help button home button
AJRCCM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published ahead of print on March 4, 2004, doi:10.1164/rccm.200211-1278OC
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow All Versions of this Article:
200211-1278OCv1
169/10/1135    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yu, S.-L.
Right arrow Articles by Lin, F.-Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yu, S.-L.
Right arrow Articles by Lin, F.-Y.
American Journal of Respiratory and Critical Care Medicine Vol 169. pp. 1135-1143, (2004)
© 2004 American Thoracic Society

Original Article

Differential Gene Expression in Gram-negative and Gram-positive Sepsis

Sung-Liang Yu, Huei-Wen Chen, Pan-Chyr Yang, Konan Peck, Min-Hui Tsai, Jeremy J. W. Chen and Fang-Yue Lin

Department of Surgery, Department of Medical Research, and Department of Internal Medicine, National Taiwan University Hospital and National Taiwan University College of Medicine; Institute of Biomedical Sciences, Academia Sinica, Taipei; and Institute of Biomedical Sciences, and Molecular Biology, National Chung Hsing University, Taichung, Taiwan

Correspondence and requests for reprints should be addressed to Fang-Yue Lin, M.D., Ph.D., Department of Surgery, National Taiwan University Hospital, No. 7, Chung Shan South Road, Taipei, 100, Taiwan, Republic of China. E-mail: fylin1{at}ha.mc.ntu.edu.tw


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Sepsis is the most common cause of death in patients in the intensive care unit. Genome-wide gene expression analysis can provide insights into the molecular alterations of sepsis. Total mRNA was extracted from the livers of 6 uninfected control mice and 60 septic mice after infusion of either live Escherichia coli or Staphylococcus aureus. Using a murine complementary DNA microarray system, changes in gene expression were monitored at six time points (uninfected, 2, 8, 24, 48, and 72 hours). Overall, 4.8% of 6,144 assessed genes were differentially regulated with a greater than twofold change across all time points. Most of the genes with altered expression were commonly present in gram-negative and gram-positive sepsis, but the expression levels of 17 genes were different between both types of sepsis at particular time points after infection. The microarray results support the hypothesis that both gram-positive and gram-negative sepsis share a final common pathway involved in the pathogenesis of sepsis, but certain genes are differentially expressed under distinct regulation. These results may provide insights into the pathogenesis of sepsis and may also help identify some altered genes that can serve as new targets for diagnostic tools and therapeutic strategies.

Key Words: bacterial infection • liver • microarray

Sepsis is the most common cause of death in patients in the intensive care unit, and often leads to the development of multiple organ dysfunction syndrome (MODS). Although there have been many attempts to prevent the development of MODS, mortality estimates for sepsis are still consistently in the range of 40–60% (1). In the modern intensive care unit, gram-positive bacteria account for up to 50% of severe sepsis, yet the pathogenesis of gram-positive sepsis is more poorly understood than that of gram-negative sepsis (2). There is an increasing amount of experimental evidence showing that fundamental differences exist in the host response to gram-positive bacteria compared with the response to gram-negative bacteria. It is generally thought that one of the fundamental differences between gram-positive and gram-negative bacteria is the way in which they initiate disease (3).

In addition, gram-positive bacteria require a highly orchestrated host response, with intracellular killing by neutrophils and macrophages. This is often not the case with gram-negative bacteria, which may be readily killed in the extracellular space by antibodies and complement factors (4). However, despite the different structural and toxicologic profiles, gram-positive and gram-negative bacteria cause a similar pattern of shock and MODS in the host (2). Tissues and organs damaged by gram-positive sepsis do not differ clinically from those affected by gram-negative sepsis (2). This indicates that both gram-positive and gram-negative sepsis may share a final common pathway that can precipitate septic shock and MODS.

Previous studies indicated that the liver is a major organ responsible for the initiation of MODS during sepsis, as it plays a central role in metabolism and in host defense mechanisms (5, 6). Kupffer cells are responsible for bacterial scavenging, bacterial product inactivation, and inflammatory mediator clearance and production. Hepatocytes, via receptors for many proinflammatory cytokines, modify their metabolic pathway toward gluconeogenesis, amino acid uptake, and increased synthesis of coagulant and complement factors and protease inhibitors. In general, the liver participates in host defense and tissue repair through hepatic cell cross-talk that controls most of the coagulation and inflammatory processes. When this control is not adequate, a secondary hepatic dysfunction may occur that sometimes leads to MODS and death (7, 8).

The complicated interactions among hepatic cells and the global changes in gene expression in the liver during sepsis are still not completely understood. Understanding of the biological and molecular mechanisms underlying sepsis induced by gram-positive and gram-negative organisms may facilitate the development of new diagnostic and therapeutic strategies in sepsis. To achieve this goal, identification and characterization of genes differentially expressed in the liver during sepsis are clearly warranted. The systematic study of gene expression patterns using complementary DNA (cDNA) microarrays may provide a powerful approach to the molecular dissection of cells and tissues by comparing expression levels of thousands of genes simultaneously (9–11). In this study, we analyzed first-time alterations in gene expression patterns at various time points within mouse liver after sepsis induction, using a murine cDNA microarray (6,144 genes, including known regulatory genes and mouse expressed sequence tags) with a colorimetric detection system (12).


