Introduction
Endotoxin is a major component of the cell wall of Gram negative bacteria. It is normally cleared from the blood by macrophages of the reticuloendothelial system predominantly localized in the liver [
1]. Acute endotoxemia, characterized by excessive levels of endotoxin in the blood, can lead to endothelial injury, hypotension, multi-organ failure and death [
1]. Endotoxin-induced tissue injury is associated with an inflammatory response and an accumulation of phagocytes at sites of damage. Neutrophils are major cellular components of an acute inflammatory reaction. These cells are rapidly recruited to areas of injury where they become activated to release reactive oxygen and nitrogen intermediates, cytokines, and lipid mediators, which aid in antigen destruction, but may also contribute to tissue injury [
2].
Neutrophils are relatively short-lived cells removed from the body under homeostatic conditions by the process of apoptosis. To exert their biological effects during inflammatory reactions, neutrophils must remain at the injured site for prolonged periods of time. In this regard, recent studies have shown that inflammatory mediators such as lipopolysaccharide (LPS), granulocyte macrophage-colony stimulating factor and leukotriene B
4 delay apoptosis and prolong neutrophil survival in culture [
3,
4]. Neutrophils migrating into rodent lungs in response to LPS challenge have also been reported to exhibit delayed rates of apoptosis [
5,
6]. Similarly, increases in cytokine and growth factor activity, and reduced neutrophil apoptosis, have been described in various inflammatory lung diseases in humans including chronic obstructive pulmonary disease, acute respiratory distress syndrome and pneumonia [
3,
7,
8].
In the present study, we characterized the effects of acute endotoxemia on the response of a unique population of neutrophils, which we have previously demonstrated are tightly adhered to the lung vasculature [
9]. Because of their location within the tissue, these cells are likely to be more relevant than circulating neutrophils, or neutrophils recovered by bronchoalveolar lavage (BAL), to the pathophysiology of endotoxin induced lung injury. We found that administering endotoxin to rats resulted in an increased number of functionally active adherent vascular lung neutrophils. Moreover, these cells exhibited decreased apoptosis and expressed the anti-apoptotic protein, Mcl-1. This was correlated with the induction of total mitogen activated protein (MAP) kinases, as well as NF-κB and cAMP response element binding (CREB) protein transcription factor activity in the cells. These findings suggest a potential mechanism mediating decreased apoptosis and increased functional activation of adherent vascular neutrophils in the lung during acute endotoxemia.
Materials and methods
Animals and treatments
Female specific pathogen-free Sprague Dawley rats (200–225 g, 6–8 weeks) were purchased from Taconic (Germantown, NY, USA). Animals were housed in microisolator cages and were maintained on sterile food and pyrogen-free water ad libitum. Acute endotoxemia was induced by i.v. injection of rats with 5 mg/kg Escherichia coli LPS (serotype 0128:B12, Sigma Chemical Co., St. Louis, MO, USA). Data from control animals treated with PBS were pooled. Each experiment used one to two animals per treatment group and was repeated at least three times.
Cell isolation
Rats were euthanized by i.p. injection of Nembutal (125 mg/kg) 0.5–48 hours after administration of endotoxin or control. Adherent vascular lung neutrophils were isolated from rats as previously described [
10]. Briefly, the lung was perfused with 50 ml of warm (37°C) Ca
+2/Mg
+2-free Hank's balanced salt solution (HBSS, pH 7.4) containing 2.5 mM HEPES, 0.5 mM EGTA and 4.4 mM NaHCO
3 at a rate of 22 ml/min and then lavaged 5–6 times with HBSS to remove alveolar macrophages and loosely adhered neutrophils. The trachea and major bronchi were then excised and the lung cut into 500 μm slices (McIlwain mechanical tissue chopper, Brinkmann Instruments, Westbury, NY, USA). Lung slices were washed in ice cold Ca
+2/Mg
+2-free HBSS with vigorous shaking at speed 7 for 3 min (Genie 2; Fisher Scientific, Pittsburgh, PA, USA), filtered (220 μm), and then incubated in ice cold HBSS for 30 min with periodic shaking. Neutrophils were recovered, after digestion of the lung tissue for 25 min with 60 U/ml collagenase D (specific activity 0.22 Wunsch unit = 800 IU, low ancillary protease activity, Boehringer Mannheim, Indianapolis, IN, USA), prepared in HBSS containing 10% fetal bovine serum (FBS) and 0.01% DNAse 1 (Sigma Chemical Co., St. Louis, MO, USA), followed by filtering (220 μm). In previous studies using transmission electron microscopy, we demonstrated that cells recovered using this methodology consisted of a homogeneous population of neutrophils derived from the lung vasculature [
9]. Giemsa staining of cytospin samples showed that the cells were 96% neutrophils. Viability was greater than 90% as determined by trypan blue dye exclusion.
