The present study was approved by the Institutional Ethics Committee of the IRCCS Ospedale Policlinico San Martino, Genoa, Italy (protocol code ILDFIBRO020, n. of Register CER Liguria: 523/2020 DB id 10931) and conducted according to the current national and international guidelines; within the study, biological samples were anonymized prior to processing. For patients the diagnosis was based on international criteria [12, 13]. After multidisciplinary discussion of the Interstitiopathy Lung Disease (ILD) group of the case, and after signing an informed consent form, explicitly authorizing the use of biological samples and any derivative thereof for research purposes, the patient underwent a transbronchial cryobiopsy with bronchoalveolar collection for diagnostic purposes [14]. Aliquots of BAL fluid unused for diagnostic purposes (15–25 mL) were therefore processed for additional research procedures. The clinical features of the patients employed in this study are summarized in Supplementary Tables 1,2. The total lung capacity (TLC), forced vital capacity (FVC), forced expiratory volume in 1s (FEV1), and diffusing capacity (DL_C⁢O) were measured according to international guidelines [12, 13].

Primary derived bronchial fibroblasts were derived from fresh ILD- bronchoalveolar lavage, 15 mL which is the small fraction normally left unused for routine diagnostic analyses. Cells of the BAL fluids were first pelleted by centrifugation. While supernatant was recovered and frozen for different types of investigations, cells were resuspended in 6 mL of complete medium (RPMI-1640 + 10% fetal calf serum (FCS); Lonza, Walkersville, MD, USA), seeded in a 24 well plate and cultured at 37 °C and 5% CO₂. Fibroblast growth factor 2 (FGF2; 3 ng/mL; Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) was added 1 day after. The medium was changed once a week with the addition of FGF2. In some cases, after 2–3 weeks, a good number of spindle-shaped cells attached to the bottom of the well could be observed. These cells were therefore cultured until confluence, detached by the use of 1$\times$ trypsin-EDTA (Euroclone S.p.A., Pero, Milan, Italy), washed, collected and expanded in 24 or 6-well plates. As quickly as possible a small aliquot of cultured fibroblasts was used for RNA extraction to preliminarily assess the expression of specific fibroblasts/myofibroblasts markers (i.e., $\alpha$-smooth muscle actin ($\alpha$-SMA), type I collagen (collagen I), fibronectin) by quantitative RT-PCR analysis. In some cases fibroblasts could be further expanded until passages 10–12, thus allowing a deeper phenotypical/functional characterization, as well as the storage of many vials in liquid nitrogen. In one particular case we could expand the fibroblasts of one patient for a very long time (9 months) until passage 36, as we previously described (SCI13D) [2]. In addition, from the same patient, but out of a second BAL sample collected one year later, we also derived new additional fibroblasts. These however differed from the previous ones, missing the chromosome 10 trisomy that characterized the earlier fibroblasts. These cells were called SCI2 and could be expanded until passage 14.

The MRC5 cell line (from Interlab Cell Line Collection “ICLC” IRCCS Ospedale Policlinico San Martino, Genoa, Italy), fibroblasts isolated from the lung tissue derived from a white male, 14-week-old embryo, was cultured in flasks in complete medium (RPMI+FCS10%) (Euroclone S.p.A). Cells reaching confluence were detached by trypsin 1X (Euroclone S.p.A), split or either frozen. Primary fibroblasts were further derived from the BAL of a patients with no evidence of pulmonary fibrosis while skin fibroblasts derived from a healthy subject were used as a further control.

We also utilized the A549 cell line (obtained from ICLC, IRCCS Ospedale Policlinico San Martino, Genoa, Italy) that, although derived from lung adenocarcinoma, is experimentally utilized as representative of alveolar basal epithelial cells [15]. We also used the human umbilical vein endothelial cell line HECV (from ICLC, IRCCS Ospedale Policlinico San Martino, Genoa, Italy). Both these cell lines were cultured as monolayer in flasks in RPMI+FCS 10%, detached when they reached 70% confluency by 1$\times$ trypsin -EDTA (Euroclone S.p.A.), and then re-cultured, frozen or employed in 3D models of cultures. All cell lines were validated by Short Tandem Repeat (STR) profiling and all the cells used were tested negative for mycoplasma.

