Anti-Inflammatory and Anti-Fibrotic Effect of Immortalized Mesenchymal-Stem-Cell-Derived Conditioned Medium on Human Lung Myofibroblasts and Epithelial Cells

Idiopathic pulmonary fibrosis (IPF) is caused by progressive lung tissue impairment due to extended chronic fibrosis, and it has no known effective treatment. The use of conditioned media (CM) from an immortalized human adipose mesenchymal stem cell line could be a promising therapeutic strategy, as it can reduce both fibrotic and inflammatory responses. We aimed to investigate the anti-inflammatory and anti-fibrotic effect of CM on human pulmonary subepithelial myofibroblasts (hPSM) and on A549 pulmonary epithelial cells, treated with pro-inflammatory or pro-fibrotic mediators. CM inhibited the proinflammatory cytokine-induced mRNA and protein production of various chemokines in both hPSMs and A549 cells. It also downregulated the mRNA expression of IL-1α, but upregulated IL-1β and IL-6 mRNA production in both cell types. CM downregulated the pro-fibrotic-induced mRNA expression of collagen Type III and the migration rate of hPSMs, but upregulated fibronectin mRNA production and the total protein collagen secretion. CM’s direct effect on the chemotaxis and cell recruitment of immune-associated cells, and its indirect effect on fibrosis through the significant decrease in the migration capacity of hPSMs, makes it a plausible candidate for further development towards a therapeutic treatment for IPF.


Introduction
Idiopathic pulmonary fibrosis (IPF) is the most common interstitial lung disease (ILD), affecting 3 million people worldwide [1]. It is highly heterogenous among patients and can manifest in various ways, often leading to misdiagnoses and delayed treatment [2,3]. IPF is caused by progressive lung tissue impairment due to extended chronic fibrosis of unknown etiology, and ultimately leads to death [4]. Current in silico approaches may provide new information regarding the development of IPF, for instance, establishing early disease diagnosis, disease severity and assessing treatment efficacy [5][6][7][8].
Unfortunately, our current knowledge is unable to provide a cure for IPF, and even symptomatic treatment can be difficult [9], especially with the acute exacerbation of IPF [10,11]. Recent developments in anti-fibrotic agents, such as Nintedanib and Pirfenidone, were shown to slow the progression of IPF when used as monotherapies or Figure 1. The effect of CM on chemokine mRNA expression in Α549 pulmonary epithelial cells, which are under pro-inflammatory stimuli. CM inhibit the mRNA expression of chemokines CCL2 (A), CCL20 (B), CXCL10 (E) and CXCL11 (F) in a dose-dependent way, but have no effect on the mRNA levels of CXCL1 (C) and CXCL8 (D). 2C: IL-1α 5 ng/mL + TNF-α 50 ng/mL; 830 μg/mL total protein of CM; 83 μg/mL total protein of CM. Results are presented as median with interquartile range. **** p < 0.0001; *** p < 0.001; * p < 0.05. N = 3.
Taken together, our results indicate that the CM has an indirect anti-inflammatory effect through the downregulation of several chemokines in A549 epithelial cells and hPSMs, ultimately leading to a possible reduced chemotaxis of immune-associated cells. and CXCL11 (F) in a dose-dependent way, but have no effect on the mRNA levels of CXCL1 (C). 2C: IL-1α 5 ng/mL + TNF-α 50 ng/mL; 830 µg/mL total protein of CM; 83 µg/mL total protein of CM. Results are presented as median with interquartile range. **** p < 0.0001; *** p < 0.001; ** p < 0.01; * p < 0.05. N = 3.

The Conditioned Medium Regulates the IL-1α-and TNF-α-Inducible mRNA Expression of Several Pro-Inflammatory Interleukins in the Pulmonary Epithelial Cell Line A549 and in hPSMs
Since we found that the CM could downregulate the expression of several chemokines implicated in the chemotaxis of immune-associated cells, we next examined its effect on the IL-1α-and TNF-α-inducible mRNA expression of several pro-inflammatory interleukins in the pulmonary epithelial cell line A549 and in hPSMs. Among the several interleukins, we chose to study the IL-1α, IL-1β, IL-4, IL-6, IL-10, IL-13, IL-17, IL-22 and TNFα, as they play major roles in the Th1, Th2 and Th17 immune responses [63].