Cytoplasmic DNA isolation and analysis
Cells (1 × 10
6) were lysed in 200 μl of buffer (5 mM Tris-Cl pH 7.4, 2 mM EDTA, 0.5% Triton X-100) on ice under DNAse-free conditions. After 30 min, cell lysates were centrifuged (16,500 × g for 20 min) and cytoplasmic DNA extracted from the supernatants by overnight precipitation in 0.1× volume 3 M sodium acetate (pH 8.0) and 2× volume 100% ethanol. After washing, DNA was dissolved in 10 μl TBE buffer (0.045 M Tris-borate, 0.001 M EDTA, pH 8.0) [
11]. Samples (5 μl) were analyzed on 1.2% agarose gels, stained with ethidium bromide and visualized under UV light.
Annexin V binding assay
Annexin V binding was detected using an Apoptosis Detection Kit (R & D Systems, Minneapolis, MN, USA). Adherent vascular lung neutrophils isolated from control animals, or 6–12 hours after administration of endotoxin, were plated for 18 hours into 4-well plates (5 × 105 cells/well). Cells were collected, washed and resuspended in PBS. Cells (100 μl) were incubated in the dark in the presence of fluorescein-conjugated annexin V (10 μl) and propidium iodide (10 μl) reagent for 20 min at room temperature. Unstained cells, cells stained with annexin V fluorescein only, and cells stained with propidium iodide only, were used as additional controls. After diluting with PBS, cells were immediately analyzed on a Coulter Epics Profile II flow cytometer (Coulter Electronics, Hialeah, FL, USA).
Measurement of chemotaxis
Chemotactic responsiveness of lung neutrophils towards bacterially-derived n-formyl-methionyl-leucyl-phenylalanine (fMLP) was quantified using a Neuroprobe 48-microwell chemotaxis chamber and 5-μm pore size Nucleopore filters [
12,
13]. Cells (2 × 10
6 cells/ml), suspended in HBSS containing 0.5% BSA, were incubated in the upper wells of the chamber at 37°C with fMLP (50 nM) or medium in the lower wells. After 45 min, the filter containing the adhered, migrated neutrophils was removed and stained with Camco Quik Stain (Baxter, McGaw Park, IL, USA). Chemotaxis was measured as the number of cells that migrated through the filter in 10 oil immersion fields.
Relative RT-PCR
Total RNA was isolated from cells using an RNeasy Miniprep kit (Qiagen Inc., Valencia, CA, USA) following the manufacturer's instructions. RNA was quantified by absorbance at 260 nm. For first strand synthesis, RNA (200 ng) in 9 μl water was denatured at 65°C for 4 min, rapidly cooled on ice, and then resuspended in a 20 μl final volume containing 50 mM Tris-HCl pH 8.3, 75 mM KCl, 3 mM MgCl2, 10 mM DTT, 1 mM of each dNTP, 20 mM random hexamers and 200 U Superscript II RNase H- RT (GIBCO BRL, Gaithersburg, MD, USA). After 1 hour incubation at 37°C, RNase H- (2 U) was added and the samples incubated for an additional 20 min. The samples were then denatured at 95°C for 5 min and chilled on ice. Initially, the linear range of amplification for cyclooxygenase-2 (COX-2) and tumor necrosis factor alpha (TNF-α) mRNA was determined using mouse COX-2 and TNF-α primers (Ambion Inc., Austin, TX, USA) with the neutrophil first strand cDNA samples as templates, following the manufacturer's instructions. The primers (5' or 3') used for COX-2 were sense primer, CAT TCT TTG CCC AGC ACT TCA C and antisense primer, GAC CAG GCA CCA AGA CCA AAG AC. For TNF-α the primers were sense primer, TCT GTC TAC TGA ACT TCG GGG T and antisense primer, TAG TTG GTT GTC TTT GAG ATC C. For COX-2 and TNF-α, 23 cycles and 26 cycles, respectively, fell within the linear range for each control and experimental sample. The 18S primer: competimer (Ambion Inc., Austin, TX, USA) ratio was adjusted to produce an amplification signal in the linear range of the COX-2 and TNF-α products, to normalize each COX-2 and TNF-α. A ratio of 1:19 was found to be appropriate to normalize both messages. Lung neutrophil cDNA was then amplified in 20 μl buffer containing 1 μl neutrophil reverse transcribed cDNA (representing 10 ng total RNA), 10 mM Tris-HCl pH 8.3, 50 mM KCl, 1.5 mM MgCl2, 0.4 μM TNF-α primer or COX-2 primer, 1:19 ratio of 18S competimers/primers, 0.1 mM each dNTP, 0.2 μl α [32P] dCTP (10 mCi/ml; > 3000 Ci/mmol) and 0.025 units of Amplitaq (Perkin Elmer Inc., Norwalk, CT, USA). Amplification was initiated by 1 min denaturation at 94°C, followed by 23 or 26 cycles (for COX-2 and TNF-α respectively) at 94°C for 15 sec, 56°C for 25 sec, and 72°C for 90 sec. The amplified PCR products (20 μl total) were mixed with 10 μl of loading buffer (95% formamide, 10 mM EDTA, 0.1% xylene cyanol, 0.1% bromophenol blue) and 5 μl aliquots were applied to a 5% denaturing polyacrylamide gel. Radioactive bands were excised from the dried gel and counted by the Cerenkov method in a scintillation counter. Amplifications for all samples were performed at the same time and run on the same gel to minimize variability.
