Main

DEX is a commonly used drug in the treatment of chronic lung disease that may follow respiratory distress syndrome of prematurity (1, 2) or hypoxemic respiratory failure in near-term and term infants. Although DEX is effective in preventing neutrophilic infiltration that results in chronic lung disease (3–6), the exact mechanism of its action is poorly understood. In addition, there are many side effects associated with dexamethasone therapy including increased incidence of sepsis, hyperglycemia, hypertension, and poor growth (1, 2, 7).

Traditionally neutrophils have been viewed as the effector cells of the acute inflammatory response (8). However, previous work using neutrophils from adults (9, 10) and our own pilot studies using neutrophils from newborns (11), indicate that the neutrophil may induce neutrophil recruitment by a mechanism involving release of the cytokine, IL-8, providing a positive feedback mechanism for acute inflammation.

Our purposes were first to determine whether DEX inhibits neutrophil migration directly or by inhibition of neutrophil-induced, neutrophil recruitment due to inhibition of IL-8 release, from neutrophils; and second, if there were developmental differences regarding the action of DEX on neutrophil migration.

METHODS

Neutrophil isolation and culture.

Fresh whole blood (16 mL) was obtained from healthy adult human volunteers or from the umbilical vein of the placentas from healthy full-term infants immediately after delivery by elective cesarean section without general anesthesia. Blood was collected in heparinized preservative-free tubes. The study was approved by the Human Subjects Review Committee of Long Island Jewish Medical Center.

Neutrophils were isolated using a modification of the method previously described by Boyum (12). The plasma, mononuclear cells, and platelets were separated from other cells by Ficoll density centrifugation (Ficoll-Paque PLUS, Pharmacia Biotechnology, Piscataway, NJ). Erythrocytes were separated from neutrophils by sedimentation in 5% Dextran and contaminating red blood cells were eliminated by lysis in a solution of 0.15 M NH4Cl, 0.01 M NaHCO3, and 0.01 M EDTA. The recovered PMN were washed three times in Hanks' solution (without calcium and magnesium), and resuspended in complete culture media (RPMI 1640 with 25 mM HEPES buffer and with L-glutamine, GIBCO, Grand Island, NY). The purity of the PMN isolation was determined by cytocentrifugation of neutrophil suspension followed by counterstaining with Diff-Quick (Baxter Healthcare Corp., McGaw Park, IL). Differential counts, which were performed by counting 200 cells, resulted in a purity of >94% neutrophils and >99.5% viability by trypan blue exclusion. Neutrophils were cultured in polystyrene multiple-well tissue culture plates at concentrations of 1 × 106 neutrophils/mL in 37°C, humidified 95% air/5% CO2. Neutrophil viability was 92–100% and 98–100% for PMNs from the newborn and adult respectively, after 18 h of culture.

IL-8 assay.

Extracellular immunoreactive IL-8 was quantitated by using a commercially available sandwich enzyme immunoassay technique (Quantikine, R&D Systems, Inc., Minneapolis, MN). The intra- and interassay coefficient of variation was 4.9 and 14%, respectively, based on assays performed in our laboratory. The minimum detectable level of IL-8 was 15 pg/mL.

Neutrophil migration assay.

A neutrophil migration assay, as previously described (13), was used to determine the biologic activity of the immunologically reactive IL-8, detected by ELISA from culture media as well as migration to exogenous, recombinant human IL-8. Each chemotaxis assay was performed in duplicate using a 48-well micro Boyden chamber (Neuro Probe, Cabin John, MD) with 3-μm pore size polycarbonate filter (Poretics, Livermore, CA). After incubation in the chemotaxis chamber at 37°C for 60 min, migrated cells on the filter were counted at 400x magnification. The neutrophil response was expressed as the average number of cells counted in a minimum of 5 high power fields after the values for the buffer control were subtracted.

Study protocols.