Briefly, cells were washed in a PBS solution (Euroclone S.p.A.), trypsinized and collected by centrifugation (400 $\times$g) and processed for mRNA extraction, according to the manufacturer’s instruction of the GeneUP™ Total RNA Kit (BiotechRabbit GmbH, Berlin, Germany). A cDNA pool for each sample was then generated by using the SuperScriptTM III First-strand synthesis system for RT-PCR Kit (Invitrogen; Milan, Italy); a sybr-green RealMasterMix SYBR ROX 2,5X (5-Prime GmbH, Hamburg, Germany) was used in an Eppendorf Mastecycler Realplex2 apparatus, applying a real time quantitative RT-PCR analysis. Each reaction was performed in triplicate for each sample, according to the following settings: a single denaturation step at 95 °C for 3 minutes; 45 cycles at 94 °C for 30 sec, 60 °C for 30 sec, 72 °C for 40 sec, and a final step at 72 °C for 7 min. Reaction specificity was counterchecked by the melting curve analysis. The expression of each target gene was normalized to the endogenous housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Primer sets for target genes were purposely designed, as for the alpha Smooth Muscle Actin ($\alpha$-SMA; forward primer: TGGAAAAGATCTGGCACCAC, reverse primer: CTCAAACATAATTTGAGTCAT), Fibronectin (FN; forward primer: TACACTGGGAACACTTACCG; reverse primer: CCAATCTTGTAGGACTGACC), fibroblast growth factor-2 receptor (FGF-2R; forward primer: AGACAGGTAACAGTTTCGGCT, reverse primer: CAGTGTCAGCTTATCTCTTGG), or derived from previously published sequences. Expression levels of target genes in the different samples/culture conditions were normalized by the levels of the corresponding gene in control samples/cultures.

Morphological features of live cultured cells were examined under an inverted Olympus CKX-41 microscope (Olympus Corp., Tokyo, Japan); images were acquired using a Nikon Digital Sight DS-5Mc equipped with the NIS-Elements F2.20 software (Nikon Corp., Tokyo, Japan). Expanded fibroblasts were also analyzed for the expression of CD105, CD90, CD73, CD45, Vimentin, type I Collagen, $\alpha$-SMA, fibronectin by cytofluorographic analysis. Phenotypical analysis was performed by staining detached cells with specific antibodies for membrane antigens such as anti-human-CD105-PE (Cat.n. 21271054), -CD73-FITC (Cat.n. 21270733), -CD90-APC (Cat.n. 21270906) or -CD45-FITC (Cat.n. 21270453) (all from Immuno-Tools GmbH, Friesoythe, Germany). After 30 min of incubation at 4 °C the cells were washed twice and resuspended in phosphate buffered saline (PBS)+ FCS 2%. For intracytoplasmic staining the cells were first fixed with 1% paraformaldehyde (PFA) and incubated in the dark at 4 °C for 15 min. After two washes with PBS+2% FCS the cells were permeabilized with a permeabilization solution (0.1% Na-citrate and 0.1% Triton in PBS 1$\times$) and then stained with the anti-human-$\alpha$-SMA antibody (clone 1A4, Sigma-Aldrich, St.Louis, MI, USA), or -procollagen 1 (SP1.D8; Developmental Studies Hybridoma Bank, University of Iowa; Iowa City, IA, USA) or -fibronectin (DP3060, ACRIS, Li-Starfish, Milan, Italy) and incubated on ice for 30 min. Cells were then washed and, after the addition of 50 µL of permeabilization solution, were stained with a specific anti-mouse or anti-rabbit secondary FITC or PE-conjugated antibody (Southern Biotechnology, Birmingham, AL 35226, USA), and incubated at 4 °C for 30 min. Cytofluorographic analysis was then performed using a FACS-Canto cytofluorimeter (Becton Dickinson; Franklin Lakes, NJ, USA).