Western blot analysis
Cells were suspended in buffer containing 50 mM HEPES pH 7.4, 10 mM KCl, 1 mM EDTA, 1 mM DTT, 1 μg/ml pepstatin A, 1 μg/ml leupeptin, 10 μg/ml soybean trypsin inhibitor, 10 μg/ml aprotinin and 0.5% NP-40 and mixed periodically using a vortex. After 10 min on ice, lysates were centrifuged (4000 × g, 5 min) and supernatants containing 10 μg of protein fractionated on 10% SDS polyacrylamide gels. The proteins were transferred to nitrocellulose paper, and incubated at room temperature for 3 hours, or overnight, at 4°C, with mouse monoclonal anti-human Mcl-1 (1:1000), p38 MAP kinase, phosphoinositide 3 kinase (PI3K), or protein kinase B-alpha (PKB-α) (1:500) antibodies (Transduction Laboratories, San Diego, CA, USA and New England Biolabs, Inc., Beverly, MA, USA), or rabbit polyclonal anti-rat p44/42 MAP kinase (1:1000), or doubly phosphorylated anti-human Thr180/Tyr182 phospho-p38 MAP kinase (1:1000) antibodies (New England Biolabs, Inc.). According to the manufacturers, these antibodies cross react with rat cells. This was followed by incubation with a 1:2000 dilution of goat anti-mouse or sheep anti-rabbit IgG horse-radish peroxidase-conjugated antibody (Transduction Laboratories and New England Biolabs) for 1 hour at room temperature. Proteins were detected using an Enhanced Chemi-Luminescence (ECL) detection system (Amersham Life Sciences, Arlington Heights, IL, USA). Protein concentrations were determined using a BCA Protein Assay Kit (Pierce, Rockford, IL, USA) with BSA as the standard.
Nuclear extracts were prepared as previously described [
14]. Briefly, cells were incubated in buffer (10 mM HEPES pH 7.4, 10 mM KCl, 2 mM MgCl
2, 2 mM EDTA) on ice for 10 min followed by incubation with 10% NP-40. After 5 min, cells were centrifuged (16,500 × g, 5 min), the pellets resuspended in buffer (50 mM HEPES pH 7.4, 50 mM KCl, 300 mM NaCl, 0.1 mM EDTA, 10% glycerol) and then placed on ice for an additional 20 min. Supernatants containing nuclear extracts were collected after centrifugation (16,500 × g, 5 min, 4°C). Binding reactions were carried out at room temperature for 30 min in a total volume of 15 μl and contained 2–5 μg of nuclear extracts, 5 μl of 5X gel shift binding buffer (20% glycerol, 5 mM MgCl
2, 2.5 mM EDTA, 2.5 mM DTT, 250 mM NaCl, 50 mM Tris-HCl, pH 7.5), 2 μg poly (dI-dC) and 3 × 10
4 cpm/μl of [
32P] labeled CREB (5'-AGA GAT TGC CTG ACG TCA GAG AGC TAG-3'), NF-κB (5'-AGT TGA GGG GAC TTT CCC AGG C-3') (Promega Gel Shift Assay Systems, Madison, WI, USA) or CCAAT/enhancer binding protein (C/EBP) (5'-TGC AGA TTG CGC AAT CTG CA-3') (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA) oligonucleotides. Probes were labeled using γ [
32P]ATP (3000 Ci/mmol, NEN, Boston, MA, USA). Protein-DNA complexes were separated on 5% or 7% non-denaturing polyacrylamide gels run at 250 V in 0.5× TBE and visualized after the gels were dried and autoradiographed. For supershift reactions, extracts were incubated with antibodies (1 μg) to NF-κB subunits (p50 or p65), CREB, phospho-CREB, C/EBP-β or C/EBP-δ for 20 min on ice prior to the addition of labeled oligonucleotide. For competitor reactions, an excess (40×) of the respective unlabeled oligonucleotide was added to the reaction mixtures prior to addition of the labeled probe.
