Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

With both the incidence and severity of asthma increasing dramatically in developed countries, understanding the events underlying inflammatory diseases of the lung is becoming more pertinent. Historically it was believed that atopic or allergic asthma was a disease characterized primarily by the presence of IgE and allergen-dependent masT cell activation (1). Recently, persistent asthma, as well as other allergic lung diseases such as extrinsic allergic alveolitis, has been recognized as a chronic inflammatory lung disease (2, 3). With persistent asthma, two distinct inflammatory patterns have been recognized, i.e., eosinophilic asthma and noneosinophilic asthma. Although the latter is the more common pattern of inflammation, myeloperoxidase (MPO) is elevated in both patterns of persistent asthma (4). Sputum interleukin-8 (IL-8) was highest in patients with noneosinophilic asthma. In addition, neutrophilic influx correlated with the presence of IL-8 (4).

With another form of asthma, sudden-onset fatal asthma, the PMN has also been implicated (5). This scenario can result in patient death within hours of the onset of an asthma attack. The ratio of PMNs to eosinophils in these patients is ~ 2 to 1 (5). Furthermore, patients diagnosed with mild cases of asthma exhibited greater numbers of PMNs relative to eosinophils during the early events of the attack (6, 7).

In other pulmonary diseases such as pneumonia, neutrophil recruitment and activation has also been associated with oxidative tissue damage (8). The implication of increased PMN recruitment is the release of their granule-associated proteins such as myeloperoxidase (MPO) into the local environment (8). Once MPO is in the microenvironment, proteases and pH changes rapidly inactivate MPO (9). Therefore, at a site of inflammation, both MPO as well as enzymatically inactive MPO (iMPO) would be present (10). The current investigators reported previously that MPO as well as iMPO affect certain peritoneal macrophage (Mø) functions (11, 12). Conversely, it has been shown by other investigators that the Mø, which constitute 85% of the cells obtained by bronchoalveolar lavage (13), directly influence a neutrophil-dependent inflammatory response via the secretion of tumor necrosis factor- (TNF-), IL-8, as well as other cytokines (14). Because alveolar Mø (AMø) function as antigen-presenting cells and a source of several cytokines, these cells are considered to be an important link between innate and adaptive immunity in the lung (15). In particular, AMø-derived TNF- and IL-8 recruit PMNs to a site of inflammation (4, 14). Therefore, the above indicate that there is "cross talk" between the Mø and neutrophil that enhances the process of inflammation.

Although the role of MPO is well documented in the cytotoxic triad, the present investigators have shown that this enzyme is also an immunoregulatory molecule (11, 12, 16). Previous studies by these investigators have shown that either MPO or iMPO stimulate murine peritoneal Mø to produce reactive oxygen intermediates (ROI) (11, 12). Furthermore, it has been demonstrated that iMPO stimulated rat peritoneal Mø to secrete the inflammatory cytokine, TNF- (17). Other investigators have shown that AMø acquire peroxidase activity in the course of pulmonary inflammation, as demonstrated by incubating AMø with peritoneal exudate PMNs (18). Additionally, it is believed that AMø internalize MPO and utilized it to aid in killing pathogens (19).

The present study was undertaken to determine if either MPO or iMPO could also stimulate AMø to secrete various substances involved in inflammation. Elucidating this process may explain the mode by which PMN-Mø interaction could exacerbate inflammatory diseases of the lung such as asthma and extrinsic allergic alveolitis. Furthermore, it may provide information regarding rapid and massive tissue damage associated with lung infection.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Age-matched female Lewis rats weighing ~ 180 g were obtained from Harlan Sprague Dawley, Indianapolis, IN. Animals were cared for and housed in a facility according to the guidelines of the Animal Welfare Act. Rats were anesthetized using halothane inhalation. Halothane has been shown not to affect the functions of AMø (20).