To determine the expression of F-actin fibers by immunofluorescence, fibroblasts were cultured in eight-well chamber slides (Nunc International, Rochester, NY, USA) in a total volume of 300 µL of culture medium and then treated with transforming growth factor-$\beta$1 (TGF-$\beta$1; 5 ng/mL; D.B.A. Italia, Segrate, Milan, Italy), or Pirfenidone (300 µg/mL; D.B.A. Italia, Segrate, Milan, Italy), or TGF-$\beta$ + Pirfenidone for 72 h. Cells were fixed in PFA 2.5%, permeabilized with Triton $\times$ 100, and stained with Alexa-Fluor 555-conjugated phalloidin (Thermo Fischer Scientific; Rodano, Milan, Italy). The expression of F-actin was then observed by fluorescence microscopy and images were acquired with a Hamamatsu C5810 color-chilled 3 CCD camera (Hamamatsu Photonics K.K., Hamamatsu City, Shizuoka, Japan). Identical camera settings (time of exposure, brightness, contrast, and sharpness) and an appropriate white balance set according to the fluorescence filter were used. Nuclei were counterstained with DAPI.

Fibroblasts were cultured in 24 well plates, washed with PBS 1$\times$ and then fixed with PFA 4% for 5 min at room temperature. Cells were rinsed with PBS and incubated overnight in the incubator at 37 °C in SA-$\beta$-gal staining solution containing: 1 mg mL^-1 X-gal, 40 mM citric acid/sodium phosphate buffer pH 6.0, 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 150 mM sodium chloride and 2 mM magnesium chloride in water. Cells were then rinsed with PBS and observed with an inverted bright field microscope and at least 5 Images for each well were acquired as described above.

Available aliquots of cells, BAL-patients originated, were expanded in complete medium (RPMI + FCS10%), detached with Trypsin 1$\times$, counted, re-seeded in 24-well plates to a density of 10⁴ cells/mL in complete medium and left to adhere for 24 h. Non-adherent cells were washed away and treatments were performed by exposing cells to complete medium supplemented with TGF–$\beta$1 (5 ng/mL; Miltenyi Biotech), Pirfenidone (150–450 µg/mL; D.B.A. Italia), or both. At the scheduled time point cells were washed with sterile PBS and incubated for 4 h at 37 °C with complete culture medium supplemented with 10% Alamar Blue™ (Invitrogen; Milan, Italy). Aliquots of the supernatants were then drawn and assessed spectrophotometrically, at 570 and 600 nm, in a Spectra MR Dynex apparatus. Cells were then rewashed in PBS and replenished with complete medium. Determinations of the absorbance were performed in duplicate for each well and treatment, for each time-point (0, 1, 4 and 7 days), according to the manufacturer’s suggestions.

Fibroblasts of different patients were also used to perform wound scratch tests; cells were seeded in 24-well plates to a density of 10⁴ cells/mL, let adhere and proliferate to semi-confluency in complete/treatment medium. A sterile 200 µL-micropipette tip was then used to generate a scratch onto the full length of the cell layer, along the diameter of each well. Cells were then washed in PBS and treatments started by the addition of TGF-$\beta$1 (5 ng/mL) and/or Pirfenidone (150, 300 or 450 µg/mL), wherever needed. The number of cells encompassed within the lesion borders, at any assessed time point and treatment, was the parameter chosen to evaluate growth conditions among the experimental settings assessed. Images of cells filling the gaps of the same lesion site were acquired after 0, 12, 24, 36 and 48 h; the boundaries of each original lesion were then superimposed to frames captured from the same plate in the same position at the different experimental time point; cells falling within the lesion boundaries were then counted using the National Institute of Health ImageJ free-software 1.53 (https://imagej.net/ij).

Using ultra-low attachment 96 well U bottom plates (Greiner Bio-One GmbH, Maybachstr 2, 72636 Frickenhausen, Germany) we further established 3D in vitro cultures of ILD-fibroblasts (Spheroids) or of ILD-fibroblasts together with epithelial/endothelial cells (Organoids). Spheroids and Organoids were then tested with Pirfenidone (D.B.A. Italia). To this aim 4000 ILD-fibroblasts or A549 or HECV (Spheroids) or either 2000 ILD-fibroblasts + 3000 A549 or HECV cells (Organoids), were re-suspended in 50 µL of complete medium (RPMI+FCS 10%), seeded in each well and then cultured in the incubator at 5% CO₂ and 37 °C for 48 h. After 48 h, 50 µL of complete medium with/without Pirfenidone were added. Pirfenidone was used at a concentration of 300 µg/mL in the final volume of 100 µL of culture medium for each well. After additional 72 h the images of each Spheroids or Organoids, treated or untreated with Pirfenidone, were acquired using a Nikon Digital Sight DS-5Mc as detailed above. Subsequently the size of untreated or treated spheroids or organoids was determined by measuring their visible area by means of the available tools of the free software Image J version 1.53.