Discussion
Recent studies have suggested that phagocytic leukocytes including macrophages and neutrophils contribute to the pathophysiology of endotoxin-induced lung injury [
22]. Whereas macrophages are normally present in the tissue, the majority of neutrophils responding to endotoxin must emigrate into the lung from the blood. The initial step in this process involves tight adherence of the neutrophils to the vasculature [
23]. In this study, we analyzed the effects of acute endotoxemia on functional and biochemical responsiveness of these cells.
Treatment of rats with endotoxin was found to stimulate neutrophil adherence to the lung vasculature. Thus, following endotoxin administration, increased numbers of adherent vascular neutrophils were recovered from the lung. A similar increase in vascular lung neutrophils has been observed in rabbits following i.v. administration of LPS [
24,
25]. Interestingly in our model, this response was biphasic occurring initially at 0.5 hours and then again at 24 hours after endotoxin administration. We speculate that this represents two distinct populations of responding cells. Sepsis is associated with a rapid release of CXC chemokines by alveolar macrophages, as well as up regulation of lung vascular intercellular adhesion molecule-1 (ICAM-1) expression [
26] and these may contribute to the initial increase in adherent vascular neutrophils in the lung. Activated neutrophils have also been reported to express increased macrophage inflammatory protein-2 (MIP-2) mRNA following endotoxin administration [
21]. It is possible that the secondary increase in adherent neutrophils in the lung vasculature is due to the generation of chemotactic factors by activated neutrophils.
Mature neutrophils have a circulating life span
in vivo of about 6–10 hours, after which time they undergo apoptosis [
27]. The present studies demonstrate that adherent vascular neutrophils from the lungs of endotoxin treated animals exhibit decreased apoptosis as evidenced by reduced cytoplasmic DNA and decreased Annexin V binding when compared to cells from untreated animals. These results are in accord with previous studies demonstrating delayed apoptosis in neutrophils recovered from BAL fluid and from peripheral blood after LPS challenge in mice and humans [
5,
28] and correlate with our findings of increased survival of these cells in culture [
29].
Acute endotoxemia caused a marked induction of Mcl-1 in freshly isolated adherent vascular lung neutrophils, which peaked between 2 and 12 hours and remained elevated for at least 48 hours. These findings indicate that prolonged neutrophil survival in the lung vasculature following endotoxin administration may be mediated, at least in part, by Mcl-1. In contrast to cells from endotoxin treated animals, adherent vascular neutrophils from control animals did not express Mcl-1 protein. This is consistent with the relatively high levels of apoptosis observed in these cells and their reduced survival in culture [
29]. Our findings are in accord with a recent study demonstrating that Mcl-1 expression in human peripheral blood neutrophils decreases with aging [
15]. As previously reported for human peripheral blood neutrophils [
30], we were unable to detect Bcl-2 in rat adherent vascular lung neutrophils, even after endotoxin treatment of animals. We were also unable to detect Bax or Bcl-X
L in rat adherent vascular lung neutrophils, which is in contrast to results with human peripheral blood neutrophils [
30,
31]. These differences may be due to unique attributes of human and rat cells, their distinct location in the vasculature, and/or to their exposure to inflammatory stimuli
in vivo.
We have previously shown that acute endotoxemia primes adherent lung neutrophils for increased production of reactive oxygen and nitrogen intermediates suggesting that these cells are functionally activated [
9]. The work reported here, demonstrates that adherent vascular lung neutrophils from endotoxin treated rats are also activated to respond to chemotactic stimuli and to express TNF-α and COX-2 mRNA and protein. The effects of endotoxin administration on these responses were rapid, occurring within 0.5 hours. Subsequently, chemotactic responsiveness and TNF-α mRNA expression declined and by 12 hours were at control levels. Similar transient increases in neutrophil chemotaxis and pulmonary TNF-α production have been described previously during experimentally induced endotoxemia [
32,
33]. These findings are consistent with the idea that chemotaxis and TNF-α production are important in early inflammatory responses of neutrophils [
34]. In contrast, protein levels for COX-2 remained elevated for 48 hours after endotoxin administration. This most likely reflects the sustained role of prostaglandins during the inflammatory process [
35,
36].