First, we determined the direct effect of DEX on neutrophil migration. Freshly isolated neutrophils (5 × 105/50 μL from adults and newborns) were mixed with DEX (final concentration of 10−10 to 10−5 M, control had no DEX) placed in the upper wells of the chemotaxis chamber. A standard solution of recombinant human IL-8 (10−8 M) was diluted in RPMI 1640 medium and added to the lower chamber. The molecular weight of IL-8 is approximately 8 kDa and therefore a 10−8 M solution corresponds to an 80 ng/mL solution. This standard concentration is a submaximal concentration for IL-8-induced chemotaxis based on previous work (14). Cells were incubated in the chemotaxis chamber for 1 h at 37°C.

Second, we studied the effect of DEX on neutrophil IL-8 release by neutrophils from adults and newborns during exposure to LPS. Neutrophils were incubated with a submaximal dose of LPS (1 ng/mL LPS based on pilot studies, L 3024, Sigma Chemical Co., St. Louis, MO) and DEX (10−9 to 10−5 M) for 6 and 18 h. After culture media were sampled, the aliquot from a individual experiment (nonpooled) was spun (3000 rpm for 12 min), and the supernatant was stored at −80°C until assayed for IL-8 concentration and chemotactic activity.

Finally, we measured the chemoattractant activity produced by neutrophils in culture media from the former experiments. Samples of culture media were placed in the lower chamber and freshly isolated adult or newborn PMNs (5 × 105 in 50 μL) were placed in the upper chamber. The experimental samples were preincubated (30 min, 37°C) with anti-human IL-8 antibody in 1:500 dilution (purified mouse monoclonal IgG antibody, R&D Systems, Inc., Minneapolis, MN) or with control IgG1 isotype antibody in the same dilution (Sigma).

Statistical analysis.

Data were analyzed by using an InStat software package (GraphPAD, San Diego, CA). Data were expressed as mean ± SEM and the differences among doses were evaluated by one-way analysis of variance. The Bonferroni t test was used for multiple comparisons between groups. Data were considered significant if p< 0.05.

RESULTS

A 1-h exposure of neutrophils to serial concentrations of DEX (10−10 to 10−5 M) did not inhibit their migration to an exogenous standard of IL-8. In addition, no differences in neutrophil migration were observed between neutrophils of the newborn and adult. The migration to IL-8 [10−8 M] without DEX was 344 ± 16 PMNs/hpf and 368 ± 21 PMNs/hfp (mean ± SE) for neutrophils from the newborn (n= 3) and adult (n= 3), respectively.

The effect of DEX, concurrent with 6 or 18 h of incubation with LPS, on neutrophil IL-8 release is shown in Figure 1A (adult) and Figure 1B (newborn). Note that IL-8 release was corrected to the number of PMNs (106); however, there were approximately 106 PMNs/mL of culture media by design. For both the adult and newborn, IL-8 media levels increased approximately 6-fold from 6 to 18 h of incubation. However, the absolute increase in IL-8 media levels were 20- to 30-fold higher for neutrophils form the newborn versus adult. In relative terms, DEX had the same inhibitory effect on IL-8 release from neutrophils of the newborn and adult; concentrations of DEX as low as 10−8 M resulted in significant reductions in IL-8 levels. The relative decrease in IL-8 release associated with varying levels of DEX was similar between the 6- and 18-h time points. For neutrophils from adults, there were 87 and 99% reductions in IL-8 release during the 18-h incubations using 10−8 and 10−7 M DEX, respectively. For neutrophils from the newborn, there were 55 and 92% reductions in IL-8 release during the 18-h incubation using 10−8 and 10−7 M DEX, respectively. Concentrations of DEX higher than 10−7 M did not further increase inhibition of IL-8 release for either the newborn or adult.

Figure 1
figure 1

A and B, The effect of DEX dosage on release of IL-8 from neutrophils of adults (A) and newborns (B) upon in vitro exposure to LPS in vitro. IL-8 release was assessed at 6 and 18 h of exposure to LPS with and without concomitant exposure to DEX. The vertical axis for the newborn is on a much greater scale than the axis for the adult.

