INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The ubiquitous transcription factor complex, nuclear factor kappa B (NF-B), is necessary for directing high level transcription of many cytokines, adhesion molecules, and other proinflammatory genes in tissue cultures; however, the extent to which NF-B controls specific biological processes in vivo is unknown. In unstimulated cells, NF-B exists in the cytoplasm as a dimer, consisting of two members of the NF-B/Rel protein family, bound to an inhibitory protein (IB). After stimulation, IB- (or IB-) is phosphorylated on two N-terminal serines, ubiquitinated, and degraded by the 26 S proteasome. Degradation of the IB subunit liberates NF-B and allows translocation to the nucleus, where DNA binding occurs at specific NF-B binding motifs. Bound NF-B then interacts with other transcription factors and the basal transcriptional machinery to regulate transcription of target genes.

Several animal models have been studied to evaluate the role of NF-B in the production of inflammatory events (1). These studies have linked in vivo NF-B activation with cytokine production and the generation of inflammation; however, an important limitation of all of these studies is that NF-B activation was measured by electrophoretic mobility shift assay (EMSA). EMSA is semiquantitative, evaluates NF-B activation at only a single point in time, and does not address the functional effects of NF-B activation in initiating gene transcription. We wanted to develop a convenient, quantitative method for evaluating NF-B activation over time to examine the consequences of NF-B activation in multiple organs in vivo. To achieve this goal, we used transgenic mice that were engineered to possess the following construct in each tissue: proximal 5' human immunodeficiency virus (HIV-1) long terminal repeat (LTR) driving the expression of Photinus luciferase complementary DNA (cDNA) (referred to as HLL mice [HIV-LTR/Luciferase]) (manuscript in preparation). The proximal HIV-LTR is a NF-B-responsive promoter (6), containing a TATA box, an enhancer region between 82 and 103 with two NF-B motifs, and three Sp1 boxes from 46 to 78. In primary cell culture, NF-B activation is absolutely required for transcriptional activity of the HIV-LTR (9, 10). In the HLL mice, we show that luciferase production and intracellular accumulation are dependent on NF-B-activated gene transcription; therefore, HLL mice provide a useful in vivo reporter-based assay system in which to analyze NF-B enhancer activity in response to a variety of inflammatory signals.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animal Model

HLL mice and nontransgenic littermates (C57B6/DBA background) weighing between 20 and 30 g were used in all experiments. Both of the parental strains are known to be lipopolysaccharide (LPS)-responsive. Escherichia coli LPS (serotype 055; B5, Sigma Chemical Co., St. Louis, MO) was given by intraperitoneal injection. Blood was obtained by retro-orbital puncture after anesthesia with methoxyflurane, and the mice were asphyxiated with carbon dioxide. Mouse tracheas were cannulated after death, and the lungs were lavaged in situ with sterile pyrogen-free physiologic saline. Saline was instilled in two 600-µl aliquots and gently withdrawn with a 1-ml syringe. For luciferase measurements, tissues were homogenized in 1 ml phosphate-buffered saline (PBS) and stored at 20° C. For other studies, the lungs and other organs were removed, quickly frozen in liquid N₂, and stored at 70° C.

Cell Counts and Differentials

Lung lavage fluid was centrifuged at 500 × g for 10 min to separate cells from supernatant. Supernatant was saved separately and frozen, and the pelleted cells were suspended in a small amount of serum-free RPMI culture medium. Total cell counts were determined on a grid hemocytometer. Differential cell counts were enumerated on cytocentrifuge slides that were stained with a modified Wright stain (Diff-Quik) by counting 200 to 400 cells in cross section.

Luciferase Assay

A volume of 10 µl of tissue homogenate was added to 200 µl lysis buffer (Luciferase Assay System; Promega, Madison, WI) and placed on ice for 20 min with frequent vortexing. After centrifugation, 100 µl luciferin was added to 20 µl cell lysate and luciferase activity was read in a luminometer.

