Up-regulation of SFTPB expression and attenuation of acute lung injury by pulmonary epithelial cell-specific NAMPT knockdown

Abstract

Although a deficiency of surfactant protein B (SFTPB) has been associated with lung injury, SFTPB expression has not yet been linked with nicotinamide phosphoribosyltransferase (NAMPT), a potential biomarker of acute lung injury (ALI). The effects of Nampt in the pulmonary epithelial cell on both SFTPB expression and lung inflammation were investigated in a LPS-induced ALI mouse model. Pulmonary epithelial cell-specific knockdown of Nampt gene expression, achieved by the crossing of Nampt gene exon 2 floxed mice with mice expressing epithelial-specific transgene Cre or by the use of epithelial-specific expression of anti-Nampt antibody cDNA, significantly attenuated LPS-induced ALI. Knockdown of Nampt expression was accompanied by lower levels of bronchoalveolar lavage (BAL) neutrophil infiltrates, total protein and TNF-α levels, as well as lower lung injury scores. Notably, Nampt knockdown was also associated with significantly increased BAL SFTPB levels relative to the wild-type control mice. Down-regulation of NAMPT increased the expression of SFTPB and rescued TNF-α-induced inhibition of SFTPB, whereas overexpression of NAMPT inhibited SFTPB expression in both H441 and A549 cells. Inhibition of NAMPT up-regulated SFTPB expression by enhancing histone acetylation to increase its transcription. Additional data indicated that these effects were mainly mediated by NAMPT nonenzymatic function via the JNK pathway. This study shows that pulmonary epithelial cell-specific knockdown of NAMPT expression attenuated ALI, in part, via up-regulation of SFTPB expression. Thus, epithelial cell-specific knockdown of Nampt may be a potential new and viable therapeutic modality to ALI.—Bi, G., Wu, L., Huang, P., Islam, S., Heruth, D. P., Zhang, L. Q., Li, D.-Y., Sampath, V., Huang, W., Simon, B. A., Easley, R. B., Ye, S. Q. Up-regulation of SFTPB expression and attenuation of acute lung injury by pulmonary epithelial cell-specific NAMPT knockdown.

Acute lung injury (ALI) and its most severe form, acute respiratory distress syndrome (ARDS), are common life-threatening conditions leading to respiratory failure (1, 2), and the mortality of ARDS remains close to 40% (3). This persistent lethality is, in part, because the etiology of ALI/ARDS is incompletely understood, and effective therapies are scarce. Further knowledge of ALI/ARDS pathophysiology is needed to develop more effective interventions. It is increasingly recognized that a critical, unmet need is to identify new genetic markers (4, 5) that would provide novel, mechanistic insights and therapeutic targets.

Our previous study identified nicotinamide phosphoribosyltransferase (NAMPT), also called pre-B-colony enhancing factor (6) or Visfatin (7), as an ALI/ARDS biomarker (8). This finding was corroborated by epidemiologic evidence from Bajwa et al. (9) that variations in NAMPT promoter polymorphisms alter the risk of developing ARDS. We demonstrated that overexpression or down-regulation of NAMPT expression aggravated or attenuated LPS + ventilator-induced ALI, respectively (8, 10). However, the underlying molecular mechanisms are not fully understood, and the therapeutic use of NAMPT inhibition against ALI has not been actively explored.

The alveolar epithelial barrier is essential in the pathogenesis and recovery from ALI (11, 12). Alveolar epithelial cells function in ion transport, surfactant production, and the secretion of inflammatory cytokines and chemokines that recruit and activate immune cells in normal lung physiology (13). However, excessive release of proinflammatory mediators alters lung physiology in ALI. Previously, we found that small interfering RNA (siRNA) inhibition of NAMPT decreased the release of inflammatory cytokines from unstimulated, as well as TNF-α- or IL-1β-stimulated, pulmonary epithelial cells (14–16). We reasoned that pulmonary epithelial cell-specific NAMPT knockdown may be a potential therapeutic strategy in ALI.

Surfactant protein B (SFTPB) is major component of pulmonary surfactant and is secreted by both alveolar type II and club lung epithelial cells (17). SFTPB is the only surfactant protein strictly required for breathing, as its absence is associated with lethal respiratory failure in mice and humans (18). Decreased SFTPB concentrations contribute to the severity of lung inflammation and injury following infection. The anti-inflammatory properties of SFTPB provide protection from oxygen-induced and LPS endotoxin-induced lung injuries. Whether NAMPT can regulate epithelial SFTPB expression under normal pulmonary physiology or during ALI has not been explored.

In this study, we first investigated whether pulmonary epithelial cell-specific knockdown of Nampt expression could attenuate LPS-induced ALI in mice using conditional Nampt knockdown mice. Secondly, we explored whether pulmonary epithelial cell-targeted expression of an anti-Nampt cDNA could attenuate LPS-induced ALI in mice. Thirdly, we determined the corresponding effects of Nampt on SFTPB expression in human lung epithelial cells and probed relevant regulatory mechanisms.

MATERIALS AND METHODS

Reagents

Roswell Park Memorial Institute (RPMI) 1640 (11875), DMEM (11965), fetal bovine serum (FBS; 14190), penicillin-streptomycin (15140), Click-It Nascent RNA Capture Kit (C10365), SuperScript Vilo cDNA Synthesis SuperMix (11754-050), TaqMan gene expression assays for human SFTPB (1090667), and human 18s rRNA (4318839) were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Escherichia coli 0111:B4 endotoxin (LPS, L4391), p38 inhibitor SB239063, JNK inhibitor SP600125, and FK866 were purchased from Sigma-Aldrich (St. Louis, MO, USA). TNF-α ELISA kit (MTA00B) and recombinant human (rh)TNF-α (210-TA) were obtained from R&D Systems (Minneapolis, MN, USA). NAMPT (pre-B cell colony-enhancing factor) antibody (A300-372A) was purchased from Bethyl Laboratories (Montgomery, TX, USA). SFTPB antibody (sc-133143) and a mouse anti-human glyceraldehyde 3-phosphate dehydrogenase (GAPDH; sc-47724) were obtained from Santa Cruz Biotechnology (Dallas, TX, USA). Acetyl-histone H3 lysine 9 (Lys9; 9649 antibody) and Simple Chromatin Immunoprecipitaion (ChIP) Enzymatic Chromatin IP Kit (9003) were purchased from Cell Signaling Technology (Beverly, MA, USA).