Analysis of invasion by spheroids in a 3D model was performed based on the procedures described by Dsouza KG and co-authors [16] with some modifications. Spheroids were preliminarily prepared as above described and cultured for 48 h in a CO₂ incubator at 37 °C. After 48 h pre-cooled collagen (PureCol EZgel; Advanced Biomatrix; Carlsbad, CA, USA) was first diluted with RPMI-1640 at a concentration of 1 mg/mL and gently mixed avoiding the formation of air bubbles. A volume of 50 µL/well of this solution was added to cover the layer of flat-bottom 96 well plates, subsequently incubated at 5% CO₂ and 37 °C for 2 h. This layer prevents contact of spheroids with the plastic of the well. Growth medium was partially removed from previously formed spheroid and each spheroid was picked up and pipetted in a small drop of collagen 1 (100 µL), previously dispensed on the bottom of a petri dish, with/without half of the needed Pirfenidone concentration. This procedure was applied to allow a better inclusion of the spheroid in the matrix. After 15 min, each spheroid was moved onto each collagen-coated 96 flat-bottom wells. Following 30 min-incubation 50 µL of complete medium with/without half concentration of Pirfenidone were added. Final concentration of Pirfenidone was therefore 300 µg/mL. Plates were then incubated at 37 °C in a CO₂ incubator and evaluation of the invasion area was determined at different time points (24, 36, 48, 72, 132 h). The area dimensions were calculated by subtracting the inner core-spheroid area (ICA) from the total area covered by invading cells (outer cell borders, OCB) and the spheroid itself, obtaining the derived growth area (OCB-ICA). The calculated (OCB-ICA)/OCB ratio was then used to express the percentage of the area occupied by growing cells.

Whenever indicated, Student-t test and the Bonferroni’s correction were applied to evaluate the statistical significance; *: 0.01 p $\le$ 0.05; **: 0.001 p $\le$ 0.01; ***: p $\le$ 0.001.

Clinical details of the patients employed in the present study are reported in Supplementary Tables 1,2. For research purposes, we received the bronchoalveolar lavage of 43 ILD patients that were out of treatment. These patients were subjected to bronchoscopy and/or to criobiopsy for diagnostic purposes. Based on clinical, histological and cytofluorographic analyses the patients were then diagnosed as IPF (N = 23), fHP (N = 3), CTD-ILD (N = 5), IPAF (N = 4), NSIP (N = 3), Sarcoidosis (N = 3) and IgG-related disease (IgG4-RD, N = 2) (see Supplementary Tables 1,2).