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Experimental Animals
The animal experiments reported in this study were performed with strict adherence to the guidelines established in the Guide for the Care and Use of Laboratory Animals, as adopted and promulgated by the National Health Research Institutes (Taipei, Taiwan). Escherichia coli (ATCC 25922) and Staphylococcus aureus (ATCC 25923) were obtained from the American Type Culture Collection (Manassas, VA) and subcultured once in brain–heart infusion broth. Overnight-cultured E. coli and S. aureus, both of which are clinical isolates and serve as quality control strains for diagnostic kits and susceptibility tests, were washed twice in physiological saline by centrifugation at 3,000 rpm at 4°C. Eighty 8-week-old BALB/c mice divided into 10 groups were used to determine the 25% lethal dose (LD25) of E. coli and S. aureus 72 hours after infection. Sixty mice were intravenously dosed via the caudal vein with the LD25 of live E. coli or S. aureus. After 2, 8, 24, 48, and 72 hours, groups of 6 uninfected control mice and 60 infected mice were killed by cervical dislocation and the livers were isolated for mRNA [poly(A) RNA] extraction and subsequent immunohistochemical analysis. To characterize this sepsis model, serum glutamic–oxaloacetic transaminase and glutamic–pyruvic transaminase levels were used to evaluate time course changes in hepatic functions. Serum glutamic–oxalacetic transaminase and glutamic–pyruvic transaminase were measured with an automatic analyzer (7250 type; Hitachi, Tokyo, Japan).

Microarray System
Mouse expressed sequence tag clones were obtained from the IMAGE consortium libraries through its distributor (ResGen Invitrogen, Huntsville, AL) (13). The cDNA microarray carrying 6,144 polymerase chain reaction-amplified cDNA fragments was prepared with an arraying machine (Wittech, Taipei, Taiwan) built in-house. Potential interindividual variability was minimized by pooling mRNA samples from two mice within each group to yield a representative sample for analysis. mRNAs were extracted from the livers of each experimental group, using an mRNA isolation kit (Qiagen, Hilden, Germany), in accordance with the manufacturer's protocol. Five micrograms of mRNA was labeled with biotin during reverse transcription. All hybridization experiments were performed in triplicate with cDNA probes prepared from distinct cohorts. The microarray images were scanned, digitized, and analyzed with a flatbed scanner (PowerLook 3000; UMAX, Taipei, Taiwan) and GenePix 3.0 software (Axon Instruments, Union City, CA). The coefficient of variation is 7.9% (see online supplement for additional details on the microarray system).

Northern Hybridization and Immunohistochemical Staining
To confirm the results derived from microarrays, nine of the differentially expressed clones were randomly selected from a cluster analysis of the microarray data, and all the inserts of the clones were individually amplified by polymerase chain reaction to serve as probes for the Northern blot. The amplified cDNA fragments were then labeled with digoxigenin-11-dUTP by random primed labeling as in our previous report (12). To correct for RNA loading, the signals were normalized with respect to glyceraldehyde-3-phosphate dehydrogenase in the same blot. Immunohistochemistry was performed similarly, as in previous reports (14). The slides were incubated with the primary antibody, diluted 1:100 for CCAAT/enhancer-binding protein (C/EBP) {alpha}, C/EBP ß, inhibitor of NF-{kappa}B {alpha} chain (I-{kappa}B {alpha} chain), and adaptin C (Santa Cruz Biotechnology, Santa Cruz, CA). After this, the slides were incubated with the biotinylated secondary antibody and peroxidase-labeled streptavidin (ABC kit; Vector Laboratories, Burlingame, CA). Negative control slides, prepared in the absence of primary antibody, were included for each staining procedure. Finally, 3,3'-diamino-benzidine was used to develop the signals (brown color), and methyl green was used for counterstaining.

Apoptosis Assay Using Terminal Deoxynucleotidyltransferase-mediated dUTP Nick-End Labeling
Apoptotic cells were detected with an ApopTag in situ apoptosis detection kit-peroxidase (Roche Diagnostics, Mannheim, Germany). The specimens were processed according to the protocol provided by the manufacturer. The cells were counterstained with methyl green. Cells that were intact and exhibited dark brown-stained nuclei were considered positive for apoptosis (15). The number of positively stained cells was measured under five randomly selected high-power fields (magnification, x400) per slide, and the significance was examined by Student t test.

