INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Platelet-activating factor (PAF), a phospholipid produced by activated platelets, leukocytes, and endothelial cells, is a potent proinflammatory autacoid with diverse physiological and pathological effects, including activation of neutrophils and eosinophils, bronchoconstriction, plasma protein extravasation, and goblet cell secretion (1). Because of these potent actions, it has been suggested that PAF could play a role in the pathogenesis of asthma. In support of this possibility, an increase in circulating PAF has been reported in asthmatic patients, especially during acute attacks (4, 5). In addition, in 5% of the Japanese population, severe asthma is positively correlated with a deficiency of PAF acetylhydrolase (6), the enzyme that degrades PAF into the inactive form, lyso-PAF. Furthermore, patients with mild asthma are reported to have reduced PAF acetylhydrolase activity in bronchoalveolar lavage fluid (7).

Mucus secretion may play an important role in airway obstruction in severe asthma. Postmortem studies of patients who died of asthma showed extensive obstruction of airways by mucous plugs (8). Furthermore, in a mouse model of allergic asthma, local delivery of antigen resulted in peripheral airway plugging with mucus (13, 14), emphasizing the potential importance of goblet cell degranulation in asthmatic responses.

PAF is known to cause mucus secretion in rat nasal airways (3), but it is unknown whether PAF causes goblet cell hyperplasia. In the present study, we examined the hypothesis that PAF stimulates goblet cell growth in airway epithelium. Tumor necrosis factor alpha (TNF) is a proinflammatory cytokine. High levels of TNF have been detected in bronchoalveolar lavage fluid from asthmatics (15), and TNF inhibits PAF acetylhydrolase, the enzyme responsible for PAF breakdown (16). Therefore, we also hypothesized that TNF could potentiate the goblet cell hyperplasia induced by PAF.

Studies were performed in guinea pigs. Because guinea pigs have goblet cells in the control state, we also studied pathogen-free rats, which normally have few goblet cells in the airways. We examined the effect of PAF plus TNF on goblet cell morphology in two ways: (1) we stained mucosubstances in the airways with Alcian blue/PAS and evaluated the areas of positive staining morphometrically; (2) we examined the gene expression of mucin MUC5 in the airways, because mucin MUC5 gene is known to be expressed in rodent airway epithelial cells (17).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals and Maintenance

Pathogen-free male Dunkin-Hartley guinea pigs (350 to 450 g body weight; Simonsen Co., Gilroy, CA) and male F344 Fisher rats (230 to 250 g body weight; Simonsen Co.) were used in this study. Guinea pigs were housed one per polycarbonate cage and rats three per cage. The animals were supplied with sterilized hardwood chip bedding and filter tops. Animal rooms were maintained at 20-22° C with a relative humidity of 20 to 50% and a 12-h light-dark cycle starting at 6:00 A.M. Food and water were provided ad libitum.

Drugs

Drugs from the following U.S. sources were used: L--phosphatidylcholine, -acetyl--o-alkyl (1-O-Alkyl-2-acetyl-sn-glycero-3-phosphocholine; PAF) (Sigma, St. Louis, MO), TNF (Sigma), ketamine hydrochloride (Ketalar; Parke-Davis, Div. of Warner-Lambert Co., Morris Plains, NJ), methohexital sodium (Brevital Sodium; Eli Lilly & Co., Indianapolis, IN), acepromazine maleate (PromAce; Fort Dodge Laboratories, Inc., Fort Dodge, IA), and NPC 15669 (Scios Inc., Mountain View, CA). PAF was dissolved in 100% ethanol and freshly diluted in sterile phosphate-buffered saline (PBS) before experiments. The final concentration of ethanol in the diluent is 0.01%. PBS with 0.01% ethanol was used as a vehicle in control groups.

Studies in Guinea Pigs and Rats

Studies were first performed in guinea pigs and showed that PAF caused goblet cell hyperplasia. Because guinea pig airways contain goblet cells and eosinophils in the control state, we also studied pathogen-free rats, a species that normally has few goblet cells or eosinophils in the airways.