Full size image

The low levels of IL-8 produced by PMNs of adults stimulated with LPS (with or without DEX exposure) were not associated with any detectable neutrophil migration. In contrast, neutrophils of the newborn produced enough IL-8 to result in appreciable neutrophil migration, especially after 18 h of incubation with LPS. Although a trend can be observed in Figure 2A representing experiments performed after 6 h of incubation similar to the 18-h results, no significant inhibition of neutrophil migration could be claimed for IL-8 antibody or DEX. However, in Figure 2B it is demonstrated that by 18 h of incubation with LPS, neutrophils from newborns produced a substance in the media, which caused high neutrophil migration activity (270 ± 33 PMNs/hpf). The neutrophil migration activity of media for the 18-h incubation with LPS was significantly reduced by 88% if DEX (10−7 M) was present. Similarly, IL-8 antibody resulted in a significant reduction (86%) in neutrophil migration activity. This effect was specific, because the control IgG antibody did not affect neutrophil migration; migration was 252 ± 27 and 246 ± 42 PMNs/hpf for media without and with control IgG antibody, respectively (p= 0.9, data not depicted on Fig. 2B). The combined effect of both DEX and IL-8 antibody on neutrophil migration activity was not statistically greater than either agent alone as shown in Figure 2B.

Figure 2
figure 2

A and B, The neutrophil migration activity by culture media of neutrophils isolated from cord blood after 6 h (A) and 18 h (B) of exposure to LPS, with and without concomitant exposure to IL-8 antibody, DEX, or both.

Full size image

DISCUSSION

Our study demonstrates that the neutrophil from the newborn, on exposure to LPS, has the ability to create a positive feedback loop for neutrophil migration, via a mechanism, which principally involves IL-8 release. DEX had no dose-related, direct effect on neutrophil migration to a standard of recombinant human IL-8, for either the newborn or adult. In our studies no direct effect of DEX was observed in a dose range of 10−10 to 10−5 M during the 1-h incubation to allow PMN migration toward the standard of IL-8. Suprapharmacologic doses of DEX 10−3 M have been reported to produce an acute direct inhibition of PMN migration but this may reflect nonspecific cell injury rather than a receptor-mediated mechanism (15).

In contrast DEX suppressed endogenous release of IL-8 from PMNs (exposed to LPS) from the adult and newborn as well as the associated chemotactic activity observed for the newborn. The chemotactic activity of culture media, for PMNs stimulated by LPSs, may be the result of several different classes of chemoattractants. The neutrophil has been shown to have the capacity to produce a number of cytokines and bioactive lipids including tumor necrosis factor α, granulocyte colony-stimulating factor, macrophage colony-stimulating factor, IL 1, IL 6, interferon-α, leukotriene B4, and platelet-activating factor (8–10). The time course of their release and the relative abilities to produce neutrophil migration is incompletely understood. We found that the culture media migration activity possibly at 6 h and definitely at 18 h of PMN exposure to LPS appeared to be almost completely abolished by IL-8 antibody, DEX, or both. The similar level of inhibition of media migration activity by IL-8 antibody and DEX (10−7 M) indicates that DEX inhibits neutrophil-induced neutrophil migration primarily via an IL-8-mediated mechanism. Leukotriene B4 and platelet-activating factor, which are produced by neutrophils, act as neutrophil chemoattractants, and have synthetic pathways inhibited by corticosteroids, may be short acting and rapidly inactivated in similar experimental conditions (16, 17). Our findings are of interest because IL-8 is one of the most potent neutrophil mediators known (14). In addition, a dominant role of IL-8 in pulmonary inflammatory conditions, such as chronic lung disease of prematurity and acute respiratory distress of adults, was shown by IL-8 antibody inhibition in bronchial lavage fluid (4, 18–20).