Statistical Analysis
To attempt to reduce the variation arising from the experimental results derived from different microarrays, the intensity values of spots from each microarray were rescaled, using the Spotfire Pro version 7.1 software program (trimmed method, trimmed value = 5%; Spotfire, Cambridge, MA). Only the signal values of spots exceeding a figure of 3,000 were meaningful in the membrane format microarray system (i.e., the intensity of a spot less than 3,000 was hardly distinguishable from background "noise," such that when the spot intensity value lay below 3,000, the value was replaced by a figure of 3,000). To focus on genes that were significantly affected by study conditions, genes with a greater than twofold change across all study-included time points were selected and normalized by mean center, setting the mean value to be equal to zero and the magnitude to be equal to one (sum of the squares of the values), through the use of the Cluster software package (Stanford University, Stanford, CA; and Massachusetts Institute of Technology, Cambridge, MA) (16). Those selected genes with a greater than twofold change were clustered into 16 groups, using the self-organization map algorithm of the software package GeneCluster 1.1 (Stanford University and Massachusetts Institute of Technology) (17). The resultant pattern, featuring either an ascending or a descending trend as a response to bacterial infection, was identified and the selected genes were clustered by means of a hierarchical cluster method (Spotfire Pro, version 7.1). Annotation of the selected genes was determined by means of the gene annotation program GeneSpring (Silicon Genetics, Redwood City, CA), and revised on the basis of their putative functions as found in the literature. To specify differentially expressed genes between the two types of sepsis, a Student t test and analysis of variance (Spotfire Pro, version 7.1) were used to analyze the difference in gene expression levels between the two types of sepsis, considering time. Significance was defined as p < 0.05.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
E. coli and S. aureus, which are the dominant pathogenic bacteria in gram-negative and gram-positive sepsis in surgical intensive care units in the United States (18) and Taiwan, were used to establish the murine sepsis model. The intravenous LD25 (expressed as colony-forming units) 72 hours after administration of was determined to be 5 x 107 and 6 x 106 for E. coli and S. aureus, respectively (see Figure E1 in the online supplement). The induction of five acute-phase proteins and the reduction of three negative acute-phase proteins were similar between the two infections (see Figure E2 in the online supplement). This implies that the doses (LD25) of E. coli and S. aureus used in this study not only induced the same mortality but also produced an equivalent inflammatory response in mice. Serum glutamic-oxalacetic transaminase and glutamic-pyruvic transaminase levels immediately increased 2 hours after both pathogen injections, and remained elevated throughout the study period (see Figure E3A in the online supplement). The amount of infiltrated leukocytes in the hepatic parenchyma significantly increased in mice with sepsis induced by both bacteria (see Figure E3B in the online supplement).

To identify hepatic mediators involved in the progression of gram-negative and gram-positive sepsis, mice were divided into six time-point groups (uninfected controls, and 2, 8, 24, 48, and 72 hours after bacterial challenge) and cDNA microarrays were used to simultaneously assess the expression profiles of 6,144 mouse genes in whole liver. Microarrays were performed in triplicate, using cDNA probes derived from different cohorts. To focus on genes that were significantly regulated in sepsis, genes with a greater than twofold change across all time points were filtered for further analysis.

After self-organization map clustering, the resultant genes were grouped into 16 clusters. Surprisingly, the great majority of affected genes had similar profiles in both types of sepsis. Two clusters of expression profiles with an ascending trend containing 63 and 103 genes, and two with a descending trend containing 56 and 74 genes, were selected from these 16 clusters (see Figures 1 and 2 , top). Overall, we noted an altered expression of 4.8% of analyzed genes during sepsis progression, and we validated their sequences individually. All the gene expression profiles from these clusters were rearranged by hierarchical cluster analysis, using the average linkage method. The expression levels are depicted in Figures 1 and 2 (bottom). To further characterize these genes, they were grouped into four groups on the basis of their cellular functions; this is summarized in Tables E1 and E2 (see the online supplement). The first group included genes involved in inflammation, cell defense, and stress response. The second group included cell structural proteins and genes encoding proteins involved in motility, nutrition transportation, and vesicle trafficking. The third group included cell surface receptors, growth factors, transcription factors, and genes encoding proteins involved in signal transduction. The fourth group included metabolic enzymes and genes encoding proteins involved in protein synthesis, protein modification, and protein turnover. Aside from genes listed in Tables E1 and E2 (see the online supplement), we identified 22 expressed sequence tags that were upregulated, and 26 expressed sequence tags that were downregulated, in both types of sepsis; this is summarized in Tables E3 and E4 (see the online supplement).



View larger version (69K):
[in this window]
[in a new window]
  		Figure 1. Dynamic gene expression analysis of septic mouse liver. Self-organization map clustering analysis of gene expression profiles with increased expression (top left, 63 genes; top right, 103 genes) was selected from 16 clusters (4 x 4 nodes). Relative expression levels of these genes are color coded (bottom left and bottom right).

 


View larger version (63K):
[in this window]
[in a new window]
  		Figure 2. Dynamic gene expression analysis of septic mouse liver. Self-organization map clustering analysis of gene expression profiles with decreased expression (top left, 56 genes; top right, 74 genes) and relative expression levels of these genes are shown (bottom left and bottom right).

 
To substantiate the results of the microarrays, we performed Northern blots for the nine genes representing the functional categories listed in Tables E1 and E2 (see the online supplement). Five upregulated genes (I-{kappa}B, TYRO-binding protein, fibrinogen-{gamma} A chain, H-2 Class II ß chain, and proteasome 26S subunit) and four downregulated genes (endothelin-converting enzyme-1, Ras-related protein-6, glutathione-S-transferase [GST] {theta}2, and NADH-ubiquinone oxidoreductase ASHI subunit) were randomly selected from each of four functional groups. To reduce the complexity of the Northern blots, only the samples of time points with the greatest differences compared with the uninfected controls were chosen. These are shown in Figure 3A . The expression of I-{kappa}B {alpha} chain increased 5.8- and 3.4-fold 2 hours after E. coli and S. aureus challenge, respectively, compared with that of uninfected controls, and rapidly returned to a normal level 24 hours later. We found that maximally increased expression of TYRO-binding protein presented at 48 hours, and maximally increased expression of fibrinogen-{gamma} A chain, H-2 Class II ß chain, and proteasome 26S subunit presented at 72 hours. The expression levels of endothelin-converting enzyme-1, Ras-related protein-6, GST {theta}2, and NADH-ubiquinone oxidoreductase ASHI subunit decreased throughout the study period. Despite some minor differences in the magnitude of change, the results of Northern blots were consistent with the microarray data. Interestingly, we found 17 genes that showed differential expression profiles between E. coli– and S. aureus–induced sepsis at some particular time points (p < 0.05); the Northern blots of five exemplary genes are shown in Figure 3B. However, the expression levels of four selected genes were not substantiated by Northern blots, and therefore the false discovery rate was presumably 22%.