Materials

Myeloperoxidase and iMPO were both supplied by Nicole Moguilevsky and Alex Bollen, Université Libre de Bruxelles, Gosselies, Belgium. Enzymatically active MPO contained 560 µg/ml protein as determined by Lowry method and 135 U/ml activity as determined by o-dianisidine assay (21). Using the same methods, stock solutions of iMPO had 3,000 µg/ml of protein. With regard to iMPO, working solutions of the iMPO contained  0.01 U/ml. Enzymatically inactive MPO was obtained by site-directed mutagenesis resulting in amino acid substitutions within the heme moiety that rendered the enzyme inactive (22). Various concentrations and times for MPO exposure were employed. Precise information is described throughout the paper. Dulbecco's modified Eagle's medium (DMEM) was purchased from Life Technologies (Rockville, MD). Gentamicin sulfate, bovine serum album fraction V essentially globulin free (BSA), dimethyl sulfoxide (DMSO), zymosan, and HEPES were purchased from Sigma (St. Louis, MO). Phosphate buffered saline (PBS) solution pH 7.2 was prepared as needed. Fetal bovine serum (FBS) was purchased from Intergen (Purchase, NY). Luminol was purchased from Eastman Kodak (Rochester, NY). All reagents were tested for endotoxin contamination using Limulus amoebocyte lysate test (LAL) (Associates of Cape Cod, Woods Hole, MA). The preparations of MPO and iMPO used contained endotoxin levels  0.3 ng/ml. This level of endotoxin did not enhance AMø functions as determined by lipopolysaccharide (LPS) titration experiments.

Alveolar Macrophage Collection

Rats were anesthetized via halothane inhalation. Subsequently, exsanguination was performed by bisecting the renal artery. The thoracic cavity was opened to visualize the lungs. The trachea was cut and a 21 G × 24 butterfly catheter placed in the opening. Lungs were lavaged using warm PBS (37°C). Approximately 10 ml of PBS were lavaged through the lungs at 1 min intervals. A total of 50 ml of PBS were used in the procedure. The cells were washed and resuspended in either DMEM with Phenol Red or media without Phenol Red (Auto-POW). The resident AMø cell number was adjusted to 1 × 10⁶ Mø/ml. A total of 100 µl of the Mø suspension were added to each well of a microtiter plate and incubated at 37°C under 5% CO₂. After incubation, the monolayers were washed extensively to dislodge nonadherent cells. By differential staining, it was determined that  99% of the cells were Mø.

Chemiluminescence Assay

Methods used have been described previously (11). Briefly, 100 µl of AMø, adjusted to 1 × 10⁶ Mø/ml, in media without Phenol Red (Auto-POW), were added to each well of a 96-well clear bottom, white-walled tissue culture plate (Whatman, Clifton, NJ). The media were supplemented with 0.6 g/dl HEPES, sodium bicarbonate 0.2 g/dl (Sigma), and 1.0 g/dl BSA. After 2 h incubation at 37°C and 5% CO₂, the cells were washed three times with Auto-POW media to remove nonadherent cells. Subsequently, the following was added to each well: 50 µl of a 80 µM solution of luminol suspended in DMSO; 50 µL of zymosan opsonized with guinea pig complement and stored at a concentration of 2 × 10⁷ zymosan particles/ml; and 100 µl of media containing the indicated treatments. The relative light units (RLU) were measured immediately in a Dynatech ML3000 plate luminometer. Results were plotted as time versus RLU. The mean of triplicate treatments ± SEM was determined. Each experiment was repeated at least three times.

Phagocytosis and Intracellular Killing Assay

The phagocytosis assay utilized in the study was similar to the one described by Lian and colleagues (23). The original procedure was modified by the current authors to accommodate the use of yeast cells. (12). Resident AMø were collected and suspended in DMEM without gentamicin. The AMø number was adjusted to 1 × 10⁶/ml, and 100 µl of the cell suspension was added to each well of a 16-well tissue culture chamber slide (Nunc Inc., Naperville, IL). Cells were allowed to attach for 2 h at 37°C in 5% CO₂, and nonadherent cells were removed by washing twice with 200 µl warm media. AMø were then directly exposed to either MPO or iMPO for a total of 10 min. Control cultures were exposed to media alone. After this 10 min incubation, cells were washed vigorously and Candida albicans suspended in culture media containing 10% FBS and 25 mM HEPES without gentamicin. The ratio of C. albicans to AMø was 5:1. Following a 60 min incubation period, the cultures were washed extensively to remove noningested yeast. Subsequently, cells were stained with acridine orange (0.1 mg/ml) for 90 sec and counterstained for 40 sec with crystal violet (1 mg/ml). Crystal violet was used to quench the fluorescence of extracellular yeast. For each treatment, a total of 400 cells were counted using a fluorescence microscope at 1,000× magnification. C. albicans that fluoresced green were scored as live and those that fluoresced orange were scored as dead.

