the mean±SEM for each group of mice and were analyzed by ANOVA with Bonferroni post hoc test for multiple comparisons. Statistical analyses of blood sampling, blood pressure, and heart rate were performed with the Steel (heterogeneity of variance) multicomparison test. Significance was set at P<0.05. Statistical analyses were performed with SAS statistical software (SAS Institute Japan, Tokyo, Japan).
Establishment of Cardiac I/R or MI in Mice
Cardiac I/R or MI was successfully induced in normal C57BL/6J mice. The levels of cardiac enzymes, such as creatine kinase, aspartate aminotransferase, and lactate dehydrogenase, increased significantly after 6 hours in the I/R group and showed markedly greater increases in the infarction group compared with the sham group (Table 1). In addition, the normalization of these enzyme levels was reduced after 24 hours in the infarction group.
Histologically, azan or hematoxylin/eosin staining showed wall thinning, coagulation necrosis, and transmural fibrosis in the risk area in the infarction group but not in the I/R group (data not shown). As shown in Table 2, no significant differences were found in heart rate or blood pressure after 24 hours compared with the preoperative values in the sham and I/R groups, whereas a decrease in blood pressure was found in the infarction group.
Histological Assessment of the Liver After Cardiac I/R or MI
The I/R and infarction groups showed a minimal, but transient, increase in alanine aminotransferase ALT. Although alanine aminotransferase may be released from the myocardium10 rather than from the liver, to exclude the effect of the transient change in hepatic venous pressure associated with cardiogenic shock, we examined histological changes in the liver after myocardial I/R or infarction. No histological abnormalities were observed in the shocked liver, as indicated by the lack of hepatocyte necrosis in acinar zone 3 in the sham, I/R, and infarction groups (Figure 1a, 1c, 1e, and 1g; hematoxylin/eosin staining). In addition, no signs of liver congestion were observed, as indicated by the lack of dilatation of the terminal hepatic venules and adjacent sinusoids in the sham, I/R, or infarction group (Figure 1b, 1d, 1f, and 1h; silver staining).
On transmission electron microscopy, no ischemic changes, such as swelling or loss of cristae in the mitochondria, a mixed irregular pattern or swelling of the rough endoplasmic reticulum, or dilatation or indistinct appearance of the sinusoids, were observed in the sham, I/R, or infarction group (Figure 2A through 2C). Based on these results, histological analysis did not demonstrate the presence of shock or congestive liver in the I/R or infarction group.
Changes in the Hepatic Gene Expression Profile After Cardiac I/R or MI
Although no histological changes were observed in the liver after cardiac I/R or MI, significant changes in gene expression were noted. Hierarchical clustering analysis, which is a nonsupervised learning method that includes 23 281 nonfiltered genes, produced clusters for the I/R or infarction group and the sham-operated group (data not shown). Because nonfiltered genes may include those that are unchanged in all samples, which generated “noise” that prevented efficient gene clustering, we filtered out these genes with different stringency and performed hierarchical clustering. Hierarchical clustering with 9165 (log-ratio variations >40th percentile) or 5156 (log-ratio variations >50th percentile) filtered genes clearly demonstrated clusters for the I/R or infarction group after 24 hours, for the I/R or infarction group after 6 hours, and for the sham group after 6 and 24 hours (supplemental Figure I). Hierarchical clustering with 773 (log-ratio variations >80th percentile) or 96 (log-ratio variations >90th percentile) filtered genes showed more detailed and clearer clusters for the I/R group after 24 hours, for the infarction group after 24 hours, for the I/R or infarction group after 6 hours, and for the sham group after 6 and 24 hours (Figure 3). Thus, by filtering out “noise” genes, more detailed and clearer clustering could be obtained, thus addressing the reliability of the analysis.11 The increased robustness (R-index) and decreased discrepancy (D-index) of clustering with filtering conditions supported this finding (supplemental Figure I; expanded Methods and Results).
Class prediction analysis, a supervised learning method based on the compound covariate predictor, was performed with various clinical parameters, including provocation (I/R or infarction), 6 hours (I/R or infarction after 6 hours), 24 hours (I/R or infarction after 24 hours), and time (sham or 6 hours, sham or 24 hours, and 6 or 24 hours). The results indicated that provocation, 24 hours, and time significantly classified these models (Table 3).
Both nonsupervised and supervised learning methods indicated differences in hepatic gene expression profiling among sham, 6 hours, and 24 hours after heart provocation, and different heart provocation (I/R or infarction) may generate differences in hepatic gene expression, especially 24 hours after provocation.