View larger version (59K):
[in this window]
[in a new window]
  		Figure 3. Northern blot analysis of altered genes in septic mouse liver. Five upregulated and four downregulated genes in both gram-negative and gram-positive sepsis are shown (A). Differential expression levels of five genes of both gram-negative and gram-positive sepsis are shown (B). To ensure relatively equal RNA loading, the blot was rehybridized with a glyceraldehyde-3-phosphate dehydrogenase (GAPDH) probe, and the intensity of the tested gene was normalized with the intensity of the GAPDH and expressed as a ratio. The sequences of these genes have been verified. Similar results were obtained in three independent experiments. Apo IVA = apolipoprotein A-IV; ECE-1 = endothelin-converting enzyme-1; GBP-2 = guanylate-binding protein-2; GST {theta}2 = glutathione-S-transferase {theta}2; I{kappa}B{alpha} = inhibitor of NF-{kappa}B {alpha} chain; MCT1 = monocarboxylate transporter 1; MHC class IIß = major histocompatibility complex class IIß; NQO = NADH-ubiquinone oxidoreductase ASHI subunit; RAB-6 = Ras-related protein-6; TYRO BP = TYRO-binding protein; SLC3 = solute carrier family 3.

 
Differential expression levels of monocarboxylate transporter-1 and Ras-related protein-4A were evident at 8 hours, and of apolipoprotein A-IV, solute carrier family 3, and guanylate-binding protein-2 (GBP-2) at 48 hours, in sepsis induced by E. coli and S. aureus. The dynamic expression profiles of these 17 genes are shown in Figure 4 , and the individual expression plots for these genes across all study-included time points are shown in Figure E4 (see the online supplement). To further characterize sepsis-affected genes that were expressed in specific cells, four antibodies (specifically directed against C/EBP {alpha}, C/EBP ß, I-{kappa}B {alpha} chain, and adaptin C) were used to carry out immunohistochemical analysis of test animal livers (Figure 5) . The protein expression level of C/EBP ß appeared to increase from 2 hours subsequent to E. coli and S. aureus infection, with the maximal induction being present 8 hours later. The highest protein expression level of I-{kappa}B was present 72 and 8 hours later for E. coli and S. aureus infection, respectively. On the other hand, protein expression levels of C/EBP {alpha} and adaptin C decreased, and maximal suppression was present between 24 and 48 hours after both bacterial injections. For all these cases, hepatocytes were more immunoreactive than endothelial cells and leukocytes. In addition, the relative magnitude of the apoptotic process in the two types of sepsis was explored in the liver biopsy specimens, using a terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling assay. In both experimental groups, the number of apoptotic cells increased, reaching maximal levels 24 hours postinjection (Figure 6) . At this time point, the number of apoptotic cells from E. coli–infected specimens was 37.6 ± 6.9 versus 4.0 ± 0.5 in the control group (p < 0.01, n = 5), and that from S. aureus–infected specimens was 10.2 ± 1.5 versus 4.0 ± 0.5 in the control group (p < 0.02, n = 5). Regardless of infection time, E. coli infection elicited greater evidence of apoptosis than infection with S. aureus. Moreover, we found that the major cell type among the apoptotic cells was the Kupffer cell, rather than the hepatocyte.



View larger version (39K):
[in this window]
[in a new window]
  		Figure 4. Differential expression levels of 17 genes in gram-negative sepsis as compared with gram-positive sepsis. The differentially expressed genes of the two types of sepsis were specified by t test/analysis of variance (Spotfire Pro, version 7.1), considering time. Significance was defined as p < 0.05.

 


View larger version (137K):
[in this window]
[in a new window]
  		Figure 5. Immunohistochemical staining using four antibodies against CCAAT/enhancer binding protein {alpha} (C/EBP {alpha}), C/EBP ß, I-{kappa}B {alpha} chain, and adaptin C. Methyl green was used for counterstaining. Original magnification, x400; scale bar, 60 µm. Arrows indicate strong immunoreactivity of the proteins. Similar results were obtained in at least three independent experiments.

 


View larger version (59K):
[in this window]
[in a new window]
  		Figure 6. Apoptosis assay of liver during sepsis progression. The terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling assay was used to identify apoptotic cells and methyl green was used for counterstaining. Apoptotic cells are indicated by arrows (top; scale bar, 60 µm), and were counted under a high-power field (bottom; original magnification, x400). Data are expressed as means ± SEM (n = 5). Statistical significance was determined on the basis of comparison with apoptotic cells from normal controls (*p < 0.01 for Escherichia coli–injected groups; {dagger}p < 0.02 for S. aureus–injected groups).

 

   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
A clear understanding of the complex nature of sepsis is critical to the development of rational strategies to manipulate this disease. The LD25 of E. coli and S. aureus used in this study not only induced the same mortality but also produced an equivalent inflammatory response in the mouse sepsis model. In reviewing the microarray data, we found that the great majority of affected genes presented similar profiles in both types of sepsis. In all, 4.8% of the genes monitored were either positively or negatively regulated to a similar extent in both cases, and they were divided into four functional categories (see the online supplement for an extended discussion).

