Background
Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive, and lethal interstitial lung disease of unknown cause [
1]. IPF is characterized by an histologic pattern of usual interstitial pneumonia (UIP) which shows a heterogeneous distribution of dense parenchyma collagen deposition and active fibroblast foci alternating with areas of normal parenchyma [
2]. An aberrant response of alveolar epithelial cells (AECs) to a repetitive damage that contributes to the loss of alveolar epithelial structures has been proposed in IPF physiopathology [
3]. Thus, an imbalance between pro-fibrotic and anti-fibrotic factors leads to an uncontrolled extracellular matrix (ECM) formation that modifies the interstitial configuration [
4]. This abnormal wound healing presents several hallmarks of accelerated aging [
5].
Interestingly, the advanced glycation end-products (AGEs), oxidative non-enzymatic products derived from modified lipids, proteins or nucleic acids, have been implicated in some diseases related to an accelerated aging process [
6‐
8], being proposed as markers of oxidative stress and aging [
9]. Oxidants such as cigarette smoke and dietary AGEs would promote the glycation process, resulting in accelerated formation of endogenous AGEs, inducing cellular dysfunction and cell death [
10]. Among dozens of AGEs described, pentosidine and Nε-Carboxymethyl lysine (CML) are the most studied. An accumulation of pentosidine has been reported in aged skin [
11] and in some pulmonary diseases [
12,
13], and a high presence of CML has been associated with cell response to oxidative stress [
14].
The involvement of AGEs in promoting ECM protein modification and cross-linking is also remarkable [
15,
16]. In this way, some studies have related a loss of tendon viscoelasticity when incubated with AGEs [
17], and an increment of arterial stiffness driven by AGEs accumulation has been described in the aortic walls [
18]. Moreover, a previous report from our group demonstrated stiffness changes in glycated 3D collagen matrices and fibroblast phenotypic transformation [
19]. This type of in vitro model would mimic the cross-links and AGEs generation. Finally, AGEs can glycate some plasmatic proteins, such as globulins or albumin, changing their physicochemical properties [
20,
21].
On the other hand, the effect of AGEs on cellular reactions has been suggested to be closely related to their receptor, RAGE [
22,
23]. However, RAGE is a member of the immunoglobulin superfamily of receptors [
24] that is highly expressed in type I AECs of healthy lungs [
25], and has been related with the differentiation of type II to type I pneumocyte cells, lung development, re-epithelialization, and maintenance of epithelial adhesion to basement membrane [
26‐
28]. RAGE can be located in the cell membrane (full-length RAGE, FL-RAGE), or soluble in the extracellular space (soluble RAGE, sRAGE), without the transmembrane and cytosolic domains [
29]. Remarkably, different roles of these two isoforms of RAGE binding the AGEs have been suggested. AGE-bound FL-RAGE activate an inflammatory signaling pathway via the nuclear factor kappa B (NF-κB) that could increase RAGE expression, enhance pro-inflammatory mediators, modify oxidative balance, activate apoptosis, and block AGEs degradation [
30]. Meanwhile, a preventive effect of their signaling pathway has been proposed when sRAGE joins AGEs [
31]. Although the involvement of AGEs/RAGEs in the pulmonary fibrotic process remains unclear, some research groups have suggested a different pattern of expression, mainly for RAGEs [
32‐
38]. Therefore, our study aims to evaluate the presence of AGEs in relation with RAGEs in IPF lungs, and the possible effect of AGEs on cell behavior.
Methods
Ethical statement
Control human lung samples were obtained from the distal area of 9 lobectomies of cancer, which showed a preserved pulmonary architecture without emphysema, infection or inflammation. IPF samples were obtained from 16 subjects who underwent surgical lung biopsy for diagnosis. The included IPF cases were discussed in the multidisciplinary Interstitial Lung Diseases Committee of the University Hospital of Bellvitge and were reported as histological UIP pattern in accordance with the American Thoracic Society/European Respiratory Society criteria [
1]. All patients provided written informed consent and the study was approved by the Ethics Committee of our center (CEIC, ref. PR202/08).
