Tissue inhibitor of metalloproteinase (TIMP)-1 is a potent inhibitor of activated matrix metalloproteinases (MMPs) such as gelatinases and collagenases. TIMP-1 is induced by transforming growth factor-β1 (TGF-β1), but details regarding signaling pathways remain unclear. T-helper-2 cytokines also have profibrotic properties and can interact with TGF-β. In the present study, we examined the effects of interleukin (IL)-13 (2,500 pM) on TGF-β1 (200 pM)-induced expression of TIMP-1 mRNA and protein in primary human airway fibroblasts obtained from 57 human subjects. IL-13 alone had no effect on TIMP-1 mRNA or protein expression. However, IL-13 synergistically augmented TGF-β1-induced TIMP-1 mRNA and protein expression (P < 0.001 vs. TGF-β1 alone). The upregulation of TIMP-1 by the combination of TGF-β1 and IL-13 involved increased transcription, with little effect on mRNA stabilization. Initial exploration of the pathways leading to the synergy determined that activation of the phosphatidylinositol 3-kinase (PI3K)-Akt pathway by IL-13 may have a negative effect on TIMP-1 production. The specific PI3K inhibitor LY-294002 in the presence of TGF-β1, IL-13, or the combination of the two caused significant increases in TIMP-1 mRNA expression, while LY-294002 increased TIMP-1 protein levels in the presence of IL-13 alone. These results suggest that IL-13 augments TGF-β1-induced profibrotic responses at both the mRNA and protein levels. Although IL-13 induced activation of PI3K-Akt, the activation did not contribute to the synergy observed with TGF-β1 plus IL-13 in TIMP-1 expression and in fact may dampen it. The mechanisms behind the synergy remain to be determined.
- phosphatidylinositol 3-kinase
tissue remodeling in chronic illnesses such as asthma is likely to be of considerable importance to long-term outcomes. One of the hallmark remodeling changes in the airways of patients with asthma is increased collagen deposition in the subepithelial basement membrane (SBM) (5, 34). Transforming growth factor (TGF)-β, a multifunctional growth factor linked to general wound repair responses through the modulation of extracellular matrix (ECM) has previously been linked to SBM thickening in people with asthma (30, 36). TGF-β enhances production of ECM components, including collagen and connective tissue growth factor (CTGF), while generally suppressing production of ECM-degradative enzymes such as matrix metalloproteinases (MMPs) (4, 8). Tissue inhibitor of metalloproteinase (TIMP)-1, a natural inhibitor of multiple MMPs, is enhanced by cytokines such as TGF-β (26). The upregulation of TIMP-1 by TGF-β involves the activator protein-1 (AP-1) pathway, but many steps of the process remain unclear (21, 40).
In addition to wound repair responses, asthma involves a T-helper (TH)-2-type inflammatory process (6, 10). Similar to TGF-β, TH-2 cytokines such as interleukin (IL)-4 and IL-13 enhance fibrotic responses. Mice with overexpressed IL-13 have increased airway fibrosis compared with wild-type mice (19, 38). At the cellular level, stimulation of fibroblasts with IL-4 has induced modest increases in collagen production from fibroblasts. However, the effects of IL-4 or IL-13 on other profibrotic elements such as TIMP-1 have not been evaluated in primary human airway fibroblasts. The signaling pathways for the known profibrotic effects of IL-4/IL-13 are controversial. Some studies support molecular effects through signal transducer and activator of transcription (STAT)-6 (25, 39). Other studies suggest IL-4/IL-13 effects through signaling molecules such as insulin receptor substrate (IRS)-phosphatidylinositol 3-kinase (PI3K)-Akt pathway (2, 12).
Although the known interactions between TGF-β and TH-2 signaling pathways are limited, a previous study from our laboratory reported that IL-13 (or IL-4), in combination with TGF-β1, induced marked upregulation of eotaxin-1 production in primary human airway fibroblasts (37). This increase in eotaxin-1 was correlated with increases in type I collagen (35). Hence, at the level of airway fibroblasts, TGF-β1 in association with IL-13 may enhance both eosinophilic and fibrotic factors. In the present study, we have demonstrated that similar synergistic increases in TIMP-1 occur in response to the combination of TGF-β1 and IL-13. This initial study outlines the mechanisms of this enhancement and the negative involvement of the PI3K-Akt pathway in that process.