Immunity
The combination of an induction of immunity-related genes (such as MHC molecules, B lymphocyte chemoattractant, complement component-2, CD36, CD39, CD82, complement-reactive protein-ductin, and recombination-activating gene-1) and a concerted reduction of antiinflammatory enzymes (11ß-hydroxysteroid dehydrogenase [HSD], 17ß-HSD, HSD-1, and HSD-4) enhances the immunity of animals against bacterial infection. The expression of antithrombin III, which is a main inhibitor of the coagulation protease, was dramatically decreased. Inhibition of coagulation protects baboons from sepsis-induced organ failure (19). These data suggest that the liver quickly reduces the expression of antithrombin III, thereby facilitating fibrin deposition and aggravating hepatic injury.

Antioxidation and Detoxification
Increasing oxidative stress is of prime importance to an organism's defense, but may have deleterious effects on tissue if it is not tightly controlled by antioxidants (20). Downregulation of the copper chaperone of superoxide dismutase, and of GST {pi}, {theta}, {theta}2, and {alpha}, might be the cause of increasing oxidative stress during sepsis. However, decreasing expression levels of these genes were accompanied by the induction of vanin, SAG (sensitive to apoptosis gene), copper transport protein (ATOX-1), and glutathione peroxidase, which are important in protecting cells from oxidative injury through diverse mechanisms (21–23). This implies that there is compensatory machinery protecting tissue from the overflowing production of oxidative pressure. Furthermore, we found that expression of 13 cytochrome P450-related genes was suppressed at a variety of levels. The cytochrome P450 enzyme system has been hypothesized as an endogenous protectant in zymosan-induced MOF (24). The progression of sepsis was also accompanied by a decrease in the expression of amine UDP-glucuronosyltransferase 2B5 and N-sulfotransferase. These two enzymes are known to have important functions in the detoxification of diverse drugs in the liver (25, 26). Thus, our data suggest that impaired detoxifying activity of the liver might be a causative factor in the pathogenesis of sepsis in the liver.

Vesicle Trafficking and Transporter
The alteration of genes involved in protein-sorting pathways, to the best of our knowledge, has never been mentioned regarding sepsis. The expression levels of vesicle trafficking-related genes were quite different: seven genes (e.g., adaptor protein-1) decreased, whereas another four genes (e.g., adaptor protein-3) increased. The expression levels of seven protein transport-related genes (SEC61{alpha}, SEC61ß, and others) increased. Taken together, the turbulent expression levels of protein sorting-related genes imply the existence of a complicated network in the secretory and endocytic pathways during sepsis. However, this remains to be investigated further. We also found that the expression of several transporter genes was altered within the period of sepsis. This was true especially of hemopexin and ATOX-1, which protected cells from oxidative injury through the mediation of heme and copper transportation (23, 27). Heme can induce the expression of heme oxygenase, which protects rats against sepsis-induced diaphragmatic dysfunction (28). The progression of sepsis is always accompanied by an alteration in hepatic amino acid transportation and glucose metabolism (29, 30). Two transport genes, SLC7 and SLC2, which encode y+L amino acid transporter-1 and facilitated glucose transporter, increased and decreased during sepsis, respectively. This fact may be responsible, in part, for the altered homeostasis of amino acids and glucose.

Cell Signaling, Cell Cycle, and Apoptosis
Induction of matrix receptor (syndecan-3) and inhibin may facilitate the inflammatory reaction in sepsis (31, 32). However, I-{kappa}B {alpha} chain and cytokine-inducible SH2-containing protein, which mediates the downmodulation of macrophage activity, were induced in the livers of septic mice (33, 34). We suggest that the induction of I-{kappa}B {alpha} chain and cytokine-inducible SH2-containing protein may be a part of an adaptive response limiting the immoderate activation of the immune system in sepsis. In addition, previous studies have shown that proinflammatory cytokines produce a reduction of C/EBP {alpha} and an induction of C/EBP ß in macrophages (35, 36), and that the induction of core-binding factor ß is essential for hematopoiesis (37). The observation of an induction of C/EBP ß and a reduction of C/EBP {alpha} in our study suggests that these two genes may play an important role in response to infections by raising the rate of hematopoiesis.

The expression profiles of cell cycle- and apoptosis-related genes during sepsis are complicated. Notch causes a profound growth arrest and downregulates IgM expression (38, 39). A reduction of Notch2 and a corresponding induction of two negative regulators of Notch (Sel1 and Itch) suggest that the regulation of Notch may be involved in the promotion of cell proliferation and elevation of immunoglobulin expression against bacterial infection. The decrease in a growth suppressor (Necdin) and the increase in two positive cell cycle regulators (Sir2 and CtIP) may push cells toward proliferation. However, increases in growth arrest-specific gene 5 and promyelocytic leukemia protein, and a decrease in the tudor repeat associator with PCTAIRE 2, lead to an antiproliferation response. In general, a stronger induction of proliferating genes in mice infected with E. coli than in mice infected with S. aureus was noticed, suggesting that the hepatocytes of gram-negative septic mice have a more potent ability to move toward a positive cell cycle progression. Ki67 antigen specifically expressed in the livers of E. coli–infected mice supported this suggestion. Expression of most apoptosis-related genes, such as those encoding caspases, Fas ligand, RAIDD (RIP-associated Ich-1/CED homologous protein with death domain), and the Bcl-2 family, was not significantly affected. Only two proapoptotic genes (transforming growth factor-ß-stimulated Clone 22 and death-associated kinase) and two antiapoptotic genes (inhibitor of apoptosis and SAG) were induced in response to bacterial infection. Perhaps the number of apoptotic cells was too few to detect changes in most apoptosis-related genes. In accordance with the microarray data, the terminal deoxynucleotidyltransferase-mediated dUTP nick-end labeling assay revealed that rare apoptotic cells were found in the livers of septic mice, and that the major type of apoptotic cell was the Kupffer cell, not the hepatocyte. Moreover, gram-negative bacteria induced more severe cell apoptosis than did gram-positive bacteria. These data are in agreement with previous observations, that increased apoptosis is not seen in parenchymal cells from the liver in sepsis (40).