To ensure the accuracy of the assay procedure, the following was done: (1) results of phagocytosis and intracellular killing were initially verified using a plate count assay (data not shown); (2) viability of C. albicans was determined by plate count prior to each experiment; and (3) yeast boiled for 30 min prior to each experiment were employed as positive controls (all stained orange). Each experiment was repeated at least three times. Representative experiments are shown in the text.

Enzyme-Linked Immunosorbant Assay

Sandwich TNF- enzyme-linked immunosorbent assay (ELISA) minikit was purchased from R&D (Cambridge, MA). The ELISA experiments utilized manufacturer's optimized procedures. Briefly, Maxisorb 96-well microtiter plates (Nunc Inc., Naperville, IL) were coated with monoclonal antibody specific for the TNF-. Wells were washed using a wash bottle, blocked with media containing BSA, and incubated with culture media supernatants from AMø exposed to MPO or iMPO for the indicated duration for each experiment. Following incubation, wells were again washed and horseradish peroxidase (HRP)-labeled, polyclonal antibody specific for the cytokine of interest was added. After incubation, wells were washed and the amount of HRP-labeled antibody detected with tetramethyl benzidine. Absorbance was read at a wavelength of 450 nm.

RNase Protection Assay

Total RNA utilized in RNase protection assay (RPA) was obtained from 1 × 10⁷ AMø using TRIzol reagent (Life Technologies) as described by the manufacturer. Isolated RNA sample was reconstituted in DEPC-treated H₂O at 1 mg/ml as determined by UV-spectrophotometric analysis. A RiboQuant multiprobe radioactive RPA kit (rCK-1) was purchased from Pharmingen (San Diego, CA). All experiments performed utilized an optimized procedure provided by the manufacturer. Briefly, 5 µl of sample, containing 10 µg RNA, were placed into tubes (Marsh Biomedicals, Rochester, NY) and dried in an Eppendorf vacuum evaporator centrifuge (Fisher, Pittsburgh, PA). The RNA was then hybridized with ³²P-labeled antisense mRNA probes (specific for TNF-, TNF-, IL-1, IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, IL-10, interferon-, and the two housekeeping genes L32 and glyceraldehyde-3-phosphate dehydrogenase [GAPDH]) and an RNase specific for single-stranded mRNA was added to digest nonhybridized mRNA. The final products were resolved on an 8 M urea 6% polyacrylamide gel. The bands on the dried gel were then visualized by autoradiography using Kodak X-Omat film. To ensure differences in cytokine bands were not a result of sample loading errors, densitometry was performed on radiograms and corrected for differences based on the L32 and GAPDH bands. Neither L32 nor GAPDH levels were affected by iMPO or MPO treatment.

Statistical Analysis of Data

One-way analysis of variance and Student-Newman-Keuls multiple comparison tests were performed to determine significance levels among the different treatment groups and controls.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Previous studies done by the present investigators have demonstrated that murine peritoneal Mø treated with MPO exhibited a dose-dependent increase in ROI (11). Employing the same system, iMPO has been shown to induce minimal ROI production (12). To determine if AMø would respond in a similar manner, chemiluminescence (CL) was performed using either MPO or iMPO. In Figure 1A, MPO induced a 6-fold increase in RLU compared with cells exposed to media alone. However, iMPO induced a 10-fold increase over control. Because these results were not consistent with previous findings obtained using peritoneal Mø, two other enzymatically inactive glycosylated proteins, enzymatically inactive horseradish peroxidase (dHRP) and mannosylated bovine serum albumin (M-BSA), were employed. The results of studies in which AMø were exposed to either dHRP or M-BSA were similar to those obtained with iMPO (Figures 1B and 1C). In another set of experiments, mannans, a known ligand of the macrophage mannose receptor (MMR) (24) were added to inhibit the stimulation of AMø by iMPO. The presence of 5 mg/ml of mannans did not diminish ROI production stimulated by iMPO (Figure 1D).