The Conditioned Medium Regulates the IL-1α-and TNF-α-Inducible mRNA Expression of Several Pro-Inflammatory Interleukins in the Pulmonary Epithelial Cell Line A549 and in hPSMs
Since we found that the CM could downregulate the expression of several chemokines implicated in the chemotaxis of immune-associated cells, we next examined its effect on the IL-1α-and TNF-α-inducible mRNA expression of several pro-inflammatory interleukins in the pulmonary epithelial cell line A549 and in hPSMs. Among the several interleukins, we chose to study the IL-1α, IL-1β, IL-4, IL-6, IL-10, IL-13, IL-17, IL-22 and TNF-α, as they play major roles in the Th1, Th2 and Th17 immune responses [63].
Altogether, our results indicate that the CM may downregulate the expression of the pro-inflammatory IL-1α, but they could also further induce a pro-inflammatory response in both the A549 epithelial cells and the hPSMs.

The Effect of the Conditioned Medium on the TGF-β-Induced Fibrotic Responses of hPSMs
Having confirmed that the CM have an anti-inflammatory effect on hPSMs, we next investigated its effect on the TGF-β-induced fibrotic responses of hPSMs, and, in particular, the mRNA expression of collagen Type I, Type III, and fibronectin and the protein expression of the total secreted collagen.

Discussion
In this study, we showed that the CM derived from an immortalized human-adiposederived mesenchymal stem cell line have a clear anti-inflammatory effect as they inhibited the mRNA and protein expression of chemokines in A549 cells and hPSMs in a dosedependent way. They also downregulated the mRNA levels of IL-1α in both of these cell types. We also observed that the CM had a mild and indirect anti-fibrotic effect on hPSMs, as they could downregulate the mRNA expression levels of collagen Type III and could decrease the TGF-β-inducible migration rate. These results suggest that the CM could have a direct effect on chemotaxis and cell recruitment, including leukocytes and myofi-

