| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Laboratory of Human Bacterial Pathogenesis, Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Hamilton, Montana; and
Laboratory of Host Defenses, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland
Correspondence: Frank R. DeLeo, Ph.D., Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, National Institutes of Health, 903 South 4th Street, Hamilton, MT 59840. E-mail: fdeleo{at}niaid.nih.gov
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-glutamyltransferase and glycolytic activity confirmed that several of these metabolic pathways were up-regulated. In contrast, seven genes encoding critical enzymes involved in fatty acid ß-oxidation, which can generate toxic lipid peroxides, were down-regulated. Our results indicate that energy metabolism and oxidative stress-response pathways are gene-regulated during PMN apoptosis. We propose that changes in PMN gene expression leading to programmed cell death are part of an apoptosis-differentiation program, a final stage of transcriptionally regulated PMN maturation that is accelerated significantly by phagocytosis. These findings provide new insight into the molecular events that contribute to the resolution of inflammation in humans.
Key Words: microarray • neutrophil • metabolism • phagocytosis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mature PMNs are fully capable of ingesting and killing microorganisms in the absence of new gene transcription [9 ]. Although previously believed to transcribe genes for only a select group of proteins (e.g., cytokines and receptors) [10 ], these short-lived cells express a surprising diversity of genes involved in numerous cellular processes [6 , 11 ]. We recently discovered that apoptosis following receptor-mediated phagocytosis in PMNs is regulated, in part, at the level of gene expression [6 ]. This observation suggests that PMNs might transcriptionally regulate additional cellular processes that ultimately facilitate resolution of inflammation. We used human oligonucleotide microarrays to identify genes differentially regulated during the initial stages of activation-induced apoptosis in PMNs. We discovered an elaborate network of differentially regulated genes encoding proteins involved in oxidative stress-related detoxification and energy metabolism pathways during the induction of programmed cell death, which form part of an apoptosis-differentiation program in human PMNs. These findings provide new insight into the molecular and cellular processes that accompany apoptosis following PMN activation. We hypothesize that the apoptosis-differentiation program is crucial for normal resolution of inflammation following bacterial infections in humans.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Phagocytosis experiments and RNA preparation/gene-expression analysis
Latex beads (LB; 2.0 µM; Polysciences, Inc., Warrington, PA) coated with immunoglobulin G (IgG), C3bi, and the combination of IgG and C3bi were prepared as described previously [6
]. For phagocytosis experiments, PMNs (107) were combined on ice with or without (untreated, time-matched controls) LB (8x107) in wells of a 12-well tissue-culture plate precoated with normal human serum (NHS). Plates were centrifuged at 350 g for 8 min at 4°C to synchronize phagocytosis. Following centrifugation, one set of control samples (0 min) was processed before incubation at 37°C, and the remaining plates were incubated at 37°C in a CO2 incubator for the duration of the assay. Phagocytosis was terminated at the indicated time points by aspirating the tissue-culture medium from wells containing PMNs followed by direct lysis with RLTTM buffer (Qiagen, Valencia, CA). RNA was purified as described previously [6
] and was subsequently used to prepare labeled cRNA target (12 µg) for analysis on Hu95A oligonucleotide arrays (Affymetrix, Santa Clara, CA). Labeling samples, hybridization, and scanning were performed according to standard Affymetrix protocols. Total PMN RNA was visualized with an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc., Wilmington, DE) to detect potential degradation of RNA species, and data from GeneChips that had 5'–3' RNA ratios of greater than 3 or a scaled noise factor of greater than 5 (GeneChip SuiteTM, Affymetrix) were discarded. Each experiment was performed with three separate donors.