Identification of Genes Differentially Expressed Between I/R and Infarction
Because the filtering process may result in loss of important genes, for identification of differentially expressed genes among different groups, we used a class comparison analysis tool (http://linus.nci.nih.gov/BRB-ArrayTools.html). Class comparison analysis (P<0.0005) among the 5 groups (ie, sham, I/R-6, I/R-24, infarction-6, and infarction-24) was performed, and genes that were differentially expressed among the 5 groups were extracted. On 1-way hierarchical clustering analysis of the extracted genes and heat map, 6 gene clusters were assigned on the basis of the gene expression patterns (Figure 4). Of the 6 groups, group 2 showed significant upregulation for I/R and infarction after 24 hours compared with the other groups. Group 3 showed upregulation for I/R, but not for infarction, after 24 hours. Group 4 showed downregulation for I/R and infarction after 24 hours compared with the other groups. Group 5 showed downregulation for infarction after 24 hours compared with the other groups. Representative genes (>3-fold difference in t value) and frequent pathways observed in each group (based on the MetaCore database) are listed in supplemental Tables I through IV.
Interestingly, in group 2, genes related to tissue remodeling, adhesion molecules, and morphogenesis were significantly upregulated. This may be related to the induction of tissue repair factors, such as antigenic factor and myocardiogenic factors, associated with I/R or infarction. In addition, genes involved in the cell cycle and apoptosis and neuron-related genes, such as retinoblastoma 1, angiopoietin-like 4, apoptotic peptidase-activating factor 1, transformation-related protein 53 (p53), and Eph receptor B1, were preferentially expressed. The expression of group 2 genes was significantly correlated with serum creatine kinase levels, suggesting that these genes reflect the severity of cardiac damage. Especially, (R=0.856, P<e−07) and apoptotic peptidase-activating factor 1 (R=0.856, P<e−07) were highly correlated with creatine kinase (supplemental Table I).
In group 3, in addition to the genes described earlier, chemokine and hormone gene pathways involved in interleukin (IL)-8 and androgen or estrogen receptor signaling were upregulated, suggesting that more tissue repair and bioreactive signaling pathways were activated. This may reflect the presence of a living myocyte I/R condition. In group 4, genes involved in lipid catabolism, immune response, proteolysis, and oxidative stress, such as apolipoprotein A-II, CD7 antigen, and reduced nicotinamide-adenine dinucleotide phosphate oxidase 1, were downregulated in the infarction and I/R groups after 24 hours. In group 5, genes involved in muscle and neurite morphogenesis, such as myosin (heavy polypeptide 11, smooth muscle) and ephrin A5, were significantly downregulated in the infarction group after 24 hours.
Effects of Hemodynamic State on Hepatic Gene Expression Profile
To exclude the possibility that changes in hemodynamic state induced alterations in hepatic gene expression, we examined the livers of mice subjected to liver I/R. For liver I/R, gentle occlusion of the hepatic artery and portal vein was applied so that the extent of liver injury was comparable with those in the myocardial I/R and infarction models (Table 1).
We analyzed the gene expression profile of the liver I/R group by using the same extracted genes as shown in Figure 4. The gene expression patterns induced in the myocardial I/R and infarction groups are clearly different from those in the liver I/R group (Figure 5), except for the group 3 gene cluster in myocardial I/R. It should be noted that the group 3 gene cluster was upregulated in the myocardial I/R group at 24 hours after provocation, whereas it was upregulated from 6 hours after provocation in the liver I/R group. Therefore, the delayed changes in hepatic gene expression in the myocardial I/R and infarction models may be due to different mechanisms resulting from liver I/R.
The assessment of liver weight revealed no differences between the myocardial I/R and infarction groups (supplemental Table V). This result supports our histological findings and indicates an absence of liver congestion in the myocardial I/R and infarction groups.