Discussion
In this study, we showed that the CM derived from an immortalized human-adiposederived mesenchymal stem cell line have a clear anti-inflammatory effect as they inhibited the mRNA and protein expression of chemokines in A549 cells and hPSMs in a dosedependent way. They also downregulated the mRNA levels of IL-1α in both of these cell types. We also observed that the CM had a mild and indirect anti-fibrotic effect on hPSMs, as they could downregulate the mRNA expression levels of collagen Type III and could decrease the TGF-β-inducible migration rate. These results suggest that the CM could have a direct effect on chemotaxis and cell recruitment, including leukocytes and myofibroblasts, and indirectly, an anti-inflammatory effect. This chemotactic and cell recruitment effect, in a second step, could affect fibrosis, given that cell recruitment, apart from inflammation, is also involved in the fibrotic process.
More specifically, a higher CM dose downregulated the IL-1α-and TNF-α-inducible mRNA expression of several chemokines in both A549 cells and hPSMs, and this mRNA downregulation was also translated to extremely decreased protein levels. We chose to investigate the effect of CM on the IL-1α-and TNF-α-inducible CCL and CXCL expression, because both of these two chemokine families exerted their pro-inflammatory actions through the recruitment of immune-associated cells, and it was also shown that CXCL chemokines played a role in the migration of fibroblasts and endothelial cells [64]. Regarding IPF, various chemokines were found to be increased or were associated with disease exacerbation. In particular, the CCL2 chemokine was found to be increased in bronchoalveolar lavage fluid (BALF) [65] and activated myofibroblasts of patients with IPF [66], suggesting that this chemokine has a major role in both the recruitment of immune-associated cells and in the process of fibrogenesis. In addition, CCL2, together with CXCL10, have been characterized as prognostic biomarkers for the disease progression and the survival outcome of patients with IPF [67], and in a murine model of chronic obstructive pulmonary disease the chemokines CCL2, CXCL1, CXCL8, CXCL10 and CXCL11 were found to be elevated and correlated with the establishment of fibrosis [68]. All of this evidence suggests that chemokines may play a double role in IPF, recruiting immune-associated cells and fibroblasts, and thus prolonging inflammation and promoting fibrogenesis.
Our results agree with other studies showing that mesenchymal stem cells play a positive role in regulating immune responses and attenuating pulmonary fibrosis. Cargnomi et al. showed that human amniotic MSCs could promote the polarization of T regulatory cells, and anti-inflammatory M2 macrophages and could also decrease the recruitment and maturation of B immune cells in a murine model of pulmonary fibrosis [69]. In another study of radiation-induced lung injuries, which utilized MSCs modified with decorin, a natural inhibitor of TGF-β, treatment with these modified MSCs resulted in reduced lymphocyte infiltration, decreased expression levels of pro-inflammatory chemokines and cytokines, and inhibition of fibrogenesis, but also increased the expression of antiinflammatory cytokines [70]. Recently, we also showed that the CM could ameliorate pulmonary fibrosis and inflammation in an in vivo model, through the reduced expression of pro-inflammatory cytokines and the decreased recruitment of immune-associated cells [48].
Regarding the CM effect on the combined IL-1α-and TNF-α-inducible mRNA expression of interleukins, we observed that the CM downregulated the IL-1α, but upregulated the IL-1β and IL-6. In previous studies, IL-1α was found to be elevated in a bleomycininduced fibrosis model [71], and autoantibodies to this cytokine were also found in the serum of patients with rapidly progressive IPF [72], suggesting that IL-1α could play a significant role in the progression of IPF, and thus treatment with CM could contribute to its downregulation. Regarding IL-1β, other studies using MSCs as treatment for pulmonary inflammation and fibrosis showed a decrease in its expression [72,73], but this difference in our results could be attributed to the fact that we used CM and not the MSCs themselves. As mentioned above, we observed an upregulation of the IL-1α-and TNF-α-inducible mRNA expression of IL-6 in both A549 cells and hPSMs when treated with CM. Various other studies and our study showed that MSCs or the CM from MSCs could ameliorate fibrosis and inflammation in vivo, and in particular, they could downregulate the expression of IL-6 [37,42,48,[74][75][76]. Again, this disagreement with our results could be attributed to the fact that this study is an in vitro study, examining the CM effect on simple cell cultures and not on the possible different cellular interactions that can be seen in in vivo studies.
As far as the anti-fibrotic effect of the CM is concerned, we observed that the CM had a mild and indirect effect on fibrosis, as it was statistically significant in downregulating the TGF-β-inducible mRNA levels of collagen Type III and the migration rate of hPSMs, but it also upregulated the mRNA levels of fibronectin and the total protein collagen production. Previous studies from other researchers and our laboratory have shown that the CM indeed have an anti-fibrotic effect on pulmonary fibrosis. Li et al. showed that the treatment with bone marrow mesenchymal stem cell (BMSC)-derived CM in rats with silica-induced pulmonary fibrosis resulted in a reduced overall collagen deposition, along with decreased mRNA levels of collagen I, collagen III, and fibronectin, TGF-β1 and hydroxyproline [77]. Utilizing the same animal model, Zhao et al. reported that the administration of BMSCs led to a decreased collagen deposition and hydroxyproline content, and it also downregulated the mRNA levels of collagen Type I and fibronectin [78]. Apart from the in vivo data, BMSCderived CM were found to inhibit the in vitro epithelial-to-mesenchymal transition. Wang et al. observed that the A549 epithelial cells treated with CM had reduced levels of the mesenchymal-associated markers α-SMA and vimentin and increased levels of the epithelial markers, E-cad and CK8 [79]. Similar results were reported utilizing other types of MSCs. We showed that ADSC-CM ameliorated pulmonary fibrosis in Bleomycin-treated mice, as it reduced the collagen deposition in their lungs [48]. Felix et al. showed that the administration of both ADSCs and their CM in Bleomycin-treated rats resulted in a significant overall improvement, with reduced levels of TGF-β and collagen I fibers [43]. All of these studies lead to the conclusion that MSCs and their CM indeed have an anti-fibrotic effect on pulmonary fibrosis, and although in this study we found increased protein levels of the total collagen production, the migration rate of hPSMs was reduced in a statistically significant manner. Therefore, one could speculate that the CM could exert their anti-fibrotic actions by decreasing the migratory capacity of hPSMs, and if the most major fibrogenic cell, the myofibroblast, could not migrate, fibrogenesis could become stalled.
Possible failure in discovering an all-encompassing single compound pharmacological drug for the successful treatment of IPF may be due to the limitation of a single mechanism of action in what seems to be quite a multimodal disease mechanism. Regenerative biologicals, such as CM, because of their multifactorial constituents, have multiple mechanism of action, as depicted in this study, with both an anti-inflammatory and anti-fibrotic effect that may be plausible additions to the treatment regime for such diseases.

