Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The Clara cell secretory protein (CCSP) is a 16-kD homodimeric protein that is primarily secreted by nonciliated bronchiolar (Clara) cells of the mammalian lung, and is one of the most abundant proteins in the airway lining fluid of mammals (1). Although the molecular function of CCSP is unknown, studies in gene-targeted CCSP-deficient (CCSP/) mice indicate that CCSP is an important modulator of lung inflammation following lung infection or injury (2). In previous studies, CCSP protein levels and mRNA steady-state abundance were decreased in the lungs during acute respiratory infection to Pseudomonas aeruginosa (P.a.) (4). The decrease in CCSP in the lung following P.a. infection coincided with the early infiltration of inflammatory cells into the lung. Interestingly, administration of exogenous CCSP to the lungs concurrent with endotoxin challenge mitigated lung inflammation, suggesting that decreased CCSP levels during infection may facilitate the induction of inflammatory cell infiltration into the lung (S. Hayashida, K. S. Harrod, and J. A. Whitsett, unpublished results).

The lung-specificity and ontogeny of CCSP transcriptional induction has previously been shown to involve complex interactions of distinct nuclear factor subsets, including thyroid transcription factor 1 (TTF-1, also Nkx2.1), hepatocyte nuclear factor 3 (HNF-3), and CCAAT enhancer binding protein (C/EBP) (5). The 5' flanking region of CCSP has striking similarity to the lung-specific genes surfactant protein (SP)-B and SP-C (9). Both SP-B and SP-C genes are trans-activated by TTF-1 and HNF-3 in lung epithelial cells in vitro (10, 11). Although the mechanisms of lung-specific and developmentally regulated transcription are currently being studied, the mechanisms of diminished lung gene expression during respiratory disease are poorly understood. In studies of acute P.a. infection of the lung, SP-B and SP-C mRNA levels were decreased in a temporal manner similar to that of CCSP mRNA abundance (4). Given the similarities in transcriptional regulation between SP-B, SP-C, and CCSP, attenuated gene expression of critical lung homeostatic genes during acute respiratory infection may involve similar transcriptional regulatory mechanisms.

To determine the regulation of CCSP transcription in the lung epithelial cells during acute infection, CCSP promoter activity was assessed in reporter gene assays in Clara cell- like human lung epithelial cells (H441 cells) during P.a. infection in vitro. Transactivation of a 2.4-kb flanking region of the CCSP promoter was markedly downregulated by P.a. The decreased promoter activity of the CCSP 5' flanking region was localized to the proximal -166 bp of the CCSP promoter. TNF- also decreased CCSP promoter activity in a manner that was also localized to the proximal -166 promoter element. Furthermore, the P.a.-mediated decrease in CCSP promoter activity is conferred through autocrine or paracrine TNF-. Importantly, acute P.a. infection decreased CCSP promoter activity in the lungs of transgenic mice. These studies suggest that the P.a.-mediated decrease in CCSP during respiratory infection occurs through autocrine/ paracrine TNF- downregulation of critical cis-active elements in the -166 bp proximal CCSP promoter element.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Reagents

All experimental reagents were obtained from Sigma Chemical Company (St. Louis, MO) unless otherwise noted. All cell culture media were obtained from Gibco BRL (Rockville, MD). Cell culture sera was obtained from Hyclone (Logan, UT). TNF- and TNF- antibody were obtained from R&D Systems (Minneapolis, MN).

Cell Culture and Transfection Experiments

H441 human adenocarcinoma cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA) and cultured under recommended conditions. H441 cells were grown to 70-80% confluence, and transfected overnight using Fugene-6 (Roche Molecular Biochemicals, Indianapolis, IN). Typically, 1-2 µg of plasmid DNA was transfected using DNA:liposome concentrations of 1:2. Cytomegalovirus (CMV)--galactosidase (-Gal) (0.5 µg) were cotransfected to normal transfection efficiency between samples. Transfection of H441 cells routinely reached 80-90% of the total cells.

Bacterial Co-Culture and TNF- Treatment

P.a. (a generous gift of Dr. J. R. Wright, Duke University, Durham, NC ) was cultured overnight in LB broth at 37°C and 225 rev/min. P.a. was isolated by centrifugation and resuspended in sterile PBS. Bacterial titers were determined by serial dilution and colony-forming assay on 2xYT agar plates overnight at 37°C. Following transfection, 1 × 10⁴-5 × 10⁵ colony-forming units (cfu) were added to transfected H441 cells in culture for 24 h. For TNF- studies, 0.5-5 ng/ml of TNF- were added to transfected H441 cells in culture for 24 h. Studies of P.a. in either the presence or absence of penicillin-streptomycin had no effect on the data reported.