Patient characteristics
Both populations, controls and IPF, had similar demographic features without statistical differences: range of age (57.60 ± 7.47 and 63.61 ± 6.67 years, respectively) and male–female ratio in both groups (Table
1). Both groups included a high proportion of ex-smokers, and a few current smokers and non-smokers, with the duration of smoking cessation of 13.20 years ± 14.13 for the control group and 15.14 years ± 11.97 for IPF patients, showing no differences between both groups (
p = 0.77). Furthermore, there were no differences between both groups in the average of pack-years (
p = 0.94). The pulmonary functional test (PFT) showed significant differences between IPF and controls in forced vital capacity (FVC) and diffusing capacity for carbon monoxide (DLCO) (
p < 0.01) (Table
1). Diabetic patients from controls and IPF were excluded from the study to avoid the AGEs formation inherent to diabetes mellitus disease.
Total | 9 | 16 | - |
Gender (Male/Female) | 7/2 | 13/3 | - |
Age | 57.60 ± 7.47 | 63.61 ± 6.67 | 0.03 |
Smoking (Current/Former/Never) | 2/5/2 | 4/10/2 | - |
Pack-years | 24.91 ± 25.18 | 24.18 ± 26.81 | 0.94 |
Smoking cessation (years) | 13.20 ± 14.13 | 15.14 ± 11.97 | 0.77 |
% FVC | 99.36 ± 14.55 | 77.92 ± 15.05 | 0.00 |
% DLCO | 89.43 ± 13.98 | 56.91 ± 18.44 | 0.00 |
Western blot analysis
Tissue samples were homogenized in radioimmunoprecipitation assay buffer (RIPA buffer) pH 7.60 with protease inhibitors (Sigma-Aldrich, USA) in Ultra-Turrax T25 basic (IKA®-Werke, Germany). Protein concentration was determined by BCA Protein Assay Kit (Thermo Fisher Scientific Inc., USA) and 30 μg of sample was loaded in Mini-PROTEAN® TGX™ precast 4–15 % polyacrylamide gel (Bio-Rad, USA) under reducing conditions. Gels were transferred onto nitrocellulose membrane in semi-dry Trans-Blot® Turbo™ Transfer System (Bio-Rad). After blocking with TBS-T 5 % BSA (Sigma-Aldrich), the membranes were incubated with rabbit polyclonal anti-AGEs 1:10000 (ab23722; Abcam, UK), rabbit polyclonal anti-CML 1:500 (ab27684; Abcam), mouse monoclonal anti-Pentosidine 1:1000 (KAL-KH012; Cosmo Bio Co., Japan) or mouse monoclonal anti-RAGE 1:1000 (ab54741; Abcam), and mouse monoclonal anti β-actin 1:2000 or mouse monoclonal anti α-tubulin 1:2500 (A1978 and T6199, respectively; Sigma-Aldrich) as a loading control during 1 h at room temperature (RT). Then the membranes were incubated with a secondary antibody: goat anti-IgG rabbit or anti-IgG mouse HRP conjugate (P0448 and P0447, respectively; Dako, Denmark) 1:1000 (for anti-AGE or anti-RAGE, respectively) in TBS-T during 1 h at RT. Membranes were washed in TBS-T 3 times for 5 min after the primary and secondary antibody incubation. Immunoblotting was detected by chemiluminescence in a LAS-3000 Imaging System (Fujifilm Holdings Corporation, Japan) with SuperSignal™ West Pico (Thermo Fisher Scientific), following the manufacturer’s recommendations. Densitometry was measured by MultiGauge image analyzer software (Fujifilm).
The antibodies used for global AGEs, pentosidine and CML, recognized any proteins which contain these regions in its structure, so the three bands observed in Western blot analysis (named AGEs/Pen/CML 1, 2 and 3) corresponding with three different molecular weights matched with three groups of AGE-modified proteins.