Adipose-Derived Mesenchymal Stem Cell Isolation and Immortalization
Adipose-derived mesenchymal stem cells were isolated and immortalized, as previously described [48]. Briefly, subcutaneous adipose tissue from a 35-year-old healthy female, who was informed and agreed to participate in this study by giving her written consent, was obtained during outpatient tumescence liposuction under local anesthesia. All principles outlined in the Declaration of Helsinki for all human experimental investigations were followed. Next, the adipose-derived stem cells (ADSCs) were immortalized as previously described [48], by transducing with the human telomerase reverse transcriptase (hTERT) gene in combination with lentiviral gene SV-40. ADSCs' phenotype was verified by confirming the expression of CD90 and CD105 and the lack of expression of Cd34 under a fluorescent microscope (Leica DM2000, Leica Microsystems GmbH, Wetzlar, Germany), as shown in Supplementary Figure S1A.

Characterization of hPSMs and ADSCs
The hPSMs and ADSCs were characterized by immunofluorescence as previously described [80]. Briefly, cells were first fixed in 4% paraformaldehyde (PFA; Sigma-Aldrich

Conditioned Medium Preparation
Conditioned media (CM) from ADSCs were collected and processed as previously described [48]. A total of 1 × 10 5 cells were plated in T-75 flasks in medium containing DMEM (PeproTech EC, Ltd., London, UK), 10% FBS (Sigma-Aldrich, St. Louis, MO, USA) and gentamycin 1% (Biosera, Nuaillé, France), at 37 • C and 5% CO 2 . Upon the cells becoming confluent, the media were changed to the same media, excluding FBS. Supernatants were collected every 72 h and fresh FBS-free medium was added. Cell supernatants were harvested and centrifuged at 2500 rpm for 10 min at 4 • C and filtered through a 0.45-micron filter (Merck-Millipore, Burlington, MA, USA) to remove cell debris and cryopreserved at −80 • C until use. Total protein of CM was calculated to be 830 µg/mL using Quawell Q5000 UV-Vis Spectometer (Quawell, San Jose, CA, USA).

Patients
Healthy pulmonary tissue, with no histopathological evidence of disease, was obtained from patients that underwent lobectomy for primary pulmonary tumor removal. The local Research Ethics Committee of the University Hospital of Alexandroupolis approved this study, and all patients gave their informed written consent prior to participation.

A549 Pulmonary Epithelial Cell Line
A549 is a cell line of alveolar basal epithelial cells that were isolated from a pulmonary adenocarcinoma of a 58-year-old Caucasian male [82]. A549 cells were cultured in DMEM (PeproTech EC, Ltd., London, UK) supplemented with FBS (Sigma-Aldrich, St. Louis, MO, USA) and antibiotics.

Stimulation of A549 and hPSMs with Recombinant Cytokines and Conditioned Medium
hPSMs at passages 2-5 were used in our studies. All experiments were performed with FBS-free media at 95% culture confluence with a stable ratio of supernatant volumeto-surface available for cell adhesion (1.5 mL:9.6 cm 2 ).

Total RNA Extraction and DNase Treatment
Total RNA from hPSMs and A549 cells were extracted using Nucleozol (MACHEREY-NAGEL, Düren, Germany) according to the manufacturer's instructions and as previously described [48]. In short, H 2 O was added to each sample, following incubation and a 15 min centrifugation at 12.000 g. Supernatants were then incubated with isopropanol and centrifuged for 10 min at 12.000 g in order to precipitate the RNA. The RNA pellet was washed twice with 75% ethanol and reconstituted with H 2 O. Total RNA concentration was measured using Quawell Q5000 UV-Vis Spectometer (Quawell, San Jose, CA, USA). Any DNA contaminations were removed using Recombinant DNase I (TaKaRa, Kusatsu, Shiga, Japan).