Fibroblasts were derived from the BAL fluids of ILD patients. In a number of cases the cells could be expanded only for a short time, while in a few cases we could grow the cells for a longer time and for numerous passages (usually 10–12 or 36 in a unique case). Briefly, cells of the BAL fluid were first pelleted by centrifugation and then seeded in a 24-well plate in complete culture medium, adding a sub-optimal concentration of FGF2 the next day. In some cases, after 8–10 days, a few fibroblasts started to grow and to form colonies: fibroblasts reaching confluence were then detached and in part processed for RNA studies. When possible, the remaining cells were re-seeded and expanded in 24 or 6-well plates. Through this procedure we could derive a small number of cells from 20 ILD BAL fluids, out of 43 received (46%). Among these 20 samples we could further derive a larger quantity of cells from 11 patients (25%; 3 IPF, 2 fHP, 1 Sarcoidosis, 3 IPAF, 1 CTD-ILD and 1 NSIP) (Supplementary Table 1). Morphological features of fibroblasts derived from some ILD patients, as compared to the control cell line MRC5, are shown in Fig. 1A. Fibroblasts derived from one IPF patient, as well as those from two fibrotic HP patients, appear flattened and enlarged with a polygonal aspect, as compared with fibroblasts derived from patients with NSIP, IPAF, CTD-ILD, or either the control cell line MRC5, that are more spindle shaped. These observations highlight that fibroblasts derived from different ILD patients may show a certain heterogeneity, possibly related to higher levels of fibrosis observed in IPF or in fHP, than in the others (see Supplementary Table 1). This evidence could further support the suggestion that these fibroblasts might be truly representative of the ongoing fibrotic process. Fig. 1B,C further display the expression of $\beta$-galactosidase ($\beta$-gal), a marker of senescence, in fibroblasts derived from one IPF case (#17) versus the control cell line MRC5 at the same passage (p5). A significant higher number of $\beta$-gal positive cells were found for fibroblasts derived from the IPF patient, as compared to the control cell line: indeed there is a general consensus that fibroblast senescence is increased and persistent in lungs of IPF patients and that it is linked to the pathogenesis of the disease [17].

Fig. 1.

Differences in morphological features or level-senescence among fibroblasts derived from various ILD patients (fibrotic vs non-fibrotic) or in comparison with the control cell line MRC5. (A) Morphological representation of fibroblasts derived from different ILD patients versus fibroblasts derived from a healthy lung-fibroblast cell line. Red arrows indicate cells with enlarged and flattened shapes, typical of myofibroblast differentiation. (B) a higher number of cells positive for $\beta$-galactosidase (red arrows) were observed in fibroblasts derived from one IPF patient than in control fibroblasts (MRC5). (C) Histograms depict the mean $\pm$ SD of B-gal positive cell of 5 images for each sample. IPF, Interstitial Pulmonary Fibrosis; fHP, fibrotic Hypersensitivity Pneumonitis; NISP, Non-Specific Interstitial Pneumonia; IPAF, Interstitial Pneumonia with Autoimmune Features; CTD-ILD, Connective Tissue Disease-Interstitial Lung Disease. White Bars correspond to 12.5 µm and 25 µm for 10$\times$ and 20$\times$ images, respectively; *: 0.01 p $\le$ 0.05.

From the RNA of fibroblasts derived from 19 patients, we further examined, by quantitative RT-PCR, the expression of specific markers, such as $\alpha$-SMA, type 1 collagen and Fibronectin EDA and EDB, which are usually up-regulated during the differentiation of fibroblasts toward myofibroblast. As shown in Fig. 2 the expression of Type I Collagen, $\alpha$-SMA and fibronectin resulted higher in ILD-fibroblasts than in fibroblasts derived from normal controls, although without reaching statistical significant. The expression of the FGF2 receptor (FGF2R) also appeared higher in fibroblasts from ILD than in those from controls (Fig. 2).

Fig. 2.

Quantitative RT-PCR determination and cytofluorographic analysis of typical markers of fibroblast/myofibroblast differentiation in cells derived from ILD patients versus controls. (A) mRNA expression of alpha-SMA, collagen type-I, fibronectin and FGF2 receptor was higher in fibroblasts from ILD patients than in fibroblasts from controls (CTR). (B) Mean Fluorescence Intensity (Mfi) of $\alpha$-SMA, collagen-I and, to a lesser extent Fibronectin, was higher in fibroblasts from ILD-patients than in the control cell line MRC5. Numbers in parenthesis relate to the ID number of representative cases of different types of ILD. FGF, Fibroblast growth factor 2; $\alpha$-SMA, alpha Smooth Muscle Actin.

Cytofluorographic analyses of fibroblasts derived from representative ILD patients have further shown that typical markers of mesenchymal/fibroblast cells were expressed in all the derived fibroblasts such as CD105, CD90 and CD73 (Supplementary Fig. 1). These fibroblasts were also positive for vimentin but negative for CD45 (data not shown), thus confirming that they are of mesenchymal and not of hematopoietic origin, such as fibrocytes. We further observed that ILD-fibroblasts display a higher expression of $\alpha$-SMA and type I Collagen than the control cell line MRC5 (Fig. 2B), in agreement with their differentiation towards myofibroblasts, typically involved in pulmonary fibrosis.