Detailed Gene Network Analysis Between the Liver and Heart in Myocardial Ischemia
Several factors can affect the liver, including humoral factors released from the ischemic myocardium, the hemodynamic state, or the autonomic nervous system. We focused on the possibility that humoral factors released from the heart may affect the liver. Cardiac gene expression profiles induced by myocardial ischemia were investigated to identify cardiac genes affecting the liver. To obtain a detailed and comprehensive gene network for the liver and heart, individual data from the liver after 24 hours were integrated with pooled data from the risk area and nonrisk area of the heart. Initially, we divided the heart and liver genes into 3 groups: heart-extracellular, liver-intracellular, and liver-extracellular. To find the network among these induced genes, published results for the interactions of individual genes were integrated with these results by using MetaCore software (GeneGo, St. Joseph, Mich). Direct interactions between individual genes were sought. Genes were excluded according to the following criteria: (1) heart-extracellular, no output signal into liver-intracellular; (2) liver-intracellular, no bidirectional signals; and (3) liver-extracellular, no input signal from liver-intracellular. As expected, the network of these differentially expressed genes involved complex interactions of individual genes; however, representative signaling pathways for MI or I/R injury were identified (Figure 6).
During MI, fibroblast growth factor, osteopontin, and heparin-binding epidermal growth factor-like growth factor (HB-EGF) were upregulated in the heart and may have been systemically secreted. Endothelial growth factor receptor and fibroblast growth factor receptor-1 may play important roles in receiving these signals in the liver. Transcription factors such as p53, myelocytomatosis oncogene, trans-acting transcription factor 1, and octamer-binding transcription factor 1 are important molecules in the regulation of these signaling pathways. Protein C, VEGF-A, and urokinase were expected to be systemically secreted from the liver (supplemental Tables VI through VIII). After infarction, genes involved in inflammation, the coagulation-fibrinolytic system, and angiogenesis showed preferential expression. After I/R, heparin-binding epidermal growth factor was upregulated in the heart and was expected to be systemically secreted. V-erb-a erythroblastic leukemia viral oncogene homolog 4 (avian) may play an important role in receiving these signals in the liver. Transcription factors such as trans-acting transcription factor 1, p53, estrogen receptor-1α, and signal transducer and activator of transcription 5A are potentially important molecules for regulation of these signaling pathways. Protein C, coagulation factor X, ciliary neurotrophic factor, and colony-stimulating factor-1 (macrophage) (CSF-1) were expected to be systemically secreted from the liver. In I/R, angiogenesis-related genes were preferentially upregulated (supplemental Tables IX through XI). On comparison of the expression profiles of the heart and liver, genes expressed at significantly higher levels in the heart than in the liver were designated as He, and those expressed at significantly higher levels in the liver than in the heart were designated as Li. Genes expressed in both the heart and liver were described as He/Li (supplemental Tables VIII and XI). In this analysis, most of the factors that were expected to be secreted from the liver induced by I/R and infarction were liver-specific. Most of these genes were not significantly upregulated in the liver I/R groups.
Serum Osteopontin Concentrations in Mice
Of the infarction-induced, cardiac-secreted factors that were expected to stimulate multiple liver genes, we quantified the serum levels of osteopontin by ELISA. Serum osteopontin concentration was significantly increased in the infarction group compared with the sham group (P=0.0012) after 24 hours (Figure 7A). In addition, the changes in osteopontin gene expression in the infarcted and reperfused heart are shown in Figure 7B.
Signaling Pathway in Primary Hepatocytes Treated With Osteopontin
To determine whether ischemia-induced, cardiac-secreted factors affected hepatic gene expression, we investigated the effects of osteopontin on primary mouse hepatocytes (supplemental Materials and Methods); 979 genes were upregulated and 734 genes were downregulated (P<0.05 and fold change >2.0 determined by class comparison analysis) by osteopontin in primary hepatocytes (GSE14843). The most frequent pathway processes observed among upregulated genes as determined with the use of MetaCore software are shown in supplemental Table XII. Osteopontin upregulated signaling pathways of protein C, angiogenesis, cell adhesion, etc, which were observed in groups 2 and 3 gene clusters in the mouse liver under conditions of myocardial ischemia (Figure 4; supplemental Tables I and II). The role of osteopontin in the postulated gene network connecting the liver and heart in myocardial ischemia is shown in Figure 6. Interestingly, many of the genes included in the postulated gene network were actually activated by osteopontin (P<0.05 or fold change >2.0 by class comparison analysis) in primary hepatocytes. Unexpectedly, osteopontin activated HB-EGF, thrombospondin 1, and fibroblast growth factor, which were released from the ischemic heart (Figure 6; supplemental Table VI) in primary hepatocytes. These results indicated that these proteins were released from the liver and from the heart under conditions of myocardial ischemia through osteopontin, and an autocrine signaling pathway may exist in the liver.