LPS-induced ALI animal lung inflammation mode

All mouse experiments were conducted in accordance with National Institutes of Health (Bethesda, MD, USA) guidelines and were approved by the University of Missouri Kansas City Animal Care and Use Committee. All mice were fed an AIN-76A diet and water ad libitum. They were housed under controlled conditions (25 ± 2°C; 12-h light/dark periods).

A Nampt gene target vector, containing a floxed exon 2 of the Nampt gene, was electroporated into a B6-White Murine ES Cell Line (SCR011; EMD Millipore, Billerica, MA, USA) and injected into C57BL/6J mouse blastocysts in the Transgenic Animal Core, at the University of Missouri (Columbia, MO, USA). The generated chimeric homozygous floxed (Nampt^F/F) mice were verified by PCR analyses of their tail DNAs. Inducible pulmonary epithelial Nampt knockdown (Nampt^PE+/−) mice were generated by mating Nampt^F/F mice with mice carrying epithelial-specific and tamoxifen-inducible transgene Cre, B6N.129S6(Cg)-Scgb1a1tm1(cre/ERT)Blh/J mice (016225; The Jackson Laboratory, Bar Harbor, ME, USA), followed by administration of tamoxifen (T5648; Sigma-Aldrich) induction by intraperitoneal injection, once a day for 5 consecutive days at a dosage of 40 mg/kg/d. Tamoxifen was dissolved in sunflower seed oil (S5007; Sigma-Aldrich) at a concentration of 12 mg/ml. Control mice were injection with the same volume of sunflower seed oil.

C57BL/6J mice of either pulmonary epithelial Nampt wild type (Nampt^PE+/+) or Nampt^PE+/−, 8–10 wk old, were anesthetized with ketamine/acepromazine (100 and 5 mg/kg, i.p.), intubated with a 20-G catheter, and administered intratracheally with either PBS or LPS (2 mg/kg/mouse, diluted in PBS; Sigma-Aldrich). After 24 h, the mice were anesthetized and intubated again before they underwent bronchoalveolar lavage (BAL) of both lungs with HBSS (1 ml/mouse). The recovered BAL fluid was used for various assays, such as total protein, cytokine, and differential cell counting, as we previously described (8, 10). Lung morphology and severity scoring were conducted according to standard procedures (19).

Construction of recombinant anti-Nampt single chain variable fragment adenovirus vectors

We expressed and purified recombinant mouse Nampt protein and obtained 2 human against mouse Nampt antibody gene clones, termed NAMPT-single chain variable fragment 1 [scFv1; antibody 1 (Ab1)] and -scFv2 (Ab2), by screening an HuScL-2TM Phage Display Naive Human scFv Library (http://www.creative-biolabs.com/). The coding sequence of Ab2, driven by the mouse surfactant protein C promoter (SPC3.7), was cloned into the adenovirus (Ad) expression vector, pAdX-PRLS-ZsGreen1 (632264; Clontech Laboratories, Mountain View, CA, USA), by In-Fusion Cloning (638916), following the manufacturer’s protocols. PCR products for the mouse SPC3.7, Nampt scFVAb2, and bovine growth hormone (bight) poly A signal were amplified from C57BL/6J liver genomic DNA, cloning vector (pCAGGS)-Nampt-scFvAb2 DNA, and pcDNA3.1-Peredox-mCherry (Plasmid 32383; Addgene, Cambridge, MA, USA) (20), respectively. Each PCR product was generated to contain a 15 base pair (bp) sequence at each end that was homologous with the vector or PCR fragment to which it would be joined by In-Fusion Cloning. The Adeno-X-LacZ construct from the Clontech System was generated to serve as a control. All recombinant expression vectors were sequence verified. Primers used for cloning are presented in Table 1. Endotoxin-free plasmids were transfected into the Adeno-X 293 Cell Line (632271; Clontech Laboratories) with the Caliphos Transfection Kit (631312; Clontech Laboratories). Recombinant Ads were isolated and purified by the Maxi Purification Kit (631533; Clontech Laboratories). The viral titers were determined by the Adeno-X Quantitative PCR Titration Kit (632252; Clontech Laboratories).

TABLE 1.

Cloning primers

Primer name	Primer sequence, 5′–3′
mSPC3.7-proF2	GTAACTATAACGGTCTCAATACTTTTTGAGAGTG
mSPC3.7-pro21-R1	CAGAACCGAAGCCCAGATAAGC
scFv-SPC3.7-F	TGGGCTTCGGTTCTGGCCACCATGGAGGTGCAGCTGTTGGAGTCT
scFv-14R	TTAATGGTGATGGTGATGATGGGCCGCCCGTTTGATTTCCAC
scFv-BGH-F	CACCATCACCATTAACGACTGTGCCTTCTAGTTGCC
bGH-PolyA R	ATTACCTCTTTCTCCCTCAGAAGCCATAGAGCCCACC

Underlined sequences denote the 15 nt homology sequence for in-fusion cloning.

In vivo Ad transduction

C57BL/6J wild-type mice, 8–10 wk old, were anesthetized with pentobarbital at a dose 60 mg/kg body weight by intraperitoneal injection, intubated with a 20-G catheter and administered intratracheally with 100 µl of either Ad-SPC-Nampt-scFv or Ad-control-insert virus solution (1 × 10⁹ infectious units). Seventy-two hours later, the mice were anesthetized and intubated for challenge with either PBS or LPS. Twenty-four hours later, the mice were euthanized. BAL collection for various assays and lung morphology and severity scoring were done as previously described (Table 2).

TABLE 2.

RT-PCR primers and products sizes

	Primer sequence, 5′–3′
Product	Forward	Reverse	Size (bp)	Accession no.
IL-6	GGGACTGATGCTGGTGACAA	TCTGCAAGTGCATCATCGTT	210	NM_031168.1
TNF-α	CCACCACGCTCTTCTGTCTAC	CCTTGAAGAGAACCTGGGAGT	303	NM_001278601
β-Actin	CAAACATGATCTGGGTCATCTTCTC	GCTCGTCGTCGACAACGGCTC	487	NM_001101

Cell culture

The A549 cell (CCL-185) and H441 cell (HTB-174) were obtained from American Type Culture Collection (Manassas, VA, USA). A549 cells, which are human adenocarcinomic alveolar basal epithelial cells, were maintained in DMEM, supplemented with 10% FBS, 2 mM glutamine, and penicillin/streptomycin (100 U/100 µg/ml). H441 cells, which are also human pulmonary adenocarcinomic epithelial cells, were maintained in RPMI, supplemented with 10% FBS, 2 mM glutamine, and penicillin/streptomycin. All cells were cultured at 37°C in a humidified atmosphere of 5% CO₂, 95% air. Cells from each primary flask were detached with 0.25% trypsin, resuspended in fresh culture medium, and seeded into 6-well plates for Western blotting or RT-PCR analyses.