Expression of Trans-dominant Negative IB-

A trans-dominant negative mutant IB- (IB-DN) was constructed with serine-adenine substitutions at the critical serines (S36, 40) in the avian IB-. Adenoviruses expressing IB-DN or -galactosidase under the control of a cytomegalovirus 70K (CMV) promoter, were made in Dr. Kerr's laboratory using standard techniques. These adenoviral vectors were injected intravenously at 5 × 10⁹ plaque-forming units (PFU). Expression of IB-DN was identified in tissue by Western immunoblots. Fifty micrograms protein from tissue homogenates was separated on a 10% acrylamide gel, transblotted, and immunodetection was done using antiserum specific for avian IB-, which does not cross react with native murine IB- or .

Measurement of Immunoreactive KC, Interleukin-6 (IL-6), and Tumor Necrosis Factor- (TNF-)

Murine KC, IL-6, and TNF- were measured in serum and tissue homogenates using commercially available enzyme-linked immunoabsorbance assays (ELISA) according to the manufacturer's instructions (KC and IL-6 [R&D, Minneapolis, MN]; TNF- [Genzyme, Cambridge, MA]).

Nuclear Protein Extractions and EMSAs

Nuclear protein extraction from tissues and EMSAs for NF-B were done as previously described (3). An oligonucleotide probe containing a consensus NF-B motif (Stratagene, La Jolla, CA) was used in these studies. Antibodies for supershift studies were purchased from Santa Cruz Biotechnology, Inc (Santa Cruz, CA).

Total RNA Extractions and Ribonuclease (RNase) Protection Assay (RPA)

Total RNA was purified by a modification of the method of Chirgwin and associates (11). Frozen lung tissue was mixed with 1 ml of TRI REAGENT (Molecular Research Center Inc., Cincinnati, OH) and ground in a tissue homogenizer. The samples were transferred to 1.5 ml eppendorf tubes, and RNA was extracted with chloroform and precipitated with isopropanol. The RNA pellet was then washed with 75% ethanol, air dried, and dissolved in 50 to 100 µl of 30% formamide/10% formaldehyde. Total RNA was quantitated by determining the light absorbance at 260 nm. RPA were done using the Riboquant multiprobe RPA system (PharMingen, San Diego, CA) according to the manufacturer's protocol. Samples were run on a 5% polyacrylamide gel, which was then dried and subjected to autoradiography. Specific bands were quantified using a laser densitometer.

Statistical Analysis

For comparison among groups, a one-way analysis of variance (ANOVA) was used with the Tukey-Kramer multiple comparisons test (p values less than 0.05 were considered significant). Correlations between variables were sought using standard linear regression techniques (InStat; Graphpad Software, Inc., San Diego, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Intraperitoneal Injection of LPS Results in Time and Dose-dependent Luciferase Activity in Multiple Organs

Initially, we treated two lines of the HLL mice, lines 20 and 27, with intraperitoneal injection of E. coli LPS at a dose of 1 µg/g of body weight. Figure 1 illustrates luciferase activity for a variety of organs at baseline values and 4 h after intraperitoneal LPS (reported as relative light units [RLU] and normalized for total protein content in each organ). Paired nontransgenic controls had very low levels of luciferase activity, with or without LPS treatment (not shown). In this study, two patterns of LPS-induced luciferase activity were observed. In the lung, liver, and kidney, low basal luciferase activity was found, but there was significantly increased luciferase activity after LPS treatment (p < 0.05). In the spleen and bone marrow, there was somewhat higher basal luciferase activity that was not significantly increased after LPS treatment.

View larger version (21K): [in this window] [in a new window]   Figure 1.   Effect of intraperitoneal endotoxin (LPS) on luciferase expression in HLL mice. Luciferase activity was normalized for total protein concentration and reported as mean (± SEM) relative light units (RLU) for various organs at baseline and 4 h after LPS injection (n = 6 for each treatment group). Significant increases in luciferase activity were found in the liver, lung, and kidney after LPS (p < 0.05). Luciferase activity was not increased in the bone marrow and spleen (p > 0.05).