Plasmids

The full-length 2.4-kb CCSP promoter (a generous gift from J. A. Whitsett, Children's Hospital, Cincinnati, OH) was subcloned by Hind III restriction digestion and ligation into the pGL3 vector encoding the firefly luciferase gene (designated ^2.4CCSP-luc). Truncated deletion constructs of the 5' flanking region of the CCSP promoter were generated from the ^2.4CCSP-luc plasmid construct by the methods of Stripp and colleagues (9). Truncated deletions construct corresponding to 1.3 kb, 0.73 kb, 0.48 kb, and 0.16 kb were generated (designated ^1.3CCSP-luc, ^0.73CCSP-luc, ^0.48CCSP- luc, and ^0.16CCSP-luc, respectively). The 1.0-kb SP-D promoter was cloned from human genomic DNA by polymerase chain reaction (PCR) amplification of specific sequences using the sequence reported by Rust and coworkers (13), and subcloned into the pGL3 plasmid immediately upstream of the luc reporter gene sequence.

Reporter Gene Assays

Luciferase activity of cell lysates from transfected cell cultures was measured by injection of 100 µL of 1 mM luciferin (Promega, Madison, WI), and data collected for 10 s at 25°C with a TD 20/20 luminometer (Turner Designs, Sunnyvale, CA). -Gal activity in cell lysates was measured by hydrolysis of 1µM o-nitrophenyl -D- galactopyranoside per minute at 37°C. Luciferase activity data were reported as light units per unit of -Gal activity for each sample.

TNF- Secretion

TNF- protein in media and cell lysates was determined by standard enzyme-linked immunosorbent assay (ELISA) procedure (Biosource International, Camarillo, CA).

CCSP Promoter Activity in CCSP-rtTa Transgenic Mice

Transgenic mice encoding the 2.4-kb CCSP promoter and reverse tretracycline transactivator gene (CCSP-rtTa) were a generous gift of Dr. Jeffrey Whitsett, Children's Hospital, Cincinnati, OH. Expression of the rtTA gene was detected by reverse transcriptase-polymerase chain reaction (RT-PCR) analysis as reported previously (13). Administration of P.a. to the lungs of CCSP-rtTa transgenic mice was performed using procedures described elsewhere (4).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

P.a. Attenuates CCSP Promoter Activity through the -166 bp Proximal Promoter Element

P.a. cultures were grown overnight and titers determined by colony forming assay on agar plates. H441 cells were grown to ~ 70% confluence, and transfected for 24 h with either the full length ^2.4CCSP-luc plasmid construct, or the truncated CCSP promoter constructs ^1.3CCSP-luc, ^0.73CCSP-luc, ^0.48CCSP-luc, or ^0.16CCSP-luc. Following transfection, 1 × 10⁴ cfu of P.a. was co-cultured with transfected H441 cells for 24 h, and luciferase activity measured in cell lysates. Luciferase activity was readily measurable in all CCSP-luc transfected H441 cells (Figure 1A). Luciferase activity in H441 cells transfected with ^0.16CCSP-luc was reduced, but still readily detectable as compared with ^2.4CCSP-luc transfected cells. Following P.a. co-culture, luciferase activity was markedly attenuated in H441 cells transfected with all CCSP-luc constructs, including ^0.16CCSP-luc transfected cells. The reduction in luciferase activity in P.a. treated samples was ~ 30% of luciferase activity in untreated, control cells with all CCSP-luc constructs.