Gene expression data were analyzed as described previously with GeneSpring expression analysis software, Version 4.04 (Silicon Genetics, Redwood City, CA) [6 ]. Briefly, genes were defined as differentially transcribed if the average expression level changed at least twofold compared with those from time-matched, unstimulated cells over the three experiments and were called "Present" in at least two experiments by GeneChip SuiteTM (Affymetrix). All of the genes included as differentially transcribed were up- or down-regulated at least twofold in one of the treatments. Receptor-specific changes in gene expression were determined as described [6 ]. Previously, we reported significant differences in gene expression between FcR- and CR-mediated phagocytosis, and these were observed mainly at 90 min following phagocytosis [6 ]. Although we observed receptor-specific differences in 8.9% of the genes identified in this study and found differences in the magnitude of the changes in expression, most were similarly regulated 3–6 h following phagocytosis by each of the receptors (FcR, CR, or FcR/CR). Receptor-mediated differences occurring between 3 and 6 h are possibly a result, in part, of differences in oxidative stress related to rate and magnitude changes in ROS production immediately following ingestion [6 ]. For the present study, genes identified as differentially transcribed following any of the three types of phagocytosis have been compiled into a single analysis.
Relative levels of differentially expressed genes were assigned based on average difference intensity (ADI) values determined by the Affymetrix software. (A) = Genes labeled absent by Affymetrix software; (VL) = very low expression = ADI < 50; (L) = low expression = ADI 50–200; (M) = moderate expression = ADI 200–500; (H) = high expression = ADI 500–1500; (VH) = very high expression = ADI > 1500. ADI is an approximation of transcript abundance and not a result of oligonucleotide primer hybridization bias.
Assay for PMN apoptosis
For detection of apoptosis in activated PMNs, phagocytosis assays were performed as described above with the following modifications. PMNs (2x106) were plated directly into 24-well plates precoated with NHS and removed by aspiration at the desired time point. We note that at all time points, there was little or no clumping of cells, and cell viability was unaffected by incubation at 37°C for up to 9 h. For example, at 6 and 9 h, there were 96.7 ± 1.5% and 96.9 ± 1.9% viable cells, respectively, after FcR/CR-mediated phagocytosis compared with 97.0 ± 1.3% and 96.7 ± 1.5% for the same times in the unstimulated cells (n=4–6 experiments by flow cytometry). This is quite similar to the percent of viable PMNs at the start of the assay after the centrifugation step (96.4±1.3%, 0 min). DNA fragmentation, a well-characterized indicator of apoptosis, was determined in PMNs following phagocytosis with a modified terminal deoxynucleotidyltransferase deoxyuridine triphosphate nick-end labeling (TUNEL) assay as described in the manufacturer’s instructions (Apo-BRDUTM apoptosis detection kit, BD Biosciences, Lexington, KY). Samples were analyzed using a FACsCalibur flow cytometer (Becton Dickinson, Mountain View, CA), and 10,000 events were collected for each sample. Alternatively, experiments were performed with RPMI-1640 medium lacking glucose (Invitrogen, Carlsbad, CA).
Taqman real-time reverse transcriptase-polymerase chain reaction (PCR) analysis
Phagocytosis experiments and RNA preparation for Taqman analysis were done with conditions identical to those used for the microarray analysis. Contaminating DNA was subsequently removed from RNA samples by treatment with DNA-Free (Ambion, Austin, TX) [6
]. Primers and probe sets were designed with Primer ExpressTM software, Version 1.5a (Applied Biosystems, Foster City, CA). TaqMan analysis of triplicate samples from a single blood donor was performed with an ABI 7700 thermocycler (Applied Biosystems) as described previously [6
].