Transfection of Nampt siRNA into A549 cells and H441 cells

NAMPT stealth siRNA was designed based on the human NAMPT cDNA reference sequence (NM_005746.1) using the Block-It RNAi Designer (Thermo Fisher Scientific) and transfected into human H441 cells, as previously described (16). To transfect NAMPT stealth siRNA into A549 and H441 cells, cells were seeded overnight in the regular growth medium (without antibiotics) so that they would be 80–90% confluent at the time of transfection. For each transfection in 48-well plates, 25 pM NAMPT stealth siRNA or scramble RNA was diluted in 12.5 μl Opti-Mem I without serum and gently mixed with 0.5 μl Lipofectamine 2000, diluted in 12.5 μl Opti-Mem I (31985-062; Thermo Fisher Scientific). After incubation for 15 min at room temperature, NAMPT stealth siRNA and Lipofectamine 2000 complexes or scrambled RNA and Lipofectamine 2000 complexes were added to each well. Cell culture plates were gently mixed by rocking back and forth. The amount of NAMPT stealth siRNA and Lipofectamine 2000 was adjusted according to the different sizes of cell culture plates. Transfected cells were incubated further at 37°C for 48 h until the treatment with LPS or TNF-α was performed.

Expression of rhNAMPT and mutant NAMPTs in H441 cells

H441 cells were transiently transfected with the pCAGGS-human (h)NAMPT and mutant NAMPT constructs (pCAGGS-H247E) using Lipofectamine 2000 for 48 h before further treatment, as previously described (14). Cell lysates were harvested for analysis.

Western blotting

Western blot analysis was performed as previously described (14). In brief, cell lysates were mixed with RIPA buffer and boiled for 5 min. Equal amounts of proteins (10 µg for NAMPT determination; 30 µg for SFTPB determination) were resolved by SDS-PAGE and transferred onto PVDF membranes (Immobilon P; EMD Millipore). The membranes were probed with specific antibodies, as described in the figure legends, followed by detection with horseradish peroxidase-conjugated goat anti-rabbit IgG or anti-mouse IgG. Bands were visualized by ECL (Pierce ECL Western Blotting Substrate, 32106; Thermo Fisher Scientific) with a FluorChem M Imager (ProteinSimple, San Jose, CA, USA) and quantified by AlphaView Software SA, v.3.4.0.0.

Isolation of RNA and RT-PCR analysis

Total cell RNA was isolated with a mirVana miRNA Isolation Kit (AM1561; Thermo Fisher Scientific) according to the supplier’s instructions. RT-PCR was performed using the SuperScript III RNA PCR Kit (18080-044; Thermo Fisher Scientific). PCR products were separated on a 1.5% agarose gel and stained by ethidium bromide (0.5 μg/ml). The band image was acquired by a FluorChemM Imager and Alpha View Software.

Nascent RNA capture and real-time PCR

Nascent RNA was captured, according to the manufacturer’s protocol of the Click-It Nascent RNA Capture Kit (Thermo Fisher Scientific). In brief, 0.25 mM ethylene uridine ribonucleotide homologs was added into the culture medium and was incorporated into the cells for 3 h. Total RNA was prepared and quantified. Ethylene uridine-labeled RNAs were biotinylated with 0.25 mM biotin azide and captured. The biotinylated RNAs were precipitated with ethanol and resuspended in DNAse/RNAse-free ultrapure water. The biotinylated RNAs mixed with Dynabeads MyOne Streptavidin T1 magnetic beads in Click-It RNA binding buffer and heated at 68°C for 5 min, followed by incubation at room temperature for 30 min while gently vortexing. The beads were immobilized using the DynaMag-2 spin magnet and were washed with Click-It Wash Buffer 1 and 2. The washed beads were resuspended in Click-It Wash Buffer 2. Magnetic beads were immobilized, and the supernatant was collected and used for cDNA synthesis using the Superscript Vilo cDNA Synthesis Kit. The real-time quantitative RT-PCR was carried out in a ViiA 7 Real-T Time PCR System (Thermo Fisher Scientific) with a 6-carboxyfluorescein-labeled succinimidyl 4-formylbenzoate (SFB) target assay and a 4,7,2′-trichloro-7′-phenyl-6-carboxyfluorescein (VIC)-labeled endogenous control PCR Master Mix Kit as previously described (21). The transcripts of human SFTPB (Hs01090667_m1) and the controls, either eukaryotic 18S rRNA (4319413E) or human β-actin (4326315E), were amplified. The relative quantification (RQ) of fold changes, along with the minimum (RQ Min) and maximum (RQ Max) range of expression, were calculated by the ΔΔC_t method using ViiA 7 Software, v.1.2.3.

ChIP assay

The ChIP assays were performed by using the Simple ChIP Enzymatic Chromatin IP Kit (9003; Cell Signaling Technology, Danvers, MA, USA), according to the manufacturer’s instruction. Antibodies that recognize acetyl-histone H3 (Lys9; 9649; Cell Signaling Technology) were used to detect specific changes in histone acetylation patterns. Immunoprecipitated DNA was reverse crosslinked, purified, and analyzed by semiquantitative PCR for 35 cycles using PCR primers (hSpB proF-5′-CCAGGAACATGGGAGTCTGG-3′ and hSpB proR-5′-TAGGAGTGGCAGCGACCTC-3′), synthesized by Integrated DNA Technologies (Coralville, IA, USA).

Statistical analyses

Statistical analyses were performed using SigmaStat (v.13.0; Systat Software, San Jose, CA, USA). Results are expressed as means ± sd of >3 samples for each group from at least 2 independent experiments. Two group comparisons were done by an unpaired Student’s t test. Three or more group comparisons were carried out using ANOVA, followed by a Holm-Sidak test. A value of P < 0.05 was considered statistically significant.