View larger version (17K): [in this window] [in a new window]   Figure 1.   CCSP promoter activity is decreased by P.a. in lung epithelial cells. (A) H441 cells were transfected with either the full-length 2.4CCSP-luc construct, or truncated deletions constructs corresponding to the proximal 1.3 kb, 0.73 kb, 0.48 kb, or 0.16 kb CCSP promoter elements (1.3CCSP-luc, 0.73CCSP-luc, 0.48CCSP-luc, or 0.16CCSP-luc). Transfected H441 cells were co-cultured with P.a. (1 × 104 cfu) for 24 h and luciferase activity measured in cell lysates. Luciferase activity was normalized to -Gal activity by co-transfection with CMV--Gal. Luciferase activity in the absence (black bars) or presence (white bars) of triplicate samples is shown (mean ± standard error). *P  0.05 as compared with untreated values of the same CCSP promoter construct. Data are representative of three separate experiments. (B) Dose-dependent decrease in CCSP promoter activity in 2.4CCSP-luc or 0.16CCSP-luc transfected H441 cells. Transfected H441 cells were co-cultured with increasing concentrations of P.a., and luciferase activity in cell lysates measured 24 h following administration. Luciferase activity (mean ± SE of triplicate samples) from 2.4CCSP-luc transfected cells (black bars) or 0.16CCSP-luc transfected cells (white bars) is shown. *Indicates statistical significance (P  0.05) of 2.4CCSP-luc transfected cells in the presence of P.a. as compared with untreated 2.4CCSP-luc transfected cells, while # indicates statistical significance of 0.16CCSP-luc transfected cells in the presence of P.a. as compared with untreated 0.16CCSP-luc transfected cells. ** Denotes statistical significance (P  0.05) from 2.4CCSP-luc transfected cells co-cultured with P.a. at 1 × 104 cfu. Data are representative of three separate experiments.

The dose-response of decreased CCSP promoter activity to co-culture with P.a. is shown in Figure 1B. Luciferase activity in ^2.4CCSP-luc or ^0.16CCSP-luc transfected H441 cells was decreased in co-cultures of 1 × 10⁴ cfu, 5 × 10⁴ cfu, 1 × 10⁵ cfu, and 5 × 10⁵ cfu of P.a. P.a. caused a dose-dependent decrease in luciferase activity in ^2.4CCSP- luc transfected H441 cells at doses of 5 × 10⁴ cfu and 1 × 10⁵ cfu. Luciferase activity in ^0.16CCSP-luc transfected H441 cells was decreased as compared with luciferase activity in untreated, transfected H441 cells when co-cultured with P.a., and showed a dose-dependent response at 5 × 10⁵ cfu of P.a. as compared with lower doses of P.a. (1 × 10⁴-1 × 10⁵ cfu). CMV--Gal transfections, as determined by -galactosidase activity, were not different in the absence or presence of P.a. in the culture media (data not shown).

TNF- Decreases CCSP Promoter Activity through the -166 bp Proximal CCSP Promoter Element

TNF- is abundantly expressed in the lungs during P.a. infection in vivo (4, 14), and has been shown to regulate SP-B and SP-C gene expression in lung epithelial cells (16, 17). To assess the role of TNF- in modulating CCSP promoter activity in lung epithelial cells, ^2.4CCSP-luc or ^0.16CCSP-luc transfected H441 cells were treated with TNF- at various concentrations for 24 h following transfection. Luciferase activity was decreased in cell lysates from either ^2.4CCSP- luc or ^0.16CCSP-luc transfected H441 cells at all doses of TNF- treatment (Figure 2A). In general, luciferase activity in ^0.16CCSP-luc transfected H441 cells was decreased as compared with luciferase activity from ^2.4CCSP-luc transfected H441 cells in untreated, as well as TNF- treated, samples. Luciferase activity in ^2.4CCSP-luc transfected H441 cells displayed a dose-dependent response to TNF- at 5 ng/ml. Luciferase activity in cell lysates from ^0.16CCSP- luc transfected H441 cells showed a dose-dependent response to TNF- at 2.5 ng/ml. CMV--Gal transfections were not different in the absence or presence of TNF- in the culture media at any concentration (data not shown).

View larger version (21K): [in this window] [in a new window]   Figure 2.   TNF- decreases CCSP promoter activity in lung epithelial cells. (A) Dose-dependent decrease in CCSP promoter activity by TNF- administration. Either 2.4CCSP-luc or 0.16CCSP- luc transfected H441 cells were treated with increasing doses of TNF-, and luciferase activity measured 24 h following administration. Luciferase activity (mean ± SE of triplicate samples) is shown, normalized to -Gal activity by co-transfection with CMV--Gal. *Indicates statistical significance (P  0.05) from untreated 2.4CCSP-luc transfected cells (black bars), while # indicates statistical significance (P  0.05) from untreated 0.16CCSP- luc transfected cells (white bars). (B) Direct comparison of decreased CCSP promoter activity at 24 h following P.a. (1 × 104 cfu; white bars) or TNF- (1.0 ng/ml; gray bars). Full-length 2.4CCSP-luc or truncated deletion constructs of the CCSP promoter (1.3CCSP-luc, 0.73CCSP-luc, 0.48CCSP-luc, or 0.16CCSP-luc) were treated with either P.a. or TNF-, and luciferase activity measured 24 h following treatment. Luciferase activity (mean ± SE of triplicate samples) are shown from three representative experiments. *Indicates statistical significance (P  0.05) in the presence of P.a. as compared with untreated transfected cells, while # indicates statistical significance in the presence of TNF- as compared with untreated transfected cells. Black bars, control.

