Elucidating the Role of Interleukin 1β and Prostaglandin E2 in Upper Airway Mucosal Wound Healing FREE

Fundamental to normal wound healing is the propensity of connective tissue to heal through a reparative process that only partially approximates the original tissue structure and function and results in scar formation. Although scarring is not the ideal outcome when wounded tissue heals, consequences of excessive matrix deposition with cicatricial contraction in the airway can be life threatening or fatal. Stenosis of the airway, often caused by what is essentially a type of hypertrophic scarring, reduces the cross-section of the lumen through which the body's air supply flows. One increasingly common mechanism of iatrogenic injury to the upper airway is laser treatment for removal of airway lesions such as amyloid deposits, hemangiomas, and previously formed fibrotic tissue.^1- 3 Lasers have also been used during laryngotracheoplasty as a welding tool, interlocking graft and adjacent normal tissue and improving integrity across the margin between graft and normal tissue.⁴ Understanding the precise cellular and molecular processes underlying mucosal damage is critical to develop less damaging methods of laser instrumentation of the upper airway and more effective reparative approaches to upper airway fibrosis and scar formation.

Acute inflammation occurs during the early stage of wound healing. After injury, inflammatory cells invade the wound and secrete a variety of soluble mediators that modulate the response of resident fibroblasts and epithelial cells. An important early inflammatory mediator is IL-1β (interleukin 1β), a cytokine secreted by a variety of cells including epithelial cells, macrophages, and neutrophils.⁵ Secretion of IL-1β within the wound bed can stimulate production of secondary inflammatory mediators such as IL-6, IL-8, and prostaglandin E₂ (PGE₂).⁶ These mediators are part of a well-integrated inflammatory response that coordinates subsequent activity in the wound bed. Although inflammation is an integral component of wound healing, excessive inflammation has been linked to abnormal wound healing outcomes, specifically fibrosis and scar formation. On the other hand, diminished inflammatory reaction to injury, as in the case of fetal wound healing, is associated with a regenerative wound healing process that lacks scar formation and fibrosis.⁷ Previous studies^8- 10 suggest that a link between inflammation and fibroblast activity in the wound bed may be partially responsible for the final outcome of wound healing and the degree of scarring.

Prostaglandin E₂ is a secondary inflammatory mediator, synthesized via cyclo-oxygenase (COX) and PGE₂ synthase (PGES) activities and degraded by multiple enzymes including dehydrogenases. It can modulate the activity of inflammatory and mesenchymal cells via 4 E-prostanoid (EP) receptors, EP1 through EP4, coupled to either intracellular calcium or cyclic adenosine monophosphate signaling pathways.^11- 15 In the lower airway (lung) mucosa, PGE₂ has an antifibrotic effect.^16- 20 Specifically, endogenous fibroblast production of PGE₂ appears to be crucial to appropriate tissue repair after injury. Prostaglandin E₂ production (COX-2, microsomal PGES type 1 [mPGES1]), degradation (15-hydroxyprostaglandin dehydrogenase [15-PGDH]), and reception (EP1-EP4) are subject to regulation by primary inflammatory cues including IL-1β.^20- 25 Up-regulation of COX-2 transcription occurs through multiple signaling pathways involving activation of nuclear factor κB, extracellular signal-regulated kinase 1/2, mitogen-activated protein kinases, and cyclic adenosine monophosphate response element, and can be augmented by cyclic adenosine monophosphate response element binding protein/p300 histone acetyltransferase activity.^26- 27 The precise integration of these regulatory mechanisms remains unclear at both the tissue and cellular levels. Furthermore, the precise integration of the PGE₂ pathway and secondary interleukin (IL-6, IL-8, and IL-10) pathways has yet to be determined.