Based on these initial studies, we further evaluated the dose- and time-dependence of LPS-induced luciferase activation (normalized for total protein) in the lung and liver, two organs thought to be important in the systemic inflammatory response to LPS, and the spleen, which showed only modest LPS-induced luciferase activity. Figure 2 illustrates luciferase activity in the lung, liver, and spleen 4 h after intraperitoneal LPS. Luciferase activity was measured after LPS doses ranging from 0.1 to 3 µg/g and normalized for total protein concentration. Lung luciferase activity was increased after administration of 1 and 3 µg/g of LPS. Liver luciferase activity, however, was stimulated by 0.3 to 3 µg/g of LPS, indicating a lower threshold for induction of luciferase in the liver than in the lung. In this experiment, spleen luciferase activity did not increase after LPS treatment, except at the highest dose (3 µg/g). These findings demonstrate that each of these organs has a different threshold for luciferase production in response to LPS treatment.

View larger version (31K): [in this window] [in a new window]   Figure 2.   Dose-response effect of LPS on luciferase expression in the lung, liver, and spleen of HLL mice (line 27). Luciferase activity was measured 4 h after intraperitoneal injection of 0, 0.1, 0.3, 1.0, and 3.0 µg/g of LPS. Luciferase activity was normalized for total protein concentration and reported as mean (± SEM) RLU for the lung (open column), liver (rising diagonal cross hatch), and spleen (falling diagonal cross hatch) (n = 5 for each treatment group).

We next evaluated the time course for luciferase activation by intraperitoneal LPS in the lungs, livers, and spleens of HLL mice. Figure 3 shows luciferase activity expressed as RLU and normalized for total protein content and for luciferase activity in paired nontransgenic mice. After LPS injection, lung luciferase activity increased by 2 h, peaked at 4 h, and decreased by 6 h. Peak luciferase activity was approximately tenfold higher than untreated HLL controls. Liver luciferase activity also increased by 2 h after LPS injection and peaked at 4 h. Peak activity was almost 100 times control values, and luciferase activity persisted at 6 h. The spleen, which had higher basal luciferase activity than the lung or liver, showed modestly increased luciferase activity from 1 to 6 h after LPS injection, peaking at 4 h. In each organ examined, luciferase activity returned to baseline by 24 h after LPS treatment. Therefore, each organ tested has a distinctive time- and dose-dependent generation of luciferase after intraperitoneal LPS.

View larger version (30K): [in this window] [in a new window]   Figure 3.   Time course for the appearance of luciferase in the lung, liver, and spleen after intraperitoneal injection of HLL mice (lines 20 and 27) with LPS (1 µg/g). Luciferase activity is expressed as mean RLU (± SEM) after normalization for total protein content and for luciferase activity in paired nontransgenic mice. Luciferase activity is shown for the lung (open column), liver (rising diagonal cross hatch), and spleen (falling diagonal cross hatch) at 0, 1, 2, 4, 6, and 24 h after LPS injection (n = 6 for each time point).

In addition to measuring luciferase activity in tissues after intraperitoneal LPS, we measured luciferase activity in alveolar macrophages obtained by lung lavage 0 to 4 h after intraperitoneal LPS (more than 97% of lavage cells are macro-phages at these time points). We found increased luciferase activity in alveolar macrophages from 1 to 4 h after intraperitoneal LPS compared with controls (data not shown).

To correlate luciferase activity with NF-B activation as measured by EMSA, tissue nuclear protein extracts were prepared from mice treated with LPS in the same time course experiment described previously (Figure 3). Figure 4 shows that NF-B activation in the lung and liver preceded increases in luciferase activity. Band A represents p65(RelA)/p50 hetero-dimers whereas band B represents p50 homodimers, as demonstrated by supershift analysis (Figure 4A). Lung NF-B activation was noted by 1 h after LPS injection, peaked at 2 h, and decreased subsequently (Figure 4A). Liver NF-B activation was increased by 1 h after LPS injection, peaked at 1 to 2 h, and decreased thereafter (Figure 4B). By EMSA, intraperitoneal LPS caused rapid activation of NF-B in the liver and peak NF-B activity in the liver that precedes peak activation in the lung. Despite the difference in timing of the peak NF-B response in the lung and liver, luciferase activity reached maximum at 4 h in both organs in these experiments. The observation that NF-B activation measured by EMSA precedes increased tissue luciferase activity is expected because luciferase activity in these mice should be dependent on NF-B-induced transcription of the luciferase transgene.