View larger version (34K): [in this window] [in a new window]   Figure 1.   ROI production from AMø in response to MPO, iMPO, dHRP, mBSA, or mannans. Resident AMø were cultured on clear bottom, white 96-well microtiter plates with the following reagents added: 50 µl media alone, 50 µl opsonized zymosan, 50 µl luminol, and 50 µl of the indicated treatments: (A) 42 µg/ml MPO or iMPO, (B) 42 µg/ml dHRP, or (C) 42 µg/ml mBSA or BSA. In D, treatments consisted of 50 µl of one of the following: media alone, 42 µg/ml iMPO, 100 µg/ml mannans, or 42 µg/ml iMPO and 100 µg/ml mannans together. Chemiluminescence was measured at 2-min intervals in luminometer, and the results were plotted as RLU versus time. Each point on the curve represents the mean ± SEM of triplicate cultures.

Once it had been established that iMPO induced a greater RB than MPO, experiments were done to determine if the same pattern prevailed for Mø-mediated phagocytosis and intracellular killing of C. albicans. Although phagocytosis was not greatly altered by the presence of either MPO or iMPO, candidic activity was markedly altered by the presence of either form of myeloperoxidase (Figures 2 and 3). These results indicated that either MPO or iMPO induced equivalent levels of Mø-mediated candidic activity. When AMø were exposed to either form of the enzyme, a dose-dependent response was observed. The lowest dose of either MPO or iMPO induced ~ 20% killing, while the highest dose induced ~ 55% candidic activity (Figures 2B and 3B).

View larger version (20K): [in this window] [in a new window]   Figure 2.   Dose-dependent effects of enzymatically active MPO on the candidic activity of AMø. AMø were cultured on 16-well chamber slides and exposed to the indicated concentration of MPO for 10 min. The media were subsequently removed and cells washed. Media containing C. albicans at a ratio of 5:1 (yeast:AMø) were added to each well for 60 min before being removed and cells washed vigorously to remove uningested yeast. Slides were then stained with acridine orange and crystal violet and the amount of (A) phagocytosis and (B) killing of C. albicans were determined on a fluorescence microscope. Ingested yeast that fluoresced orange were scored as dead, and those that fluoresced green were scored as live. Each experiment was performed in triplicate and each graphed value represents the mean ± SEM of four 100-cell counts per treatment. Expressed statistic significance represents treated versus control. **P < 0.01, ***P < 0.001.

View larger version (22K): [in this window] [in a new window]   Figure 3.   Dose-dependent effects of iMPO on the candidic activity of AMø. AMø were cultured on 16-well chamber slides and exposed to the indicated concentration of iMPO for 10 min. The media were then removed and cells washed. Media containing C. albicans at a ratio of 5:1 (yeast:AMø) were added to each well for 60 min before being removed and cells washed vigorously to remove uningested yeast. Slides were then stained with acridine orange and crystal violet and the amount of (A) phagocytosis and (B) killing of C. albicans were determined on a fluorescence microscope. Ingested yeast that fluoresced orange were scored as dead, and those that fluoresced green were scored as live. Each experiment was performed in triplicate, and each graphed value represents the mean ± SEM of four 100-cell counts per treatment. Statistic significance represents treated versus control. ***P < 0.001.