Pathogen-free male Dunkin-Hartley guinea pigs were anesthetized with ketamine (60 mg/kg, intramuscularly) and acepromazine maleate (0.5 mg/kg, intramuscularly) and were allowed to breathe spontaneously. PAF (10⁵ M, 100 µl) or vehicle (PBS, 100 µl) was instilled intratracheally via a 22-gauge Angiocath catheter (Becton Dickinson, Sandy, UT) through the mouth, while the laryngeal area was visualized using a high-intensity illuminator (FiberLite; Dolan Jenner Industries, Inc., Lawrence, MA). Guinea pigs were euthanized 6, 24, 48, or 72 h after PAF instillation. To study the effect of the leumedin NPC 15669 (a drug that prevents leukocyte migration) on PAF-induced responses, a group of guinea pigs (n = 4) were pretreated with NPC 15669 (100 mg/kg, intraperitoneally) 2 h before PAF was instilled intratracheally (10⁵ M, 100 µl) and again 24 h later with a second dose. The animals were euthanized 48 h after PAF instillation.

Pathogen-free male F344 Fisher rats were anesthetized with methohexital sodium (Brevital, 25 mg/kg, intraperitoneally) and allowed to breathe spontaneously. PAF (10⁵ M) in vehicle (200 µl), PAF (10⁵ M) plus TNF (1 µg/ml) in vehicle (200 µl), or vehicle alone (200 µl) was instilled intratracheally via the mouth, similarly to the guinea pigs. In eight rats, PAF (10⁵ M) or vehicle was instilled once, and the rats were euthanized 24 h later. In another eight rats, PAF (10⁵ M) or vehicle was instilled twice with 48-h intervals, and the rats were euthanized 48 h later. In the remaining rats, PAF (10⁵ M) alone, TNF alone, PAF plus TNF (1 µg/ml), or vehicle was instilled three times with 48-h intervals, and the animals were killed 48 h after the last instillation.

Tissue Preparation

After various times of exposure, the guinea pigs and rats were anesthetized deeply with sodium pentobarbital (65 mg/kg, intraperitoneally), the heart was exposed, a blunt-ended needle was inserted from the apex of the left ventricle into the aorta, and the systemic circulation was perfused with 1% paraformaldehyde in PBS. The lungs were fixed by perfusion via airways at 5 to 6 cm H₂O pressure and also via the pulmonary artery. The trachea and lungs were removed and fixed with 4% paraformaldehyde for 24 h. After fixation, the guinea pig tracheal segments, beginning 5 mm below the larynx and 14 mm long, were cut across, and then cut longitudinally along the membranous portion into two pieces. In rats, cross sections of trachea were used because of the smaller size of the trachea; two segments, beginning 5 mm below the larynx, and 4 to 5 mm long were cut from each trachea. The right main stem bronchus was dissected and cut as cross sections. The right caudal lobe was used to study the intrapulmonary bronchi (eighth or ninth generation; see Reference 18).

The tissues from guinea pigs and rats were dehydrated with graded concentrations of ethanol (50%, 70%, 85%, 95%, 100%) and embedded in methacrylate JB-4 (Polysciences, Inc., Warrington, PA). Then the tissues were sectioned (4 µm in thickness), stained with Alcian blue and periodic acid-Schiff (PAS) for acidic and neutral intracellular mucins and counterstained with hematoxylin. In separate sections, Luna's stain was used to detect eosinophils (3), which contained red granules and blue bilobate nuclei. Slides were observed at ×400 magnification, using an Axioplan microscope (Zeiss Inc., Thornwood, NY) equipped with a Plan-NEOFLUAR ×40/0.75 objective lens. Eosinophil numbers were counted in 20 consecutive high-power fields (HPF, ×400 magnification) of the epithelial layer (basement membrane to cell apices).

Morphometric Analysis of Intraepithelial Mucosubstances

The percentage of Alcian blue/PAS-stained mucosubstance area in the epithelium was determined using a semi-automatic image analysis system. The stained slides were examined with a light microscope (Axioplan; Zeiss) that was connected to a video camera (3CCD; Sony Corp. of America, Park Ridge, NJ) and a control unit (DXC7550MD; Sony Corp.), and then a color digital image capture board (IMAXX; PDI, Redmond, WA) and a color monitor (Multisync XV17; NEC Corp., Tokyo, Japan). Images of the airway epithelium were recorded from 10 consecutive high-power fields with a phase contrast lens at ×400. Analysis was performed on a Power Macintosh 9500/120 computer (Apple Computer, Inc., Cupertino, CA) with images displayed on the color monitor, using the public domain NIH Image program (developed at the U.S. National Institutes of Health and available from the Internet by anonymous FTP from or on floppy disk from the National Technical Information Service, Springfield, Virginia; part number PB95-500195GEI). The area of epithelium and Alcian blue/PAS-stained mucosubstances within the epithelium were manually circumscribed and analyzed using the NIH Image program. The data are expressed as the percentage of the total area of epithelium occupied by Alcian blue/PAS stain.