We next evaluated proliferative response of ILD-fibroblasts to FGF, TGF$\beta$ or TGF$\beta$ + Pirfenidone. As shown in Fig. 3A, FGF weakly stimulated proliferation of fibroblasts derived from the different patients. Pirfenidone was capable of inhibiting TGF$\beta$-induced proliferation (Fig. 3B–E) with an inhibition range of 20–32% when Pirfenidone was used at 150 µg/mL. The response to pirfenidone displayed, however, a dose-dependent mode, particularly evident for fibroblasts from the fHP (#25) and CTD-ILD (#34) cases, reaching almost 50% of inhibition (Fig. 4A). On the contrary fibroblasts of the NSIP case (#41) were almost unresponsive and a partial inhibition (23%) was observed at the highest concentration of 450 µg/mL. The clear dose-response effect exerted by Pirfenidone is evident in the representative images of Fig. 4B referred to ILD patient #34. Moreover, the treatment with Pirfenidone down-modulated the TGF$\beta$-induced expression of actin fibers in fibroblasts derived from two ILD patients (#25, #34), as depicted in Fig. 5.

Fig. 3.

Evaluation of the proliferative response to FGF2 or to TGF$\beta$ or TGF$\beta$+Pirfenidone in fibroblasts derived from different ILD patients. (A) FGF2 induced a weak stimulation of the proliferation. (B–D) Proliferation of ILD-fibroblasts induced by TGF$\beta$ resulted inhibited by the addition of Pirfenidone (PIRF) at a concentration of 150 µg/mL. For each graph the depicted number refers to patients’ ID from which the fibroblasts were derived (Supplementary Table 1). SCI13D and SCI2 were derived from two BALs of the same patient (#25), received at different times. (E) Histograms represent the summary of the single cases presented in panel (B–D), further confirming that, as a whole, the inhibitory effect exerted by Pirfenidone was significant after 4 as well as after 6 days of treatment. *: 0.01 p $\le$ 0.05.

Fig. 4.

Determination of dose-dependency effect of Pirfenidone. (A) Evaluation of the growth of fibroblasts derived from 3 different ILD patients treated with increasing doses of Pirfenidone (150, 300 and 400 µg/mL). Histograms represent the mean $\pm$ SD of the 3 cases (fHP #25; CTD-ILD #34; NSIP #41) each assessed in duplicate. (B) Images represent cultures of fibroblasts derived from one representative patient (#34) exposed to different concentrations of Pirfenidone. The white bar corresponds to 12.5 µm.

Fig. 5.

Evaluation of F-actin expression in fibroblasts derived from two Pirfenidone-treated or -untreated ILD patients. Pirfenidone consistently reduced the expression of F-actin fiber fluorescence in fibroblasts derived from two ILD patients with/without TGF$\beta$ stimulation. The white bar corresponds to 25 µm.

The effect of pirfenidone on migration of ILD-fibroblasts was further evaluated by the use of a wound scratch assay [18] thus measuring fibroblasts polarizing and migrating in the wound space, with and without Pirfenidone treatment. Fig. 6A shows the migration of fibroblasts from one representative case of 5 tested following the stimulation with TGF$\beta$ (5 ng/mL) with/without Pirfenidone treatment (150 µg/mL), at different time-points (Fig. 6A). As shown in Fig. 6 Pirfenidone inhibited the ability of fibroblasts derived from different ILD-patients (#17, #25, #34, #38 and #41) to migrate across the lesion borders.

Fig. 6.

Determination of the migratory capacity of ILD-fibroblasts in a 2D system through a wound scratch test assay. (A) The migratory capacity of ILD-fibroblasts induced by TGF$\beta$, treated or untreated with Pirfenidone, was determined by acquiring images of the cells filling-in the lesion gap (yellow lines) at the indicated time points (0, 12, 24 and 48 h), as here shown for fibroblasts from one representative patient (#17). (B) Histograms represent the counted cells within the lesion’s boundaries for fibroblasts derived from ILD-patients at several time-points and under different treatments. Experiments were performed in duplicate for each time point and treatments and the values are the mean of two separate experiments. The white bar corresponds to 12.5 µm.