Among the candidate hepatic-secreted factors under conditions of myocardial ischemia (Figure 6A; supplemental Table VIII), we quantified the levels of VEGF-A in the supernatants of primary hepatocytes treated with osteopontin. The concentration of VEGF-A measured by ELISA was significantly increased in the supernatants of primary hepatocytes treated with osteopontin (n=6) compared with the mock group (n=7; P=0.0042; supplemental Figure II). Thus, important factors for tissue remodeling could be released from the liver through humoral factors, such as osteopontin, that are released from the heart under conditions of myocardial ischemia.
We performed a quantitative RTD-PCR with TaqMan probes. In the I/R group, protein C, coagulation factor X, CNTF, and CSF-1 were upregulated in the liver. In the infarction group, protein C, urokinase, and VEGF-A were upregulated in the liver (supplemental Figure IIIA). In the hepatocytes treated with osteopontin, protein C, coagulation factor X, ciliary neurotrophic factor, CSF-1, urokinase, and VEGF-A were upregulated compared with the mock group (supplemental Figure IIIB). These results were consistent with those of cDNA microarray analyses performed in this study.
The liver is an essential organ that synthesizes many bioactive proteins, including acute-phase inflammatory proteins (eg, C-reactive protein and IL-6) and coagulation factors. Therefore, it has been speculated that the liver may be involved in systemic reactions that modify the pathophysiology of ACS, although this possibility has not been addressed in detail.
In this study, we examined the gene expression profiles of the livers of mice affected by myocardial I/R or infarction. Marked changes in hepatic gene expression were observed after 24 hours, despite the lack of histological changes in the liver. These changes were essentially restored to normal after 3 to 7 days (data not shown). These findings may not be due to hemodynamic changes during myocardial I/R or infarction. Instead, inflammatory mediators or humoral factors released from the affected heart may be responsible for the observed alterations in hepatic gene expression. This was further confirmed by investigation of signaling pathways in primary hepatocytes induced by osteopontin, a candidate humoral factor released from the ischemic myocardium in vitro.
To exclude the possibility that these changes in gene expression were due to systemic hypotension during I/R or infarction, we performed an additional experiment involving liver I/R to examine whether a pattern of gene expression similar to that in the myocardial I/R and infarction groups could be observed in the liver. Hepatic gene expression in the liver I/R group was completely different from those in the myocardial I/R and infarction groups, with the exception of a small gene cluster (group 3). Although the group 3 gene cluster was upregulated in both the liver I/R and myocardial I/R groups at 24 hours after provocation, peak expression was delayed in the myocardial I/R group compared with the liver I/R group. A recent report of extended observations of cytokine expression in murine hepatic I/R injury indicated that the levels of expression of tumor necrosis factor-α, IL-1β, and IL-6 peaked within 4 hours and returned to baseline at 24 hours.12 In contrast, in the myocardial I/R and infarction models, these cytokines peaked ≈24 to 48 hours and decreased at 7 days.13 These findings were consistent with those of this study (data not shown). Therefore, the delayed peak of hepatic gene expression observed in this study may be correlated with the extent of inflammation in the myocardium after destruction of myocytes, rather than changes in the hemodynamic state of the liver. The lack of histological changes in the liver in the myocardial I/R and infarction models supported these suggestions, although the influence of hemodynamic state on hepatic gene expression should be carefully considered.
Interestingly, genes related to tissue remodeling, adhesion molecules, and morphogenesis were significantly upregulated in the livers of mice that were subjected to I/R or infarction. This may be related to the induction of tissue repair factors such as angiogenic or myocardiogenic factors in the heart undergoing I/R or infarction. In support of this notion, in addition to the genes upregulated during infarction, chemokines and hormonal factors, including IL-8, androgen, and estrogen receptor genes, were upregulated during I/R. These findings may reflect the presence of living myocytes and the greater release of tissue repair and bioreactive factors during I/R than during infarction.
A recent study that included a sequential analysis of ischemic mouse heart with quantitative RT-PCR demonstrated expression of IL-1β, IL-6, monocyte chemoattractant protein-1, macrophage inflammatory protein-1, and granulocyte-CSF at 6 and 24 hours.13 These results were essentially consistent with those of our microarray analysis of pooled RNA extracted from heart specimens (data not shown).
In this study, the hepatic RNA samples were not pooled but were used to analyze the hepatic gene expression profiles individually. This strategy was successful, in that our microarray results were consistent with those produced from pooled or nonpooled liver specimens. Moreover, it facilitated the statistical evaluation of differentially expressed genes among the various groups and revealed dynamic changes in hepatic gene expression through clustering analysis.