Counting of Goblet Cells and Pregoblet Cells

We counted goblet cells and total epithelial cells (total number of nuclei) from 10 captured images (see above section) in guinea pig tracheal epithelium. The linear length of the basal lamina under each analyzed region of epithelium was determined by tracing the contour of the digitized image of the basal lamina, using the NIH image program. The data are expressed as number of cells per mm basal lamina. In rats, especially after one or two instillations of PAF, a number of "developing" goblet cells are formed. These cells show Alcian blue/PAS-positive staining, the granules are small, and the cells are not packed with granules. We call these developing goblet cells "pregoblet" cells, a stage before these cells become complete goblet cells (cells packed with granules). Therefore, we counted goblet cells, pregoblet cells, and total epithelial cells (which also include ciliated cells, nonciliated secretory cells, and basal cells). Goblet cells are tall, cuboidal, goblet to low columnar in shape, with abundant Alcian blue/PAS-stained granules filling most of the cytoplasm between the nucleus and the luminal surface. Pregoblet cells are defined as cells with smaller mucus-stained areas (< 1/3 height in epithelium from basement membrane to luminal surface) or with sparsely and lightly Alcian blue/PAS-stained, small granules.

Probe Preparation

The complementary DNA (cDNA) for rat MUC5 was generously provided by Dr. Carol Basbaum. A 320-bp cDNA fragment of rat MUC5 was subcloned into the Xba/HindIII site of the transcription vector, pBluescript-SK() (Stratagene, La Jolla, CA). To prepare RNA probes for in situ hybridization, this recombinant plasmid containing the rat MUC5 cDNA fragment was linearized and transcribed in vitro with the T7 or T3 polymerase to obtain antisense or sense probe, respectively. The probes for in situ hybridization were generated in the presence of sulfer-35-uridine triphosphate ([³⁵S]UTP). After transcription, the cDNA template was digested with deoxyribonuclease (DNase), and the radiolabeled RNA was purified via a Sephadex G-25 Quick Spin Column (Boehringer Mannheim, Indianapolis, IN) and precipitated in an ethanol/ammonium acetate solution. Before use, RNA probes were washed with 70% ethanol and diluted in 10 mM DTT.

In Situ Hybridization

Airway tissues from rats were fixed with 4% paraformaldehyde for 1 h and washed in PBS. The tissues were cryoprotected with 30% sucrose overnight and then embedded in optimal cutting temperature (OCT) compound (Sakura Finetek U.S.A., Inc., Torrance, CA) and frozen. Frozen sections (5 µm) were cut and placed on positively charged glass slides (Superfrost Plus; Fisher Scientific, Pittsburgh, PA). Sections cut in close proximity were used for hybridization with sense and antisense probes. Alternate sections were used for Alcian blue/PAS staining.

The specimens were refixed in 4% paraformaldehyde, rehydrated in 0.5× standard saline citrate (SSC), and then acetylated in triethanolamine with acetic anhydride. Hybridization was carried out with 2,500 to 3,000 cpm/µl of antisense or sense probe in 50% deionized formamide, 0.3 M NaCl, 20 mM Tris, 5 mM EDTA, 1× Denhardt's solution, 20 mM dithiothreitol, 10% dextran sulfate, 0.5 mg/ml yeast transfer ribonucleic acid (tRNA), and 0.5 mg/ml sonicated salmon sperm DNA at 55° C overnight. Posthybridization treatment consisted of washes with 2× SSC, 1 mM EDTA, 10 mM -mercaptoethanol at room temperature, incubation with ribonuclease (RNase) solution (20 µg/ml) for 30 min at room temperature, and further washes in 0.1× SSC, 1 mM EDTA, 10 mM -mercaptoethanol at 55° C for 2 h and then in 0.5× SSC at room temperature. Specimens were dehydrated, air-dried, and covered with Kodak NBT nuclear track emulsion (Eastman Kodak, Rochester, NY) for autoradiography. After exposure for 7 to 21 d at 4° C, the slides were developed, fixed, and counterstained with hematoxylin.