Inflammatory mediators induce their biological effects by binding to specific receptors on target cells. This initiates biochemical signaling pathways leading to increased survival and functional activation. Members of the MAP kinase signaling cascade have been shown to delay apoptosis in a number of cell types [
37‐
39]. Proteins belonging to this family are also involved in cell signaling leading to functional responsiveness of inflammatory cells including chemotaxis and mediator production [
40]. Treatment of rats with endotoxin induced expression of total p38 and p44/42 MAP kinase in adherent vascular lung neutrophils within 0.5 hours, a response which persisted for 48 hours. Our findings that these MAP kinases are upregulated within 0.5 hours in response to endotoxin suggests that they may contribute to increased chemotactic responsiveness and TNF-α production. In this regard, inhibition of p38 MAP kinase activity has been reported to block chemotaxis and TNF-α release by lung neutrophils, epithelial cells and eosinophils [
40‐
42]. Interestingly, significant phospho-p38 MAP kinase was also detected in neutrophils from control animals. Phospho-p38 MAP kinase is known to promote apoptosis in human peripheral blood neutrophils [
37] and a similar pathway may be involved in adherent vascular lung neutrophil apoptosis in control animals. Alternatively, endotoxin administration to animals may induce phosphatases that degrade phospho-p38 MAP kinase [
43]. In contrast to phospho-p38 MAP kinase, we were unable to detect phospho-p44/42 MAP kinase in rat adherent vascular neutrophils even after endotoxin administration. This may be due to an inability of the antibody to recognize phosphorylated residues on these proteins in primary rat neutrophils. Abraham
et al[
44] have recently reported p42 MAP kinase activation in mouse lung neutrophils in response to endotoxin administration. Differences between our results may be due to unique attributes of rat and mouse models.
Inositol lipids generated via PI3K and downstream targets such as PKB-α have been implicated in the regulation of cell survival, as well as in induction of MAP kinases and TNF-α [
45‐
47]. We found that endotoxin treatment of the animals caused a significant increase in expression of both PI3K and PKB-α at 12 hours, which is consistent with reduced apoptosis in these cells. However our findings that PI3K levels remained elevated for 48 hours, while PKB-α decreased, suggest that PI3K regulates signaling pathways in adherent vascular neutrophils that are distinct from PKB-α. Our data also suggest that in lung neutrophils, PI3K and PKB-α are not co-ordinately regulated with chemotaxis, or expression of TNF-α or MAP kinases.
The transcription factors NF-κB and CREB have been reported to activate COX-2 in LPS-stimulated monocytes and macrophages
in vitro[
48,
49]. Moreover, endotoxin-induced increases in neutrophil TNF-α, as well as interleukin-1β and MIP-2
in vivo appear to be mediated by NF-κB and CREB [
21]. We found that activated NF-κB and CREB were present in the nucleus of adherent vascular lung neutrophils from endotoxemic, but not control rats, suggesting that these transcription factors play a role in increased responsiveness of the cells. These findings are in accord with previous studies on mouse lung neutrophils during endotoxemia [
21]. Our observation that nuclear binding activity appeared in the cells rapidly after endotoxin administration (within 0.5 hours) indicates that NF-κB and CREB are involved in early functional activation. A similar role has been suggested for these transcription factors in the response of intraparenchymal mononuclear cells to hemorrhage [
50]. The fact that NF-κB and CREB activation occurs prior to expression of PI3K and PKB-α indicates that these transcription factors may regulate these proteins in lung neutrophils. Interestingly, a secondary increase in NF-κB was observed in cells isolated 24 hours after endotoxin administration. This may reflect the activity of a second distinct neutrophil population responding to endotoxin.
Administration of endotoxin to the animals also induced C/EBP nuclear binding activity. In contrast to NF-κB and CREB, this was most pronounced 12 hours post-treatment of the animals. It is possible that NF-κB and CREB mediate the early phases of neutrophil activation, while C/EBP proteins function to maintain this response [
50]. The observation that the temporal pattern of expression of activated C/EBP and PKB-α are similar suggests co-ordinated regulation of these proteins [
51]. A decrease in activation of transcription factors like NF-κB and CREB and an increase in C/EBP may also contribute to reduced activity of pro-apoptotic genes, resulting in prolonged longevity of neutrophils in the lung vasculature. Recent studies have demonstrated upregulation of C/EBP in lung injury induced by LPS
in vivo[
52,
53]. Our findings are consistent with this report and suggest that this transcription factor may be a marker of tissue injury.