We analyzed the network connecting the heart-extracellular genes and liver-intracellular genes induced after I/R injury or infarction by using expression data from pooled heart samples and averaged the expression data for individual liver samples. The results suggested that factors secreted from the heart altered gene expression in the liver. By detailed analysis of signaling pathways, we identified 9 candidate genes (eg, Osteopontin, HB-EGF, Reticulon 4) that were upregulated in the heart and were expected to be systemically secreted and to regulate gene expression in the liver (Figure 6). Moreover, we identified the factors that were expected to be secreted from the liver induced by these signaling pathways, such as protein C, coagulation factor X, CNTF, CSF-1, and angiogenesis-related genes. These factors were expected to be systemically secreted from the liver and to modulate the pathophysiology and outcome of ACS. It has been reported that protein C prevents myocardial I/R injury,14 VEGF enhances capillary density and improves cardiac function,15 and urokinase is essential for cardiac functional recovery after acute myocardial infarction.16
Of the factors that were expected to be secreted from the heart, we confirmed that infarction increased the serum osteopontin concentration after 24 hours. Osteopontin is essential for the development of myocytes, tissue repair, and angiogenesis, and its downstream products, eg, polo-like kinase, were upregulated in the liver. To confirm these findings, we examined the signaling pathways in primary hepatocytes treated with osteopontin. Osteopontin activated signaling pathways of protein C, angiogenesis, and cell adhesion (supplemental Table XII) by inducing the expression of protein C, urokinase, VEGF-A, CSF-1, factor X, and ciliary neurotrophic factor (CNTF) in primary hepatocytes, which was confirmed by RTD-PCR or ELISA (supplemental Figures II and IIIB). Moreover, many other genes involved in the postulated gene network associating the liver and heart (Figure 6A and 6B) were actually activated in primary hepatocytes treated with osteopontin, confirming this signaling pathway. These results suggest that humoral factors play important roles in signal transduction from the ischemic myocardium to the liver.
Although our results addressed humoral factors from the heart that may affect hepatic gene expression, the effects of other factors, such as autonomic nerves, should also be considered. Because the liver has rich sympathetic and parasympathetic innervation,17–19 it is possible that sympathetic hyperactivity affects hepatic gene expression. Although hydralazine has been reported to activate sympathetic nerves,20,21 we observed no differences in gene expression in the hydralazine-treated group compared with the sham-operated group (data not shown). Therefore, autonomic nerves seemed to have little effect on hepatic gene expression determined in this study.
In conclusion, we reported new insights into the pathophysiology of ACS, which may facilitate identification of the mechanisms by which an acute coronary event causes systemic reactions. Further studies are needed to determine whether early therapeutic targeting of the liver during an acute coronary event has any beneficial effect on the clinical outcome in these patients.
Although we confirmed that the serum osteopontin concentration was increased during myocardial ischemia, other proteins that could potentially be secreted from the heart and liver were not assayed. Further studies are needed to determine whether these proteins, including osteopontin, actually affect hepatic gene expression as observed in this study.
We thank Dr Yoh Zen (Department of Human Pathology, Kanazawa University Graduate School of Medical Science, Kanazawa, Japan) for consultation on the pathology of the liver.
Tousoulis D, Antoniades C, Bosinakou E, Kotsopoulou M, Tsoufis C, Marinou K, Charakida M, Stefanadi E, Vavuranakis M, Latsios G, Stefanadis C. Differences in inflammatory and thrombotic markers between unstable angina and acute myocardial infarction. Int J Cardiol. 2007; 115: 203–207.
Busch G, Seitz I, Steppich B, Hess S, Eckl R, Schömig A, Ott I. Coagulation factor Xa stimulates interleukin-8 release in endothelial cells and mononuclear leukocytes: implications in acute myocardial infarction. Arterioscler Thromb Vasc Biol. 2005; 25: 461–466.
van't Veer LJ, Dai H, van de Vijver MJ, He YD, Hart AA, Mao M, Peterse HL, van der Kooy K, Marton MJ, Witteveen AT, Schreiber GJ, Kerkhoven RM, Roberts C, Linsley PS, Bernards R, Friend SH. Gene expression profiling predicts clinical outcome of breast cancer. Nature. 2002; 415: 530–536.
van de Vijver MJ, He YD, van't Veer LJ, Dai H, Hart AA, Voskuil DW, Schreiber GJ, Peterse JL, Roberts C, Marton MJ, Parrish M, Atsma D, Witteveen A, Glas A, Delahaye L, van der Velde T, Bartelink H, Rodenhuis S, Rutgers ET, Friend SH, Bernards R. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med. 2002; 347: 1999–2009.