While the inflammatory response to injury and the role of IL-1β and PGE₂ have been well studied in the lower airway, they remain less well characterized at higher airway levels such as in the tracheal mucosa. Interleukin 1β up-regulation of COX-2 activity and PGE₂ secretion can increase mucin gene expression and protein secretion by epithelial cells. Endogenous PGE₂ production is also thought to play a role in regulating the effects of hyperoxic injury to the tracheal epithelium through mechanisms that remain unclear.^28- 29 This study was designed to examine the role of inflammatory mediators in the larger wound healing process as it occurs in the upper airway. Specifically, it was designed to address the following questions: (1) Does laser injury to upper airway mucosa result in up-regulation of inflammatory genes? (2) Is up-regulation of inflammatory genes in the mucosal tissue reflected in the presence of inflammatory mediators in mucosal secretions? (3) How do mucosal fibroblasts in the upper airway contribute to the inflammatory response? The answers to these questions are meant to address existing gaps in our understanding of the cellular and molecular processes underlying repair of the upper airway mucosa and to extend previous studies using secretion analysis as a tool to monitor the mucosal response to injury or trauma and subsequent healing.

We selected the New Zealand white rabbit for this study. This animal model has been previously shown to be applicable for the study of subglottic stenosis (SGS) and mucosal airway wound healing.^25,30 Four experiments were performed, using a total of 38 animals.

All animal experiments were conducted under approved protocols compliant with institutional animal care and use committee regulations. Animals were prepared for surgery as follows. The neck fur was clipped and the skin prepared with povidone-iodine and 70% alcohol scrubs. With the animals under general anesthesia (ketamine hydrochloride, 35 mg/kg, and xylazine hydrochloride, 5 mg/kg) and local infiltration of the skin with lidocaine (0.2 mL, 1% solution), a vertical midline neck incision was made. The soft tissues of the neck were dissected in the midline, after which the strap muscles were separated. The overlying pretracheal fascia was incised and the subglottis entered via a midline cricoidotomy extending through the first and second tracheal rings, and the posterior subglottic mucosa was exposed. The posterior subglottis was injured with a carbon dioxide laser at the following settings: 1-second continuous pulse; beam diameter, 2 mm; power settings, 5, 8, and 12 W; and either 2 pulses per airway (25% coverage, posterior subglottis only) or 4 pulses per airway (40% coverage, posterior and lateral subglottis). The cricoid and tracheal rings were then placed back into anatomic position and the skin was loosely closed with 5.0 silk sutures (Ethicon Inc, Somerville, NJ). Control airways were prepared by performing sham surgeries as detailed in this paragraph, with the exception of the laser injury step.

Upper airway secretions were collected throughout the duration of the study as previously described³⁰ by means of absorbable gelatin foam sponge (Gelfoam; Pfizer Inc, New York, NY) swabs. The secretions were extracted from the swabs by means of sterile saline to yield approximately 1000 μL of supernatant fluid. The entire protocol was performed at 4°C to minimize the risk of denaturation. Interleukin 1β and PGE₂ were assayed by means of enzyme-linked immunosorbent assay kits following the supplier's recommended protocol for each (PGE₂ Parameter Kit; R&D Systems, Minneapolis, Minn; and BD Opt EIA IL-1β; BD Biosciences, San Diego, Calif). After standardization to total original secretion weight, 2-sided t tests were used to assess differences in marker expression between time after injury and the normal, uninjured state. Marker levels in upper airway secretions were normalized to preoperative baseline levels for both sham-operated and laser-injured airways. For the purposes of secretion analysis, a total of 6 sham-operated airways were used.

Twelve animals were selected for a short-term study of upper-airway mucosal tissue and secretions (24 and 48 hours). Airways were manipulated as detailed in the preceding section, and the posterior mucosa was injured by two 5-W pulses. Animals were killed at 24 or 48 hours after injury by intracardiac administration of pentobarbital sodium (50 mg/kg) with the animal under anesthesia, and the airways were dissected. The injured subglottic mucosa was removed from the underlying cartilage layer and frozen at −80°C for subsequent analysis. An immediately inferior 5 × 5-mm piece of tracheal mucosa was similarly collected and preserved. Total RNA was extracted, digested with DNase I (MessageClean Kit; GenHunter Co, Nashville, Tenn) to reduce chromosomal DNA contamination, and reverse transcribed into complementary DNA; expression analysis for selected genes was conducted as described subsequently. For the purposes of tissue analysis, control tissue was derived from uninjured posterior mucosa of sham-operated animals killed at 21 days postoperatively. This tissue was deemed suitable to provide baseline levels of gene expression in the context of inflammatory genes.