MATERIALS AND METHODS
Informed consent was obtained from all individuals in accordance with the “Guiding Principles for Research Involving Animals and Human Beings” of the American Physiological Society.
Cytokines, antibodies, inhibitors, and other materials.
Dulbecco’s modified Eagle’s medium (DMEM) was purchased from GIBCO-BRL (Rockville, MD), and fetal bovine serum (FBS) was obtained from Gemini (Woodland, CA). Recombinant human IL-13, TGF-β1, and TIMP-1 antibodies (MAB970 and BAF970) were obtained from R&D Systems (Minneapolis, MN). TRIzol reagent was obtained from Invitrogen/Life Technologies (Carlsbad, CA). Phospho-Akt (Ser473) rabbit polyclonal antibody and Akt polyclonal antibody were obtained from Cell Signaling Technology (St. Louis, MO). Actinomycin D and cycloheximide were purchased from Sigma (St. Louis, MO). LY-294002 (VWR, West Chester, PA) was used at 10 or 25 μM in 0.1% DMSO (18, 27, 28). Hybond enhanced chemiluminescence (ECL) nitrocellulose membrane, ECL Western blotting detection reagents, and high-performance chemiluminescence film were obtained from Amersham Biosciences (Arlington Heights, IL).
Endobronchial biopsies were obtained from healthy volunteers (controls) and patients with asthma ranging in severity from mild to severe as previously described (36). Briefly, 29 patients with severe asthma who were taking high-dose glucocorticoids (GCs), 10 patients with mild to moderate asthma who were taking no or low doses of inhaled GCs, and 18 healthy volunteers with no evidence for any respiratory disease were studied. Biopsies were obtained from third- to fifth-order subcarinae, placed in DMEM with 10% FBS and penicillin-streptomycin-gentamicin, processed as previously described, and studied at passage 3 or 4 (37). For most studies, fibroblasts (40,000/well) were plated in 24-well plates. After 24-h serum starvation, FBS (0.5%) and TGF-β1 (200 pM), IL-13 (2,500 pM), or the combination of TGF-β1 (200 pM) plus IL-13 (2,500 pM) was added. The supernatants and cells were harvested at 48 h. Time-course and dose-response studies were conducted as outlined in results, with supernatants and cells harvested at the described time points for later analysis of TIMP-1 protein and mRNA. Western blot analysis was performed in cells from six-well plates.
Enzyme immunoassay for TIMP-1.
Standard sandwich ELISAs were used to quantify TIMP-1 concentration in the supernatant. The TIMP-1 antibody MAB970 was used as the capture antibody, and biotinylated BAF970 was used as the detection antibody. Assay sensitivity ranged from 10 to 20 pg/ml.
Real-time PCR for the measurement of TIMP-1 gene expression.
TIMP-1 mRNA expression in fibroblasts was determined by performing reverse transcription (RT) followed by real-time quantitative PCR. Total RNA was extracted from fibroblasts using TRIzol reagent. RT was performed as described previously (37). The TIMP-1 primers and probe, labeled with 5′-reporter dye 6-carboxyfluorescein (FAM) and 3′-quencher dye 6-carboxy-N,N,N′,N′-tetramethylrhodamine (TAMRA), were designed using primer express software (Applied Biosystems, Foster City, CA). The following are the sequences for the TIMP-1 primers and probe (GenBank accession no. NM_003254): forward primer (nt 150–168): 5′-CACCCACAGACGGCCTTCT-3′; reverse primer (nt 214–239): 5′-CTGGTATAAGGTGGTCTGGTTGACTT-3′; and probe (nt 172–195): 6FAM-ATTCCGACCTCGTCATCAGGGCCA-TAMRA.
