Dissociated Presenilin-1 and TACE Processing of ErbB4 in Lung Alveolar Type II Cell Differentiation

Abstract

Neuregulin (NRG) stimulation of ErbB4 signaling is important for type II cell surfactant synthesis. ErbB4 may mediate gene expression via a non-canonical pathway involving enzymatic cleavage releasing its intracellular domain (4ICD) for nuclear trafficking and gene regulation. The accepted model for release of 4ICD is consecutive cleavage by Tumor necrosis factor alpha Converting Enzyme (TACE) and γ-secretase enzymes. Here, we show that 4ICD mediates surfactant synthesis and its release by γ-secretase is not dependent on previous TACE cleavage. We used siRNA to silence Presenilin-1 (PSEN-1) expression in a mouse lung type II epithelial cell line (MLE12 cells), and both siRNA knockdown and chemical inhibition of TACE. Knockdown of PSEN-1 significantly decreased baseline and NRG-stimulated surfactant phospholipid synthesis, expression of the surfactant proteins SP-B and SP-C, as well as 4ICD levels, with no change in ErbB4 ectodomain shedding. Neither siRNA knockdown nor chemical inhibition of TACE inhibited 4ICD release or surfactant synthesis. PSEN-1 cleavage of ErbB4 for non-canonical signaling through 4ICD release does not require prior cleavage by TACE.

1.0 Background

Respiratory distress syndrome (RDS), formerly known as hyaline membrane disease, is a common problem in preterm infants born before 28 weeks. This disease is caused primarily by deficiency of pulmonary surfactant in immature lungs and is more common the earlier the infant is born [1]. Despite the beneficial effects of prenatal glucocorticoids and postnatal surfactant replacement therapies, RDS remains one of the significant causes of morbidity and mortality in premature infants [2], [1]. Pulmonary surfactant is a mixture of surface active phospholipids and proteins (termed surfactant protein B (SP-B) and SP-C) which is produced in alveolar type II epithelial cells [3]. The development of surfactant synthesis is under multifactorial control, in which paracrine mesenchyme-type II cell communication mechanisms play a central role. We have shown that the growth factor Neuregulin (NRG-1), which is secreted by fibroblasts, and its target receptor ErbB4, which is expressed by type II cells, play a prominent role in stimulation of type II cell maturation and surfactant synthesis [4], [5]. NRG is expressed in the midtrimester human fetal lung [6] and increases in fetal lung at the onset of surfactant synthesis [7].

ErbB4 is a member of the ErbB receptor tyrosine kinase family, which also includes the epidermal growth factor receptor ErbB1, ErbB2, and ErbB3 [8]. The ErbB receptors are transmembrane tyrosine kinase proteins and act as important regulators of cell proliferation and differentiation during fetal organ development including lung development [9]. ErbB4 signal transduction is a complex process that involves both canonical and non-canonical signaling pathways. Binding of NRG to its extracellular ligand-binding site causes ErbB4 to form homo or heterodimers with other ErbB receptors linked by disulfide bonds in the extracellular domain [10]. These receptor dimers then undergo auto phosphorylation on tyrosine residues within the intracellular domain. In the canonical signal pathway tyrosine phosphorylation activates signal cascades through specific intracellular signaling pathways such as the phosphatidylinositol-3 (PI3) Kinase/Akt pathway to ultimately influence gene expression [11]. However, within the ErbB family, ErbB4 is unique in that it may undergo proteolytic processing to initiate non-canonical signaling [12]. The accepted model for the non-canonical pathway involves two sequential cleavage processes. The first step in this pathway is performed by a transmembrane metalloprotease Tumor necrosis factor alpha Converting Enzyme (TACE) which releases the ErbB4 ectodomain by a cleavage that produces two fragments: a 120 kDa ectodomain fragment which is released into the extracellular space and an 80 kDa membrane-associated fragment that contains the ErbB4 transmembrane domain and the entire cytoplasmic region, including the tyrosine kinase domain. Ectodomain cleavage of ErbB4 in cells occurs at a low constitutive or basal level [13] that can be increased by neuregulin or other ErbB4 ligands [14]. The ectodomain cleavage of ErbB4 is sensitive to metalloprotease inhibitors [13] and does not occur in cells genetically deficient in TACE. The ErbB4 m80 fragment (membrane associated fragment) that remains following ectodomain cleavage is further processed by γ-secretase that cleaves at the transmembrane domain to release a soluble intracellular s80 fragment (4ICD) into the cytosol from which it translocates to the nucleus in association with chaperone proteins [14] [15]. Gamma secretase is an enzyme complex which consists of 4 components: presenilin 1 (PSEN-1) or Presenilin 2 (PSEN-2), Nycastrin, anterior pharynx defective-1 (APH-1) and the Presenilin-Enhancer 2. PSEN-1 or PSEN-2 are the active enzymatic component; other components behave as scaffolding molecules and essential cofactors [16], [17] [18], [19]. Studies with transgenic mice show that PSEN-1 and PSEN-2 have functionally distinct phenotypes in several organs including the lung [20], [21].