We identified 17 genes whose expression levels were different between E. coli– and S. aureus–induced sepsis. GBP-2 is one of the most abundant antiviral proteins induced by IFN-{gamma} (41), and C-type lectins are important in the innate immune system of mammals (42, 43). The specific upregulation of GBP-2 and C-type lectin in E. coli–induced sepsis may play an undefined role in the immune defense of the host against gram-negative and gram-positive pathogens. It has been demonstrated that calpactin is involved in the formation of pedestals between enteropathogenic E. coli and epithelial cells, mediating the adhesion of pathogens with host cells (44). The induction of calpactin may help E. coli to attach to hepatocytes and disturb the calcium influx, resulting in the modified signaling of downstream genes. Adenylate cyclase catalyzes the hydrolysis of ATP to cAMP that activates protein kinase A, leading to modulation of inflammatory reactions. A reduction of adenylate cyclase activity is responsible for the pathogenesis of cardiac dysfunction in sepsis (45). The specific downregulation of adenylate cyclase by S. aureus may account, in part, for the different pathogenesis of both types of sepsis. Finally, although the liver is made up mainly of hepatocytes, Kupffer cells, and sinusoidal endothelial cells, as well as inflammatory cells in sepsis, our microarray analysis did not distinguish which cells caused the change in gene expression, but provided us with some useful hints concerning which genes are worthy of further investigation.

In conclusion, we examined the dynamic gene expression profiles of E. coli–induced sepsis and S. aureus–induced sepsis. There are complex alterations of genes involved in different cellular functions during sepsis progression, and a number of these genes represent novel observations deserving further investigation to elucidate their respective roles in sepsis. Most of the changes observed are generally present in both types of sepsis, but the expression levels of 17 genes are distinguishable between sepsis caused by different pathogens at particular time points after bacterial infection. This information may provide broader insights into the pathogenesis of sepsis and help to identify some altered genes that could serve as new markers for diagnosis and potential targets for therapies in sepsis.


   FOOTNOTES
 
Supported by the National Science Council through grant NSC 90-2314-B002-392 (Taiwan, Republic of China).

S-L.Y. and H-W.C. contributed equally to this work.

This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org

Conflict of Interest Statement: S-L.Y. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; H-W.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; P-C.Y. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; K.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M-H.T. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; J.J.W.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; F-Y.L. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

Received in original form November 5, 2002; accepted in final form February 28, 2004