View larger version (48K): [in this window] [in a new window]   Figure 4.   EMSA for NF-B binding activity in nuclear protein extracts for (A) lung and (B) liver. Tissues were harvested at 0, 1, 2, 4, 6, and 24 h after a single intraperitoneal injection of 1 µg/g of LPS. The specificity of protein binding in bands A and B for the NF-B motif is shown by cold and nonspecific competition. Addition of 50 ng of unlabeled oligonucleotide (cold, C) containing an intact NF-B binding site successfully competed for protein binding and eliminated both bands, but addition of 50 ng unlabeled nonspecific oligonucleotide (NS) did not affect binding. Supershifts were done using the 2-h lung sample, to further identify the banding pattern. Band A contains p65 (RelA), because addition of antibody resulted in a supershift and elimination of this band. Addition of antibody to p50 resulted in a supershift and decreased intensity of both bands (but most prominently band B). Antibody to cRel had no effect on the banding pattern. EMSAs were repeated for each tissue with similar results.

In HLL Mice, Inducible Luciferase Activity Is Dependent on NF-B Activation

To confirm that inducible luciferase activity in tissues from HLL mice is dependent on NF-B activation, we specifically inhibited NF-B activation in the liver by expressing a trans-dominant inhibitor of NF-B activation in liver tissue and assessing the effect on LPS-induced luciferase activity. Several studies have shown that a mutation in IB-, which removes or substitutes critical serine residues involved in signal-induced phosphorylation, creates an IB- protein which behaves as a trans-dominant inhibitor of the NF-B complex (termed IB-DN) (12). Because the serine targets for phosphorylation are missing, the IB-DN can efficiently bind NF-B in the cytoplasm, but cannot be inactivated or degraded in response to physiological or pharmacological signals. This provides an effective means by which to sequester NF-B in a bound, cytoplasmic state. Replication-deficient adenoviruses with an avian IB-DN construct were made and 5 × 10⁹ PFU were administered intravenously to HLL mice. Expression of the IB-DN in these mice was detected exclusively in the liver by immunoblotting, using a specific antibody to avian IB- (Figure 5). Intravenous administration of adenoviral vectors (expressing IB-DN or -gal) alone did not cause increased luciferase in any organ at 24 to 96 h after injection (data not shown), which indicates that this method of intravenous gene delivery is not, by itself, associated with activation of NF-B. Table 1 shows the results of an experiment in which adenoviral vectors expressing IB-DN or -gal were injected intravenously at 5 × 10⁹ PFU in HLL mice, followed by intraperitoneal LPS (3 µg/ g) 48 h later. Four hours after intraperitoneal LPS, the group that was treated with adenovirus-IB-DN had significantly lower LPS-induced luciferase activity in the liver compared with mice that received adenovirus--gal followed by LPS (p < 0.05). Interestingly, LPS-induced luciferase activity was similar in both treatment groups in all other organs tested, indicating that inhibition of luciferase production was specific to the liver, which was the only site where we found transgene expression in mice intravenously injected with adenoviral vectors. Mice treated with LPS alone had luciferase activity that was similar to mice pretreated with adenovirus--gal in all organs (data not shown). These findings show that NF-B activation is required for LPS-induced expression of luciferase in HLL mice and that luciferase can be measured as a surrogate marker for NF-B activation in these mice.

View larger version (13K): [in this window] [in a new window]   Figure 5.   Western blot for IB-DN in various organs 48 h after intravenous injection of 5 × 109 PFU of adenoviral vectors expressing avian IB-DN. Expression of the IB-DN is identified exclusively in liver homogenate using a specific antibody to avian IB-.

Intraperitoneal LPS Results in Increased Messenger RNA (mRNA) Expression for NF-B-dependent Cytokines and Elevated Concentrations of NF-B-dependent Cytokines in Serum and Tissue Homogenates