Nuclear localization of ErbB4 is the preferred mechanism of ErbB4 signaling in several regulatory processes during development [22]. The accepted model for γ-secretase activity in ErbB4 processing is that ectodomain cleavage by TACE is a prerequisite step [12]. In this study we sought to more specifically define the sequential interactive relationship of PSEN-1 and TACE for ErbB4 processing controlling lung alveolar type II cell surfactant production. Our focus on PSEN-1 was motivated by the more severe developmental phenotype of alveolar maturation in the PSEN-1 knockout mouse compared to PSEN-2 knock out and our previous work showing that importance of PSEN-1 signaling for fetal type II cell maturation [22], [23]. We studied the effect of PSEN-1 knockdown and TACE knockdown in MLE12 cells and evaluated the effects on ErbB4 cleavage in association with the expression of the SP- B and SP-C mRNA and protein and synthesis of the major surfactant phospholipid disaturated phosphatidylcholine (DSPC).

2.0 METHODS AND MATERIALS

2.1 Materials

Dulbecco’s Modified Eagle’s (DME) low glucose medium was purchased from Sigma-Aldrich (St. Louis, MO). Fetal bovine serum (FBS) was from BD Biosciences (Lot # ANB 18202A), L- glutamine, Pen/Strep and an siRNA cocktail of three siRNA sequences targeting PSEN-1 (Psen 1Mss208049, Psen 1Mss208050, Psen 1Mss208051) were purchased from Invitrogen (Carlsbad, CA). TACE inhibitor DAPI-1 was purchased from Peptides International (Louisville, KY). siRNA cocktail of three siRNA sequences targeting TACE, Silencer Negative Control scrambled siRNA and Silencer GAPDH siRNA were purchased from Ambion (St Louis, MO). Glass 100mm cell culture dishes were purchased from Pyrex/Corning (St Louis, MO). The murine lung epithelial cell line MLE12 was purchased from the American Type Culture Collection (Manassas, VA), Neuregulin 1β was produced using an expression vector kindly provided by Kermit Carraway III (UC Davis, CA) and purified by Dr. Ann Kane, Phoenix Laboratory (Tufts Medical Center, Boston, MA)., BCA Protein Assay Kit and RIPA buffer were obtained from Pierce/Thermo Scientific (Logan, UT), protease inhibitor cocktail from Sigma Aldrich. Invitrolon PVDF filter paper and NuPage 4–12% Bis-Tris pre-cast gels (1.0 mmX 12 wells) were obtained from Novex/Invitrogen. Tris-Glycine-SDS 10X running buffer, transfer buffer (10X Tris-buffered saline washing buffer) and TBS-Tween-20 (10X) were from Boston Bio Products (Ashland, MA). Methanol 100%, Kodak film (X-OMAT Blue XB) and Restore Plus Western Blot Stripping Buffer were from Thermo Fisher Scientific (Palm Beach, FL). Anti-Presenilin 1 N-terminal (1–65) rabbit polyclonal antibody (cat#529591) was from Calbiochem/EMD Millipore (Billerica, MA). Anti-prosurfactant protein C (rabbit polyclonal antibody, ab90716), anti-prosurfactant protein B (rabbit polyclonal antibody ab15011), monoclonal antibody to beta actin (HRP conjugated, cat# 20272), Anti-TACE antibody (rabbit polyclonal antibody, cat# ab2051) were all from Abcam (Cambridge, MA). Peroxidase-conjugated Affinipure Goat Anti-Rabbit IgG (H+L), (111-035-144) was from Jackson ImmunoResearch. Anti-ERBb 4 antibody (cat# sc-283) was obtained from Santa Cruz (Santa Cruz, CA). Chloroform (spectrophotometric grade) was obtained from Sigma Aldrich (St Louis, MO). Methanol was from Fisher Scientific; Silica gel H thin layer chromatography sheets were from Analtech (Newark, DE). Osmium tetroxide 05500-1G, carbon tetrachloride and dipalmitoylphosphatidyl choline (P-5911) were from Sigma Aldrich (St Louis, MO). Ultima-Gold scintillation fluid was from Perkin Elmer (Waltham, MA). For centrifugation we used a Beckman J-6M.

2.2 Cell Culture

MLE-12 cells were used as a model for type II alveolar epithelial cells. MLE12 cells exhibit characteristics of alveolar type II cells, including the expression of SP-B and SP-C and formation of microvilli and multivesicular bodies. They have a strong response to fetal fibroblast-conditioned media (FCM) and NRG with increased DPSC synthesis [7], [24]. MLE-12 cells were grown in DMEM containing 10% FBS, 2% pen/strep and 2% L-glutamine. Media were changed every second day.