Reverse transcriptase PCR
Total mRNA was extracted from IPF and control lung samples using TRIzol® reagent procedure (Invitrogen, UK) following the manufacturer’s instructions. Samples were quantified, and Reverse transcription reaction was done with 1 μg by iScript™ cDNA Synthesis Kit (Bio-Rad) in a thermal cycler (Bio-Rad). PCR was performed with DNA polymerase (Biotools B&M Labs, Spain) following the manufacturer’s advice; cDNA was mixed with RAGE primers (164 bp): sense 5′-CAGGACCAGGGAACCTACAG-3′ and antisense 5′-CATGTGTTGGGGGCTATCTT-3′; and β-actin primers (302 bp) were used as a housekeeping gene: sense 5′-GCACTCTTCCAGCCTTCCTTCC-3′ and antisense 5′-TGCTTGCTGATCCACATCTGCT-3′ (Sigma-Aldrich). Amplified samples ran in 2 % agarose gel electrophoresis with ethidium bromide. Then the gel was revealed within an UV chamber and images were developed in instant film to be digitalized afterwards.
Immunohistochemistry
Control and IPF tissue samples were fixed using a 4 % formaldehyde solution in phosphate buffered saline (PBS) and embedded in paraffin. Then the samples were cut in 4 μm sections for the immunohistochemical procedures. Slides were deparaffinized and rehydrated, and then endogenous peroxidases were blocked by incubating samples with 3 % H2O2 for 10 min. AGE-BSA immunostaining was performed with polyclonal rabbit anti-AGEs antibody (Abcam), following the manufacturer’s instructions of Vectastain™ Elite Avidin-Biotin Complex Kit (ABC Kit) (Vector Laboratories, USA). Briefly, antigen retrieval was performed with boiling Tris-EDTA buffer pH 9. Subsequently, slides were blocked in an incubation solution (PBS, 0.2 % Triton X-100 and 0.2 % bovine gelatin) with 20 % of normal goat serum for 1 h at RT. Then, samples were incubated with primary antibody diluted 1:5000 in the incubation solution overnight at 4 °C. After that, slides were incubated in goat anti-rabbit IgG antibody 1:400 from Vectastain™ Kit during 1 h at RT. Later, slides were incubated with ABC Kit for 1 h at RT. Slides were washed with PBS 3 times for 5 min each, between incubations. A brown color was revealed with 3,3′-Diaminobenzidine tetrahydrochloride hydrate (Sigma-Aldrich) in PBS, and sections were counterstained with Harris’ haematoxylin (Casa Álvarez S.A., Spain) and cover-slipped with DPX (Merck Millipore, Germany).
RAGE immunostaining was made with mouse monoclonal anti-RAGE antibody (Abcam), following manufacturer’s instructions. In this case, antigen retrieval was made with boiling citrate buffer pH 6 (Dako). The antibody was diluted 1:200 in the incubation solution with 1 % of normal goat serum overnight at 4 °C. After that, slides were incubated in goat anti-mouse IgG2a HRP-conjugated antibody 1:200 (NB7516; Novus Biologicals, USA) diluted in the incubation solution with 1 % of normal goat serum. Revealing and counterstaining were performed following the same protocol explained above. A negative control was made by incubating control slides without primary antibody in all the immunohistochemical assays. The strong brown staining was considered as a positive signal. Images were evaluated by two expert pathologists blinded to the sample.
Cell viability assay
In order to evaluate the AGEs effect in cell viability, primary fibroblasts from IPF lungs, A549 cell line (ATCC, Manassas VA, USA), and human airway epithelial (HAE) cell line (CRL-4011™, ATCC) were treated with AGE-BSA.
AGE-BSA was prepared following the protocol of Khan et al. [
39]. Briefly, 50 mg/mL BSA (Sigma-Aldrich) was incubated with 1 M D-Ribose (Sigma-Aldrich) in PBS pH 7.4 for 20 days at 37 °C. In addition, BSA was incubated without D-Ribose as a control. After that, the solutions were dialyzed and filtered through 0.22 μm membranes (Merck Millipore) to remove D-Ribose debris; and then the presence of AGEs were checked and quantified by fluorescence (wavelength emission 440 nm/excitation 370 nm) and BCA Protein Assay Kit.