Ready-for-use human GAPDH primers and probes (labeled with VIC) were obtained from Applied Biosystems (GenBank accession no. NM_002046, part no. 4310884E). Real-time PCR was performed on the ABI Prism 7700 sequence detection system (Applied Biosystems). The 25-μl PCR contained 15 ng of cDNA template, 900 nM primers, and 200 nM probe. GAPDH was evaluated using the same PCR protocol as TIMP-1. The specificity of PCR for both TIMP-1 and GAPDH was verified by no signal in no template controls and no amplification fibroblast RNA samples. TIMP-1 mRNA was indexed to GAPDH using the formula 1/(2ΔCT) × 100% and then converted to the percentage over medium control.
In vitro PCR-based nuclear run-on assay.
Nuclear run-on was performed on nuclei (CelLytic nuclear extraction kit; Sigma) isolated from fibroblasts stimulated with media alone, TGF-β1, and TGF-β1 plus IL-13 for 48 h according to methods described previously (9, 29). The nuclei were stored in 200 μl of storage buffer [50 mM Tris·HCl, pH 8.3, 40% glycerol (vol/vol), 5 mM MgCl2, and 0.1 mM EDTA]. Nuclei (200 μl) were split into two aliquots and incubated at 30°C in 20% glycerol, 30 mM Tris·HCl, 2.5 mM MgCl2, 150 mM KCl, 1 mM DTT, and 40 U of RNase inhibitor (Applied Biosystems). ATP, CTP, GTP, and UTP (0.5 mM each) were added to one aliquot (+NTPs), while no NTPs were added to the second aliquot (−NTPs). After 2.5-h incubation, RNA was isolated, DNase I digestion was performed, and RT and real-time PCR products were analyzed as described to determine the quantity of newly formed mRNA, calculated as follows: These rates were then compared between the stimuli.
Effects of actinomycin D and cycloheximide on TIMP-1 mRNA expression and stability.
The roles of transcription and de novo protein synthesis in TIMP-1 expression were analyzed by adding either 5 μM actinomycin D (Act D) or 100 μg/ml cycloheximide (CHX) at the time of stimulation. RNA was harvested at 6, 24, and 48 h for analysis. To determine the effects of IL-13 on TIMP-1 mRNA stability, fibroblasts were cultured in the presence of TGF-β1 or the combination of TGF-β1 plus IL-13. After 24 or 36 h of stimulation, Act D was added and RNA was harvested at 0, 2, 4, and 8 h. Total RNA was extracted for quantification of TIMP-1 mRNA using real-time PCR, and the amount of mRNA was compared at each time point for the two conditions.
Western blot analysis of Akt.
Western blot analysis was performed for phosphorylated or total Akt as described previously (27, 37). After serum starvation, TGF-β1, IL-13, or the combination of TGF-β1 plus IL-13 was added for 0.25, 1, and 6 h, and cells were lysed in lysis buffer with 1% protease inhibitor cocktail (Sigma). Total protein (30 μg) (measured by bicinchoninic acid; Pierce, Rockford, IL) was mixed with sample buffer, heated for 10 min, loaded, and run on a 10% SDS-PAGE gel. The separated proteins were transferred to a Hybond ECL nitrocellulose membrane, blocked at room temperature for 1 h with 5% milk, and incubated at 4°C with rabbit anti-phospho-Akt polyclonal antibody (1:1,000 dilution) (Ser473) overnight. The blots were washed, incubated at room temperature for 2 h with HRP-conjugated anti-IgG F(ab′)2 antibody (1:5,000 dilution), and developed using an ECL Western blot detection system. Membranes were stripped and reblotted with total Akt polyclonal antibody (1:1,000 dilution). Quantification was performed by evaluating the ratio of the density of the total Akt bands to that of the phospho-Akt bands for all conditions.
Effects of PI3K inhibitor on Akt phosphorylation and TIMP-1 mRNA/protein expression.