2.3 Transfection with siRNA

MLE-12 cells were transfected with siRNA using the transfection reagent Dharmafect 2. To knock down presenilin 1, three pre-designed Presenilin-1 siRNA sequences that target three different regions of Presenilin mRNA were used. To knock down TACE we used a cocktail of 3 siRNAs targeting the TACE gene. The protocol for transfection of MLE-12 cells was adapted from the manufacturer’s guidelines; all steps were done using RNAse- free pipette tips and RNase spray for decontaminating the work area. MLE-12 cells were plated into 6 well plates. The transfection process was initiated when the cells were 30–40% confluent. 5μl siRNA in 95 μl serum free DMEM/well and 1μl transfection reagent Dharmafect 2 in 99 μ serum free DMEM/well) were incubated for 5 minutes at room temperature, then mixed and incubated for 20 minutes at room temperature. At the end of the incubation, 800 μl of antibiotic free media was added to the mixture. A total of 1mL of transfection media was added to each well of cells and allowed to incubate at 37°C. After 48 hours, the transfection medium was aspirated and replaced with fresh transfection media prepared as described above. A second transfection step was done in the same manner after 48 hours from the first transfection and cultures continued for another 24 hours resulting in a total exposure time of 72 hours. When transfecting the cells for the second time the media were changed to serum free media, and some cells were treated with NRG (3.3 nM) for 24 hours. Each experiment included the control conditions of scrambled siRNA and GAPDH siRNA. After 72 hours of total exposure time, the MLE-12 cells in the 6-well plates were harvested to extract protein.

2.4 Chemical blockade of TACE

MLE12 cells were plated in 6 well plates. When the cells were 50–60% confluent they were first treated for 24 hours with the TACE inhibitor TAPI. In preliminary experiments increasing concentrations (50nM, 100nM, 150nM, 200nM, 300nM) were used to determine the minimal effective dose. Thereafter cells were treated for 24 hours with 200nM TAPI alone, TAPI plus NRG, NRG only or media only. Cells were then harvested.

2.5 Protein extraction and quantification

Cells were harvested by first washing with PBS 3 times. 500 μl RIPA buffer: protease inhibitor mixture (1:50) was added to each well twice. Using a cell scraper, cells were scrapped off plates (on ice), aspirated, incubated on ice for 60 minutes with vortexing every 10 minutes, centrifuged at maximum speed for 15 minutes and the supernatant transferred to new tubes and stored at −20°C. Protein concentration in each sample was determined in duplicate (BCA protein assay, Thermo Scientific).

2.6 Western blot analysis

The protein samples were heated at 70°C for 10 minutes with sample reducing buffer to denature the protein. Proteins were loaded into pre-cast NuPage 4–12% Bis-Tris gels and separated by gel electrophoresis. Subsequently the proteins were transferred to a PVDF membrane. The membrane was blocked with 5% dry milk in TBST for 2 hours at room temperature or at 4°C overnight. The blots were probed with primary antibody against PSEN-1 or TACE, and then stripped and probed for SP-B, SP-C and beta Actin. Primary antibodies were incubated overnight at 4°C and secondary antibodies were incubated for 2 hours at room temperature. Blots were developed using Western Lightning Plus ECL and detected with a Kodak film X-OMAT Blue XB. Densitometry was done for all blots and beta actin was used as an internal control for protein loading.

2.7 Choline incorporation into DSPC

MLE-12 cells were cultured and transfected in 24–well plates as described above except that the cells were treated with 0.5μCi/ml [³H] choline for the final 24 hours of transfection. At the end of the 72 hours of transfection, the cells were harvested, the lysates sonicated and protein concentrations measured. DSPC was isolated by lipid extraction, osmium tetroxide treatment and then liquid chromatography as we have described [25], [7]. The isolated DSPC samples were placed in scintillation fluid and DPMs counted using a beta scintillation counter. The results were calculated as DPM per μg protein and presented as percentage of the experimental specific control value.