Upper airway (subglottic/tracheal) fibroblasts were obtained from human tissues produced during standard medical procedures, with institutional review and approval for work with human-derived materials. For these experiments, a total of 6 upper-airway fibroblast cultures were used: 1 fibroblast population derived from fetal trachea (23 weeks' gestation), 2 fibroblast populations derived from adult normal trachea, and 3 fibroblast populations derived from pathologic specimens of SGS. Cells were cultured in Dulbecco modified Eagle medium (Invitrogen, Grand Island, NY) supplemented with 10% fetal bovine serum (Invitrogen) and 10-U/mL penicillin G sodium, 10-U/mL streptomycin sulfate, and 0.025-μg/mL amphotericin B (Antibiotic-Antimycotic; Life Technologies, Rockville, Md) in a humidified atmosphere containing 5% carbon dioxide at 37°C. Culture medium was changed twice a week and cells were subcultured as they became confluent. All cells were used before passage 10 (approximately 15 total population doublings).

Rabbit-specific primers were designed for (1) 18S ribosomal RNA (Gene Bank No. AY150553); 5′ sequence, AAGCCATGCAT GTCTAAGTACGCA; 3′ sequence, CAAGTAGGA GAGGAGCGAGCGACC; (2) IL-1β (Gene Bank No. M26295); 5′ sequence, CGGCAGGT CTTGTCAGTCGTT; 3′ sequence, TGCAGAGGACG GGTTCTTCTT; and (3) PTGS2 (COX-2) (Gene Bank No. U97696); 5′ sequence, CCATGGGTGTG AAAGGCAAGA; 3′ sequence, TGGGTGAAG TGCTGGGCAAAG. The messenger RNA (mRNA) collected and extracted from rabbit tissues was reverse transcribed into complementary DNA and quantitatively amplified (reverse transcriptase–polymerase chain reaction [PCR]). Briefly, the reverse transcription reaction included 500 ng of DNA-free total RNA pooled from each group, random primers, and a DNA polymerase (SuperScript II; Invitrogen) and was incubated at 25°C for 10 minutes, 48°C for 30 minutes, and 95°C for 5 minutes in a thermocycler (PE 2700; Applied Biosystems, Foster City, Calif). For PCR amplification, PCR reagents (SYBR Green; Applied Biosystems) were used. The PCR reaction (in triplicate) included 5 μL of 10× PCR buffer, 6 μL of 25mM magnesium chloride, 4 μL of each deoxynucleotide triphosphate (blended with 2.5mM deoxyadenosine triphosphate, deoxyguanosine triphosphate, and deoxycytidine triphosphate, and 5mM deoxyuridine triphosphate), 2.5 μL of each gene-specific primer (5μM), 0.5 μL of uracil-N-glycosylase (0.5 U) (AmpErase; Roche Diagnostics, Indianapolis, Ind), 0.25 μL of DNA polymerase (1.25 U) (AmpliTaq Gold; Roche Molecular Systems Inc, Pleasanton, Calif), and 5 μL of complementary DNA in a final volume of 50 μL. The conditions for PCR (TaqMan; Applied Biosystems) were as follows: 50°C for 2 minutes, 95°C for 12 minutes, 40 cycles at 95°C for 15 seconds, and 60°C for 1 minute in a sequence detection system (ABI PRISM 7700; Applied Biosystems). The system-specific sequence detection software (Applied Biosystems) was used for instrument control, automated data collection, and data analysis. Relative quantitation (fold difference) of the expression levels of each transcript for each group was calculated by means of the 2-ΔΔCt method, which produces a value inversely related to the relative abundance of the mRNA. Values were then standardized to mRNA values derived from control tissue obtained as described earlier and expressed as fold increase.