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Cain BS, Meldrum DR, Harken AH, McIntyre RC. The physiologic basis for anticytokine clinical trails in the treatment of sepsis. J Am Coll Surg 1998;186:337–350.[CrossRef][Medline]
  2. Sriskandan S, Cohen J. Gram-positive sepsis: mechanisms and differences from gram-negative sepsis. Infect Dis Clin North Am 1999;13:397–412.[CrossRef][Medline]
  3. Opal SM, Cohen J. Clinical gram-positive sepsis: does it fundamentally differ from gram-negative bacterial sepsis? Crit Care Med 1999;27:1608–1616.[CrossRef][Medline]
  4. Andersson J, Nagy S, Bjork L, Abrams J, Holm S, Andersson U. Bacterial toxin–induced cytokine production studied at the single-cell level. Immunol Rev 1992;127:69–96.[CrossRef][Medline]
  5. Wang P, Chaudry IH. Mechanism of hepatocellular dysfunction during hyperdynamic sepsis. Am J Physiol 1996;270:R927–R938.
  6. Ring A, Stremmel W. The hepatic microvascular responses to sepsis. Semin Thromb Hemost 2000;26:589–594.[CrossRef][Medline]
  7. Dhainaut J, Marin N, Mignon A, Vinsonneau C. Hepatic response to sepsis: interaction between coagulation and inflammatory processes. Crit Care Med 2001;29:S42–S47.[CrossRef][Medline]
  8. Marshall JC. Inflammation, coagulopathy, and the pathogenesis of multiple organ dysfunction syndrome. Crit Care Med 2001;29:S99–S106.[CrossRef][Medline]
  9. Chen JJ, Wu R, Yang PC, Huang JY, Sher YP, Han MH, Kao WC, Lee PJ, Chiu TF, Chang F, et al. Profiling expression patterns and isolating differentially expressed genes by cDNA microarray system with colorimetry detection. Genomics 1998;51:313–324.[CrossRef][Medline]
  10. Abdellatif M. Leading the way using microarray: a more comprehensive approach for discovery of gene expression patterns. Circulation 2000;86:919–920.
  11. Lockhart DJ, Winzeler EA. Genomics, gene expression and DNA arrays. Nature 2000;405:827–835.[CrossRef][Medline]
  12. Chen JJ, Peck K, Hong TM, Yang SC, Sher YP, Shih JY, Wu R, Cheng JL, Roffler SR, Wu CW, et al. Global analysis of gene expression in invasion by a lung cancer model. Cancer Res 2001;61:5223–5230.[Abstract/Free Full Text]
  13. Lennon G, Auffray C, Polymeropoulos M, Soares MB. The I.M.A.G.E. Consortium: an integrated molecular analysis of genomes and their expression. Genomics 1996;33:151–152.[CrossRef][Medline]
  14. Mutasa HC, Pearson EC. Use of light microscopic immunotechniques in selecting preparation conditions and immunoprobes for ultrastructural immunolabelling of lactoferrin. Histochem J 1988;20:558–566.[CrossRef][Medline]
  15. Negoescu A, Lorimier P, Labat-Moleur F, Drouet C, Robert C, Guillermet C, Brambilla C, Brambilla E. In situ apoptotic cell labelling by the TUNEL method: improvement and evaluation on cell preparation. J Histochem Cytochem 1996;44:959–968.[Abstract]
  16. Iyer VR, Eisen MB, Ross DT, Schuler G, Moore T, Lee JC, Trent JM, Staudt LM, Hudson J Jr, Boguski MS, et al. The transcriptional program in the response of human fibroblasts to serum. Science 1999;75:83–87.
  17. Tamayo P, Slonim D, Mesirov J, Zhu Q, Kitareewan S, Dmitrovsky E, Lander ES, Golub TR. Interpreting patterns of gene expression with self-organizing maps: methods and application to hematopoietic differentiation. Proc Natl Acad Sci USA 1999;96:2907–2912.[Abstract/Free Full Text]
  18. Cohen J, Abraham E. Microbiological findings and correlations with serum tumor necrosis factor-{alpha} in patients with severe sepsis and septic shock. J Infect Dis 1999;180:116–121.[CrossRef][Medline]
  19. Welty-Wolf KE, Carraway MS, Miller DL, Ortel TL, Ezban M, Ghio AJ, Idell S, Piantadosi CA. Coagulation blockade prevents sepsis-induced respiratory and renal failure in baboons. Am J Respir Crit Care Med 2001;164:1988–1996.[Abstract/Free Full Text]
  20. Malmezat T, Breuille D, Capitan P, Mirand PP, Obled C. Glutathione turnover is increased during the acute phase of sepsis in rats. J Nutr 2000;130:1239–1246.[Abstract/Free Full Text]
  21. Pitari G, Malergue F, Martin F, Philippe JM, Massucci MT, Chabret C, Maras B, Dupre S, Naquet P, Galland F. Pantetheinase activity of membrane-bound Vanin-1: lack of free cysteamine in tissues of Vanin-1 deficient mice. FEBS Lett 2000;483:149–154.[CrossRef][Medline]
  22. Swaroop M, Bian J, Aviram M, Duan H, Bisgaier CL, Loo JA, Sun Y. Expression, purification, and biochemical characterization of SAG, a ring finger redox-sensitive protein. Free Radic Biol Med 1999;27:193–202.[CrossRef][Medline]
  23. Kelner GS, Lee M, Clark ME, Maciejewski D, McGrath D, Rabizadeh S, Lyons T, Bredesen D, Jenner P, Maki RA. The copper transport protein Atox1 promotes neuronal survival. J Biol Chem 2000;275:580–584.[Abstract/Free Full Text]
  24. Carcillo JA, Korzekwa KR, Jones GS, Parise RA, Gillespie DG, Whalen MJ, Kochanek PM, Branch RA, Kost CK Jr. The cytochrome P450 suicide inhibitor, 1-aminobenzotriazole, sensitizes rats to zymosan-induced toxicity. Res Commun Mol Pathol Pharmacol 1998;102:57–68.[Medline]
  25. Banoglu E. Current status of the cytosolic sulfotransferases in the metabolic activation of promutagens and procarcinogens. Curr Drug Metab 2000;1:1–30.[Medline]
  26. Mackenzie PI, Mojarrabi B, Meech R, Hansen A. Steroid UDP glucuronosyltransferase: characterization and regulation. J Endocrinol 1996;150:S79–S86.[Abstract]
  27. Sung L, Shibata M, Eskew JD, Shipulina N, Morales PJ, Smith A. Cell-surface events for metallothionein-1 and heme oxygenase-1 regulation by the hemopexin–heme transport system. Antioxid Redox Signal 2000;2:753–765.[Medline]
  28. Taille C, Foresti R, Lanone S, Zedda C, Green C, Aubier M, Motterlini R, Boczkowski J. Protective role of heme oxygenases against endotoxin-induced diaphragmatic dysfunction in rats. Am J Respir Crit Care Med 2001;163:753–761.[Abstract/Free Full Text]
  29. Inoue Y, Bode BP, Souba WW. Hepatic Na+-independent amino acid transport in endotoxemic rats: evidence for selective stimulation of arginine transport. Shock 1994;2:164–172.[Medline]
  30. Baue AE. Multiple organ failure, multiple organ dysfunction, and the systemic inflammatory syndrome: where do we stand? Shock 1996;2:238–297.
  31. Luo J, Kato M, Wang H, Bernfield M, Bischoff J. Heparin sulfate and chondroitin sulfate proteoglycans inhibit E-selectin binding to endothelial cells. J Cell Biol 2001;80:522–531.
  32. Eckmann L, Smith JR, Housley MP, Dwinell MB, Kagnoff MF. Analysis by high-density cDNA arrays of altered gene expression in human intestinal epithelial cells in response to infection with the invasive enteric bacteria Salmonella. J Biol Chem 2000;275:14084–14094.[Abstract/Free Full Text]
  33. Ogle CK, Kong F, Guo X, Wells DA, Aosasa S, Noel G, Horseman N. The effect of burn injury on suppressors of cytokine signalling. Shock 2000;14:392–399.[Medline]
  34. Johnson TS, O'Leary M, Justice SK, Maamra M, Zarkesh-Esfahani SH, Furlanetto R, Preedy VR, Hinds CJ, ElNahas AM, Ross RJ. Differential expression of suppressors of cytokine signalling genes in response to nutrition and growth hormone in the septic rat. J Endocrinol 2001;169:409–415.[Abstract]
  35. Tengku-Muhammad TS, Hughes TR, Ranki H, Cryer A, Ramji DP. Differential regulation of macrophage CCAAT-enhancer binding protein isoforms by lipopolysaccharide and cytokines. Cytokine 2000;12:1430–1436.[CrossRef][Medline]
  36. Begbie M, Notley C, Tinlin S, Sawyer L, Lillicrap D. The Factor VIII acute phase response requires the participation of Nf{kappa}B and C/EBP. Thromb Haemost 2000;84:216–222.[Medline]
  37. Adya N, Castilla LH, Liu PP. Function of CBFß/Bro proteins. Semin Cell Dev Biol 2000;11:361–368.[CrossRef][Medline]
  38. Sriuranpong V, Borges MW, Ravi RK, Arnold DR, Nelkin BD, Baylin SB, Ball DW. Notch signaling induces cell cycle arrest in small cell lung cancer cells. Cancer Res 2001;61:3200–3205.[Abstract/Free Full Text]
  39. Morimura T, Miyatani S, Kitamura D, Goitsuka R. Notch signaling suppresses IgH gene expression in chicken B cells: implication in spatially restricted expression of Serrate2/Notch1 in the bursa of Fabricius. J Immunol 2001;166:3277–3283.[Abstract/Free Full Text]
  40. Oberholzer C, Oberholzer A, Clare-Salzer M, Moldawer LL. Apoptosis in sepsis: a new target for therapeutic exploration. FASEB J 2001;15:879–892.[Abstract/Free Full Text]
  41. Anderson SL, Carton JM, Zhang X, Rubin BY. Genomic organization and chromosomal localization of a new member of the murine interferon-induced guanylate-binding protein family. J Interferon Cytokine Res 1999;19:487–494.[CrossRef][Medline]
  42. Weis WI, Taylor ME, Drickamer K. The C-type lectin superfamily in the immune system. Immunol Rev 1998;163:19–34.[CrossRef][Medline]
  43. Ofek I, Mesika A, Kalina M, Keisari Y, Podschun R, Sahly H, Chang D, McGregor D, Crouch E. Surfactant protein D enhances phagocytosis and killing of unencapsulated phase variants of Klebsiella pneumoniae. Infect Immun 2001;69:24–33.[Abstract/Free Full Text]
  44. Goosney DL, DeVinney R, Finlay BB. Recruitment of cytoskeletal and signaling proteins to enteropathogenic and enterohemorrhagic Escherichia coli pedestals. Infect Immun 2001;69:3315–3322.[Abstract/Free Full Text]
  45. Wu L, Yang S, Yang R, Hsu H, Hsu C, Dong L, Liu M. G protein and adenylate cyclase complex-mediated signal transduction in the rat heart during sepsis. Shock 2003;19:533–537.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Proc. Natl. Acad. Sci. USAHome page
S. Alper, R. Laws, B. Lackford, W. A. Boyd, P. Dunlap, J. H. Freedman, and D. A. Schwartz
Identification of innate immunity genes and pathways using a comparative genomics approach
PNAS, May 13, 2008; 105(19): 7016 - 7021.
[Abstract] [Full Text] [PDF]