Reactive oxygen intermediates have been shown to stimulate the production of inflammatory cytokines, including TNF- (25). Studies were conducted to see if the ROI production by AMø correlated with TNF- secretion. However, before these studies, experiments were done to determine what role, if any, the small amounts of LPS present in myeloperoxidase samples contributed to TNF- secretion. Since the working preparations of MPO used contained  0.1 ng/ml of LPS, AMø were exposed to various levels of LPS in media supplemented with FBS and the titer of TNF determined by ELISA (data not shown). These results indicate that the AMø did not respond to the level of LPS present in the working concentrations of MPO or iMPO. Once this was determined, AMø were exposed to 126 µg/ml of MPO, iMPO, or MPO and iMPO simultaneously for 3, 6, and 9 h. After incubation for the stated times, media were collected and the titer of TNF- secretion was ascertained. The results of these studies indicated that peak TNF- secretion was obtained after 6 h incubation (Figure 4). In addition to various time intervals, other experiments were performed utilizing various concentrations of either MPO or iMPO. Figure 5A shows TNF- secretion due to MPO stimulation was not dose dependent. Only at the highest dose of MPO (126 µg/ml) was there a significant response (~ 550 pg/ml) compared with media controls. Figure 5B illustrates that iMPO did induce a dose response in AMø with 126 µg/ml stimulating secretion of ~ 2,200 pg/ml of TNF- compared with controls, which averaged 500 pg/ml. A total of 14 and 42 µg/ml of iMPO stimulated 1,000 and 1,500 pg/ml of TNF-, respectively. Because both MPO and iMPO would be found in the microenvironment, AMø were exposed to equal amounts of both forms of the enzyme. Results of these studies indicate that, rather than an additive effect, there was an inhibition of TNF- titers obtained when compared with MPO or iMPO alone (Figure 4).

View larger version (32K): [in this window] [in a new window]   Figure 4.   The effects of MPO and iMPO on TNF- secretion from AMø over time. AMø were incubated on 96-well plates in the presence of media alone, 126 µg/ml MPO, 126 µg/ ml iMPO, or 126 µg/ml of both MPO and iMPO. After the indicated time, media were removed and assayed for the presence of TNF- by ELISA. Experiments were repeated three times using at least four wells per treatment and are presented as mean ± SEM. Statistic significance is expressed as treated versus control. ***P < 0.001; no significance was found between the MPO and iMPO + MPO treatments at 6 h; significant difference (P < 0.01) existed between MPO and iMPO + MPO treatments at both 3 and 9 h. For all time points, iMPO was significantly different (P < 0.01) than iMPO + MPO.

View larger version (17K): [in this window] [in a new window]   Figure 5.   Dose-dependent effects of MPO or iMPO on TNF- secretion by AMø. AMø were cultured on 96-well plates in the presence of 0, 14, 42, or 126 µg/ml (A) MPO or (B) iMPO. After 6 h, media were collected, and the quantity of TNF- was determined by ELISA. Experiments were repeated three times using at least four wells per treatment and are presented as mean ± SEM. Statistic significance is expressed as treated versus control. ***P < 0.001.

If secretion of TNF- was upregulated by addition of MPO alone, iMPO alone, or both forms of the enzyme added simultaneously, it was likely that mRNA levels for this cytokine would be increased likewise. An RPA determined that either MPO or iMPO enhanced mRNA expression for IL-1, IL-1, and TNF- by 3 h; however, once again, iMPO was a more potent inducer of mRNAs than MPO for the cytokines assayed. Transcripts for the other cytokines were not detectable at the time points analyzed. Also, these studies confirmed that when MPO was combined with iMPO, the mRNA levels for each of the cytokines studied were reduced when compared with iMPO alone (Figure 6).

View larger version (103K): [in this window] [in a new window]   Figure 6.   Alterations in cytokine mRNA levels from AMø exposed to MPO or iMPO. AMø were cultured in 6-well plates with control media (lane 2) or 42 µg/ml MPO (lane 3), 42 µg/ml iMPO (lane 4), or 42 µg/ml of both MPO+iMPO (lane 5) for 3 h. Cells were then lysed and total cell RNA was collected. Lane 1 contains a positive control RNA sample. An RPA was performed with 32P-labeled antisense nucleotide sequences for interleukins-1, -1, -2, -3, -4, -5, -6, and -10, TNF-, TNF-, interferon-, L32, and GAPDH. Resulting protected probes were electrophoresed and the gels dried and exposed to X-ray film. Experiments were repeated three times with a representative shown here.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Neutrophils are one of the first cells to arrive at a site of inflammation (26). In cases of chronic as well as acute inflammatory diseases of the lung, it has been reported that the number of PMN and Mø are elevated (2, 3, 7). Also, Mø-derived cytokines such as granulocyte/Mø colony-stimulating factor, IL-1, IL-8, and TNF are also increased (1, 4, 7, 14). These cytokines cause PMN recruitment and degranulation (14, 27). Degranulation of PMNs would provide a continuous supply of both MPO and iMPO in the lung tissue microenvironment. Within 10 min of PMN activation, MPO is secreted into the microenvironment (9). After MPO is secreted, ~ 40% of the enzyme is inactivated (10). Therefore, iMPO is a major contributor to the total amount of enzyme in the extracellular environment.