Measurement of glucose, lactate, adenosine 5'-triphosphate (ATP), reduced glutathione, and -glutamyltransferase activity
Phagocytosis experiments were performed with conditions identical to those used for the microarray analysis. Aliquots of culture medium were removed at the indicated times, and glucose and lactate concentrations were determined with kits purchased from R-Biopharm, Inc. (Marshall, MI). PMNs were solubilized in sodium citrate lysis buffer (20 mM Tris-Cl, pH 7.5, 1 mM sodium citrate, 5 mM manganese chloride, 5 mM 2-mercaptoethanol, 10% glycerol, 1% Triton X-100, 2 mM pheylmethylsulfonylfluoride, 0.1 mg/ml pepstatin A, 0.1 mg/ml leupeptin) for 30 min on ice. Lysates were clarified by centrifugation (11,700 g, 2 min, 4°C) and stored immediately at -80°C until analyzed. Data for glucose use and lactate production experiments were normalized to the first experiment performed. Intracellular ATP was determined from PMN lysates with a kit purchased from Sigma Cell Culture. -Glutamyltransferase activity was measured with a kit from Sigma Cell Culture according to the manufacturer’s instructions. Reduced glutathione was measured with a kit from Roche Diagnostics (Indianapolis, IN).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
12,500 human genes for changes in gene expression 3–6 h following phagocytosis (Fig. 1B)
. We identified 404 genes that were differentially transcribed (195 genes up-regulated and 209 down-regulated) 3 h following phagocytosis (Fig. 1B)
. Six hours after phagocytosis and during progression of apoptosis, 751 genes were differentially expressed, and the majority of these genes (456) were down-regulated (Fig. 1B)
. In total, 867 unique genes were differentially regulated between 3 and 6 h after phagocytosis (Fig. 1B)
. One hundred twenty-seven of these genes have been directly implicated in apoptosis or were cell fate-related, and many were reported previously [6
]. Of the remaining 740 genes, 291 (39.5%) were up-regulated and 449 (60.5%) were down-regulated.
Regulation of genes encoding key mediators of detoxification/redox metabolism pathways
To facilitate further analysis, differentially expressed genes were categorized by function (Fig. 1C
and 1D)
. The greatest number of induced and repressed genes encoded proteins involved in metabolism and vesicle trafficking (Figs. 1 B
and 1C
,and 2
). Importantly, 27 genes, which together, comprise 6 distinct metabolic pathways, were differentially regulated (Figs. 2
and 3
). Eight genes involved in glutathione metabolism were up-regulated, suggesting a response to oxidative stress during this period of time (Fig. 3A)
. Consistent with these findings, PMNs undergoing activation-induced apoptosis contained significantly more reduced glutathione (19.0±5.0%) and had higher -glutamyltransferase activity (20.2±9.9%) than did unstimulated PMNs (P=0.02 and 0.049 vs. unstimulated PMNs, respectively; Fig. 3A
, inset). Genes encoding thioredoxin and thioredoxin reductase, also important components in cellular detoxification and maintaining redox balance, were up-regulated (Fig. 3B)
. Detoxification and redox balance mediated by glutathione and thioredoxin metabolism are required for controlling oxidant damage within cells and are critical for reducing damage to nearby tissues in the event of cell lysis [13
, 14
]. Genes encoding heme oxygenase-1 and biliverdin [reduced nicotinamide adenine dinucleotide phosphate (NADPH)]-flavin reductase, enzymes required for heme catabolism, another important detoxification pathway, were up-regulated (Fig. 3C)
. Catabolism of heme reduces the potential for the production of toxic metabolites [15
] and generates bilirubin, which has potent antioxidant properties [16
]. Furthermore, redox cycling of bilirubin-biliverdin in the presence of NADPH and biliverdin (NADPH)-flavin reductase is cytoprotective at nanomolar concentrations of bilirubin [17
]. We note that each of these three detoxification-redox systems has an absolute requirement for NADPH and thus, consumes energy. Taken together, these data demonstrate that metabolic pathways critical to oxidative stress response in human PMNs are regulated at the level of gene expression during apoptosis.
|
|
-importin homologue, ß-importin, RanBP, and HRB2, were down-regulated (Fig. 2)
. Moreover, several key mediators of the ubiquitin-proteasome pathway were down-regulated (Fig. 3D)
. The repression of these pathways may function to conserve energy resources that are already limited or be part of a general down-regulation of cellular processes.