To confirm the effects of PI3K inhibitor LY-294002 on PI3K-Akt pathway and TIMP-1 expression after optimization of inhibition experiments, 10 or 25 μM LY-294002 or 0.1% DMSO was added to cell culture medium 1 h before stimulation with 200 pM TGF-β1, 2,500 pM IL-13, or the combination of TGF-β1 plus IL-13. Cells were lysed at 1 h after addition of stimulus and analyzed for phosphorylation of Akt as described above. In separate experiments, the RNA and supernatants were collected after 48 h in the presence or absence of 10 μM LY-294002 after stimulation with TGF-β1, IL-13, or combined TGF-β1 and IL-13 for TIMP-1 mRNA and protein determination as described above.
To determine whether the increases in TIMP-1 were due to varying effects of TGF-β1, IL-13, or the combination of the two and PI3K inhibitor LY-294002 on proliferation, stimuli were added to nearly confluent fibroblast cultures and the cells were harvested for cell number analysis at 48 h. The CyQuant cell proliferation assay kit (Molecular Probes, Eugene, OR) was used to determine the cell number per well in accordance with the manufacturer’s directions. The results are expressed as cell number per condition.
Data were checked for normality of distribution. Normally distributed data are presented as means ± SE. Log transformation was performed with variables that were heavily skewed to the right. When more than two groups were compared, ANOVA was used, while Student’s t-test was used to compare two sets of data. Paired analysis was used for small group comparison of changes. When data were log transformed, they were graphed as means ± SE converted from the log-transformed data to the original scale. Although the response patterns to TGF-β1 and IL-13 were generally consistent, there was variability from subject to subject in the absolute amount of basal TIMP-1 produced. Therefore, each stimulated condition (TGF-β1, IL-13, or TGF-β1+IL-13) is represented as a percentage of the unstimulated baseline. P ≤ 0.05 was considered statistically significant.
Effects of TGF-β1 and IL-13 on TIMP-1 protein and mRNA levels.
As expected, TGF-β1 increased TIMP-1 protein more than medium alone did. IL-13 alone had no effect on TIMP-1 production. However, TGF-β1 plus IL-13 caused a further increase in TIMP-1 production that was greater than that of either stimulus alone (Fig. 1A). There were no statistical differences in TIMP-1 levels under any of the conditions between subject groups (Table 1). Paired analysis of the changes with the combination of TGF-β1 plus IL-13 compared with TGF-β1 alone also showed significant increases across all three groups (healthy volunteers, P < 0.02; mild to moderate asthma, P < 0.02; and severe asthma, P < 0.001). Because statistical differences did not exist between study groups, cultures from the various groups were used interchangeably. However, there remained variability in the response such that ∼60–70% of individuals studied showed synergy for TIMP-1 expression in the presence of TGF-β1 plus IL-13.
Similar to protein, TGF-β1 significantly increased TIMP-1 mRNA expression, while IL-13 alone had no effect. The combination of TGF-β1 plus IL-13 synergistically and significantly increased TIMP-1 mRNA expression over that of TGF-β1 alone (Fig. 1B).
Dose response of IL-13 on TGF-β1-induced TIMP-1 expression.
To determine the appropriate concentration of IL-13 to use with TGF-β1 to evaluate TIMP-1 production, dose responses were constructed in four separate primary fibroblast cultures. IL-13 [0–8,300 pM (100 ng/ml)] was added to 200 pM (5 ng/ml) TGF-β1. The maximum response occurred when 2,500 pM IL-13 was combined with 200 pM TGF-β1 (Fig. 2). This dose of IL-13 was used for the remaining studies.
Time course for production of TIMP-1 mRNA and protein after TGF-β1 and/or IL-13.
Figure 3A shows that IL-13 decreased TIMP-1 mRNA by 30% compared with control medium at 2 h. However, both TGF-β1 and TGF-β1 plus IL-13 increased TIMP-1 mRNA expression as early as 2 h after stimulation. There was a progressive increase over time that began to plateau at 24 h. By 48 h, the TIMP-1 mRNA level after stimulation with TGF-β1 plus IL-13 was greater than that after TGF-β1 alone. Figure 3B demonstrates that TIMP-1 protein was increased by TGF-β1 alone and by TGF-β1 plus IL-13 at 24 h, but there were no differences in amount. Similar to the initial study (n = 57), at 48 h, the combination of TGF-β1 plus IL-13 increased the TIMP-1 protein level to a greater degree than TGF-β1 alone.