Statistical Evaluation

All data are expressed as means ± SEM. For statistical analysis, the two-way or one-way analysis of variance (ANOVA) followed by Dunnett t test, or linear regression test or Student t test was used as appropriate. A probability of less than 0.05 was considered a statistically significant difference.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PAF-induced Goblet Cell Hyperplasia: Guinea Pig Trachea

In the naive state (untreated animals), the tracheal epithelium of guinea pigs contained goblet cells (Table 1); vehicle-treated control guinea pigs showed no change in the number of goblet cells, total epithelial cells, or the percentage of area of epithelium stained with Alcian blue/PAS for up to 72 h (p > 0.05; Table 1 and Figure 1, upper panel, open bars). After a single intratracheal instillation of PAF (10⁵ M, 100 µl; Figure 1, upper panel, solid bars), there was a significant increase in number of goblet cells and mucus-stained area, detectable as early as 24 h after instillation (Table 1; Figure 1, upper panel). After exposure to PAF, there was a time-dependent increase in mucous area (linear regression, r = 0.6, p < 0.05, n = 17), which appeared to reach a plateau at 72 h after instillation. The number of total epithelial cells per mm basal lamina 72 h after PAF instillation was slightly but not statistically significantly increased (p > 0.05, Table 1).

View larger version (16K): [in this window] [in a new window]   Figure 1.   Effect of PAF on the percentage of Alcian blue/PAS- stained area of epithelium (upper panel ) and on the recruitment of eosinophils (lower panel ) in guinea pig trachea. One hundred µl of control vehicle (open bars) or of PAF (105 M; solid bars) was instilled intratracheally, and the guinea pigs were euthanized 6, 24, 48, or 72 h later. Data are means ± SEM (for n = 4-7 guinea pigs). *p < 0.05, **p < 0.01, and ***p < 0.001, significantly different from vehicle control group.

In control guinea pigs, a significant number of eosinophils was present in the airway epithelium, and the number of eosinophils did not increase after exposure to vehicle up to 72 h (p > 0.05, Figure 1, lower panel ). However, instillation of PAF (10⁵ M, 100 µl) caused a further time-dependent eosinophil recruitment into the airway epithelium (Figure 1, lower panel; linear regression, r = 0.84, p < 0.0001, n = 17). Maximal eosinophil recruitment occurred at 48 h. Figure 2 shows typical light micrographs of goblet cells and eosinophils in the airways in response to PAF.

View larger version (109K): [in this window] [in a new window]   Figure 2.   Light photomicrographs of guinea pig tracheal epithelium in JB-4 sections (4 µm thick) after instillation of 100 µl of control vehicle (left panels) or 100 µl of PAF (105 M) (right panels). Guinea pigs were euthanized 24 h after instillation. Top panels were stained with Alcian blue/PAS to identify goblet cells (arrows), and bottom panels were stained with Luna's stain to identify eosinophils (arrowheads). Note the PAF-induced hyperplasia of goblet cells in Alcian blue/PAS-stained cells (also see Table 1). Eosinophils are stained with red granules and blue nuclei and are distributed within and under the epithelium. Bar = 50 µm.

Because PAF caused both goblet cell hyperplasia and eosinophil recruitment, we studied the effect of pretreatment of guinea pigs with NPC 15669, a leumedin that inhibits leukocyte recruitment (19). NPC 15669 prevented the recruitment of eosinophils into the airways (p < 0.001), but had no effect on PAF-induced goblet cell hyperplasia (p > 0.05, Figure 3). After NPC 15669 treatment, the number of goblet cells (59 ± 9 per mm basal lamina) was not statistically different (p > 0.05) from that (61 ± 11 per mm basal lamina) in the animals given PAF alone.

View larger version (15K): [in this window] [in a new window]   Figure 3.   Effects of leumedin NPC 15669 on the PAF-induced increase in the percentage of Alcian blue/PAS-stained mucous area of epithelium (upper panel ) and on the recruitment of eosinophils (lower panel ) in guinea pig trachea. Vehicle (100 µl, open bars) or PAF (105 M, 100 µl; solid bars) was instilled intratracheally. A group of guinea pigs were pretreated with NPC 15669 (100 mg/kg, intraperitoneally) 2 h before PAF was instilled intratracheally (105 M, 100 µl) and again 24 h later with a second dose (hatched bars). The animals were euthanized 48 h after PAF instillation. Leumedin NPC 15669 prevented the PAF-induced eosinophil recruitment but not the goblet cell hyperplasia. Data are means ± SEM (for n = 4-5 rats). *p < 0.05 and ***p < 0.001, significantly different from vehicle control group or ### p < 0.001, significantly different from PAF group, by ANOVA test followed by Dunnett t test.