We measured mRNA expression of a variety of chemokines and cytokines in lung tissue after intraperitoneal LPS in HLL mice using multiprobe RPA (PharMingen, San Diego, CA). Figure 6A illustrates lung tissue mRNA expression of a panel of chemokines. Several chemokines, whose expression is thought to be regulated by NF-B, demonstrated increased mRNA production after intraperitoneal LPS (15). RANTES (regulated upon activation, normal T-cell expressed and secreted), eotaxin, macrophage inflammatory protein-1 (MIP-1), MIP-2, interferon-gamma-inducible protein-10 (IP-10), and monocyte chemotactic protein-1 (MCP-1) were upregulated after intraperitoneal LPS injection, but with various kinetics. In the untreated control, only RANTES mRNA was detected. Eotaxin, MIP-1, MIP-1, MIP-2, IP-10, and MCP-1 were induced from 1 to 6 h after IP LPS and returned to baseline values by 24 h. Gene expression of MIP-2, a CXC chemokine, was increased in the lung by 1 h after intraperitoneal LPS, remained increased at 2 h, and decreased by 4 to 6 h. Of the non-NF-B dependent chemokines tested, lymphotactin and T-cell activation gene-3 (TCA-3) were not detected, and MIP-1 was inducible with similar kinetics to MIP-1. L32 and reduced glyceraldehyde-phosphate dehydrogenase (GAPDH) were included as constitutively expressed messages, to assess equality of total RNA in each sample.

View larger version (25K): [in this window] [in a new window]   View larger version (23K): [in this window] [in a new window]   Figure 6.   RPA for the time course of expression of a panel of (A) chemokines or (B) cytokines in lung tissue of mice treated with 1 µg/g of LPS. Bands were quantitated by laser densitometry and the spectrum of chemokine or cytokine mRNA signals is shown as a linear histogram. The y-axis shows optical density (O.D.) of the bands in arbitrary units and the x-axis shows distance across the gel in millimeters. (A) Intraperitoneal LPS-induced expression of mRNA for RANTES, eotaxin, MIP-1, MIP-1, MIP-2, IP-10, MCP-1, and the constitutive mRAs, L32, and GAPDH, is shown. Lymphotactin and TCA-3 were not detected. (B) Expression of mRNA for TNF-, LT-, TNF-, IL-6, IFN-, TGF-1, TGF-2, L32, and GAPDH is shown.

Gene expression of a panel of cytokines in lung tissue after intraperitoneal LPS is shown in Figure 6B. TNF- and IL-6 mRNA were induced at 2 h after intraperitoneal LPS and were decreased by 6 h. Interferon- (IFN-) was inducible by intraperitoneal LPS at 2 h after treatment, and TNF- mRNA was minimally increased by intraperitoneal LPS. Lymphotoxin  (Lt-) mRNA was present in untreated controls and not markedly increased by LPS treatment. Transforming growth factor-1 (TGF-1) and TGF-2 were expressed at baseline and were minimally altered by intraperitoneal LPS. Of the chemokines and other cytokines tested, presumed NF-B-dependent genes were prominently upregulated by intraperitoneal LPS, including TNF-, TNF-, IL-6, RANTES, eotaxin, MIP-1, MIP-2, IP-10, MCP-1, and IFN- (15). Expression of non-NF-B-dependent genes, with the exception of MIP-1, was not upregulated after intraperitoneal LPS. Upregulation of mRNA expression of these NF-B-dependent cytokines by intraperitoneal LPS treatment occurs at early time points (within 4 h) and correlates with early NF-B activation by gel shift and with upregulation of NF-B-dependent luciferase in lung tissue.

We measured concentrations of TNF-, IL-6, and KC in serum after intraperitoneal LPS (1 µg/g). These cytokines are NF-B-dependent in vitro (21, 22, 26) and may mediate some LPS-induced effects. Significantly elevated cytokine levels were detected in mouse serum after LPS injection (Figure 7A), but no differences were noted between HLL transgenic mice and nontransgenic littermates. TNF- concentrations peaked 1 to 2 h after LPS, with rapid decline by 4 h. IL-6 production had a later onset, with concentrations peaking at 2 to 4 h after LPS, decreasing by 6 h, and returning to baseline values by 24 h. KC, a CXC chemokine and neutrophil chemoattractant, also had a delayed rise in serum compared with TNF-. Peak KC levels occurred 4 h after LPS injection, remained significantly elevated at 6 h, and returned near baseline values by 24 h. At 4 h after intraperitoneal LPS, serum KC and IL-6 both increased in a dose-dependent manner (Figure 7B). Interestingly, serum TNF- peaked and receded before peak NF-B-dependent luciferase activity in the lung and liver, whereas serum concentrations of IL-6 and KC correlated better with NF-B-dependent luciferase protein production in these organs.