Effects of TGF-β1 and/or IL-13 on fibroblast proliferation.
CyQuant data showed that none of the observed stimuli affected fibroblast proliferation at 48 h. The cell number, expressed as a percentage of cell number in the medium control, was not different among the groups (TGF-β1, 99.51 ± 8.69%; IL-13, 97.24 ± 11.28%; and TGF-β1+IL-13, 97.25 ± 7.98%).
Role of transcription and de novo protein synthesis in TIMP-1 mRNA expression induced by TGF-β1, IL-13, or TGF-β1 plus IL-13.
Cell viability was not affected by Act D or CHX treatment, because <5% of Trypan blue was taken up at 6, 24, and 48 h under any conditions. Figure 4A shows that Act D inhibited the increases in TIMP-1 mRNA with TGF-β1 or TGF-β1 plus IL-13 stimulation at 6 and 24 h. Similar but smaller effects of Act D on TIMP-1 mRNA levels were observed in cells treated with medium or IL-13 alone. Act D had minimal effects on TIMP-1 mRNA at 48 h. CHX lowered TIMP-1 mRNA levels at 6 h for TGF-β1 and the combined TGF-β1 and IL-13 but did not inhibit TIMP-1 mRNA expression at later time points under any conditions.
Effects of IL-13 on the rate of TGF-β1-induced TIMP-1 mRNA transcription.
PCR-based nuclear run-on assays demonstrated that the transcription rate of TIMP-1 was significantly greater in the presence of IL-13 and TGF-β1 than with TGF-β1 or control medium alone. This may further indicate that at 48 h, in contrast to combined IL-13 and TGF-β1, TGF-β1 alone no longer increased TIMP-1 transcription above control levels (Fig. 4B).
TIMP-1 mRNA stability in the presence of TGF-β1 or TGF-β1+IL-13.
The rate of decline in TIMP-1 mRNA levels after 24- or 36-h stimulation with TGF-β1 or TGF-β1 plus IL-13 was without significant differences at 2, 4, and 8 h (Fig. 4C).
Role of PI3K-Akt pathway in the synergistic effects of IL-13 on TGF-β1-induced TIMP-1 expression.
IL-4Rα signaling is thought to involve an IRS-1 site that, when activated, leads to activation of the PI3K-Akt pathway. Therefore, the phosphorylation of Akt was evaluated by Western blot analysis of lysed fibroblasts. TGF-β1 did not induce Akt phosphorylation beyond that of the control medium. However, IL-13 induced phosphorylation of Akt within 0.25 h of stimulation. The increase in Akt phosphorylation was partially maintained through 1 h and decreased (but still present) at 6 h. Although the phospho-Akt induced by TGF-β1 plus IL-13 was slightly weaker than that of IL-13 alone, it was stronger than medium or TGF-β1 alone (Fig. 5, A and B).
To further determine the functional involvement of the PI3K-Akt pathway in TIMP-1 production, the specific PI3K inhibitor LY-294002 was added to confirm the inhibitory effects on Akt phosphorylation (Fig. 5C). After confirmation of PI3K-Akt inhibition, TIMP-1 mRNA and protein were evaluated after stimulation with TGF-β1 and/or IL-13 for 48 h in the presence of 10 μM LY-294002. LY-294002 increased TIMP-1 mRNA expression two- to fourfold over that of cells not treated with inhibitors under all four conditions (Fig. 6A). However, as noted by others, LY-294002 inhibited cell proliferation [percentage of cells for LY-294002 compared with no inhibitor controls: 75.37 ± 13.63% (medium alone), 79.25 ± 6.88% (IL-13), 78.90 ± 5.70% (TGF-β1), and 84.5 ± 10.46% (TGF-β1+IL-13)]. Therefore, to evaluate the effects on protein levels, TIMP-1 protein concentration was divided by the average percentage of cell numbers in the presence of inhibitor vs. control medium. LY-294002 increased TIMP-1 protein levels compared with no inhibitor for control medium or IL-13. Addition of LY-294002 to TGF-β1 only marginally increased TIMP-1 protein (P = 0.072), while there was no increase in TIMP-1 protein after the addition of LY-294002 to the combination of TGF-β1 plus IL-13 (Fig. 6B).