2.8. RT-PCR

Real-time PCR was used to determine the mRNA levels for PSEN-1, Sftpb and Sftpc. PSEN-1 primers: Forward sequence 5′ TCA/AGA/AAG/CGT/TGC/CAG/C 3′, Reverse sequence 5′ CGT/GGC/GAA/GTA/GAA/CAC/GA 3′, Sftpb primers: Forward sequence 5′ AGG/ATG/CCA/TGG/GCC/CT 3′, Reverse sequence 5′ TCA/GTG/TCC/TGT/AGT/GGC/CAT/T 3′, Sftpc primers: Forward sequence 5′ CCA/CTG/GCA/TCG/TTG/TGT/ATG 3′, Reverse sequence 5′GTA/GGT/TCC/TGG/AGC/TGG/CTT/A 3′, all were purchased from Applied Biosystems (Woburn, MA). The 50 μl reaction master mix contained 25 μl Taq polymerase, 1,25 μl Multiscribe and RNA inhibitor mix, 8 μM each of forward and reverse primer, 5μM probe and 1 μg of RNA sample. Amplification and detection of specific products were done with the ABI PRISM 7900 sequence detection system from Applied Biosystems. The amplification protocol consisted of an initial denaturation and enzyme activation at 95°C for 10 minutes, followed by 45 cycles at 95°C for 15 seconds and 60°C for 1 minute. In order to normalize the PSEN-1 and Sftp levels, actin was used as an internal control. Samples were run in triplicate. The differences in the Ct values of the PSEN-1 siRNA transfected cells compared to the cells transfected with a scrambled (scr) siRNA sequence were expressed as DDCT and presented as % of the controls (= scr siRNA). The message levels of Sftpb and Sftpc were calculated accordingly.

2.9 Conditioned Media

MLE12 cells (3 × 10⁵) were plated and treated for 72 hours with siRNA or TAPI as described above. Conditioned media was collected probed for the presence of the shed ectodomain of ErbB4 using western blot analysis.

2.10 Statistics

Statistical analysis was done using non-parametric ANOVA with post-hoc analysis or two-tailed t-tests (Graph Pad Software, San Diego, CA) as appropriate with a level of significance of P=0.05. All data are expressed as Means ± SEM.

3.0 RESULTS

3.1 PSEN-1 knockdown decreases SP-B and SP-C expression in MLE12 cells

To determine if PSEN-1 regulates SP-B and SP-C expression, we down regulated PSEN-1 using siRNA in MLE12 cells and measured the effect on SP-C and SP-B mRNA using qRT-PCR and protein levels using western blot analysis. PSEN-1 mRNA was reduced to 38% ± 17 (mean ± SE, n= 8) compared to the scrambled control; PSEN-1 protein was reduced to 50% ± 12 (mean ± SE, n=5, p= 0.02) of the scrambled control (Figure 1A). At this level of PSEN-1 knockdown, SP-B mRNA was reduced to 67% ± 11 (mean ± SE, n= 8) compared to the scrambled control. The level of SP-B protein was reduced to 60% ± 9.7 (mean ± SE, n=5, p= 0.01) of the scrambled control (Figure 1B). SP-C mRNA was reduced to 57% ± 6 (mean ± SE, n= 8) compared to the scrambled control. SP-C protein was decreased to 63% ± 4.2 (mean ± SE, n=5, p= 0.017) in response to PSEN-1 protein knock down (Figure 1C). The specificity of PSEN-1 knockdown was checked by assaying GAPDH protein, which showed no significant decrease at the highest level of PSEN-1 knockdown (data not shown here). These results suggest that PSEN-1 is required for optimal SP-B and SP-C expression.

Figure 1.

Representative western blots and densitometry quantification showing PSEN-1 knockdown in MLE12 cells and the effect on SP-B and SP-C levels. Densitometry results are expressed as % of the scrambled (SCR) siRNA condition (controls). Bars are means ± SEM of N=5 experiments. *=P<0.05. (A) PSEN-1 knockdown in PSEN-1 siRNA treated cells, GAPDH siRNA treated cells and SCR siRNA treated cells. Values are expressed as % of the SCR condition (controls). (B) SP-B expression in PSEN-1 siRNA, GAPDH siRNA and SCR siRNA treated MLE12 cells. (C) SP-C expression in PSEN-1 siRNA, GAPDH siRNA and SCR siRNA treated MLE12 cells.

3.2 PSEN-1 knockdown suppresses the stimulatory effect of neuregulin on SP-B and SP-C protein expression

Others and we have previously shown that NRG stimulates surfactant production through the ErbB4 receptor [4], [5]. To determine if PSEN-1 down-regulation affects the stimulatory effect of NRG, we knocked down PSEN-1 in MLE12 cells as described above and exposed these cells to 3.3 nM NRG during the final 24 hours. Western Blot analysis of scrambled control cells treated with NRG showed significantly increased expression of SP-B and SP-C to 150.7 ± 9; 148% ± 10 respectively; means ± SE, n=5, p= 0.01 and p=0.002, respectively (Figure 2A and 2B). PSEN-1 knockdown completely abrogated the stimulatory effect of NRG on SP-B and SP-C protein levels (Figure 2C and 2D). These data suggest that the ability of NRG to up-regulate SP-B and SP-C is dependent upon adequate PSEN-1 expression.

Figure 2.