PAF-induced Goblet Cell Hyperplasia in Rat Airways

Because the airways of control guinea pigs contain goblet cells, we also studied pathogen-free rats, a species whose airways normally contain few goblet cells (Table 2), especially in the trachea (Figures 4 and 5) and few eosinophils (0-1 per 10 mm epithelium; data not shown). In tracheal epithelium, after one or two instillations of PAF (10⁵ M, 200 µl, n = 4), instillation of PAF resulted in mostly pregoblet cells but also increased the number of goblet cells (Table 2). When PAF (10⁵ M, 200 µl, n = 4) was instilled three times with 48-h intervals, more mature goblet cells were found (Table 2). The number of total epithelial cells was slightly but not statistically significantly increased (Table 2). In main stem bronchial epithelium, PAF instilled three times increased the number of goblet cells from 11 ± 2 to 44 ± 7 per mm basal lamina (p < 0.01) and the number of total epithelial cells from 168 ± 9 to 206 ± 12 per mm basal lamina (p < 0.05), while the number of pregoblet cells was not changed (from 17 ± 1 to 18 ± 2 per mm basal lamina, p > 0.05). In intrapulmonary bronchial epithelium, PAF instilled three times increased the number of goblet cells from 7 ± 3 to 63 ± 8 per mm basal lamina (p < 0.001) and the number of total epithelial cells from 170 ± 2 to 203 ± 10 per mm basal lamina (p < 0.05), while the number of pregoblet cells was not changed (from 21 ± 2 to 24 ± 23 per mm basal lamina, p > 0.05). PAF also increased the Alcian blue/PAS-stained mucous area; the percentage of stained area in intrapulmonary bronchi was greater than in the trachea (p < 0.01, Figure 4). Figure 5 shows typical light micrographs of goblet cells at the three anatomic levels in response to PAF. PAF did not recruit significant numbers of eosinophils or neutrophils in rats (p > 0.05, data not shown).

View larger version (20K): [in this window] [in a new window]   Figure 4.   Effect of PAF on the percentage of Alcian blue/PAS- stained mucous area of epithelium of pathogen-free rats at various levels of airways. The percentage of Alcian blue/PAS-stained area of the epithelium was measured in airways after control vehicle (open bars) or PAF (105 M; solid bars) was instilled intratracheally in 200 µl on three occasions, and the animals were euthanized 48 h after the last instillation. Data are expressed as means ± SE of the mean (n = 4-5 rats per group). *p < 0.05, **p < 0.01, significantly different from vehicle control group; #p < 0.05, significantly different from the control trachea group; ## p < 0.05, significantly different from PAF-treated trachea group, respectively.

View larger version (84K): [in this window] [in a new window]   Figure 5.   Light photomicrographs of airways in pathogen-free rats in JB-4 sections (4 µm thick) after 200 µl of control vehicle (left panels) or of PAF (105 M) (right panels) was instilled intratracheally three times with 48-h intervals between instillations; the animals were euthanized 48 h after the last instillation. PAF caused an increase in number of goblet cells in trachea, main stem bronchi, and intrapulmonary bronchi. Arrows indicate goblet cells. Bar = 50 µm.

Effect of TNF on PAF-induced Goblet Cell Hyperplasia in Rat Trachea

PAF (10⁵ M, 200 µl) given three times with 48-h intervals caused significant goblet cell hyperplasia and increased the mucous staining area in rat trachea (p < 0.01). TNF (1 µg/ml, 200 µl) alone caused no significant increase in the number of goblet cells or in the percentage of mucous area (Table 2 and Figure 6, p > 0.05). However, when PAF and TNF were administered together, a marked increase in number of goblet cells and percentage of mucous area occurred (Table 2 and Figure 6, p < 0.001); the effect was considerably greater than that caused by PAF alone (p < 0.01) (Table 2 and Figure 6). PAF plus TNF increased the number of total epithelial cells (p < 0.001), an effect that was greater than PAF alone (p < 0.01) (Table 2).

View larger version (16K): [in this window] [in a new window]   Figure 6.   Effect of TNF on PAF-induced increase in the percentage of Alcian blue/PAS-stained area of epithelium in pathogen-free rats. Tissues were stained with Alcian blue/PAS, and the percentage of Alcian blue/PAS-stained area of epithelium was measured after three instillations of 200 µl control vehicle, PAF (105 M), TNF (1 µg/ml), or TNF plus PAF with 48-h intervals and euthanized 48 h after the last instillation. Data expressed as means ± SEM (n = 4-5 rats per group). **p < 0.01, ***p < 0.001, significantly different from vehicle control group and ## p < 0.01, significantly different from PAF group.