Fibrotic fibroblasts were isolated from IPF patients that underwent surgical lung biopsy and grew in DMEM (Gibco™, Thermo Fisher Scientific, USA) with 10 % FBS. A549 was cultured in F12K medium (Lonza, Switzerland) with 10 % FBS. HAE cell line was cultured in Bronchial Epithelial Cell Growth Medium (BEGM, Lonza).
Quick Cell Proliferation Colorimetric Assay Kit (MBL international, USA) was performed in 96-well plates following the manufacturers’ recommendations. Cells were seeded at 1 × 104 cells/well and were incubated in medium with 2 % FBS for 24 h before the experiment. Then cells were treated with different concentrations of AGE-BSA for 3 h. A well without cells was assessed for all conditions as reference value. Reagent was added to each well for an additional 2 h and the plate was shaken and read in the Thermo Scientific Multiskan® EX (Thermo Fisher Scientific) at 450 nm, and absorbance was adjusted to the measurement in the reference value for each condition.
3D cell culture into glycated matrix
Three-dimensional culture based on glycated collagen matrix was performed to evaluate the behavior of fibrotic fibroblast cells under the influence of ECM cross-links and the consequent AGEs production. The 3D collagen matrices were produced using native type I collagen from bovine dermis at 4 mg/mL (Cosmo Bio Co., Japan), and were glycated adding ribose at 5 and 15 mM (Sigma-Aldrich), following a previously standardized protocol [
19].
Fibroblasts were added and mixed with the collagen before polymerization of the gel. Cells were seeded into 96-well plates at 15 × 103 cells/well. After matrix polymerization, a culture medium with a different concentration of ribose was added. Phenotype changes were evaluated by Western blot detection of alpha-smooth muscle actin (α-SMA) 1/500 (A5228; Sigma-Aldrich) on days 1, 7 and 14, following the procedure described above.
Statistical analysis
Data of experimental groups were compared and analyzed with IBM SPSS Statistics 23 (IBM, USA). Differences between the two groups were analyzed by Student’s t-test or Mann–Whitney U-test when comparing two parametric or non-parametric samples, respectively. To evaluate the differences among experimental groups, one-way ANOVA was assessed. Results are expressed as means ± SD. The p-value < 0.05 (*) or < 0.01 (**) were considered statistically significant.
Discussion
In the last few years, IPF has been proposed as a result of an accelerated aging process of the lung [
5]. Some findings support ECM as a target of oxidative stress in the lung and the subsequently produced AGEs, which are biomarkers of an in vivo aging process that may promote fibrogenesis [
9,
41]. In this line, our results demonstrate that AGEs are increased in IPF samples at the same time that RAGEs are decreased. Specifically, two well-known AGEs, CML and pentosidine-modified proteins, which have been related to aging, tissue stiffness and AECs apoptosis [
11,
14,
18] are significantly overexpressed in IPF samples. Some pathogenic features of IPF may be implicated in the AGEs formation; collagen deposition, aging, oxidative compounds derived from smoking, dust or diet, and decreases of soluble RAGE are processes that could enhance the accumulation of AGEs in lung fibrosis [
15].
At the same time, the higher amount of AGEs formation could influence the perpetuation of the fibrogenic process. In vitro studies have shown that AGEs could induce cell toxicity and death [
39], delaying wound healing in epithelial cells [
42]; whereas enhancing collagen and transforming growth factor (TGF)-β1 synthesis [
43]. Additionally, AGEs induce epithelial-mesenchymal transition (EMT) in epithelial cells from rat kidney [
44], although no EMT effect has been found in type II AECs from the rat lung [
45]. Actually, some studies have even suggested that blocking the AGEs might attenuate pulmonary fibrosis [
46]. However, there is scarce information about the effect of AGEs in human AECs and fibroblasts from lung parenchyma. In the present study, we demonstrated, as an initial approach, that alveolar type II epithelial cell line viability (A549) and HAE cell line were more sensitive to the presence of AGEs, while fibrotic fibroblasts had a low response to high dosage of AGEs; suggesting a different effect of AGEs, depending on cell type. Although the A549 cell lines may be suitable for initial exploratory IPF studies, data derived from such cells must acknowledge the limitations associated with these tools [
47].