The present study utilized rat AMø and recombinant human MPO and iMPO. Although a homologous system would have been preferable, it is not feasible to obtain the amounts of rat MPO needed. In addition, the iMPO employed in this study was a mutated form of MPO with the histidine removed from the heme moiety (22). The similarity of MPO to iMPO was determined by two different methods: (1) gel electrophoresis and (2) ELISA, using both monoclonal and polyclonal antibodies. Also, no differences existed between the carbohydrate moieties of the two forms of this enzyme (22).

With reference to the amounts of enzyme employed in these studies, the following should be kept in mind. Other investigators have reported 20-45 U/ml of MPO at a site of inflammation (10). Others have reported 16-29 µg/ml of iMPO in an inflamed joint (28). Since, to date, the amount of both MPO and iMPO at a site of inflammation is not known, the above estimated amounts could be grossly misleading. In the present study, 33 µg/ml of iMPO was biologically active and is close to, if not within, physiologic parameters. With respect to MPO, only the lowest dose, 56 U/ml, was close to physiologic range. Although test results did not indicate that this amount of MPO enhanced phagocytosis and TNF secretion, this concentration of MPO did induce an RB and a slight upregulation of cytokine mRNA. Therefore, these results could indicate the physiologic relevance of this concentration of MPO.

Because it has been shown that tissue damage associated with inflammation in the lung can be attributed to ROI production (8), initial studies were done to determine if either MPO or iMPO would enhance the RB of AMø. The results of this study as well as others indicated that either MPO or iMPO increased Mø-derived ROI production (8). In the present study, iMPO enhanced the RB more than MPO (Figures 1A and 1B). One possible explanation for this is that MPO would rapidly interact with the newly released products of the RB resulting in the formation of various reactive oxygen species (29). These products would oxidize cell surface receptors, thereby limiting the stimulatory effects of the MPO (30). Since iMPO does not have enzymatic activity, continuous stimulation of the cell would occur.

Although phagocytosis and the RB are highly correlated, these events can be separated. Neither the percentage of AMø phagocytizing C. albicans nor the mean number of C. albicans ingested was markedly different between control cultures and those stimulated with iMPO. There was a slight increase in phagocytosis by AMø exposed to MPO; however, it did not correlate with the relative increase in yeast cell killing that was observed. Also, this increase was not consistent in all replicate experiments. Hence, exposure of AMø to either MPO or iMPO did not greatly enhance the average number of C. albicans ingested by each Mø (Figures 2A and 2B). Within the time frame studied, these data would indicate that there was not oxidation of the receptors involved in the phagocytic process.

Although it has been reported that ROI play a major role in Mø-mediated candidic activity, these studies usually involved the use of either an Mø cell line or peritoneal Mø. The current study demonstrates that iMPO is more potent than MPO with respect to the production of ROI in AMø. This is in agreement with a previous report by the present investigators using endothelial cells (16). However, it is opposite to that which we have observed with murine peritoneal Mø (unpublished data). Therefore, differences in ROI induction but not in candidic activity could reflect inherent physiologic differences in the above types of Mø. Because AMø are in direct contact with the environment, the threshold for their activation is higher or else severe lung damage would result (31). As the level of activation changes, receptor expression and affinity are altered (32). Also, previous studies indicated that mannans (a known ligand of the MMR) effectively blocked the proinflammatory effects of either MPO or iMPO (20). This observation was not reproduced in the current study using AMø. The fact that AMø respond differently from cell lines or other Mø types is supported by the fact that mannans had no effect on the production of ROI by AMø. Because it has been reported that MPO enters Mø via the MMR (33), the present studies involving mannans indicate that the binding of either MPO or iMPO was probably not due to the enzyme's interaction with the AMø MMR. Because there are at least three other scavenger receptors that bind mannosylated compounds (34), engagement of any one of these receptors could activate the RB more efficiently than the various pathways involved in the intracellular killing of C. albicans . Thus, one could observe lesser candidic activity in the presence of a greater RB.