A direct comparison of the CCSP promoter truncated deletion constructs to P.a.-mediated and TNF--mediated regulation of CCSP promoter activity is shown in Figure 2B. As shown previously, both P.a. co-culture and TNF- treatment reduce CCSP promoter activity as compared with untreated, CCSP-luc transfected H441 cells. The decreased activity of the CCSP promoter by P.a. co-culture or TNF- treatment was observed for all of the truncated deletion constructs tested. Likewise, both P.a. co-culture and TNF- treatment attenuated CCSP promoter activity of both the full length 2.4 kb CCSP promoter as well as the -166 bp proximal CCSP promoter element. Interestingly, the decrease in CCSP promoter activity to both P.a. and TNF- were strikingly similar.

P.a. Co-Culture Does Not Induce Cytotoxicity or Apoptosis of Transfected H441 Cells

To assess whether confounding factors may influence the P.a.-mediated or TNF--mediated decrease in CCSP promoter activity, cytotoxicity was assessed by trypan blue exclusion staining of ^2.4CCSP-luc transfected H441 cells. Transfected H441 cells in the absence of P.a. displayed less than 3% of the cells staining for trypan blue (Figure 3A). Trypan blue staining of transfected H441 cells in the presence of varying concentrations of P.a. varied from 10- 14% of the total cells analyzed. Importantly, cytotoxicity of transfected H441 cells to P.a. co-culture did not exhibit a dose-response relationship over a range of 1 × 10⁵-5 × 10⁶ cfu. In studies of TNF- treatment of ^2.4CCSP-luc transfected H441 cells, TNF- at a dose range of 5, 10, 25, or 50 ng/ml did not alter trypan blue staining of transfected H441 cells (Figure 3B), similar to findings with P.a. co-culture. Co-culture of P.a. or TNF- treatment did not induce cell injury in CCSP-luc transfected H441 cells.

View larger version (17K): [in this window] [in a new window]   Figure 3.   Cytotoxicity is not altered by P.a. or TNF- administration to transfected H441 cells. Cytotoxicity of transfected H441 cells was assessed by trypan blue exclusion 24 h following P.a. or TNF- administration. Transfected H441 cells were collected at 24 h following P.a. (A, dose range 1 × 105-5 × 106 cfu/ml) or TNF- (B, dose range 5 ng/ml-50 ng/ml), stained with 0.4% trypan blue in PBS. Viable (trypan blue excluded) and non-viable (trypan blue stained) cells were counted by hemocytometer for each sample. Data represent mean ± SE of viable (black bars) and nonviable (white bars) cells counted in four hemocytometer grid areas. (C) Transfected H441 cells were treated with P.a. (1 × 105 cfu/ml) or TNF- (5.0 ng/ml) for 24 h, and apoptosis assessed by FACS analysis of annexin V staining. Mean percentages of total cells (10,000 events) staining for annexin V are shown. Mean ± SE represent triplicate samples from each treatment group. Percent of apoptotic cells was not different between any treatment groups.

Early apoptotic events were assessed in ^2.4CCSP-luc transfected H441 cells following P.a. co-culture or TNF- treatment by annexin V staining and fluorescent-activated cell sorter analysis. Annexin V staining of total cells did not exceed 5% in untransfected or transfected H441 cells, or in the presence of P.a. or TNF-, as measured by fluorescein isothiocyanate- conjugated annexin V antibody (Figure 3C). Overall, apoptotic staining was not markedly changed by P.a. co-culture or TNF- treatment of ^2.4CCSP-luc transfected H441 cells.

P.a. or TNF- Decreases CCSP Promoter Activity, but Not SP-D Promoter Activity, in H441 Cells

A 1-kb 5' flanking region of the human SP-D promoter was generated by PCR amplification from H441 cell genomic DNA, and cloned into the pGL3 plasmid immediately upstream of the luc gene (designated ^1.0SP-D-luc). To assess whether P.a.- or TNF--mediated downregulation of CCSP promoter activity was conferred to other genes expressed by Clara cells in vivo, CCSP and SP-D promoter activity were assessed in parallel experiments following P.a. co-culture or TNF- treatment. As shown previously, CCSP promoter activity was markedly decreased 24 h following P.a. co-culture (1 × 10⁴ cfu) or TNF- treatment (5 ng/ml) (Figure 4). The SP-D promoter displayed high activity as assessed by luciferase activity when transfected into H441 cells. At 24 h following P.a. co-culture, SP-D promoter activity was unchanged as compared with untreated ^1.0SP-D-luc transfected H441 cells. In contrast, SP-D promoter activity was augmented ~ 40% by TNF- 24 h following treatment. CCSP promoter activity, but not SP-D promoter activity, was decreased by P.a. or TNF- in lung epithelial cells.