Cells were allowed to reach confluency under standard cell culture conditions, lysed on ice in 62.5mM Tris (pH 6.6), 10% glycerol, 1% sodium dodecyl sulfate, 0.5mM phenylmethylsulfonyl fluoride, 2-μg/mL aprotinin, 10-μg/mL leupeptin, and 5-μg/mL pepstatin, homogenized and boiled for 5 minutes. Equal aliquots of protein (50 μg) were electrophoresed on sodium dodecyl sulfate–polyacrylamide gels by means of a mini-gel system (BioRad Laboratories, Hercules, Calif) and transferred onto membranes of polyvinylidene fluoride (a modified nitrocellulose). Membranes were blocked with 5% milk in phosphate-buffered saline with 0.1% Tween-20 for 1 hour at room temperature. Membranes were then probed with primary antibodies: polyclonal rabbit anti–human EP1, EP2, EP3, or EP4; anti–COX-2; monoclonal mouse anti-mPGES1; and monoclonal anti–15-PGDH (Cayman Chemical, Ann Arbor, Mich), followed by a 1-hour incubation with horseradish peroxidase–conjugated goat anti–rabbit or goat anti–mouse IgG (Pierce, Rockford, Ill) in 1% milk in phosphate-buffered saline/Tween. Protein bands were visualized with a chemiluminescent kit (Pierce). To standardize for protein loading, β-actin was identified by means of a monoclonal primary antibody (Sigma-Aldrich Corp, St Louis, Mo) and secondary goat antimouse (Jackson ImmunoResearch, West Grove, Pa).

Fibroblast production and secretion of IL-1β and PGE₂ was assayed with human-specific, commercially available enzyme-linked immunosorbent assay kits (PGE₂ Parameter Kit; R&D Systems; and BD Opt EIA IL-1β; BD Biosciences). The amount of each secreted inflammatory mediator was quantified and standardized to the total of amount of protein present within the cell layer.

Fibroblasts were grown on glass coverslips under normal growth conditions. Medium was removed, and cells were fixed with paraformaldehyde and permeabilized with Triton-X 100. Translocation of nuclear factor κB from the cytoplasm to the nucleus was assessed by means of indirect immunofluorescence. Fibroblasts grown on coverslips were serum starved for 24 hours and stimulated for multiple periods with varying doses of IL-1β. Cells were fixed, permeabilized, and stained with primary antibodies for p65 and p50 (Santa Cruz Biotechnology Inc, Santa Cruz, Calif), followed by a secondary antibody coupled to fluorophore (Alexa 488; Molecular Probes–Invitrogen, Carlsbad, Calif). Digital images were captured with an inverted microscope equipped with phase-contrast objective (Nikon Inc, Melville, NY) connected to a CCD (charge-coupled device) camera (Diagnostic Instruments Inc, Sterling Heights, Mich), and prepared and labeled with Adobe Photoshop 7.0 software (Adobe Systems Inc, San Jose, Calif).

The IL-1β and PGE₂ secretion levels in laser-injured and sham-operated animals were expressed as fold changes compared with basal preoperative levels. Comparisons were made between standardized values for a given time between laser-injured and sham-operated airways. Results are expressed as averages, with error bars representing standard errors of the mean. For quantitative PCR analysis, up-regulation of IL-1β and COX-2 genes is expressed as fold increase above levels in control tissue, after normalization to an internal control gene (18S rRNA). Data are expressed as averages, with error bars representing standard deviations. In all cases, statistical significance was assigned for P<.05, derived by means of 2-tailed t tests.