TH-2 cytokines such as IL-4 and IL-13 are thought to be involved in airway remodeling; however, the signaling pathways involved in the process are not clear. The present study demonstrates that IL-13 has no direct effect on the profibrotic factor TIMP-1 but has significant synergistic effects on TIMP-1 expression when present in conjunction with TGF-β1. These results suggest that at least part of the profibrotic influence of IL-13 is to enhance the effects of well-described fibrotic factors such as TGF-β1.
Both TH-2 cytokines and TGF-β have been identified in the biopsies of patients with asthma, especially those of patients with more severe disease and those with a thickened SBM (36). In addition, TIMP-1 levels are increased in the lavage fluid of patients with severe asthma with regard to the presence of eosinophils, which are known to produce TGF-β (20, 22). TIMP-1 is a profibrotic factor that is strongly induced by TGF-β and has the capacity to modify cellular activities and modulate matrix turnover (8, 40). Little is known about the effects of TH-2 cytokines on TIMP-1 expression, and the studies outlined in this article suggest that IL-13 alone has no effect on TIMP-1 expression. However, in the presence of TGF-β1, IL-13 synergistically augments TIMP-1 mRNA and protein expression in primary human airway fibroblasts. Similar enhanced profibrotic effects of TH-2 cytokines in combination with TGF-β have been observed in murine and human fibroblasts (16, 37). In the present study, although the response was consistent and significant, there were both responders and nonresponders to the synergistic effects of TIMP-1 expression induced by TGF-β1 plus IL-13. Similarly, murine studies have noted substantial differences in IL-4 fibrotic responses dependent on genetic background (C57Bl/6 vs. CBA/J) (11), which suggests that some of the variability in the human system could also be genetically linked. Because of this variability, to evaluate the mechanism behind the synergistic response, only responders were chosen for mechanistic studies.
TIMP-1 production in response to TGF-β1 or TGF-β1 plus IL-13 appears to be transcriptionally regulated, with almost complete inhibition of TIMP-1 mRNA expression by Act D at 6 and 24 h. The diminished inhibition at 48 h was likely due to the relatively short half-life of Act D in culture, with most previous studies having used short time courses for Act D (1, 23). Using a real-time-based nuclear run-on assay, we found that 2.5 h poststimulation was optimal for observing new transcript synthesis. However, differences in heterogeneous nuclear RNA (hnRNA) processing and reinitiation of transcription cannot be ruled out with the methodology we used. Figure 4B suggests that IL-13 augmentation of TGF-β1-induced TIMP-1 mRNA involved an increase in the rate of transcription. Interestingly, at 48 h poststimulation, TGF-β1 did not increase TIMP-1 transcription over that of control. While further study is needed, this may be one of the mechanisms by which TGF-β1+IL-13 synergistically increased TIMP-1 mRNA and/or protein. In contrast, TIMP-1 mRNA stability did not differ significantly in the presence of TGF-β1 vs. TGF-β1 plus IL-13, suggesting that posttranscriptional regulation does not play a major role in the upregulation of TIMP-1 expression. Interestingly, Fig. 4C shows that TIMP-1 mRNA levels increased after 4-h Act D treatment. The reasons for this are not clear, but Act D at the doses used may have partially lost its inhibitory effect on TIMP-1 transcription by that time point. Although Fig. 4A suggests that new protein synthesis had a marginal impact on TIMP-1 mRNA at 24 and 48 h, a significant effect on TIMP-1 mRNA was observed at 6 h. It is conceivable that the early inhibitory effect of CHX diminished TIMP-1 expression by interfering with production of rapidly cycling signal pathway molecules involved in initial responses, but further study is needed to confirm this hypothesis.