View larger version (14K): [in this window] [in a new window]   Figure 4.   Downregulation of CCSP, but not SP-D, promoter activity by P.a. or TNF- in lung epithelial cells. H441 cells were transfected with either 2.4CCSP-luc or 1.0SP-D-luc, and treated for 24 h with either P.a. (1 × 104 cfu/ml) or TNF- (5.0 ng/ml). Luciferase activity of cell lysates from each group was determined, and normalized to -Gal activity by co-transfection with CMV--Gal. Values represent the mean luciferase activity ± SE from triplicate samples. *Indicates statistical significance (P  0.05) of luciferase activity in 2.4CCSP-luc transfected cells following P.a. or TNF- as compared with untreated controls. **Denotes statistical significance (P  0.05) of luciferase activity from 1.0SP-D-luc transfected cells after TNF- treatment as compared with untreated 1.0SP-D- luc transfected cells. Data are representative of three separate experiments.

Neutralizing Antibody to TNF- Blocks Both TNF-- or P.a.-Mediated Downregulation of the CCSP Promoter

TNF- activity was inhibited utilizing a specific neutralizing antibody (anti-TNF- Ab) against human TNF-. TNF- treatment of either ^2.4CCSP-luc or ^0.16CCSP-luc transfected H441 cells decreased luciferase activity 24 h following administration (Figure 5A). In the presence of anti-TNF- Ab, decreased CCSP promoter activity was reversed in both ^2.4CCSP-luc and ^0.16CCSP-luc transfected H441 cells. Attenuation of the TNF--mediated downregulation of the truncated -166 bp proximal CCSP promoter was accomplished by 100 ng/ml of anti-TNF- Ab. In ^2.4CCSP- luc transfected H441 cells, 225 ng/ml of the anti-TNF- Ab was necessary to reverse the TNF--mediated downregulation of the full length 2.4 kb CCSP promoter. Anti- TNF- Ab had no effect on CMV--Gal transfection or cell cytotoxicity (data not shown).

View larger version (17K): [in this window] [in a new window]   Figure 5.   Neutralization of TNF- attenuates the TNF-- and P.a.-mediated decrease in CCSP promoter function. 2.4CCSP-luc (black bars) or 0.16CCSP-luc (white bars) transfected H441 cells were treated with TNF- (A) or P.a. (B) for 24 h in the presence of a neutralizing antibody against human TNF- (TNF- Ab). Values denote mean ± SE luciferase activity in 2.4CCSP-luc transfected or 0.16CCSP-luc transfected samples (n = 3 per group). *Indicates statistical significance (P  0.05) of 2.4CCSP-luc transfected cells in the presence of P.a. or TNF- as compared with untreated transfected cells, and #indicates statistical significance of 0.16CCSP-luc transfected cells in the presence of P.a. or TNF- as compared with untreated transfected cells. Data represent three separate experiments performed on different days.

To assess the role of TNF- in the P.a.-mediated decrease in CCSP promoter function, anti-TNF- Ab was administered concurrently with P.a. to ^2.4CCSP-luc or ^0.16CCSP-luc transfected H441 cells. As shown previously, P.a. administration decreased promoter activity in both ^2.4CCSP-luc and ^0.16CCSP-luc transfected H441 cells (Figure 5B). Administration of anti-TNF- Ab attenuated the P.a.-mediated decrease in CCSP promoter activity in both ^2.4CCSP-luc transfected and ^0.16CCSP-luc transfected H441 cells. Five hundred (500) ng/ml of anti-TNF- Ab was able to partially reverse the decrease in CCSP promoter function by P.a. in studies of the full-length 2.4 kb CCSP promoter, but had no effect on the -166 bp proximal CCSP promoter. Anti-TNF- Ab at 1,000 ng/ml was able to fully reverse the decrease in CCSP promoter activity by P.a. in both ^2.4CCSP-luc and ^0.16CCSP-luc transfected H441 cells.

TNF- Protein Secretion in Lung Epithelial Cell Cultures to P.a.