RESULTS

Epithelial cell-specific knockdown of Nampt attenuated the lung inflammation and the decrease of SFTPB in LPS-induced ALI mice

As inflammation is a key factor in expression of SFTPB, and our previous studies found that knockdown of Nampt gene expression could inhibit expression of inflammatory cytokines from pulmonary epithelial cells, we reasoned that epithelial cell-specific knockdown of Nampt may function to attenuate lung inflammation and affect the SFTPB expression in vivo. To test this hypothesis, we subjected both Nampt^PE+/− mice, in which 1 allele of the Nampt gene is specifically knocked out (Fig. 1A) in pulmonary epithelial cells, as described in Materials and Methods, and Nampt^PE+/+ ALI mice induced by LPS. Phenotypic parameters in lung tissues and BAL were analyzed. RT-PCR analyses of IL-6 and TNF-α mRNA levels in lung tissues indicated that they were significantly lower in LPS-treated Nampt^PE+/− than in LPS-treated Nampt^PE+/+ mice with IL-6 (25.4 ± 6.24 vs. 74.6 ± 5.49, n ≥ 5/group, P < 0.01) and TNF-α (32.3 ± 7.19 vs. 67.7 ± 6.89, n ≥ 5/group, P < 0.01), respectively (Fig. 1B). Histologic evaluation by hematoxylin and eosin (H&E) staining found that neutrophilic infiltration, interstitial edema, and alveolar wall damage were reduced in sections of lungs from Nampt^PE+/− mice compared with those from Nampt^PE+/+ mice (Fig. 1C). In addition, total lung injury scores of Nampt^PE+/− mice were significantly lower than those of Nampt^PE+/+ mice in the LPS-induced lung injury group (4.21 ± 1.12 vs. 7.12 ± 0.98, n ≥ 5/group, P < 0.01; Fig. 1D). Consistent with these findings in lung tissues, total BAL cell counts (1.07 × 10⁶ ± 0.45 vs. 2.24 × 10⁶ ± 0.45, n ≥ 5/group, P < 0.01), total BAL neutrophil count (0.97 × 10⁶ ± 0.28 vs. 2.13 × 10⁶ ± 0.34, n ≥ 5/group, P < 0.01), BAL protein concentration (552.4 ± 193.7 vs. 1009.3 ± 356.8, n ≥ 5/group, P < 0.01), and TNF-α levels (115.1 ± 40.0 vs. 190.4 ± 55.7, n ≥ 5/group, P < 0.01) in the BAL were significantly lower from Nampt^PE+/− mice than those from Nampt^PE+/+ mice (Fig. 1E–H). Western blot analysis of Nampt expression demonstrated the lower protein levels (0.55 ± 0.19 vs. 1.00 ± 0.18, P < 0.01), whereas higher protein levels of SFTPB (1.43 ± 0.22 vs. 1.00 ± 0.21, n ≥ 5/group, P < 0.01) were found in BAL from Nampt^PE+/− mice than those from Nampt^PE+/+ mice (Fig. 1I, J). These results suggested that lung epithelial cell-specific Nampt knockdown attenuates lung inflammation and a rescued inflammation-induced decrease of SFTPB expression in mouse lung tissues.

Figure 1.

Attenuation of LPS-induced ALI in Nampt^PE+/− mice. A) Genotypes of mouse lung and several other organs show lung-specific cutting of the Nampt gene in Nampt^PE+/− mice. PCR fragments were separated on 2% agarose electrophoresis and visualized by ethidium bromide. The 200 bp desired band is indicated by the white arrow. B) PCR quantifications of IL-6 and TNF-α expression in lung tissue from Nampt^PE+/+ mice and Nampt^PE+/− mice that were treated with either PBS or LPS. C) Representative H&E images demonstrate that LPS-induced Nampt^PE+/+ mice had reduced amounts of inflammation, edema, and alveolar damage compared with LPS-induced Nampt^PE+/+ mice. Original scale bars, 100 μm; original magnification, ×200. D) Lung injury scores of Nampt^PE+/− mice were significantly lower than those of Nampt^PE+/+ mice in LPS-induced ALI models. Lung injury score was defined as percent alveoli damage × cell density; n ≥ 5/group. E, F) LPS-treated Nampt^PE+/− mice had a decreased total BAL cells count (E), total BAL polymorphonuclear neutrophil count (F), G, H) BAL protein concentration (G), and BAL TNF-α level (H) than LPS-treated Nampt^PE+/+ mice. I) Representative Western blot images of BAL Nampt and SFTPB levels. Equal volume of BAL supernatant from either PBS- or LPS-treated Nampt^PE+/+ mice and Nampt^PE+/− mice was separated by SDS-PAGE and immunodetected by either anti-Nampt or anti-SFTPB antibody. J) Protein quantification. Protein bands of both NAMPT and SFTPB from BAL were quantified as described in the Materials and Methods. NAMPT or SFTPB levels from wild-type control mice were arbitrarily set at 1. Values are means ± sd of 6 independent experiments.*P < 0.05 vs. PBS control mice, ^#P < 0.05 vs. wild-type treated with LPS group.

Administration of SPC-Nampt antibody gene Ad reduced LPS-induced mouse lung inflammation