Using quantitative PCR (TaqMan), we found that IL-1β and COX-2 mRNA levels increased at 24 and 48 hours after mucosal injury (Figure 1) compared with baseline values. It should be noted that control values were obtained by amplifying complementary DNA obtained from sham-operated airways at 21 days after injury. These controls were chosen for several reasons. First, they represent an appropriate control for laser-injured mucosa analyzed at 21 days after injury. Second, in contrast to virgin airways, they should contain an elevated inflammatory response that will partially control for the damage induced by the anterior approach to the subglottis. Compared with these control values, IL-1β levels in the injured subglottic mucosa were approximately 20- and 17.5-fold higher at 24 and 48 hours, respectively, after injury. In the immediately adjacent tracheal mucosa, IL-1β levels increased by approximately 10- and 8-fold over baseline values at 24 and 48 hours after injury, respectively. Although these increases are statistically significant, they are lower than in the injured subglottic mucosa and demonstrate the localization and gradation of the injury response. Significant differences between subglottic and tracheal mRNA levels (P<.05) are indicated on the graphs (daggers indicate tracheal mRNA levels significantly lower than their subglottic counterparts). In the case of COX-2 at 48 hours, the difference between mRNA levels in the subglottic and tracheal mucosa did not reach statistical significance (P = .059). At 21 days after laser-induced injury, IL-1β levels in subglottic mucosa declined to approximately 3-fold higher than baseline levels (not statistically significant). Similar data were obtained for COX-2 levels in the injured subglottic mucosa and the adjacent tracheal mucosa. The COX-2 levels at 24 and 48 hours after injury were 9- and 7-fold higher than baseline levels, declining to a 4-fold increase by 21 days (not statistically significant). The COX-2 levels in the adjacent tracheal mucosa were approximately 3-fold higher than baseline levels (not statistically significant).

The IL-1β and PGE₂ levels in mucosal secretions obtained from this animal group were found to increase after injury compared with baseline preoperative values (data not shown). The PGE₂ levels increased immediately postoperatively and decreased during the next 48 hours, while IL-1β levels began to rise at 24 hours after injury and remained elevated during the next 24 hours. Statistical significance was not reached at all time points because of the limited number of animals. However, these preliminary data were used as a starting point for subsequent secretion analysis.

Carbon dioxide laser settings used to injure the posterior subglottic mucosa were varied with respect to 2 variables: power and extent of area covered. Specifically, power settings were increased from 5 to 12 W, and extent of covered area was increased from two 2-mm spots covering approximately 25% of the subglottic area to four 2-mm spots covering approximately 50% of the subglottic area. At the 5-W power setting, IL-1β levels in secretions increased immediately postoperatively and remained elevated above the levels in secretions derived from sham-operated airways, with a decreasing trend over time (Figure 2). Although statistical significance was not reached at this power setting, there was a clear trend toward higher IL-1β levels in laser-injured compared with sham-operated control airways. At a higher power setting, 8 W, IL-β levels in secretions from laser-injured airways were significantly higher than in their sham-operated counterparts at 1 and 3 days postoperatively. Interestingly, further increases in laser power settings (to 12 W) or increases in area covered did not result in further elevation in IL-1β levels present in secretions. In general, it should be noted that in sham-operated airways, IL-1β levels did not rise above basal preoperative levels.

In contrast to IL-1β, PGE₂ levels present in mucosal secretions increased substantially after sham operations, peaking to approximately 13-fold above basal preoperative levels at day 7 after injury (Figure 3). The PGE₂ levels present in 5-W laser-injured airways closely paralleled those of sham-operated airways, with only a marginally higher peak of approximately 20-fold at 7 days after injury. As the power setting increased from 5 to 8 to 12 W, PGE₂ levels in laser-injured airways appeared to peak at earlier times, a trend indicating a possible change in the temporal expression pattern for this mediator with increasing extent of injury. Statistically significant higher levels of PGE₂ were present in airways injured by an 8-W laser with 50% area coverage, compared with sham-operated airways, immediately and at 1 day after surgery.

Interleukin 1β has previously been shown to trigger cytoplasmic-nuclear translocation of nuclear factor κB p65/p50 heterodimers, resulting in increased COX-2 transcription. In this study, IL-1β induced a dose-dependent translocation of p65 from the cytoplasmic pool to the nuclear pool in upper airway mucosal fibroblasts. In contrast, p50 was found to be localized to both the cytoplasm and the nucleus. Interleukin 1β–dependent translocation of p65 occurred within a short time frame, beginning at 15 to 30 minutes after stimulation (Figure 4). This response pattern was observed for all 3 mucosal fibroblast phenotypes—normal adult, fetal, and SGS.