PAF-induced MUC5 Gene Expression in Rat Airways

In control rats and in vehicle-treated rats, no signals were detected with the antisense probe of MUC5 in the trachea (n = 4, Figures 7f and 7g). When PAF (10⁵ M, 200 µl) was instilled into the trachea once and the rats were euthanized 24 h later, only low levels of expression were detected and only in 2 to 3 epithelial cells per cross section (n = 3, data not shown). However, when PAF (10⁵ M, 200 µl) was instilled three times at 48-h intervals and the animals were euthanized 48 h after the last instillation, more intense expression of MUC5 occurred in the tracheal epithelium (n = 4, Figures 7b and 7c); MUC5 expression was found in cells that stained for mucins (in subsequent sections, Figure 7a). TNF alone did not induce MUC5 expression (data not shown). Sections examined with the MUC5 sense probe showed no expression (Figures 7d and 7e). Signals in other cell types, including smooth muscle and connective tissue, were not detectable.

View larger version (90K): [in this window] [in a new window]   Figure 7.   MUC5 mucin gene expression in rat trachea after three intratracheal instillations of PAF (105 M, 200 µl), with 48-h intervals. The rats were euthanized 48 h after the last instillation. Section a shows goblet cells in rat tracheal epithelium stained with Alcian blue/PAS after exposure to PAF. The remaining sections show results from in situ hybridization after incubation of mucin gene MUC5 antisense probe (b, c, f, g) or sense probe (d, e). The exposures are shown in bright field (left side) and dark field (right side). In control animals exposed only to vehicle, no expression was seen with the MUC5 antisense probe (f, g). In PAF-treated animals, the antisense probe (b, c) shows MUC5-positive expression in cells that correspond to the epithelial cells staining with Alcian blue/PAS (a). In PAF-treated animals, the sense probe showed no expression (d, e). Bar = 50 µm.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study shows that PAF causes goblet cell hyperplasia in airways of guinea pigs and rats. Guinea pigs were sensitive to PAF instillation: A single instillation of PAF resulted in goblet cell hyperplasia, present by 24 h and maximal at approximately 72 h. PAF instillation also resulted in eosinophil recruitment. Pretreatment with NPC 15669, a drug that inhibits leukocyte recruitment (19), prevented PAF-induced eosinophil recruitment but had no effect on PAF-induced goblet cell hyperplasia, suggesting that eosinophil recruitment was not involved in the hyperplasia.

Because guinea pig airways contain goblet cells in the control state, we also studied pathogen-free rats, animals whose airways contain few or no goblet cells. PAF caused a dose- dependent goblet cell development: One or two instillations of PAF produced mostly pregoblet cells; three instillations produced more mature goblet cells. Goblet cell hyperplasia occurred throughout the airways, but the effect was greater in peripheral airways. This could be due to differences in the amount of PAF delivered to various airways, or it could be due to intrinsic differences in responses of the airways at different levels.

Because the enzyme responsible for the breakdown of PAF (PAF acetylhydrolase) is inhibited by TNF (16), we examined the effect of TNF on PAF-induced effects on goblet cells. When PAF and TNF were administered together, goblet cell hyperplasia was considerably greater than with PAF alone. TNF alone had no effect on goblet cell growth. In addition to TNF, other inflammatory mediators (e.g., interleukin-1 [IL-1], lipopolysaccharide [LPS]) also inhibit PAF acetylhydrolase (16). Thus, the presence of other inflammatory mediators could exaggerate effects of PAF and produce pathologic states of goblet cell hyperplasia.

Mucin genes are believed to be expressed during goblet cell growth (17). MUC5 exists in the airway epithelium (17). Murine overexpression of IL-4 results in goblet cell hyperplasia in airways, which is associated with the expression of MUC5 but not MUC2 (22). In the present studies, instillation of PAF into the airways resulted in expression of MUC5 in cells that stained with Alcian blue/PAS. We suggest that an upregulation of MUC5 messenger RNA (mRNA) and subsequent mucin protein synthesis is responsible, at least in part, for the accumulation of mucus glycoprotein in nonciliated airway epithelial cells after stimulation with PAF.

There is strong evidence that goblet cell hyperplasia and degranulation play major pathophysiologic roles in patients with chronic bronchitis who have airway obstruction (23), in patients dying of acute severe asthma (8), and in cystic fibrosis patients (24). PAF stimulates goblet cell growth and may therefore play a role in the mucous plugging in airways that occurs in these diseases.