We next assessed the cytotoxic response to Pirfenidone of ILD-fibroblasts in 3D cultures. We took advantage of ultra-low attachment 96 U-bottom well plates to establish spheroids (only ILD-fibroblasts) or organoids (epithelial or endothelial cells + ILD-fibroblasts) and to further determine the activity of Pirfenidone. Fig. 7A,B shows that the formation of spheroids of ILD-fibroblasts was inhibited after Pirfenidone treatment (300 µg/mL) for 72 h with the exception of fibroblasts derived from one NSIP case (#41). A range of inhibition of 32% was detected among the responsive cases, with a more evident inhibition value for fibroblasts derived from the IPF case #17 (65%). Pirfenidone was also active in spheroids established with the epithelial adenocarcinoma A549 cells (41% of inhibition) but not with the endothelial cell line HECV (Fig. 7A,B). We then determined Pirfenidone cytotoxicity in a model of organoids composed of fibroblasts and A549 or HECV cells together: Pirfenidone inhibited the growth of A549+fibroblasts organoids (mean of 3 different added fibroblasts tested: 40% of inhibition), while a very weak effect was detectable for organoids formed by the endothelial HECV cells plus ILD-fibroblasts (Fig. 7C,D). It can be of further interest to note that the response behavior of the cells cultured in 2D was similar to cells cultured in 3D models (Supplementary Fig. 2). Ineffectiveness of the pirfenidone treatment on fibroblasts from the NSIP patient #41 was also confirmed, although a weak anti-proliferative effect (23%) was detectable when Pirfenidone was added at the concentration of 450 µg/mL and fibroblasts were tested in the 2 model, as described before.

Fig. 7.

Evaluation of the Pirfenidone activity on ILD-fibroblasts, or epithelial (A549) or endothelial (HECV) cells, in a 3D model of in vitro growth. (A) Spheroids composed of fibroblasts derived from different ILD patients, or A549 or HECV cell lines, were tested with Pirfenidone (300 µg/mL): pirfenidone significantly inhibited proliferation of ILD-fibroblasts with the exception of the NSIP case. The epithelial cell line A549 resulted inhibited but not the endothelial cell line HECV. (B) Images depict representative spheroids for each sample type. The white bar corresponds to 25 µm. (C) Organoids composed by A549 or HECV cells and ILD-fibroblasts were treated with pirfenidone (300 µg/mL): the drug resulted effective as a trend, although without statistical significance, for A549 cells+ILD-fibroblasts (3 cases assessed) but not for HECV cells+ILD-fibroblasts (2 cases assessed). Histograms values are the mean $\pm$ SD of 6 spheroids-areas or organoids-area per condition, evaluated as indicated in the methods section. (D) Images display representative organoids for each sample type. The white bar corresponds to 25 µm; *: 0.01 p $\le$ 0.05; **: 0.001 p $\le$ 0.01. ***: p $\le$ 0.001.

Spheroids derived as above described were further tested for cell invasiveness. We evaluated the migratory mobility of the cells by determining the enlargement of the area of fibroblast migration with time and treatment as a percentage of the total occupied area (Fig. 8; see also Supplementary Fig. 3). Cell spreading, well evident already from 36–48 h, was significantly reduced by Pirfenidone treatment, as depicted in Fig. 8A,B, particularly in fibroblasts derived from the one CTD-ILD case tested (#34).

Fig. 8.

Evaluation of the efficacy of Pirfenidone treatment on invasiveness of ILD-fibroblasts grown in 3D. (A) The invasiveness of ILD-fibroblasts composing the spheroids resulted inhibited following pirfenidone treatment (300 ug/mL) at different time points (24, 36, 48, 72 and 132 h). Images are representative of fibroblasts from two ILD patients. The white bar corresponds to 25 µm. (B) Histograms depict the mean $\pm$ SD of the percentage of enlargement area of different spheroids (N is indicated in each graph) per each treatment/time point, and refer to the two different ILD cases tested (#34 and #41); *: 0.01 p $\le$ 0.05; **: 0.001 p $\le$ 0.01.