cDNA Synthesis and Real-Time PCR
Next, 200 ng of the DNAse treated RNA was reverse transcripted using the Prime-Script 1st strand cDNA Synthesis Kit (TaKaRa, Kusatsu, Shiga, Japan) according to the manufacturer's instructions. Following cDNA synthesis, 25 ng cDNA was then amplified using qRT-PCR Sybr Green (Kapa Biosystems, Wilmington, NC, USA) and 100 nM of genespecific primers, shown in Table 1, in SaCycler-96 RUO (Sacace Biotechnologies, Como, Italy). A two-step amplification protocol was performed for all studied genes and the gene expression of each studied gene was normalized against GAPDH gene expression in the same sample using the 2-∆∆Ct method. Human DuoSet ® ELISAs (R&D Systems, Minneapolis, MN, USA) were used to estimate the protein concentrations of CCL2, CCL10, CXCL1, CXCL8, CXCL10 and CXCL11 chemokines in hPSMs and A549 cells' supernatants, according to the manufacturer's in-structions and as previously described [61]. Briefly, flat 96-well plates were coated overnight with capture antibody for each chemokine, and the following day, plates were incubated with the recommended blocking buffer for 2 h. Next, duplicates of each supernatant and known concentrations of chemokine samples were added in wells, incubated for 2 h, and then, biotinylated detection antibody for each chemokine was added for another 2 h. Streptavidin-horseradish peroxidase was then added for 20 min and the following addition of tetramethylbenzidine with H 2 O 2 produced different optical densities (OD) of color which were measured at 450 nm on a microplate reader (Diareader EL×800; Dialab, Wr. Neudorf, Austria). The chemokines concentration was calculated using a linear standard curve according to the manufacturer's instructions.

Measurement of Total Protein Collagen Production
Collagen was measured with the Sircol assay (Sircol; Biocolor, Carrickfergus, UK), as previously described [80]. Briefly, ice-cold Isolation & Concentration Reagent was added to hPSMs' supernatants, and samples were incubated overnight at 4 • C. The next day, samples were centrifuged at 13,000× g for 10 min, and Sirius Red Dye was then added. After a 30 min incubation, samples were centrifuged at 13,000× g for 10 min, supernatant was discarded, and the collagen pellet was dissolved in 0.5 M NaOH alkali reagent. Next, the optical densities (ODs) of the samples and controls of known collagen concentration were measured at 540 nm in a microplate reader (Diareader ELx800; Dialab, Wr. Neudorf, Austria). The collagen concentration was calculated using a linear standard curve according to the manufacturer's instructions.

Wound Healing Assay
The effect of CM on the TGF-β-induced migration of hPSMs was assessed in vitro with the wound-healing scratch assay, as previously described [80]. A narrow gap devoid of cells was created on confluent hPSMs cultures, and next, we measured the rate of gap closure. This process resembles wound healing and is dependent on myofibroblast migration, hyperplasia, and proliferation. hPSMs were cultured in 6-well plates, and when confluence was reached, 1 mechanical wound per well was performed with a 10-µL pipette tip. The wound was vertical to predrawn lines on the bottom of the well so that we could define 3 stable points per well at the junctions of those lines with the wound. Images were recorded at those fixed wound points with an inverted Olympus microscope equipped with a camera. hPSMs were treated with various cytokines (as previously described) for 24 h, and images were taken at time points 0 and 24 h. The average percentage of gap closure after 24 h in treated wells was divided by the average percentage of gap closure in untreated wells to assess the treatment effect.

Statistics
Statistical analyses were performed using Prism Software 9 (GraphPad Software, San Diego, CA, USA). Results are presented as median with interquartile range (IQR). Comparison of values among sample groups was performed with ordinary one-way ANOVA. Statistical significance was set at p < 0.05.

Conclusions
In conclusion, we showed that ADSC-derived CM could directly affect the chemotaxis and the cell recruitment of immune-associated cells, as they strongly reduced the expression of various chemokines, and in a second phase, they could also indirectly affect fibrogenesis, as the migration capacity of hPSMs was decreased, and cell recruitment was implicated in the fibrosis process. More studies are needed to further elucidate the anti-inflammatory and anti-fibrotic mechanisms of the CM and to possibly identify the responsible soluble mediators, through which the CM exerts its therapeutic actions.