We constructed an Ad expressing a Nampt scFV antibody (Ad-SPC-anti-Nampt-scFv), driven by the lung epithelium-specific human SPC promoter using the Adeno-XTM Adenoviral System 3. We first tested whether Ad-SPC-anti-Nampt-scFv (Ad-Nampt-scFv) can regulate SFTPB expression in A549 and H441 cells. Fluorescence images of H441 cells and A549 cells transduced with the Ad-SPC-Nampt-scFv virus indicated the high efficiency of Ad transduction in the cells (Fig. 2A). Western blot analysis showed that NAMPT expression was not significantly changed, whereas SFTPB was increased in Ad-SPC-Nampt-scFv-transduced H441 cells compared with the controls (1.64 ± 0.22 vs. 1.00 ± 0.23, P < 0.01; Fig. 2B, C). We then subjected wild-type C57BL/6J mice to either Ad-control or Ad-SPC-Nampt-scFv by intratracheal administration for 3 d before being challenged with PBS or LPS for 24 h. The mice were then anesthetized for BAL collection and then lung tissue isolation for various assays. Histologic evaluation by H&E staining found that neutrophilic infiltration, interstitial edema, and alveolar wall damage were reduced in sections of lungs from Ad-SPC-anti-Nampt-scFv-treated mice compared with those from the control mice (Fig. 2D), and total lung injury scores of Ad-SPC-anti-Nampt-scFv-treated mice were significantly lower than those of the control mice in the LPS-induced lung injury group (5.62 ± 1.23 vs. 8.56 ± 1.12, n ≥ 5/group, P < 0.01; Fig. 2E). Total BAL cell counts (3.82 × 10⁶ ± 0.84 vs. 6.84 × 10⁶ ± 2.64, n ≥ 5/group, P < 0.01; Fig. 2F), total BAL neutrophil counts (3.46 × 10⁶ ± 0.72 vs. 6.08 × 10⁶ ± 2.21, n ≥ 5/group, P < 0.01; Fig. 2G), and BAL protein concentration (741.5 ± 147.6 vs. 1158.1 ± 341.4, n ≥ 5/group, P < 0.01; Fig. 2H) from Ad-SPC-Nampt antibody-treated mice were significantly lower than those from control virus-injected mice. Western blot analysis of SFTPB expression demonstrated the lower protein level of SFTPB (0.43 ± 0.14 vs. 1.00 ± 0.18, n ≥ 5/group, P < 0.01) in BAL with equal protein from the Ad-control insert virus plus LPS-treated mice than those from the Ad-control insert virus plus PBS-treated mice, whereas the SFTPB protein level was higher (0.73 ± 0.16 vs. 0.43 ± 0.14, n ≥ 5/group, P < 0.01) in BAL from the Ad-Nampt-scFv antibody virus plus LPS-treated mice than those Ad-control insert virus plus LPS-treated mice, indicating that the Ad-Nampt-scFv antibody virus can rescue the LPS-induced inhibition of SFTPB expression in the mice (Fig. 2I, J). These results indicate that epithelial cell-specific expression of the anti-Nampt gene has a therapeutic potential for attenuation of lung inflammation.

Figure 2.

Administration of SPC-Nampt antibody gene Ad reduced LPS-induced murine ALI. A) Representative fluorescence images of H441 and A549 cells that were infected with the Ad-SPC-Nampt antibody gene virus. Original scale bars, 200 µm; original magnification, ×100. B) Representative Western blot images of NAMPT, SFTPB, and GAPDH expression in H441 cells that were infected with the Ad-SPC-Nampt antibody gene virus. Equal amounts of cell lysate proteins from each sample were separated by 15% SDS-PAGE and immunodetected by anti-Nampt, anti-SFTPB antibody, and anti-GAPDH antibody, respectively. C) Protein quantification. Protein bands were quantified as described in the Material and Methods. Both NAMPT and SFTPB levels were normalized to GAPDH levels. NAMPT and SFTPB levels in control cells were arbitrarily set at 1. Values are means ± sd of 3 independent experiments. *P < 0.01 vs. controls. D, E) H&E staining (D) and lung injury scores (E). Wild-type mice were administered intratracheally with either the Ad-control (Ad-Cntl) insert or Ad-anti-NAMPT-scFv. Three days later, the mice were challenged intratracheally with either PBS or LPS for 24 h before their BAL and lung tissue samples were collected for various analyses. Lung H&E staining and lung injury scores in these mice were performed as described in the Material and Methods. Original scale bars, 100 µm; original magnification, ×200; n ≥ 5/group. *P < 0.05 vs. PBS groups, ^#P < 0.05 vs. Ad control group. F–H) BAL total cell counts (F), BAL neutrophil counts (G), and BAL protein concentration (H) were counted and quantified as previously described. I) Representative Western blot images of BAL SFTPB levels. Equal volume of BAL supernatant from PBS- or LPS-treated mice after either the Ad-control insert or Ad-anti-NAMPT-scFv administration was separated by SDS-PAGE and immunodetected by anti-SFTPB antibody. J) Protein quantification. Protein bands of BAL SFTPB were quantified as described in the Material and Methods. SFTPB levels from the Ad-control group were arbitrarily set at 1. Values are means ± sd of 6 independent experiments. *P < 0.05 vs. Ad-control insert plus PBS group, ^#P < 0.05 vs. Ad-control insert plus LPS group.

TNF-α-induced increase of NAMPT and inhibition of SFTPB protein expression in H441 cells

Knockdown of NAMPT gene expression in H441 cells significantly up-regulated SFTPB expression at the basal level and abated TNF-α-mediated attenuation of SFTPB expression. We have previously demonstrated that inflammatory cytokines can stimulate NAMPT expression in A549 cells (an alveolar type II cell line), but whether they have a similar effect on NAMPT expression in H441 cells (a lung club cell line) was unknown. We treated H441 cells with different doses of LPS (10 ng/ml to 1 µg/ml) or rhTNF-α (25–100 ng/ml) and quantified the NAMPT protein level in each sample. As shown in Fig. 3, TNF-α treatment (Fig. 3C, D) for 24 h significantly increased the NAMPT protein level in a dose-dependent manner in H441 cells, whereas no significant changes in expression were observed with LPS treatment (Fig. 3A, B). As LPS challenged Nampt^PE+/− mice, or mice treated with SPC-anti-Nampt-scFv cDNA displayed significantly higher BAL SFTPB levels than their controls (Figs. 1I and 2B), we embarked on the exploration of whether NAMPT has a direct impact on SFTPB expression in H441 cells using a “loss of function” approach. The knocking down of NAMPT with NAMPT siRNA for 48 h in H441 cells resulted in a significantly decreased NAMPT expression (0.63 ± 0.11 vs. 1.00 ± 0.12, n = 3, P < 0.05) and a significantly increased SFTPB expression (1.60 ± 0.13 vs. 1.00 ± 0.11, n = 3, P < 0.01) in the siRNA treatment group compared with the control group at the protein level (Fig. 3E–G). H441 cells transfected with Nampt siRNA or the corresponding scrambled control were then treated with or without LPS or TNF-α. The results showed that TNF-α significantly decreased SFTPB expression (0.26 ± 0.15 vs. 1.00 ± 0.11, n = 3, P < 0.01) in H441 cells. Down-regulation of NAMPT rescued inhibition of SFTPB expression in H441 cells with TNF-α (0.88 ± 0.12 vs. 0.26 ± 0.15, n = 3, P = 0.01) treatment (Fig. 3E–G). These findings indicate that knockdown of NAMPT could increase SFTPB expression and abate TNF-α-induced inhibition of SFTPB in the epithelial cells.