Cytokine-induced gene responses are controlled by integration of signals from various signaling pathways on the promoter elements of target genes. Although several studies have supported the integration of IL-4/IL-13 and TGF-β signaling pathways (11, 32, 37), the mechanisms behind these interactions have not been discerned. IL-4/IL-13 classically signal through an IL-4Rα-associated JAK-STAT pathway (14). As previously reported, the synergistic increases in eotaxin-1 likely occur through effects on STAT-6 binding sites in its promoter, but initial investigations of that pathway have not demonstrated obvious differences between IL-13 or TGF-β1 plus IL-13 (37, 41). In contrast to eotaxin-1, the TIMP-1 promoter does not contain STAT-6 binding sites (8). This suggests that the enhancement of TIMP-1 production by IL-13 is less likely to occur directly through a STAT-6 mechanism.
In addition to STAT-6, the IL-4Rα molecule has been reported to signal through an IRS-PI3K-Akt pathway (3, 24). Activation of this pathway has been considered important in a variety of cell functions and it is conceivable that the downstream serine-threonine kinase, Akt, or associated pathways, such as MKK3/6 may indirectly modify signaling pathways through the AP-1 binding on the TIMP-1 promoter (7, 31). In addition to stimulatory effects, PI3K activity has been shown to have negative regulatory effects as well (3, 15, 17, 23). Consequently, we also evaluated the role of PI3K-Akt pathway in IL-13 upregulation of TGF-β1-induced TIMP-1 expression. In human airway fibroblasts, IL-13 increased phosphorylation of Akt as early as 15 min, which was maintained through 1 h and decreased at 6 h. Interestingly, Fig. 5, A and B, suggests that the presence of TGF-β1 may have slightly reduced IL-13-induced phosphorylation of Akt.
Surprisingly, inhibition of PI3K (by LY-294002) demonstrated profound two- to fourfold upregulation of TIMP-1 mRNA expression under all conditions. LY-294002 was used because long culture times were required, and this PI3K inhibitor is more stable than the alternative, wortmannin. These results infer that basal activation of the PI3K-Akt pathway appears to exist in fibroblasts, which then negatively regulates TIMP-1 expression. Addition of IL-13 further increases PI3K-Akt pathway activation, suggesting that activation of the IRS-1/2 site serves as an inhibitory “brake” on other IL-4Rα pathways, such as STAT-6. In fact, our human studies are similar to a recent report in which the insulin/IL-4Rα (I4R) motif that mediates IRS association with the IL-4Rα chain was mutated to inhibit IRS binding. In this murine model, preventing the binding of IRS-2 to the I4R motif abolished Akt phosphorylation and greatly increased allergen-associated, IL-4Rα-mediated lung responses, including IgE and mucus production (3). In our system, the combination of TGF-β1 with IL-13 similarly limited the activation of Akt (as suggested by Western blotting), thereby enhancing TIMP-1 production. Further manipulation of the PI3K-Akt pathway, including the use of Akt−/− system and small interfering RNA of Akt may help researchers to discern the role of Akt in TIMP-1 production (13, 33).
Interestingly, PI3K inhibition in the presence of TGF-β1 plus IL-13 did not significantly augment the TIMP-1 protein level. We speculate that the combination may have modified the PI3K-Akt pathway sufficiently to limit the effects of the additional PI3K inhibitor at the protein level. Specifically, there may be other steps involved in the translation of TIMP-1 that are active only in the presence of TGF-β1+IL-13+PI3K inhibition.
In conclusion, the current study has demonstrated that IL-13 synergistically increased TGF-β1-induced TIMP-1 mRNA and protein expression, primarily at a transcriptional level. While this synergy was not due to activation of the PI3K-Akt pathway, the results suggest a negative regulatory role for the PI3K-Akt pathway in IL-4Rα signaling for TIMP-1 expression. Further studies are needed to define the relationship of the PI3K-Akt pathway to STAT-6 and other signaling mechanisms, along with the relationship between those signaling pathways and TIMP-1 expression.
This work was supported by National Institutes of Health Grants HL-64087, AI-40600, and RR-00051 and the American Lung Association affiliates of Colorado, Oklahoma, and Alaska.
We thank Lynn Reed for expert technical support.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society