Consistent with translocation of the p65/p50 complex to the nucleus, stimulation of mucosal fibroblasts by IL-1β resulted in a dose-dependent up-regulation of COX-2 protein levels as detected by Western blotting (Figure 5A and B). In unstimulated fibroblasts, COX-2 protein levels were virtually undetectable, as would be expected given that this is an inducible enzyme. Up-regulation of COX-2 protein levels was conserved in normal adult, fetal, and SGS-derived fibroblasts with respect to both qualitative and quantitative patterns. Additional analysis indicated that while basal levels of mPGES1 protein are relatively low in upper airway fibroblasts, they are dramatically up-regulated after stimulation with IL-1β. Media from all experiments were collected and analyzed for PGE₂ secretion by fibroblasts via a high-sensitivity commercial enzyme-linked immunosorbent assay kit. Administration of IL-1β caused an increase in endogenous PGE₂ secretion by normal fetal and adult upper airway fibroblasts as well as by SGS-derived fibroblasts (Figure 5C; all P<.05). Under basal cell culture conditions, all 3 fibroblast populations produced detectable levels of PGE₂ in comparable concentrations. On stimulation with IL-1β, PGE₂ production was increased; interestingly, SGS fibroblasts secreted much higher amounts of PGE₂ than did the other 2 cell types. These results are further discussed in the “Comment” section.

Previous work in our laboratory has indicated that EP receptor levels are differentially regulated during fetal and adult tissue repair.²⁵ In this study, we analyzed the protein levels for all 4 EP receptors in the 3 chosen fibroblast populations and found that IL-1β does not alter receptor levels in these mucosal fibroblasts (representative Western blots are shown for EP2 and EP4; data for other receptors not shown) (Figure 6A). Breakdown of PGE₂ is an important component of the PGE₂ pathway and has previously been described to contribute to tissue PGE₂ levels.^23- 24 Treatment of upper airway fibroblast cultures with IL-1β did not change the intracellular protein levels of 15-PGDH (Figure 6B).

Connective tissue repair results from interactions between participating cells and soluble mediators. Fibroblasts have long been identified as one of the main participants in the development of excessive fibrosis, and their responses are regulated by growth factors, cytokines, chemokines, and other mediators.³¹ Pathologic conditions are often associated with changes in the temporal, quantitative, or qualitative profile of soluble signals within the wound bed.^17- 18 Specifically, changes in the inflammatory profile have been associated with profound alterations in the overall outcome of wound healing.^7- 10,32

In the lower airway mucosa, inflammatory responses to injury are related to the development of fibrosis. Decreased COX-2 activation in and PGE₂ secretion by mucosal fibroblasts in response to lipopolysaccharide and/or IL-1β are associated with the development of chronic lung fibrosis, suggesting that, in the lung, PGE₂ plays a protective antifibrotic role.^16- 20 In contrast, in the skin, PGE₂ can play either proinflammatory or anti-inflammatory and fibrotic roles, based on which PGE₂ (EP) receptor is involved, with EP2 and EP4 being associated with anti-inflammatory effects.²⁵

To date no study, to our knowledge, has addressed the role of PGE₂ in particular and inflammatory mediators in general in the development of upper airway fibrosis. Data presented herein illustrate several important findings with respect to the role of IL-1β and PGE₂. First, IL-1β and COX-2 expression is up-regulated relatively quickly and persists for at least 48 hours after mucosal injury. Second, up-regulation of these 2 inflammatory genes decreases, but does not completely disappear, in the latter stages of mucosal repair (21 days after injury). This raises questions as to whether more extensive injury is capable of causing long-term activation of inflammatory pathways. Third, activation of inflammatory genes is localized to the anatomic borders of the injured mucosa. Although IL-1β and COX-2 are up-regulated in the tracheal mucosa immediately adjacent to the injured subglottic mucosa, this up-regulation is substantially lower than that measured at the site of injury. This finding is important because it illustrates that, in this experimental design, maximal activation of inflammatory mediators occurs in the posterior mucosa as a result of laser-induced damage and not in a generalized response due to the anterior approach to the subglottis (ie, incisional surgery/wounding of the cricothyroid membrane). Furthermore, it supports the use of IL-1β and PGE₂ detected in mucosal secretions as surrogate markers of injury and healing responses and, as such, serves to establish the validity of secretion analysis as a tool for monitoring the course of upper airway wound healing noninvasively.