View larger version (21K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   Figure 7.   Time course (A) and dose-response (B) for the appearance of KC, IL-6, and TNF- in serum after treatment of mice with intraperitoneal LPS. (A) Detection of immunoreactive TNF- (triangles), IL-6 (squares), and KC (circles) in serum is shown as the mean (± SEM) serum concentration at 1, 2, 4, 6, and 24 h after intraperitoneal injection of 1 µg/g of LPS (n = 5 each time point). (B) Immunoreactive IL-6 (squares), and KC (circles) in serum 4 h after intraperitoneal injection of 0, 0.1, 0.3, 1.0, and 3.0 µg/g of LPS (± SEM) (n = 5 for each treatment group).

Because KC and related CXC chemokines are thought to be important in attracting neutrophils to the lung in response to inflammatory stimuli, we measured KC concentration in lung homogenates (Figure 8A). These measurements were made by ELISA from tissue homogenates obtained 4 h after intraperitoneal LPS. KC showed a dose-dependent increase in tissue concentration after intraperitoneal LPS. The LPS-induced increase in lung KC concentration mirrored serum measurements, showing increasing KC concentrations to 1 µg/g of LPS and no further increase at 3 µg/g of LPS. Comparison of tissue KC levels to NF-B-dependent luciferase activity revealed a linear relationship between lung KC and lung luciferase activity (Figure 8B).

View larger version (13K): [in this window] [in a new window]   View larger version (21K): [in this window] [in a new window]   Figure 8.   (A) Dose response for the appearance of KC in lung tissue after intraperitoneal LPS. Immunoreactive KC in lung tissue homogenates 4 h after intraperitoneal LPS is reported as ng/mg total protein concentration (± SEM, n = 3 for each treatment group). (B) Correlation between luciferase activity and immunoreactive KC in lung tissue after treatment of mice with intraperitoneal LPS. Lung luciferase activity is shown as mean RLU (± SEM), and lung KC is shown as mean concentration per milligram of total protein (± SEM). The dose of LPS (0, 0.1, 0.3, 1.0, and 3.0 µg/g) used to acquire each data point is labeled.

Intraperitoneal LPS Results in Neutrophilic Lung Inflammation

Intraperitoneal LPS resulted in a time- and dose-dependent accumulation of neutrophils in the alveolar space, as detected by lung lavage. In lung lavage from control animals and mice killed 6 h after intraperitoneal LPS, less than 1% of lung lavage cells were neutrophils. By 24 h, approximately 8% of lavaged cells were neutrophils in mice treated with 1 µg/g or 3 µg/g intraperitoneal LPS (p < 0.05 compared with untreated control animals) (data not shown). Intraperitoneal LPS at a dose of 0.3 µg/g did not result in neutrophilic alveolitis at 24 h. Total lavage cell counts were similar in all treatment groups (data not shown). In these mice, neutrophilic alveolitis occurred after LPS doses that also increased lung luciferase activity, suggesting that local NF-B activation in the lungs is required for signaling neutrophil immigration into the air spaces after intraperitoneal LPS.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Because production of many inflammatory cytokines and adhesion molecules in vitro is regulated by NF-B, we wanted to develop a model for quantifying the functional consequences of NF-B activation in multiple organs in the setting of systemic inflammation induced by intraperitoneal LPS. Therefore, we used a recently generated line of transgenic mice expressing Photinus luciferase cDNA under the control of a NF-B-dependent promoter. We thought that NF-B-dependent luciferase production would reflect NF-B activity over time and allow a more detailed analysis of the role of NF-B activation in regulating inflammatory events.

We found that luciferase activity in tissues of HLL mice was relatively low in untreated mice, except in the forebrain. After intraperitoneal LPS, luciferase activity peaked at 4 h in the lung, liver, and kidney. On further analysis, we showed that both the lung and liver have distinct thresholds for LPS-induced luciferase activation. In these organs, NF-B activation, as measured by EMSA, preceded elevations in luciferase activity, which would be expected because NF-B activation is required for the production of luciferase in HLL mice. In addition, we showed that luciferase activity could be measured in select cell types by determining that luciferase activity in lung lavage macrophages was increased by 1 h after intraperitoneal LPS.