The effect of NRG on SP-B and SP-C expression in MLE12 cells following PSEN-1 knock down. (A) SP-B expression in SCR siRNA treated MLE12 cells in the absence and presence of NRG. (B) SP-C expression in SCR siRNA treated MLE12 cells in the absence and presence of NRG. (C) SP-B expression in PSEN-1 siRNA treated MLE12 cells in the absence and presence of NRG. (D) SP-C expression in PSEN-1 siRNA treated MLE12 cells in the absence and presence of NRG. Bars are means ± SEM of N=5 experiments. *=P<0.05, **=P<0.005 compared to no NRG treatment.

3.3 Release of the ErbB4 cytoplasmic fragment (4ICD) is decreased in MLE12 cells treated with PSEN-1 siRNA

The ErbB4 cytoplasmic domain (4ICD) is thought to be important to control gene expression of surfactant proteins. To determine if PSEN-1 regulates the level of 4ICD, we knocked down PSEN-1 in MLE12 cells and measured 4ICD levels using Western blot analysis. Cells treated with PSEN-1 siRNA showed significantly decreased levels of the 4ICD product (60% ± 2.5; mean ± SE, n=5, p=0.03) compared to SCR siRNA treated cells. Control samples (SCR siRNA) treated with NRG showed increased 4ICD levels (130% ± 8 of non-NRG treatment; mean ± SE, n=5, p=0.02). Thus, NRG up-regulated PSEN-1 processing of ErbB4, increasing the level of the 4ICD fragment, but in PSEN-1 knock-down cells NRG did not increase the 4ICD fragment. (Figure 3).

Figure 3.

Representative western blots and densitometry quantification showing the effect of PSEN-1 knockdown on the level of free ErbB4 cytoplasmic domain (4ICD) at baseline and following NRG treatment. Mean ± SEM, N=5. *=P<0.05 compared to SCR cells treated with NRG.

3.4 PSEN-1 knockdown decreases DSPC synthesis in MLE12 cells

DSPC is the major phospholipid component and major surface active component of surfactant. We therefore assessed the effect of PSEN-1 knockdown on DSPC synthesis in MLE12 cells. Cells treated with PSEN-1 siRNA showed significantly decreased DSPC synthesis compared to scrambled siRNA treated samples (Figure 4). NRG increased the level of DSPC in SCR-treated cells, and this stimulatory effect was lost following PSEN-1 knockdown (Figure 4). These results indicate that PSEN-1 acts to regulate baseline and NRG-induced DSPC synthesis in MLE12 cells, similar to its effect on SP-B and SP-C.

Figure 4.

The effect of PSEN-1 knockdown on choline incorporation into DSPC. Bars show the mean ± SEM of [³H]-choline incorporated into DSPC, expressed as disintigrations per minute (DPM) per microgram of protein × 10³ in MLE12 cells treated with PSEN-1 siRNA compared to SCR treated cells, in the absence or presence of NRG. *=P<0.05 compared to SCR with NRG; **=P<0.005 compared to SCR without NRG.

3.5 Release of the ErbB4 ectodomain is not changed in MLE12 cells treated with PSEN-1 siRNA

To assess the effect of PSEN-1 on expression of the ErbB4 ectodomain we collected conditioned media from cells treated with PSEN-1 siRNA. Conditioned media from NRG-treated cells showed increased free ErbB4 ectodomain levels compared to untreated cells. Conditioned media from cells treated with PSEN-1 siRNA showed no change in the levels of the ErbB4 ectodomain (96% ± 5; mean ± SE, n=5) compared to SCR siRNA treated cells. The free ErbB4 ectodomain level was increased in SCR cells treated with NRG to 120% ± 8.3 (mean ± SE, n=5), and in PSEN-1 siRNA treated cells the level of the 120 kD ErbB4 ectodomain was increased to 135% ± 5.2 (mean ± SE, n=5) (Supplementary Figure 1). These observations suggest that, in contrast to its effect on levels of the 4ICD cytoplasmic domain, PSEN-1 has no detectable effect on the release of the ErbB4 ectodomain.

3.6 TACE knockdown does not affect SP-B and SP-C expression In MLE12 cells

The necessary first step in proteolytic processing of ErbB4 leading to ultimate release of the 4ICD fragment and non-canonical ErbB4 signaling is thought to be cleavage by TACE, a trans-membrane metalloproteinase. To determine if TACE activity is required for NRG stimulation of surfactant production, we used siRNA to knock down TACE in MLE12 cells and measured SP-B and SP-C expression levels by western blot analysis. TACE protein was reduced to 53% ± 3 (mean ± SE, n=4, p=0.003) of the scrambled control (Figure 5A). MLE12 cells treated with TACE siRNA showed no change in SP-B and SP-C (Figure 5B). Furthermore, TACE knockdown did not inhibit the stimulation of SP-B and SP-C by NRG. Cells treated with NRG showed higher expression of SP-B and C in the SCR treatment condition (140.7% ± 9, 120% ± 7 respectively; mean ± SE, n=4; p= 0.02, p=0.003 respectively) and this stimulatory effect appeared to be maintained in TACE knockdown samples (Figure 5C and 5D). These results indicate that NRG stimulation of surfactant production is largely independent of TACE activity in MLE12 cells.