In an experimental setting, analysis of injured mucosal tissue is a readily available and useful method to characterize the wound healing process by using genomic and proteomic tools. However, tissue analysis is not a viable tool for the analysis of injured human airways. Therefore, there exists a great need for minimally invasive techniques that allow for a detailed and precise characterization of the cellular and molecular processes ongoing within the healing tissue. Such techniques could be used to predict the long-term outcome of the tissue repair process after various injuries. We have previously demonstrated that it is possible to detect changing levels of inflammatory mediators (IL-1β and PGE₂) in the healing rabbit airway by analyzing mucosal secretions.³⁰ Herein we have extended these results with a focus on correlating the extent of mucosal injury with qualitative and quantitative changes in the secretion profile of IL-1β and PGE₂ and have demonstrated that laser injury to the posterior subglottic mucosa alters the secretion profile in a manner distinct from that of sham-operated animals. Our initial secretion analysis, in particular, demonstrated that, in an acutely injured airway, there is a correlation between the activation of inflammatory genes and a change in the secretion profile for these 2 soluble mediators. As such, we believe that this analytical tool may prove useful in characterizing the events occurring within the healing mucosa. It is important that neither IL-1β nor PGE₂ levels directly and clearly rise in a correlative manner with the degree of mucosal injury. Rather, there appears to be a more complex relationship between the levels of these 2 mediators and the extent of injury. Increasing laser power or extent of area covered appears to actually diminish the up-regulation of IL-1β observed at lower settings. Sham surgery alone induced a substantial increase in PGE₂ levels in secretions. Greater laser injury to the posterior mucosa appeared to shift the secretion profile toward earlier times. Together, these data suggest that key inflammatory mediators in mucosal secretions can potentially serve as surrogate markers of injury and predictors of healing outcome, but additional experiments are needed to transform secretion analysis into a truly predictive and reliable clinical tool. Nevertheless, these preliminary data help to establish the viability of this noninvasive approach in assessing some of the ongoing processes occurring within the healing tissue.

There exist multiple cell sources of inflammatory mediators during wound healing. Although they are not a dedicated inflammatory cell type, fibroblasts have been shown to participate in the inflammatory process. In the lower airway in particular, fibroblast production of PGE₂ has been linked to the development of lower-airway/pulmonary fibrosis.²⁰ The present study was aimed at replicating this analysis in the upper airway. Specifically, our results suggest that fibrotic laryngotracheal airway wound healing may differ from its counterpart in the lower airway.

Nuclear factor κB (p65/p50) translocation to the nucleus, in response to IL-1β, up-regulates COX-2 transcription and stimulates PGE₂ production.²⁷ Our data indicate that IL-1β stimulation triggers a rapid, dose-dependent translocation of p65 into the cell nucleus of upper airway fibroblasts that correlates with robust up-regulation of COX-2 and mPGES1, enzymes responsible for PGE₂ synthesis. Testing of the endogenous PGE₂ synthetic pathway in normal fetal and adult upper airway fibroblasts, as well as fibroblasts derived from pathologic (SGS) specimens, indicates that PGE₂ synthesis is appropriately regulated by primary inflammatory mediators such as IL-1β. Moreover, SGS-derived fibroblasts appear to produce more PGE₂ in response to IL-1β than the other 2 cell types. This is in contrast to data derived from lower airway studies, suggesting that mucosal healing in the respiratory tract may not be completely conserved. It should be noted that these data are somewhat preliminary, as only a few human-derived airway fibroblast cultures were available for analysis; however, the results were consistent for the 2 normal adult and the 2 SGS-derived samples. Subsequent studies will need to rely on greater sample size to begin to more quantitatively compare these fibroblast phenotypes. Nevertheless, this represents an important first step in characterizing upper airway mucosal fibroblast participation in the mucosal inflammatory and wound healing process.