To assess production of TNF- in H441 cells by P.a., TNF- protein levels were assessed in cell lysates and in the culture media 6 h following the administration of P.a. to ^2.4CCSP-luc transfected H441 cells. TNF- protein was not detectable in cell lysates in any treatment group (data not shown), or in the culture media from untreated H441 cells (Figure 6). Transfection of H441 cells with ^2.4CCSP-luc did not induce TNF- secretion. At 6 h following P.a., TNF- protein was readily detectable by ELISA in the culture media of H441 cells following P.a. Co-culture of P.a. markedly increases TNF- secretion in ^2.4CCSP-luc transfected H441 cells before the decrease in CCSP promoter activity.

View larger version (17K): [in this window] [in a new window]   Figure 6.   P.a.-induced TNF- secretion by transfected lung epithelial cells. TNF- protein secretion was assessed in 2.4CCSP-luc transfected H441 cells in the presence of P.a. co-culture. TNF- protein levels secreted from transfected H441 cells were assessed by ELISA in the culture media. *Denotes statistical significance (P  0.05) from untreated 2.4CCSP-luc transfected H441 cells. Values represent mean ± SE of triplicate samples.

CCSP Promoter Regulation In Vivo by P.a.

Transgenic mice encoding the 2.4 kb CCSP promoter conferring regulation of the reverse tetracycline transactivator gene (CCSP-rtTa) were infected with 1 × 10⁸ cfu of P.a., and rtTa gene expression assessed at 24 and 48 h following infection. This dose (10⁸ cfu) has been shown previously to decrease CCSP mRNA and protein in the lungs of mice (4). RT-PCR analysis of rtTa mRNA abundance was determined in total RNA from the lungs of uninfected and P.a.-infected CCSP-rtTa mice. RT-PCR analysis of -actin mRNA was used for determinations of unrelated mRNA expression and comparisons between groups. Steady-state abundance of -actin mRNA in the lung was not altered by P.a. infection as compared with that of uninfected CCSP-rtTa mice (Figure 7). Expression of rtTa was readily detectable in the lungs of uninfected CCSP-rtTa mice similar to previous reports (18). In the lungs of P.a.-infected CCSP-rtTa mice, rtTa mRNA abundance was markedly decreased at 24 h following infection, as compared with rtTa mRNA abundance in the lungs of uninfected mice. The appearance of faint PCR products of rtTA could be observed upon overexposure of ethidium-stained agarose gels (data not shown). At 48 h following infection, rtTa mRNA abundance was less dramatically decreased as compared with levels of rtTa expression from mice at 24 h following infection. This is similar to data reported previously that CCSP mRNA levels are more dramatically decreased at 24 than at 48 h following P.a. infection (4). Collectively, these data suggest that CCSP promoter activity is decreased in the lungs of mice following acute P.a. infection.

View larger version (28K): [in this window] [in a new window]   Figure 7.   CCSP promoter activity in CCSP-rtTa transgenic mice following P.a. infection. CCSP-rtTa transgenic mice were administered either 1 × 108 cfu of P.a. or vehicle intratracheally, and rtTa mRNA abundance assessed by RT-PCR analysis of total lung RNA at 24 and 48 h after infection. PCR products shown were visualized by ethidium-stained agarose gel electrophoresis. Steady-state levels of -actin mRNA are shown as controls. Faint bands of rtTa PCR products can be seen in 24-h, P.a.- infected samples by overexposure of the agarose gel to UV.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study, CCSP promoter activity is markedly decreased in lung epithelial cells by P.a. co-culture or TNF- treatment. Decreased promoter activity by P.a. or TNF- was observed in all truncated deletion constructs of the full-length CCSP promoter, including the -166 bp proximal CCSP promoter element. Regulation of CCSP promoter activity by P.a. or TNF- was dose-dependent in both ^2.4CCSP-luc and ^0.16CCSP-luc transfected lung epithelial cells. Cytotoxicity and induction of apoptosis were not apparent in transfected lung epithelial cells following P.a. or TNF- administration. Downregulation of promoter activity was specific to CCSP promoter element, as human SP-D promoter activity was not changed or increased by P.a. or TNF- administration, respectively. Neutralization of TNF signaling attenuated both P.a. or TNF--mediated downregulation of CCSP promoter function. Lastly, CCSP promoter activity was markedly decreased in the lungs of CCSP-rtTa transgenic mice following P.a. infection. These data support the concept that P.a. infection decreases CCSP through downregulation of CCSP promoter function, and involves both cis-active elements in the proximal 166 bp promoter region and autocrine or paracrine signaling by TNF-.