Figure 5.

Representative western blot and densitometry quantification of TACE knockdown in MLE12 cells and the effect on surfactant protein expression. (A) TACE knockdown in TACE siRNA, GAPDH siRNA and SCR siRNA treated cells. Densitometry values were normalized to actin and compared to SCR siRNA treatment. Bars show mean ± SEM of N=4 experiments. (B) Representative western blots of SP-B and SP-C expression in TACE siRNA, GAPDH siRNA and SCR siRNA treated MLE12 cells. (C) Western blots and densitometry of SP-B expression in TACE siRNA and SCR siRNA treated cells without and with NRG treatment. Densitometry values were normalized to actin and compared to the respective SCR value. Bars are means ± SEM of N=4 experiments; *=P<0.05. (D) Western blots and densitometry of SP-C expression in TACE siRNA and SCR siRNA treated cells without and with NRG treatment. Densitometry values are normalized to the respective SCR value. Bars are means ± SEM of N=4 experiments; **=P<0.005.

3.7 Levels of shed ErbB4 ectodomain, but not 4ICD is decreased in MLE12 cells treated with TACE siRNA

To assess the role of TACE in the release of both the ectodomain and the 4ICD components of ErbB4, we measured the ectodomain level in conditioned media and the 4ICD level in cell lysates following TACE knockdown via siRNA. Western blot analysis of cells treated with TACE siRNA showed no change in 4ICD levels compared to SCR treated cells (98% ± 2.5 of SCR cells; mean ± SE, n=4). NRG treatment increased the level of 4ICD in both SCR treated and TACE siRNA treated cells (130.6% ± 2.1, 150% ± 2.3, means ± SE, n=4 respectively) compared to untreated scrambled control cells (Figure 6A). To further demonstrate that TACE inhibition blocked proteolytic release of the ErbB4 ectodomain, conditioned medium from TACE siRNA treated cells was collected after 24 hours of treatment, and media assayed for presence of the released ectodomain fragment of ErbB4. Western blot analysis of the conditioned media from cells treated with TACE siRNA showed decreased ErbB4 ectodomain levels (40% ± 1.1; mean ± SE, n=5, p=0.003) compared to scrambled siRNA treated cells. NRG stimulated the level of released ectodomain in scrambled siRNA treated cells (150% ± 8.7, mean ± SE, n=4, p=0.0173) compared to cells not treated with NRG. The stimulatory effect of NRG on ectodomain shedding was significantly lost in TACE siRNA treated samples (Figure 6B). These observations indicate that TACE specifically regulates ectodomain cleavage, but not cytoplasmic cleavage regulated by PSEN-1 leading to 4ICD release, contrary to current models of ErbB4 non-canonical signaling.

Figure 6.

The effect of TACE knockdown on the levels of the 4ICD cleavage fragment (A) and the released ectodomain fragment (B) in MLE-12 cells. Figure shows representative western blots and densitometry values normalized to SCR without or with NRG treatments respectively. Bars are means ± SEM of N=4 experiments; **= P<0.005 compared to SCR condition, T-test with Bonferroni correction for multiple comparisons.

3.8 Chemical inhibition of TACE activity

To further validate the results obtained from TACE knockdown, we used TAPI, a specific and well-characterized chemical inhibitor of TACE activity. Dose-response experiments in MLE12 cells showed that 200 nM was sufficient to reduce active TACE levels (measured by western blot) to 30% ± 1.4 (mean ± SE, n=4, p= 0.0004) compared to media treated cells (Supplementary Figure 2).

3.9 TACE inhibition by TAPI does not affect SP-B, 4ICD expression

We confirmed the absence of a role of TACE in ErbB4-mediated regulation of SP-B by treating MLE12 cells with 200nM TAPI without and with NRG. MLE12 cells treated with TAPI showed no change in SP-B 99% ± 1.4 (mean ± SE, n=4) (Figure 7A). These results indicate that baseline SP-B production is largely independent of TACE activity, at least in MLE12 cells. MLE12 cells exposed to NRG alone or NRG plus TAPI showed higher levels of SP-B protein. SP-B was significantly increased in media plus NRG treated cells to 214% ± 14 (means ± SE, n=4, p=0.0004) and in NRG plus TAPI treated cells to 197% ± 11 (means ± SE, n=4, p=0.004) (Figure 7B). These results confirm that unlike PSEN-1, TACE is not involved in NRG-stimulated mediation of ErbB4 control of SP-B expression in MLE12 cells. Analysis of the conditioned media showed that TAPI-treated cells shed significantly less ErbB4 ectodomain (50% ± 3.4; mean ± SE, n=4, p=0.001) compared to untreated cells. NRG treatment stimulated ectodomain release in control cells (143.8% ± 5; mean± SE, n=4, p=0.01). As expected, this stimulatory effect was absent in conditioned media from TAPI treated cells (Figure 8A). The presence of the 4ICD fragment in cell lysates from TAPI treated cells was not changed (85.3% ± 5.8; mean ± SE, n=3, p=0.17) compared to media treated cells (Figure 8B). These data indicate that TACE inhibition by TAPI prevents TACE activity consistent with TACE knockdown using siRNA method.