Prostaglandin E₂ degradative enzymes such as 15-PGDH can be regulated at both tissue and cell levels by exogenous stimuli, resulting in increased tissue levels of PGE₂.^23- 24,33 In addition, PGE₂ receptor expression levels appear to be altered subsequent to tissue injury; these changes may play a role in the inflammatory and wound healing processes.^{11,14- 15,25} In our experimental system, the degradative and receptive components of the PGE₂ signaling pathway appear to be constitutively expressed and not regulated by inflammatory mediators such as IL-1β. Data presented earlier in this report indicate that mucosal fibroblasts express detectable levels of 15-PGDH and EP receptors and that protein levels are not altered by administration of exogenous IL-1β. It was somewhat surprising that, although IL-1β is a potent activator of fibroblast PGE₂ synthetic enzymes, it fails to significantly alter the expression profile of either the degradative enzyme 15-PGDH or the EP receptors. This phenomenon implies that not all arms of the PGE₂ signaling pathway (synthetic, degradative, receptive) are regulated by the same primary trigger, and that it is possible that, at the tissue level, IL-1β regulation of COX-2 and mPGES1 may be integrated with regulation of the EP receptors by other primary triggers such as growth factors.

This study addresses the cellular and molecular processes that contribute to the upper airway mucosal wound healing response subsequent to carbon dioxide laser injury. The results indicate the following: First, activation of inflammatory genes (IL-1β and COX-2) occurs early after injury, is relatively localized to the injury site, and partially resolves during the wound healing process. Second, this activation correlates with changes in the secretion profile of inflammatory mediators (IL-β and PGE₂), in a manner at least partially dependent on the extent of mucosal injury. Third, mucosal fibroblasts represent a putative source of PGE₂ present in mucosal secretions after injury. In contrast to their lower airway counterparts, mucosal fibroblasts derived from fibrotic (SGS) specimens do not appear to have an altered PGE₂ synthetic mechanism.

Correspondence: Patricia A. Hebda, PhD, Department of Pediatric Otolaryngology/Rangos Research Center, Children's Hospital of Pittsburgh, 3460 Fifth Ave, Pittsburgh, PA 15213 (hebda@pitt.edu).

Submitted for Publication: June 3, 2006; final revision received December 8, 2006; accepted December 30, 2006.

Author Contributions: Drs Sandulache and Hebda had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Sandulache, Li-Korotky, Dohar, and Hebda. Acquisition of data: Sandulache, Chafin, Li-Korotky, Otteson, and Hebda. Analysis and interpretation of data: Sandulache, Li-Korotky, Dohar, and Hebda. Drafting of the manuscript: Sandulache. Critical revision of the manuscript for important intellectual content: Chafin, Li-Korotky, Otteson, Dohar, and Hebda. Statistical analysis: Sandulache and Hebda. Obtained funding: Sandulache, Dohar, and Hebda. Administrative, technical, and material support: Chafin, Li-Korotky, Otteson, Dohar, and Hebda. Study supervision: Li-Korotky and Hebda.

Financial Disclosure: None reported.

Funding/Support: This study was supported by the American Society of Pediatric Otolaryngology (pilot grant; Dr Hebda), the Children's Hospital of Pittsburgh Research Advisory Committee (seed grant; Dr Hebda), and the Eberly Family Endowed Chair in Pediatric Otolaryngology Research and the Lester A. Hamburg Endowed Fellowship in Pediatric Otolaryngology (Dr Li-Korotky).

Previous Presentation: This study was presented in part at the Annual Meeting of the Association of Pediatric Otolaryngology; May 21, 2006; Chicago, Ill.

Acknowledgment: We thank Catarina Wong and Chia-Yee Lo for skilled technical assistance and Ryan Branski, PhD, for extensive discussion and interpretation of results.