In HLL mice, we demonstrated that LPS-induced increases in luciferase activity are dependent on NF-B activation by using a trans-dominant NF-B inhibitor. By specifically expressing IB-DN in the liver, we were able to suppress LPS-induced luciferase expression in this organ without affecting luciferase activity in other organs. Interestingly, serum levels of IL-6 and KC were similar in mice treated with IB-DN + LPS and those treated with adenovirus--gal + LPS or LPS alone (data not shown). These findings suggest that NF-B activation in liver is not critical for determining NF-B-dependent luciferase activity in other organs in this model. In addition, high level liver NF-B activation is either not required for maximal serum levels of IL-6 and KC, or the liver is not the major source of these cytokines in circulation after intraperitoneal injection of LPS.

In addition to increased NF-B activation and NF-B-dependent luciferase production, intraperitoneal LPS resulted in upregulation of mRNA expression of an array of cytokines in lung tissue thought to be NF-B-dependent. Transcriptional regulation of these cytokines is a complex series of events related to the interaction of NF-B with other transcription factors, as well as non-NF-B-related factors. In these experiments, expression of these potentially NF-B-dependent cytokines occurred at early time points (by 4 h) that correlated reasonably well with increased lung luciferase activity.

We measured serum concentrations of KC and IL-6, two presumed NF-B-dependent cytokines. Serum levels of IL-6 and KC were time- and dose-dependent, with peak concentrations 2 to 4 h after intraperitoneal doses of 1 to 3 µg/g of LPS. Interestingly, serum levels of TNF- peaked by 1 h and receded by 4 h, markedly preceding peak tissue luciferase activity. In lung homogenate, KC concentration was dependent on LPS dose. Because KC and related chemokines bind to heparin and heparin sulfates in extracellular matrix, KC is sequestered in the lung, which has a large volume of extracellular matrix (27). Lung KC and related CXC chemokines may be critical for creating a chemotactic gradient favoring neutrophil influx into the lungs, a process that is thought to be important in the induction of LPS-induced lung injury (30). In these studies and in previous work (1), we have found that the accumulation of chemokines in the lung substantially precedes neutrophilic alveolitis after intraperitoneal injection of LPS. This finding may be explained by the formation of chemotactic gradients favoring immigration of neutrophils toward the air space at 24 h but not at 4 to 6 h after LPS (30). In HLL mice, we found that lung luciferase was linearly related to lung KC content, suggesting that local production of chemokines provides the link between NF-B activation and LPS-induced lung inflammation that we have previously reported (1).

We measured neutrophil accumulation in the alveolar spaces at 24 h after intraperitoneal LPS. At doses of intraperitoneal LPS that increased lung luciferase and maximized lung KC content (1 to 3 µg/g), neutrophil accumulation in the air spaces was found. However, a lower dose of intraperitoneal LPS (0.3 µg/g) did not result in neutrophilic influx into the alveolar space at 24 h, even though this dose of intraperitoneal LPS did cause increased NF-B-dependent luciferase activity in the liver. This finding suggests that local NF-B activation and inflammatory mediator production in the lungs may be required for subsequent neutrophilic alveolitis.

Recently, NF-B activation has been implicated as an important factor in humans with acute respiratory distress syndrome (ARDS), which is characterized by neutrophilic lung inflammation and diffuse alveolar damage, and can result from systemic inflammation. Schwartz and colleagues (31) reported that NF-B is activated in alveolar macrophages from patients with ARDS to a significantly higher degree than in alveolar macrophages from critically ill patients with other diseases. In addition, NF-B activation may be important in the pathogenesis of sepsis. Bohrer and colleagues (32) reported that in peripheral blood monocytes of patients with sepsis, NF-B activation correlates with mortality. Specifically, all patients in that study who died with sepsis had increased NF-B activation (greater than twice baseline) in the first 6 d, whereas all patients who survived had NF-B activation that remained less than twice the baseline value at each time point during the 14-d study period. These and other data suggest that the pathobiology of sepsis and ARDS is related to exuberant production of NF-B-dependent proinflammatory molecules, leading to inflammatory cell influx, cell activation, and tissue injury in response to bacterial products such as LPS. Further definition of the role of NF-B in this complex series of events should increase our understanding of the systemic inflammatory response to LPS, which may lead to novel treatment strategies.