Figure 7.

The effect of TACE inhibition by TAPI on SP-B levels. (A) Representative western blot and densitometry quantification of SP-B protein levels in MLE12 cells treated as control or with 200 nM TAPI. Inhibition of TACE by TAPI did not affect SP-B levels. (B) Representative western blot and densitometry quantification of the effect of TACE inhibition by TAPI on NRG stimulation of SPB protein levels. Bars represent mean ± SEM of N=4 experiments; **=P<0.005, ***=P<0.0005 compared to the respective controls.

Figure 8.

The effect of TACE inhibition by TAPI on the levels of released ectodomain fragment (A) and 4ICD cleavage fragment (B) from ErbB4. Figure shows representative western blots and densitometry values normalized to SCR without or with NRG treatments respectively. Bars are means ± SEM of N=4 experiments; * = P<0.05, **= P=0.005 compared to respective controls. Bars are means ± SEM of N=4 experiments.

4.0 DISCUSSION

In spite of advances in the treatment and prevention of RDS including the use of prenatal glucocorticoids and neonatal treatment with postnatal surfactant replacement, RDS remains one of the significant causes of morbidity and mortality in premature infants [26], [27]. Therefore, it is important to investigate the endogenous pathways regulating the control of surfactant synthesis in the alveolar type II cell. We have shown the important role of ErbB4 receptor and its ligand NRG in surfactant synthesis [7], [4]; [5]; [22], but the specific signaling mechanism(s) in type II cells that promote surfactant synthesis is not well understood. Our previous work suggests that ErbB4 proteolytic processing and trafficking mechanisms are involved in its control of type II cell maturation and surfactant production [22]. In this study we defined the significance of ErbB4 processing by PSEN-1 and TACE for stimulation of type II cell surfactant DSPC synthesis and SP-B and SP-C expression to provide a basis for the development of novel therapeutic approaches to prevent and treat RDS. In the process we found that proteolytic processing of ErbB4 by PSEN-1 does not require previous processing by TACE. These are novel findings for both ErbB4 signaling and type II cell biology.

Previous studies of the proteolytic processing of ErbB4 in NRG-induced signaling indicate this is a two step process beginning with cleavage at the cell surface by TACE with shedding of the ectodomain, followed by cleavage at the inner cell membrane by γ-secretase to release the 4ICD fragment which traffics to the nucleus [12]. There are two separate enzymes that may make up the active component in the γ-secretase complex, PSEN-1 and PSEN-2, which are transcribed from different genes. While there is some overlap in their function, studies show that γ-secretase complexes containing PSEN-1 or PSEN-2 have functionally distinct phenotypes [20]. However, deletion of either PSEN-1 or PSEN-2 in mice results in pulmonary phenotypes, [23]. While the pulmonary phenotypes have not been completely characterized, it is known that newborn mice with PSEN-1 knockout have lungs with poorly developed saccular structures and undergo neonatal death with poorly expanded lungs. PSEN-2 knockout mice survive birth but develop progressive alveolar wall thickening, alveolar and airway fibrosis, and hemorrhage. Double knockouts show an exacerbation of the fetal/neonatal abnormalities of alveolar development [20]; [21]. Mice with an inactivating mutation in TACE also have poorly developed saccular structures and undergo neonatal death with poorly expanded lungs [28]. However, no study has addressed the interdependence of TACE and PSEN-1 processing of ErbB4 in the ErbB4 signaling pathway controlling type II cell surfactant expression.

We focused on PSEN-1 because of the more severe developmental phenotype of alveolar maturation in the PSEN-1 knockout mouse. We used siRNA to silence PSEN-1 expression in MLE12 cells. MLE12 cells treated with siRNA targeting PSEN-1 showed a reduction in the level of ErbB4 4ICD cytoplasmic cleavage product and no change in the ectodomain shedding component of ErbB4. Moreover, these cells exhibited significant decreases in the surfactant proteins SP-B and SP-C and in the synthesis of the major phospholipid component of surfactant, DSPC. Further, cells treated with PSEN-1 siRNA showed a reduction in the released ErbB4 4ICD cytoplasmic cleavage product and no change in the shedding of the ectodomain component of ErbB4. NRG treatment afforded the opportunity to evaluate the initiation of this ErbB4 processing pathway. Our data indicate that cells treated with NRG showed increased 4ICD levels in association with stimulated expression of SP-B, SP-C and DSPC synthesis. The fact that these stimulatory effects were lost with PSEN-1 knockdown provides important evidence of the importance of PSEN-1 processing in ErbB4-mediated type II cell surfactant synthesis that strengthens our conclusions from previous studies [22]. Thus our data indicate that ErbB4 signaling in lung alveolar type II cells utilizes the pathway of the PSEN-1- dependent γ-secretase proteolytic cleavage for promoting surfactant synthesis.