Hedenfalk I, Duggan D, Chen Y, Radmacher M, Bittner M, Simon R, Meltzer P, Gusterson B, Esteller M, Kallioniemi OP, Wilfond B, Borg A, Trent J, Raffeld M, Yakhini Z, Ben-Dor A, Dougherty E, Kononen J, Bubendorf L, Fehrle W, Pittaluga S, Gruvberger S, Loman N, Johannsson O, Olsson H, Sauter G. Gene-expression profiles in hereditary breast cancer. N Engl J Med. 2001; 344: 539–548.
Gabrielsen A, Lawler PR, Yongzhong W, Steinbrüchel D, Blagoja D, Paulsson-Berne G, Kastrup J, Hansson GK. Gene expression signals involved in ischemic injury, extracellular matrix composition and fibrosis defined by global mRNA profiling of the human left ventricular myocardium. J Mol Cell Cardiol. 2007; 42: 870–883.
Giesen PL, Peltenburg HG, de Zwaan C, Janson PC, Flendrig JG, Hermens WT. Greater than expected alanine aminotransferase activities in plasma and in hearts of patients with acute myocardial infarction. Clin Chem. 1989; 35: 279–283.
Vandervelde S, van Luyn MJ, Rozenbaum MH, Petersen AH, Tio RA, Harmsen MC. Stem cell-related cardiac gene expression early after murine myocardial infarction. Cardiovasc Res. 2007; 73: 783–793.
Loubele ST, Spek CA, Leenders P, van Oerle R, Aberson HL, Hamulyák K, Ferrell G, Esmon CT, Spronk HM, ten Cate H. Activated protein C protects against myocardial ischemia/reperfusion injury via inhibition of apoptosis and inflammation. Arterioscler Thromb Vasc Biol. 2009; 29: 1087–1092.
Zhang J, Ding L, Zhao Y, Sun W, Chen B, Lin H, Wang X, Zhang L, Xu B, Dai J. Collagen-targeting vascular endothelial growth factor improves cardiac performance after myocardial infarction. Circulation. 2009; 119: 1776–1784.
Heymans S, Luttun A, Nuyens D, Theilmeier G, Creemers E, Moons L, Dyspersin GD, Cleutjens JP, Shipley M, Angellilo A, Levi M, Nübe O, Baker A, Keshet E, Lupu F, Herbert JM, Smits JF, Shapiro SD, Baes M, Borgers M, Collen D, Daemen MJ, Carmeliet P. Inhibition of plasminogen activators or matrix metalloproteinases prevents cardiac rupture but impairs therapeutic angiogenesis and causes cardiac failure. Nat Med. 1999; 5: 1135–1142.
Johansson M, Elam M, Rundqvist B, Eisenhofer G, Herlitz H, Jensen G, Friberg P. Differentiated response of the sympathetic nervous system to angiotensin-converting enzyme inhibition in hypertension. Hypertension. 2000; 36: 543–548.
Acute coronary syndrome (ACS) is accompanied by systemic changes in inflammation, coagulation, and metabolism, which may affect the outcome and prognosis of ACS. These systemic reactions are not explained by cardiac events alone. Several lines of evidence suggest that patients with fatty liver disease have a high risk of developing cardiovascular diseases, and it is possible to speculate that the liver is involved in a systemic reaction that modifies the pathogenesis of ACS. However, the relation between liver and myocardial ischemia in the acute ischemic phase has not been elucidated so far. In this investigation, we simultaneously analyzed the gene expression profiles of the liver and heart during acute myocardial ischemia in mice and observed the presence of humoral factors that intervened between the heart and liver. These humoral factors were released from the heart and influenced the liver to secrete important tissue remodeling factors. One of these humoral factors, osteopontin, a widely expressed glycoprotein, was increased in the ischemic heart and altered the gene expression of hepatocytes to produce important tissue remodeling factors (such as vascular endothelial growth factor-A). Our observations suggest that hepatic gene expression is potentially regulated by humoral factors of cardiac origin provoked by myocardial ischemia, and we provide direct evidence that the liver is involved in a systemic reaction that accompanies ACS. Our findings provide potential new insights into the pathophysiology of ACS.
The online-only Data Supplement is available at http://circgenetics.ahajournals.org/cgi/content/full/CIRCGENETICS.108.795484.