Figure 3.

A, C) Increased NAMPT protein expression in TNF-α-induced H441 cells and NAMPT siRNA attenuation of TNF-α-mediated inhibition of SFTPB expression. H441 cells were treated without or with different doses of LPS (A) or TNF-α (C) for 24 h. An equal amount of total cell lysate protein (10 µg) from each sample was separated by 15% SDS-PAGE and immunodetected by Western blot analysis using anti-human NAMPT or GAPDH antibodies. B, D) Protein bands were quantified and NAMPT levels were normalized to GAPDH levels. NAMPT levels in control cells were arbitrarily set at 1. Values are means ± sd of 3 independent experiments. *P < 0.05 vs. control cells. H441 cells were transfected with either scrambled RNA (Sc RNA) or NAMPT siRNA for 48 h before treatment, without or with either LPS (1 µg/ml) or TNF-α (25 ng/ml), for 24 h. E) Cell lysates of NAMPT (10 µg), SFTPB (30 µg), and GAPDH (10 µg) protein were separated by 15% SDS-PAGE and immunodetected by Western blot analysis. Representative Western blot images of NAMPT, SFTPB, and GAPDH expression in H441 cells. F, G) Protein bands were quantified and NAMPT (F) and SFTPB (G) levels were normalized to GAPDH levels. NAMPT and SFTPB levels in Sc RNA cells without LPS treatment were arbitrarily set at 1. Values are means ± sd of 3 independent experiments. *P < 0.05 vs. Sc RNA control cells, ^#P < 0.05 vs. Sc RNA cells within the same treatment group.

Inhibition of NAMPT expression induced SFTPB mRNA via transcriptional regulation in H441 cells

To elucidate further molecular mechanisms underlying the regulation of NAMPT on SFTPB expression, we determined whether NAMPT inhibition augments the steady-state SFTPB mRNA level using RT-PCR. NAMPT siRNA treatment significantly increased the steady-state SFTPB mRNA level (RQ 1.6; RQ Min 1.5; RQ Max 1.7) compared with the nontransfected (1.0; 0.9; 1.1) and scrambled siRNA (1.0, 0.8, 1.2) controls (Fig. 4A). An RNA capture approach was also performed and demonstrated that NAMPT siRNA treatment significantly increased SFTPB gene transcription (RQ 4.3; RQ Min 2.9; RQ Max 6.4) compared with the nontransfected (1.0; 0.8; 1.2) and scrambled siRNA (1.5; 0.8; 2.6) controls (Fig. 4B). Additionally, we found that NAMPT siRNA treatment enhanced histone acetylation at Lys9, as assessed by ChIP assays (Fig. 4C, D), which suggests that NAMPT knockdown augments SFTPB gene transcription via enhancement of histone acetylation.

Figure 4.

Inhibition of NAMPT expression induces SFTPB mRNA via transcriptional regulation in H441 cells. A) SFTPB mRNA levels were increased in H441 cells following siRNA knockdown of NAMPT. RNA was isolated from H441 cells transfected with control scrambled RNA and NAMPT siRNA for 48 h and subjected to quantitative RT-PCR. The relative quantification ± sd results represent 2 independent experiments, each analyzed in 5 technical replicates (n = 10). *P < 0.05. B) The effect of NAMPT expression on the SFTPB gene transcription. RNA capture was performed on H441 cells transfected with either control scrambled RNA or NAMPT siRNA for 48 h. The relative quantification ± sd of SFTPB mRNA levels was determined by quantitative RT-PCR analysis (n = 6 technical replicates). *P < 0.05. C). The effect of NAMPT expression on histone H3 hyperacetylation of the SFTPB promoter. An aliquot of the cells, cultured as in panel B, was harvested at 48 h, and the acetylation status of histone H3 in nucleosomes was assessed by a ChIP assay with an anti-acetylated histone H3 (K9) antibody and a primer pair specific for the SFTPB promoter. Input DNA was amplified as a control. D) Quantification of ChIP assays (n = 4). *P < 0.005.

NAMPT inhibited SFTPB expression via both its nonenzymatic activity and enzymatic activity

NAMPT is a pleiotropic protein that is not only involved in the mammalian salvage pathway of NAD synthesis via its enzymatic activity (19) but also in the regulation of inflammatory cytokine expression in pulmonary epithelial cells via its nonenzymatic and activator protein 1 (AP-1)-dependent mechanism (14). To examine whether NAMPT regulates SFTPB expression via its enzymatic activity, we transfected H441 cells with either wild-type pCAGGS-hNAMPT or the mutant pCAGGS-hNAMPT-H247E construct, which has very low NAMPT activity (14). The results (Fig. 5A–C) showed that overexpression of the mutant H247E NAMPT similarly and significantly inhibited the SFTPB protein levels (0.77 ± 0.12 vs. 1.00 ± 0.13, n = 3, P < 0.01) as wild-type NAMPT (0.46 ± 0.18 vs. 1.00 ± 0.13, n = 3, P < 0.01). The pCAGGS vector control had no effect. The pretreatment of H441 cells with FK866, which is the intracellular NAMPT enzymatic inhibitor, and then stimulation of the cells, with or without TNF-α (Fig. 5D–F), showed that SFTPB expression was increased after treatment with FK866; however, treatment of FK866 did not significantly rescue TNF-α-induced inhibition of SFTPB. As knockdown of NAMPT with siRNA could rescue TNF-α-induced inhibition of SFTPB, whereas inhibition of NAMPT enzymatic activity with FK866 did not have that effect, and overexpression of mutant NAMPT had a similar phenomenon of inhibiting SFTPB expression with wild-type NAMPT, we speculated that NAMPT inhibited SFTPB expression mainly via both nonenzymatic and enzymatic activity at baseline; however, the nonenzymatic activity of NAMPT plays a more important role in regulating TNF-α-induced SFTPB inhibition.

Figure 5.