Previous studies have shown that CCSP mRNA and protein are markedly decreased in the lungs of mice following acute P.a. infection (4). CCSP deficiency has been associated with increased lung inflammation and injury following acute viral and bacterial infection in gene-targeted animal models, suggesting that CCSP may have important immunomodulatory function in the lung (3, 4, 18). The mechanisms of decreased CCSP in lung following infection or injury have not been elucidated. In the present study, CCSP promoter activity was markedly decreased in lung epithelial cells following P.a. or TNF- administration. The H441 cell line has been a beneficial model for studying Clara cell gene expression of CCSP in vitro (5, 7, 8, 19, 20). Cell-specific and developmental regulation of the CCSP gene in H441 lung epithelial cells has been further substantiated by in vivo studies in transgenic mice (7, 8). Herein, P.a. infection in vivo attenuated CCSP promoter-conferred gene expression in the lungs of transgenic mice, corroborating the in vitro findings in lung epithelial cells. Thus, the findings herein provide strong evidence that the regulation of CCSP promoter function by P.a. and TNF- are likely mechanisms of regulation during acute P.a. infection in the lung.

In the current study, the P.a. and TNF- mediated downregulation of the CCSP promoter occurred in both the full length 2.4 kb promoter element, as well as regulation through the proximal -166 bp CCSP promoter element. Clara cell-specific and developmental transactivation of the CCSP promoter in the lung and in cultured cell systems is regulated, at least in part, through the proximal promoter element studied herein (7, 8, 20). A number of DNA-binding transactivating nuclear factors have been implicated in the regulation of the proximal CCSP promoter region, including TTF-1, HNF-3, and C/EBP (5, 20). Furthermore, interactions between TTF-1 and HNF-3 nuclear factors in the proximal CCSP promoter appear to potentiate the transactivating capabilities of each individual nuclear factor (9). Thus, regulation of any one nuclear factor within the proximal cis-acting promoter element may have important influence on the transactivation of CCSP gene expression. The current study suggests that decreased CCSP promoter activity by P.a. or TNF- is regulated through the -166 bp proximal promoter region in lung epithelial cells.

The Clara cells of the mammalian lung can express a number of genes important in lung function or homeostasis, including the surfactant genes, SP-A, SP-B, and SP-D, as well as CCSP (1). In the current study, P.a.- and TNF--mediated modulation of CCSP promoter function was specific to CCSP, and not SP-D promoter activity of a 1-kb SP-D 5' flanking promoter element. A 285 bp 5' flanking region proximal to the SP-D transcriptional start site has been shown to be sufficient to restrict SP-D expression to lung epithelial cells (12). The regulation of the SP-D promoter by P.a. or TNF- has not been previously explored. Following administration of P.a. to the lung, SP-D mRNA steady-state levels were increased coinciding with decreased CCSP mRNA levels (Hayashida, unpublished results). In the current study, P.a. did not alter SP-D promoter activity, whereas TNF- caused a moderate induction of SP-D promoter function. Interestingly, SP-D promoter regulation by P.a. was discordant with SP-D promoter regulation by TNF-. These findings suggest that regulation of the CCSP promoter by P.a. and TNF- may be distinct from other Clara cell-specific gene regulation, and does not involve ubiquitous downregulation of lung epithelial cell gene expression. Indeed, He and associates suggested recently that SP-D promoter regulation is likely distinct from other lung specific genes, including CCSP (21).

In the current study, modulation of CCSP promoter activity by P.a. was attenuated by the addition of a neutralizing antibody against TNF-. In a mouse model of acute P.a. infection of the lung, TNF- is markedly induced early during the course of infection (4, 14). Induction of TNF- following P.a. coincides with the decrease in CCSP and SP-C mRNA levels in the lung (4). In previous studies of LPS-mediated lung injury, inhibition of TNF- signaling in the distal lung epithelium was able to reverse the decrease in SP-C mRNA, suggesting that TNF- may have an important role in regulation of lung gene expression during acute injury or infection (22). TNF- has been shown to decrease SP-C promoter activity and SP-B and SP-C mRNA in cultured lung epithelial cells and in the lungs of mice (15, 16). The findings in the current study suggest that TNF- may, in part, attenuate CCSP promoter activity in the lung during acute P.a. infection.