In our studies of 4ICD levels in response to NRG treatment following PSEN-1 knockdown, the effect of NRG stimulation on the level of 4ICD was not completely removed. This could reflect some cleavage activity by PSEN-2 or possibly the fact that with our knockdown the PSEN-1 protein level was reduced to 50%, possibly allowing a low residual PSEN-1 activity. However at the same time the NRG effect on the levels of SPB, SPC and DSPC were all significantly affected by the PSEN-1 knockdown. This fact, plus the fact that the change in 4ICD with NRG was small, we interpret as evidence that any residual 4ICD production was not sufficient to allow NRG stimulation of surfactant proteins and DSPC. This conclusion is also supported by our previous work in which a mutant ErbB4 construct that lacked a gamma secretase binding site transfected into fetal lung type II cells did not increase baseline or NRG-stimulated SP-B mRNA expression [22].

γ-Secretase cleavage and nuclear transport of ErbB4 are important components in regulation of fetal lung type II cell surfactant protein production [22]. Because the accepted model of ErbB4 proteolytic processing includes cleavage by TACE followed by cleavage by γ-secretase, we then sought to demonstrate that shedding of the ectodomain through TACE cleavage is also a necessary first step for ErbB4 signaling in controlling surfactant synthesis. We used both siRNA to knock down TACE and chemical inhibition of TACE enzyme activity in MLE12 cells to study the importance of TACE processing of ErbB4 for SP-B and SP-C production. For TACE inhibition we used TAPI, a hydroxamate inhibitor of TACE activity [29], which targets the zinc active motif on the enzyme. The results of these experiments showed that knockdown of TACE either by siRNA or TAPI does not inhibit SP-B and SP-C expression in MLE12 cells. More significantly we found that TACE inhibition did not suppress the stimulatory effect of NRG on surfactant protein levels. To further probe the role of TACE we studied the effect of TACE inhibition on γ-secretase mediated cleavage of ErbB4 in response to NRG by probing for the 4ICD fragment of ErbB4 in whole cell lysates of TACE siRNA treated and TAPI treated cells. We found that even after TACE knockdown or inhibition NRG stimulation continued to increase 4ICD levels. On the other hand the shedding of the ectodomain of ErbB4 was decreased in the conditioned media of these cells, as would be expected with inactivation of TACE. These findings suggest that the two cleavage steps in ErbB4 are not linked events that must both occur for ErbB4 processing leading to signaling by the 4ICD component. However, we cannot rule out the possibility that the low remaining TACE activity was sufficient to prime PSEN-1 cleavage. Nevertheless, together with our previous studies of the significance of nuclear localization of ErbB4 in NRG-mediated control of type II cell differentiation [22] these data indicate that in this system proteolytic processing by TACE and by γ-secretase have separate functions for ErbB4 signaling. In conclusion we find that ErbB4 signaling in type II epithelial cells utilizes the pathway of the PSEN-1- dependent γ-secretase proteolysis. Our results emphasize the importance of PSEN-1 for ErbB4 signaling in control of alveolar type II cell differentiation. Our work also introduces novel insight into the mechanism of ErbB4 cleavage processing to produce signaling by the 4ICD component by showing that TACE activity is not a necessary antecedent for γ-secretase cleavage of ErbB4 in the control of surfactant DSPC synthesis and SP-B and SP-C expression. This dissociation between TACE and γ-secretase processing of ErbB4 in lung type II cells may have important translational impact as investigators seek to modulate ErbB4 signaling. In particular, this study may provide significant clues for how to target ErbB4 biology in developing translational approaches to benefit babies with RDS.

RESEARCH HIGHLIGHTS.

• PSEN-1 knockdown decreases SP-B and SP-C and DSPC levels in MLE12 cells

• PSEN-1 knockdown suppresses neuregulin stimulation of SP-B, SP-C and DSPC levels

• PSEN-1 knockdown decreases ErbB4 cytoplasmic domain but not ectodomain release

• TACE inhibition does not affect PSEN-1 processing of ErbB4 or SP-B and SP-C levels

NON-STANDARAD ABBREVIATIONS

SP-B

Surfactant Protein B

SP-C

Surfactant Protein C

NRG-1

Neuregulin-1

TACE

Tumor Necrosis Factor Alpha Converting Enzyme

DSPC

Disaturated Phosphatidyl Choline