One surprising finding in this study was that both neutrophils from the adult and newborn exhibited the same degree of migrational activity to exogenous IL-8 in a chemotaxis chamber. This result is different from the many other previous studies, including our own, which demonstrated that in vitro, the newborn has a lower neutrophil chemotactic activity than the adult, when cells are exposed to a variety of chemoattractants in vitro (13, 14). Our only explanation at this point is that in our experiment we used a culture media as opposed to previous work, which used a physiologic salt solution to suspend neutrophils as well as the chemoattactant. A second surprising finding was that under our study conditions there was a much greater IL-8 release from neutrophils of the newborn compared with the adult, upon stimulation with LPS. One may speculate that the in vitro conditions primed the PMNs from the newborn to respond to LPS to a greater degree or that LPS receptors were increased for the newborn. In our pilot studies we did not find that at baseline, before incubation with LPS, there were any differences between age groups for IL-8 release; for both groups IL-8 release was undetectable by immunoassay. Whatever the mechanism, it is of interest that exogenous factors may be able to modulate IL-8 release from stimulated neutrophils.

The mechanism of how DEX inhibits neutrophil-induced neutrophil chemotaxis may be an important avenue for future research in the area of inflammatory diseases. It is surprising that there is very little work on the molecular mechanism of DEX's action, particularly in neutrophils. Based on work using other inflammatory cells, particularly the macrophage, production of IL-8 may be part of a general cytokine release mechanism, controlled by NF-κB and activator protein-1. These transcription factors lead to the release of proinflammatory cytokines, including TNF-α, IL-1, IL-6, and IL-8, that can be inhibited by glucorticoids (21, 22). The inhibitory mechanism of glucorticoid on proinflammatory cytokine production may involve several mechanisms that may vary according to effector cell type. Glucorticoids appear to either directly inhibit NF-κB, enhance the production of a specific inhibitor of NF-κB, or inhibit transcription of a group of proinflammatory cytokines, including IL-8 by direct binding of the activated, cytoplasmic glucocorticoid receptor to glucocorticoid responsive elements in the promoter regions of responsive genes (17, 23, 24). In addition there is evidence that the cytoplasmic glucorticoid receptor can also function in an DNA-independent manner to inhibit cytokine gene expression through protein-protein interactions between glucocorticoid receptor and the transcription factors NF-κB and activator protein-1, rendering the factors inactive (25).

DEX has been shown to be efficacious in the treatment of chronic lung disease in newborns. However, the timing of therapy, the duration of therapy, and the elimination of serious side effects remains problematic (1, 26–32). Under our in vitro study conditions, DEX produced maximal inhibition of IL-8 release and neutrophil migration at 10−7 M concentration. In vivo experiments in adults also demonstrated that PMN chemotaxis is inhibited at serum DEX concentrations of approximately 1.3 × 10−7 M (15). However, DEX concentrations measured by liquid chromatography were reported in the 10−6 M range for extremely low birth weight neonates who received the commonly used bolus doses of i.v. DEX (0.28–0.40 mg/kg) (33). The 10−6 M range of DEX observed in the above clinical setting is ten times higher than our in vitro maximal inhibitory DEX concentration. Preliminary results for two recent clinical trials also suggests that low-dose DEX therapy in ventilator-dependent, low birth weight infants is as effective as the higher doses (34, 35). Taken together, this information suggests that infants with chronic lung disease may need lower doses of DEX than routinely used, which may result in fewer side effects.

In summary, the neutrophil of the newborn has the ability to amplify acute neutrophilic inflammation, principally by the release of IL-8. This positive feedback mechanism may play an important role in the development of chronic lung disease in the newborn, or play a role in host defense against microbial pathogens. DEX appears to have an indirect effect on neutrophil migration to a site of inflammation, in part by inhibiting neutrophil-induced neutrophil recruitment. Our in vitro model may be useful for studying the molecular mechanism underlying DEX's ability to inhibit proinflammatory cytokines, so more selective and safer therapy can be developed to treat chronic lung disease in the newborn. DEX's molecular action involves control of transcription for a group of proinflammatory cytokines including IL-8, which have been implicated in the pathogenesis of BPD. Using IL-8 release from the neutrophil as a reflection of DEX's activity, our study suggests that a 10-fold lowering of the standard DEX dose may adequately reduce lung inflammation, thereby reducing DEX's serious adverse effects in the treatment of BPD.