Home page
 FASEB J.Home page
H.-W. Chen, S.-F. Su, C.-T. Chien, W.-H. Lin, S.-L. Yu, C.-C. Chou, J. J. W. Chen, and P.-C. Yang
Titanium dioxide nanoparticles induce emphysema-like lung injury in mice
FASEB J, November 1, 2006; 20(13): 2393 - 2395.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
J.-L. Vincent and E. Abraham
The Last 100 Years of Sepsis
Am. J. Respir. Crit. Care Med., February 1, 2006; 173(3): 256 - 263.
[Abstract] [Full Text] [PDF]

Home page
 Hum ReprodHome page
H.-W. Chen, J. J.W. Chen, S.-L. Yu, H.-N. Li, P.-C. Yang, C.-M. Su, H.-K. Au, C.-W. Chang, L.-W. Chien, C.-S. Chen, et al.
Transcriptome analysis in blastocyst hatching by cDNA microarray
Hum. Reprod., September 1, 2005; 20(9): 2492 - 2501.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
D. Angus, A. Ishizaka, M. Matthay, F. Lemaire, W. MacNee, and E. Abraham
Critical Care in AJRCCM 2004
Am. J. Respir. Crit. Care Med., March 15, 2005; 171(6): 537 - 544.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow All Versions of this Article:
200211-1278OCv1
169/10/1135    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yu, S.-L.
Right arrow Articles by Lin, F.-Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yu, S.-L.
Right arrow Articles by Lin, F.-Y.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Proc. Am. Thorac. Soc. Am. J. Respir. Cell Mol. Biol.
Copyright © 2004 American Thoracic Society