An alternate theory to why no differences were noted in candidic activity between MPO and iMPO would be as follows: MPO should exhibit more candidic activity than iMPO because MPO would readily form a cytotoxic triad with the products of the RB. Since iMPO could not participate in the cytotoxic triad (due to its enzymatic inactivity), more ROI would be needed to kill an equivalent number of C. albicans. This could explain why equivalent levels of candidic activity were reported between MPO and iMPO, despite greater levels of RB being observed from AMø exposed to iMPO (Figures 2B and 3B).

Lung inflammation and various cytokines are highly correlated with asthma. Patients diagnosed with mild to moderately severe asthma have demonstrated elevated levels of the following "proinflammatory cytokines": TNF-, IL-1, and IL-8 (2, 4). AMø have been shown to be the primary source of TNF- in the lung (35). In particular, TNF- has been implicated in the pathogenesis of asthma (2). TNF- has been shown to induce PMN efflux into alveolar tissue by upregulating intercellular adhesion molecule-1, endothelial leukocyte adhesion molecule-1, and vascular cell adhesion molecule-1 (2, 14). This cytokine also induces IL-8, a substance that is a chemotactant for PMNs and induces their degranulation (1). The continued presence of PMNs would thus provide a continuous supply of MPO and iMPO, which would stimulate AMø to secrete more TNF-. Other investigators have stated that TNF- initiates a cytokine cascade and can account for many, if not all, of the steps of inflammation (36). Therefore, "cross talk" between the PMN and AMø via MPO and/or iMPO would help to perpetuate the inflammatory response.

With regard to TNF- secretion, obvious differences in the ability of neutrophil-derived myeloperoxidase to activate AMø were observed. Enzymatically inactive MPO induced significantly more TNF- than MPO (P < 0.001). It is known that ROI can function as second messengers within a cell (25). Therefore, secreted ROI could function in an autocrine and paracrine manner to enhance secretion of TNF- through their action as second messengers. Conversely, ROI, as a component of the cytotoxic triad, are known to oxidize receptors (30), which could lead to inhibition of cytokine secretion. Thus, the initial RB may enhance TNF- secretion, but receptor oxidation prevents continued enhanced signal transduction. Receptor oxidation should not be as extensive with iMPO due to its enzymatic inactivity, providing a possible explanation for why iMPO induced higher titers of TNF- than MPO (Figure 4). Receptor oxidation by MPO and ROI also provides a possible explanation for the reduction in TNF- production by AMø exposed to both MPO and iMPO; however, further research is needed to fully explain this effect. The fact that iMPO induced more TNF- production than MPO was verified by the RPA experiments (Figure 6). An increase in both IL-1 and IL-1 mRNA levels was also observed when MPO or iMPO treatments were applied, a finding supporting the proinflammatory effects of these enzymes. The pattern of upregulated mRNAs also confirmed that there is not a synergistic effect with simultaneous treatment of AMø with both MPO and iMPO.

In conclusion, investigators previously attributed asthma to mast cell activation. However, recognition of the PMN as a major contributor to the pathology of lung inflammation has only recently been reported. The results of the present study have described the importance of PMN-Mø interaction in the exacerbation of inflammation associated with the lung. These results indicate that (1) either neutrophil-derived MPO or iMPO induce the production of ROI and the secretion of TNF- by rat AMø in vitro; (2) substances that block binding of the MMR do not appear to affect ROI production stimulated by either MPO or iMPO, suggesting an alternate stimulatory pathway for MPO and iMPO in AMø; (3) either MPO or iMPO stimulated upregulation of certain cytokine mRNAs, and (4) MPO/iMPO costimulation does appear to inhibit production of TNF- and inflammatory cytokine mRNA upregulation compared with either MPO or iMPO alone. All of the studies done with MPO and iMPO imply that iMPO is a more potent immunoregulator of AMø than MPO. The fact that either MPO or iMPO can alter AMø functions could partially explain the perpetuation of inflammation associated with asthma and other inflammatory diseases of the lung.