In addition, immunohistochemical results showed increased AGEs associated with the hyperplasic AECs, which suggest that AGEs could induce pro-fibrotic effects such as loss of epithelium. On the other hand, our results indicate the presence of myofibroblast transformation from fibrotic fibroblasts, in part because of the presence of AGEs at the surrounding ECM. Previous work of our group has demonstrated that 3D collagen matrices generated under glycation present an increase of AGEs, which is time-dependent and appears 5 days after glycation begins [
19]. The AGEs formation is not only dependent on ribose concentration, but also on the presence of serum and glucose in the media, which would explain the fact that fibroblast-myofibroblast transformation appears in glycated 3D matrices, with and without ribose addition. However, a potential limitation for the interpretation of this observation is that other non-controlled collateral collagen cross-link reactions described at the glycated process of this in vitro model and not associated with AGEs formation could also influence in the myofibroblast transformation.
On the other hand, since a higher presence of RAGE has been found in healthy pulmonary tissue, in comparison to other organs, there are many studies that support its outstanding role in pulmonary homeostasis [
25]. Our results demonstrated that the protein expression of RAGE (FL-RAGE and sRAGE) was decreased in all fibrotic lungs compared with control lungs. According to this observation, Quiesser et al. suggested an important function of RAGE in cellular adhesion and spreading to the basal membrane, playing a structural role in the maintenance of the alveolar epithelium [
32]. It has also been demonstrated that RAGE stimulates elastin expression and plays a supporting role in respiratory mechanics [
48]. Furthermore, RAGE may belong to a family of cell adhesion molecules [
49], making links with basal lamina components such as type IV collagen or laminin [
28,
50]. Thus, hypothetically, this RAGE down-regulation noticed in IPF lung samples could reduce the interaction of AECs to ECM and facilitate the basal membrane disruption, resulting in the occupation of alveolar spaces. Additionally, our results showed that the soluble isoform, sRAGE, is the greatest RAGE variant in the control lungs, with a sRAGE/FL-RAGE ratio of 1.67 in agreement with other studies performed in human lungs [
51]. Interestingly, it has been described that only 7 % of RAGE transcript encodes for the main alternative splicing soluble variant [
29], so a greater part of sRAGE might come from the cleavage of FL-RAGE by metalloproteinases [
52], which are overexpressed in lung tissue and BAL fluids from IPF patients [
53]. Consequently, it could cause the loss of joins between the type I AECs and the basement membrane, avoiding the normal re-epithelialization in response to damage [
54]. In support of this, our results showed a sRAGE/FL-RAGE ratio in IPF lungs of 4.17, 2.5 fold increased in respect to the control lungs, in the possible context of the MMPs’ raising in the fibrotic process.
Regarding our immunohistochemical analysis of RAGEs, a predominance location around the cell membrane of AECs was observed in control lung tissues, as was previously described [
55]. In contrast, RAGE was totally absent in the hyperplasic AECs and fibroblast foci from IPF lungs. Although this decrease of RAGE in fibrotic lung samples could be explained by the loss of AECs, some studies suggest that RAGE might be decreased from the beginning of AECs damage [
33,
56]. Likewise, in our study we have observed that the loss of RAGE was not restricted only to protein expression, but also to the genetic expression. Hence, RAGE reduction is not only associated with Type I AECs’ disappearance as a consequence of tissue remodeling, but also the areas of preserved parenchyma show a decreased RAGE expression.
Although solid evidence demonstrates a complex function of RAGE in the lung, the implication grade and the role in lung physiology is still under discussion. Furthermore, the decrease of RAGE expression in IPF, especially sRAGE, suggests a potential utility to be tested in the future as biomarker [
57] and therapeutic target [
31,
57‐
59].
Acknowledgment
The authors want to thank Dr. M. Muñoz-Esquerre from University Hospital of Bellvitge for helping in control patient recruitment.
Supported by: ISCIII (PS12/02455), SEPAR, SOCAP and FUCAP.