FK866 treatment increased SFTPB expression in H441 cells. A) Representative Western blot images of NAMPT, SFTPB, and GAPDH expression in H441 cells. H441 cells were transfected with pCAGGS, pCAGGS-hNAMPT, or pCAGGS-hNAMPT-H247E. Cell lysates of NAMPT (10 µg), SFTPB (30 µg), and GAPDH (10 µg) protein were separated by 15% SDS-PAGE and immunodetected by Western blot analysis. B, C) Protein bands were quantified, and NAMPT (B) and SFTPB (C) levels were normalized to GAPDH levels. Values are means ± sd of 3 independent experiments. *P < 0.05 vs. pCAGGS cells, ^#P < 0.05 vs. pCAGGS-hNAMPT cells. D) Representative Western blot images of NAMPT, SFTPB, and GAPDH expression in H441 cells. H441 cells were pretreated with different doses of FK866 for 6 h before being treated, without or with TNF-α (25 ng/ml), for 24 h. Cell lysates of NAMPT (10 µg), SFTPB (30 µg), and GAPDH (10 µg) protein were separated by 15% SDS-PAGE and immunodetected by Western blot analysis. E, F) Protein bands were quantified, and NAMPT (E) and SFTPB (F) levels were normalized to GAPDH levels. Values are means ± sd of 3 independent experiments. *P < 0.05 vs. cells without FK866 or rhTNF-α treatment, ^#P < 0.05 vs. cells pretreated with FK866 but without rhTNF-α treatment.

JNK pathway associated with NAMPT-mediated inhibition of SFTPB expression in H441 cells

We previously found that NAMPT increased AP-1 binding to the IL-8 promoter to activate transcription in epithelial cells via the p38 MAPK and the JNK pathways (14). We wondered if NAMPT-mediated inhibition of SFTPB operated through similar pathways. We transfected H441 cells with either wild-type pCAGGS-hNAMPT or mutant-type pCAGGS-hNAMPT-H247E and subjected them to treatment with the p38 pathway inhibitor SB203580 or the JNK pathway inhibitor SP600125 for 6 h. The results (Fig. 6A, B) showed that the JNK inhibitor significantly attenuated the NAMPT-induced inhibition of SFTPB in H441 cells transfected with either pCAGGS-hNAMPT (2.16 ± 0.25 vs. 1.00 ± 0.23, n = 3, P < 0.01) or pCAGGS-hNAMPT-H247E (1.76 ± 0.25 vs. 1.00 ± 0.18, n = 3, P < 0.01). Inhibition of p38 did not alter NAMPT-induced regulation of SFTPB. The induction of SFTPB by JNK inhibition appeared to be dependent on NAMPT expression, as treatment with SP600125 did not increase SFTPB precursor levels in H441 cells transfected with the pCAGGS empty vector (data not shown). These data suggested that wild-type NAMPT or mutant NAMPT mediated inhibition of SFTPB expression, in part, via the JNK pathway.

Figure 6.

JNK pathway is involved in the NAMPT-inhibited SFTPB in H441 cells. A) Representative Western blot images of NAMPT, SFTPB, and GAPDH expression in H441 cells. H441 cells were transfected with pCAGGS-hNAMPT or pCAGGS-hNAMPT-H247E for 48 h before being treated, without or with either p38 inhibitor or JNK inhibitor, for 6 h. Cell lysates of SFTPB (30 µg) and GAPDH (30 µg) protein were separated by SDS-PAGE and immunodetected by Western blot analysis. Protein bands were quantified and SFTPB levels (B) were normalized to GAPDH. Values are means ± sd of 3 independent experiments. *P < 0.05 vs. control cells without inhibitor treatment.

DISCUSSION

In this study, we have demonstrated that pulmonary epithelial cell-specific knockdown of Nampt gene expression by a genetic manipulation or anti-Nampt antibody cDNA expression significantly attenuated LPS-induced mouse lung inflammation, which was evidenced by lower BAL neutrophil infiltrates, total protein, TNF-α and IL-6 levels, as well as lower lung injury scores compared with control mice. One unique feature was that BAL SFTPB levels were significantly higher in mice with decreased Nampt expression compared with control mice. We further found that down-regulation of NAMPT increased the expression of SFTPB, as well as rescued the TNF-α-induced inhibition of SFTPB, whereas overexpression of NAMPT inhibited SFTPB expression in H441 cells. These findings suggested that NAMPT suppressed SFTPB expression, not only by reducing the TNF-α and oxidative stress agents, which have been found to inhibit SFTPB expression, but also by acting on pulmonary epithelial cells and inducing signaling pathways directly. We next found that inhibition of Nampt up-regulates SFTPB expression by enhancing histone acetylation to increase its transcription. Additional data indicate that this effect was mainly mediated by NAMPT nonenzymatic function via JNK pathway.

Two critical pathophysiological processes in ARDS are alveolar barrier dysfunction and overwhelming inflammation (22). Knockdown of Nampt gene expression specifically in pulmonary epithelial cells significantly curbed LPS-induced up-regulation of inflammatory cytokines, TNF-α and IL-6, in BAL of Nampt^PE+/− mice relative to Nampt^PE+/+ mice (Fig. 1). This may, in part, explain the therapeutic efficacy of Nampt by attenuating inflammatory reactions in the pathogenesis of ALI. One unique and novel observation was that in LPS-challenged Nampt^PE+/− mice or mice treated with SPC-anti-Nampt-scFv cDNA, BAL SFTPB levels were significantly higher than their controls (Figs. 1I and 2B). SFTPB is the only surfactant-associated protein absolutely required for postnatal lung function and survival. Complete deficiency of SFTPB in mice and humans results in lethal, neonatal respiratory distress syndrome and is characterized by a virtual absence of lung compliance, highly disorganized lamellar bodies, and greatly diminished levels of SFTPC mature peptide (17). SFTPB-deficient mice are susceptible to hyperoxic lung injury. They are associated with increased severity of pulmonary edema, hemorrhage, and inflammation, lung permeability, and protein leakage into the alveolar space (23). Gregory et al. (24) and Schmidt et al. (25) reported that SFTPB was significantly decreased in BAL of ARDS patients relative to normal controls. A variant polymorphism of the SFTPB gene has been associated with ARDS (26, 27). Ge et al. (28) demonstrated that in SFTPB humanized mice, the human SFTPB C allele of the 1580C/T polymorphism (single nucleotide polymorphism rs1130866) was more susceptible to bacterial pneumonia than the SFTPB T allele in vivo. Pneumonia is a frequent cause of ARDS. On the other hand, administration of phospholipids with the active SFTPB peptide was sufficient to restore pulmonary function and prevent alveolar capillary leak after oxygen exposure, demonstrating the protective role of SFTPB during oxygen-induced lung injury (29). Recruitment of inflammatory cells and elaboration of proinflammatory cytokines in BAL fluid were reduced in SFTPB-overexpressing mice compared with SFTPB^+/− mice, suggesting that SFTPB inhibited endotoxin-induced lung inflammation (30). Thus, our findings that pulmonary epithelial cell-specific knockdown of Nampt gene expression by either a genetic manipulation or anti-Nampt antibody cDNA expression significantly up-regulated SFTPB expression in LPS-induced ALI in mice relative to their wild-type control mice may, at least in part, underlie their protective molecular mechanism. To shed further insight, we have provided 6 lines of corroborative and luminary evidence that Nampt has a direct impact on SFTPB gene expression in cultured A549 cells and H441 cells. First, knockdown of NAMPT gene expression using a well-established NAMPT siRNA (31) in H441 cells significantly up-regulated SFTPB expression at the basal level and abated TNF-α-mediated attenuation of SFTPB expression (Fig. 3E). Second, inhibition of NAMPT by a known small chemical inhibitor, FK866 (32), up-regulated SFTPB expression in H441 cells (Fig. 5D, E). Third, an overexpression of NAMPT in pCAGGS-hNAMPT cDNA-transfected H441 cells significantly down-regulated SFTPB expression (Fig. 6A, B). Fourth, an overexpression of a NAMPT mutant, NAMPT-H247E, whose enzymatic activity is abolished (14), also significantly down-regulated SFTPB, which suggested that the enzymatic activity of NAMPT may not be wholly necessary for its mediation of SFTPB expression (Fig. 5A–C). Fifth, pharmacological inhibition experiments indicated that the JNK pathway was involved in the regulation of NAMPT on SFTPB expression (Fig. 6). Sixth, epigenetic studies and the RNA capture experiment found that NAMPT inhibition was correlated with the increased histone acetylation and the enhanced transcription of SFTPB expression (Fig. 3). These converging lines of evidence support that therapeutic use of pulmonary epithelial-specific knockdown of NAMPT gene expression to ALI may be, at least in part, through its regulation of SFTPB expression, whose up-regulation is operative to attenuate LPS inflammation and ALI.