Energy metabolism is regulated at the level of gene expression during apoptosis in PMNs
As up-regulation of glutathione and thioredoxin metabolism and heme catabolism pathways require reducing potential in the form of NADPH, we next examined changes in the expression of genes that contribute to energy metabolism. Thirty genes encoding proteins that regulate energy metabolism in eukaryotic cells were differentially expressed during the early stages of activation-induced apoptosis (Fig. 4
). Genes encoding seven enzymes that regulate fatty acid catabolism/ß-oxidation to produce acetyl-CoA were down-regulated. These enzymes included two fatty acyl-CoA synthetases and the rate-limiting enzyme in fatty acid catabolism, carnitine palmitoyltransferase 1, which together, facilitate transport of fatty acids from the cytosol into the mitochondria for catabolic synthesis of acetyl-CoA (Fig. 4)
. Genes encoding acyl-CoA oxidase, acetyl-CoA acyltransferase, and 2,4-dienoyl-CoA reductase, enzymes required for ß-oxidation of fatty acids to produce acetyl-CoA in peroxisomes and mitochondria, respectively, were also down-regulated (Fig. 4)
. The observation that genes encoding several important mediators of acetyl-CoA synthesis were down-regulated indicates reduced demand for and/or availability of acetyl-CoA during PMN apoptosis (Fig. 4)
. Limited availability of acetyl-CoA might be a factor in the induction of apoptosis, as ATP synthesis would presumably be restricted. In addition to altering energy metabolism, down-regulation of fatty acid ß-oxidation would diminish the potential for production of lipid peroxides and other peroxyl radicals [18
, 19
].
|
In addition to the up-regulated enzymes of the HMS, up-regulation of glucosamine 6-phosphate isomerase and a lactate/pyruvate transporter suggested that glycolysis was increased. Up-regulation of glucosamine 6-phosphate isomerase is important, as it provides glycolysis with hexosamines derived from the catabolic breakdown of macromolecules, glycoproteins, and glycosaminoglycans [20 ]. The observation that genes encoding both subunits of hexosaminidase were up-regulated provides strong support for the idea that hexosamines contribute to glycolysis (Fig. 4) . Genes encoding glycogenin and hexokinase, key regulators of glycogen synthesis and glycolysis, respectively, were down-regulated (Fig. 4) . Our finding that the gene encoding hexokinase, a rate-limiting enzyme in glycolysis, was down-regulated seems at variance with our observation that expression of other glycolytic pathway mediators was up-regulated (Fig. 4) . Down-regulation of hexokinase gene expression was confirmed by Taqman real-time PCR (Fig. 5 ). One explanation for this interesting finding is provided below.
|
Glycolysis accompanies apoptosis in PMNs following phagocytosis
To determine whether glycolysis was up-regulated (as suggested by the gene-expression data), we measured glucose consumption and lactate production during activation-induced apoptosis in PMNs (Fig. 6
). PMNs undergoing apoptosis consumed significantly more glucose from the culture media than did unstimulated cells (Fig. 6A)
. Lactate production coincided precisely with glucose depletion, indicating that nearly all of the glucose used by PMNs during apoptosis was converted to lactate (Fig. 6A
and 6B) . Increased glycolysis during activation-induced apoptosis was not a result of phagocytosis or ROS production per se, as these events were completed several hours before the induction of apoptosis [6
]. Although it is counterintuitive that the gene encoding hexokinase was down-regulated during increased glycolysis, recent reports have demonstrated that hexokinase inhibits apoptosis [21
]. Thus, it is likely that very stringent regulation of hexokinase promoted glycolysis and apoptosis simultaneously. This hypothesis is supported by our discovery that genes encoding serine/threonine kinase Akt and three phosphoinositide 3-kinases were down-regulated (ref. [6
], and data not shown), as the Akt/phosphoinositide 3-kinase pathway increases hexokinase activity, preventing cytochrome c release and apoptosis [22
, 23
]
|