Previous studies have implicated TNF- in altered lung gene expression both in vitro and in vivo. Transgenic overexpression of TNF- specifically in the lungs of mice induced a severe alveolitis with altered alveolar type II morphology (23). Administration of TNF- to the lungs of mice markedly decreased SP-C mRNA abundance coinciding with mild lung histopathology (15). In both in vitro and in vivo studies, TNF- decreased SP-C promoter activity through cis-active elements confined to the proximal 320 bp promoter element adjacent to the SP-C transcriptional initiation site (15). Likewise, decreased rabbit SP-B gene expression in lung epithelial cell cultures by TNF- also involves DNA elements in the proximal 236 bp promoter element (24), similar to that of SP-C (15), and the findings presented herein regarding CCSP. Collectively, these studies suggest that TNF- may play an important role in lung function and homeostasis through transcriptional regulation of critical lung proteins.

In the present study, acute P.a. infection in the lungs of CCSP-rtTa transgenic mice decreased CCSP promoter activity early during the course of infection. The CCSP promoter has been used extensively to confer gene expression of both mammalian and nonmammalian genes in the epithelium of large and small airways of mice (8, 13, 25). Likewise, transgenic technology has been beneficial for the study of CCSP promoter regulation in vivo, providing a critical mechanism for validation of promoter regulation studies in vitro (7, 8). In the current study, CCSP promoter activity was decreased in the lung epithelium following P.a. infection, consistent with the decrease in CCSP promoter activity in lung epithelial cells following co-culture with P.a. Importantly, the decrease in CCSP promoter activity in vivo to P.a. infection at 24 and 48 h is consistent with the decrease in CCSP mRNA abundance during P.a. infection reported previously (4). Although not directly tested in the current study, induction of TNF- during acute P.a. infection has been demonstrated before the decrease in CCSP mRNA abundance (4, 14). Thus, the findings in transgenic mice in the current study are consistent with the regulation of CCSP promoter activity by P.a. in vivo, and may involve, in part, autocrine or paracrine TNF- signaling in the lung.

The present findings are distinct from recent studies, having shown an induction of CCSP mRNA by TNF- and interferon- using a different cultured human bronchial epithelial cell line (26, 27). In the studies of Yao and colleagues, CCSP mRNA levels were increased in BEAS-2B cells following TNF- treatment, in part, through post-transcriptional mechanisms (27). Both promoter activity and transcriptional regulation were not altered by TNF-. The differences between the findings of Yao and colleagues and the findings presented herein include the use of different lung epithelial cell lines and the dose of TNF- administration. At doses of TNF- similar in both studies, CCSP mRNA steady-state levels are decreased in BEAS-2B cells (27). The H441 human Clara cell-like in vitro model has been used extensively for the study of cell-specific and developmental regulation of CCSP promoter activity and gene expression (5, 19, 20). Furthermore, the in vitro findings presented herein are supported by the in vivo studies in the CCSP-rtTa transgenic mice. In the present study, the decrease in CCSP promoter activity by P.a. or TNF- in lung epithelial cells corresponds with the observed decrease in CCSP mRNA and protein levels in the lungs of mice following acute P.a. infection (4).

The mechanisms of lung-specific gene regulation during infection and injury are central to the elucidation of dysfunction and homeostasis in the diseased lung. A number of lung-specific proteins are markedly decreased following acute viral or bacterial infection, including SP-B, SP-C, CCSP, and aquaporin 5 (AQP-5) (4, 28). Furthermore, decreased abundance of these proteins is regulated, at least in part, at the level of gene expression (4, 28). Importantly, gene-targeted animal models of SP-B, CCSP and AQP-5 have shown that these proteins are critical for normal lung function and homeostasis (2, 29, 30). Thus, the decreased expression of these, as well as other genes, likely contributes to the decline in lung function and homeostasis during acute infection.

In the present study, P.a. and TNF- markedly decreased CCSP promoter activity in lung epithelial cells in a dose-dependent manner. Modulation of CCSP promoter function by either P.a. or TNF- was confined to the proximal CCSP promoter element. Interestingly, autocrine or paracrine secretion of TNF- by lung epithelial cells during P.a. co-culture may, in part, constitute the mechanism by which P.a. regulates CCSP promoter function in lung epithelial cells. CCSP promoter activity was decreased in the lungs of transgenic mice following acute P.a. infection. During acute P.a. lung infection, CCSP protein levels and mRNA abundance are markedly decreased. Ablation of CCSP function during acute infection exacerbates lung inflammation and injury. Thus, decreased CCSP levels during acute infection may contribute to lung pathogenesis. The transcriptional regulation of critical lung proteins, such as CCSP, is likely central to understanding the lung dysfunction and altered homeostasis that occurs during acute respiratory infection.