How down-regulation of Nampt gene expression up-regulates SFTPB expression is not fully elucidated in this study. However, our observation that NAMPT inhibition was correlated with the increased histone acetylation and the enhanced transcription of SFTPB expression (Fig. 4) has provided an important clue for the underlying signal transduction pathway and a solid foundation for further exploration. Nampt is a key enzyme in cellular NAD synthesis (33). The sirtuin (Sirt) protein family (Sirt1–7) requires NAD for its deacetylase activity and regulates a variety of biologic processes (34): DNA damage response, cellular senescence, and inflammation. It is conceivable that decreased cellular NAD synthesis, resulting from Nampt inhibition, led to the decreased deacetylase activities in Sirt proteins and thereby, the increased histone acetylation, resulting in the increased transcription of SFTPB expression. Acetylation of internal lysine residues of core histone N-terminal domains has long been recognized to be correlatively associated with transcriptional activation in eukaryotes (35). Although we have not provided a direct evidence on our assumption, Ota et al. (36) reported that a histone deacetylase inhibitor, Trichostatin A, restored SFTPC expression by increasing histone H4 acetylation in the SFTPC promoter region in vitro. A similar mechanism may be operative for Nampt inhibition-mediated up-regulation of SFTPB expression. Additionally, our other study revealed that expression of Sirt3–5 in liver mitochondria was decreased significantly in Nampt^−/− mice vs. Nampt^+/+ mice (37). This observation indirectly adds more credence to the above Nampt-Sirt-SFTPB expression axis hypothesis, although it would be of interest to prove directly and experimentally that Nampt inhibition inhibits the deactylase activities and gene expression of the Sirt family. This plausible mechanistic rendition also needs a reconciliation with the finding that an overexpression of a NAMPT mutant, NAMPT-H247E, whose enzymatic activity is abolished, also significantly down-regulated SFTPB. Previously, we reported regulation of inflammatory cytokine expression in pulmonary epithelial cells by Nampt via a nonenzymatic and AP-1-dependent mechanism, in part, via the JNK pathway (14). This study also found that the JNK pathway was involved in regulation of NAMPT on SFTPB expression. It warrants further study to characterize the full repertoire of signal transducers in this process.

ARDS represents a complex syndrome with considerable morbidity and mortality, for which there exists no targeted treatment strategies, although recent advances in clinical care have improved outcomes (38). Thus, it still calls for necessary pursuits for more novel therapeutic approaches, as well as for investigation into new mechanisms that might point us in the direction of more effective treatment and preventive strategies for this devastating syndrome. Our study fits well with this strategy to develop potential new and effective targeted therapy for ARDS. Targeted inhibition of Nampt expression, specifically in pulmonary epithelial cells, instead of in all lung tissue or the whole body, may improve its therapeutic efficacy for ARDS and reduce side-effects often associated with systematic drug delivery (39). With further mechanistic exploration and optimization, epithelial cell-specific knockdown of Nampt may be translated into a potential novel and viable therapeutic modality for human ALI and other lung diseases, such as bronchopulmonary dysplasia, in the not-too-distant future.

Glossary

Ab1/2

antibody 1/2

Ad

adenovirus

ALI

acute lung injury

AP-1

activator protein 1

ARDS

acute respiratory distress syndrome

BAL

bronchoalveolar lavage

bp

base pair

ChIP

chromatin immunoprecipitation

FBS

fetal bovine serum

GAPDH

glyceraldehyde 3-phosphate dehydrogenase

H&E

hematoxylin and eosin

Lys9

lysine 9

Nampt

nicotinamide phosphoribosyltransferase

Nampt^PE+/−

pulmonary epithelial Nampt knockdown

Nampt^PE+/+

pulmonary epithelial Nampt wild type

rh

recombinant human

RQ

relative quantification

scFv

single chain variable fragment

SPC3.7

surfactant protein C promoter

SFTPB/C

surfactant protein B/C

siRNA

small interfering RNA

Sirt

sirtuin

AUTHOR CONTRIBUTIONS

D. P. Heruth, D.-Y. Li, V. Sampath, R. B. Easley, and S. Q. Ye conceived of and designed the study; G. Bi, L. Wu, P. Huang, S. Islam, W. Huang, and B. A. Simon analyzed and interpreted the research; and G. Bi, P. Huang, S. Islam, D. P. Heruth, L. Q. Zhang, and S. Q. Ye drafted the manuscript.