Although glycolysis is important for promoting cell survival and blocking apoptosis in multiple cell types [24 , 25 ], we found that increased glycolysis coincided with apoptosis in human PMNs. Therefore, to determine whether the availability of glucose altered activation-induced apoptosis, we measured apoptosis in the presence and absence of glucose (Fig. 6D) . The ability of unstimulated PMNs to undergo apoptosis spontaneously was unaltered by the availability of glucose; however, the lack of glucose partially inhibited activation-induced apoptosis in several (four of six) individuals tested (Fig. 6D) . Individual variability likely accounted for glucose-dependent differences in PMN apoptosis, as the time for the initial deflection of apoptosis in the absence of glucose varied somewhat among individuals. The reduction of apoptosis in the absence of glucose was not a result of reduced PMN phagocytosis, as FcR/CR-mediated phagocytosis was similar in the presence and absence of glucose (data not shown). Previous studies have shown that apoptosis in glucose-deprived cells is dependent on expression of c-myc [26 ], a gene whose expression was not detected in PMNs by our analysis (data not shown). Therefore, it is likely that glycolysis facilitated apoptosis directly or mediated processes that accompany apoptosis following phagocytosis in PMNs.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
ACKNOWLEDGEMENTS |
|---|
Received October 8, 2002; revised October 8, 2002; accepted November 4, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Related Articles
This article has been cited by other articles:
|
|
 |
|
  S. D. Kobayashi, J. M. Voyich, A. R. Whitney, and F. R. DeLeo Spontaneous neutrophil apoptosis and regulation of cell survival by granulocyte macrophage-colony stimulating factor J. Leukoc. Biol., December 1, 2005; 78(6): 1408 - 1418. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Voyich, K. R. Braughton, D. E. Sturdevant, A. R. Whitney, B. Said-Salim, S. F. Porcella, R. D. Long, D. W. Dorward, D. J. Gardner, B. N. Kreiswirth, et al. Insights into Mechanisms Used by Staphylococcus aureus to Avoid Destruction by Human Neutrophils J. Immunol., September 15, 2005; 175(6): 3907 - 3919. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Borjesson, S. D. Kobayashi, A. R. Whitney, J. M. Voyich, C. M. Argue, and F. R. DeLeo Insights into Pathogen Immune Evasion Mechanisms: Anaplasma phagocytophilum Fails to Induce an Apoptosis Differentiation Program in Human Neutrophils J. Immunol., May 15, 2005; 174(10): 6364 - 6372. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Maiuri, G. Tajana, T. Iuvone, D. De Stefano, G. Mele, M. T. Ribecco, M. P. Cinelli, M. F. Romano, M. C. Turco, and R. Carnuccio Nuclear Factor-{kappa}B Regulates Inflammatory Cell Apoptosis and Phagocytosis in Rat Carrageenin-Sponge Implant Model Am. J. Pathol., July 1, 2004; 165(1): 115 - 126. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Zhang, Y. Kluger, Y. Nakayama, R. Poddar, C. Whitney, A. DeTora, S. M. Weissman, and P. E. Newburger Gene expression in mature neutrophils: early responses to inflammatory stimuli J. Leukoc. Biol., February 1, 2004; 75(2): 358 - 372. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. B. Robey Caveat experimentor: reporting expression profiling data involving multiple, functionally redundant isoforms J. Leukoc. Biol., September 1, 2003; 74(3): 307 - 308. [Full Text] [PDF]
|
|
|
|
|
|
S. D. Kobayashi, J. M. Voyich, G. A. Somerville, J. M. Musser, H. L. Malech, and F. R. DeLeo Response to the Letter submitted by R. Brooks Robey J. Leukoc. Biol., September 1, 2003; 74(3): 309 - 310. [Full Text] [PDF]
|
|
|
|
 |
|
 B. Zhang, J. Hirahashi, X. Cullere, and T. N. Mayadas Elucidation of Molecular Events Leading to Neutrophil Apoptosis following Phagocytosis: CROSS-TALK BETWEEN CASPASE 8, REACTIVE OXYGEN SPECIES, AND MAPK/ERK ACTIVATION J. Biol. Chem., August 1, 2003; 278(31): 28443 - 28454. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. D. Kobayashi, J. M. Voyich, K. R. Braughton, and F. R. DeLeo Down-Regulation of Proinflammatory Capacity During Apoptosis in Human Polymorphonuclear Leukocytes J. Immunol., March 15, 2003; 170(6): 3357 - 3368. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, 
versus unstimulated PMNs (
), one-way ANOVA with Dunnett’s post-test (GraphPad InStat, Version 3.0, GraphPad Software, San Diego, CA). **, P= 0.03 for • versus unstimulated PMNs (
