The hiPSC line BU3NGST was kindly provided by Prof. Darrell Kotton, Boston University, Center for Regenerative Medicine. This cell line is a dual-reporter construct composed of fluorochrome-encoding cassettes targeted to the endogenous NKX2.1 and SFTPC loci (BU3 NKX2.1^GFP; SFTPC^tdTomato) (15). hiPSCs were maintained in mTeSR1 (StemCell Technologies), on Matrigel (Corning) coated cell culture plates at 37°C/5% CO₂ in a cell culture CO₂ incubator. Cells were subcultured by using ReLeSR (StemCell Technologies) or Gentle Cell Dissociation Reagent (StemCell Technologies) (16, 17).

BU3NGSTs were differentiated into NKX2.1⁺ lung progenitor cells and iAT2s as described previously by Jacob et al. (16, 17). In short, hiPSCs were checked for their pluripotency via Alkaline Phosphatase staining (ES Cell Characterization Kit, CHEMICON International) or immunofluorescence staining of TRA 181 and SSEA 4 (ES cell characterization Kit, CHEMICON International). Induction of definitive endoderm was conducted via STEMdiff Definitive Endoderm Kit, (StemCell Technologies). On day 14 of differentiation, lung progenitor specification was evaluated by immunofluorescence staining of NKX2.1 (Invitrogen) and Albumin (ALB, R&D Systems). NKX2.1^GFP+ lung progenitor cells were enriched by GFP signal for NKX2.1 based on a previously described protocol. The sorting was performed by FACS cell sorting at MACSQuant Tyto Cell Sorter (Miltenyi Biotec). For data evaluation FlowJo Version 7.2.1 and v10 was used. Purified lung progenitors were seeded in Matrigel (Corning) domes at a cell density of 50 cells/µL and passaged every second week. To increase SFTPC^tdTomato+ cells CHIR withdrawal and addback was performed. At day 45 of differentiation, iAT2s were enriched by flow cytometry (MACSQuant Tyto Cell Sorter, Miltenyi Biotec) using tdTomato signal for SFTPC expression and subsequently cultured as 3D alveolar organoids. Differentiated SFTPC^tdTomato+ iAT2s in 3D Matrigel were grown in CK+DCI medium, with media changes every 48 – 72 h. Alveolar organoids were passaged every 14 days.

Primary human lung fibroblasts from ILD patients, in our study referring to cases of pulmonary fibrosis,and non-CLD (P4) for co-culture experiments (ILD fibroblasts) and MS based secretome analysis (ILD and non-CLD fibroblasts) were isolated according to a published protocol (18) and obtained through the CPC-M bioArchive at the Comprehensive Pneumology Center in Munich, Germany.

All patients underwent surgery at the LMU Hospital and the Asklepios Pulmonary Hospital Munich-Gauting. Tissue from ILD patients (n = 3, Table 1 ) was provided through lung transplantation. Control fibroblasts were derived from lung tissue resections of age-matched non-CLD patients (female n = 1, = 3, male n = 2).

The study was approved by the local ethics committee of the Ludwig-Maximilians University of Munich, Germany (Ethic vote #333-10) and written informed consent form was obtained for all study participants.

Human fetal lung fibroblasts (IMR-90, P8) for control co-cultures were obtained from ATCC (Catalog # CCL-186™) and grown in Dulbecco’s Modified Eagle Medium: Nutrient F-12 (DMEM/F12; Gibco) with 20% fetal bovine serum (FBS SUPERIOR, Sigma) and 1% penicillin-streptomycin (Pen Strep, Gibco).

Cells were seeded in 6-well plates. at a density of 1x10⁵ cells in 2 mL media (DMEM/F12, 20% FBS, 1% penicillin/streptomycin) per well until reaching 80% confluency. Once confluent, each well was washed three times with a 15-minute incubation per wash with 1 mL of FBS-free culturing medium (DMEM/F12, 1% penicillin/streptomycin) to eliminate remaining FBS. Fibroblasts were cultured for 48 h in FBS-free medium, supernatants were collected and stored at -80°C for further analysis.

Primary lung ILD fibroblasts and IMR-90 (control fibroblast cell line) were grown in cell culture flasks until 70% confluency. A single cell suspension was prepared using 0.25% EDTA-Trypsin (Gibco). iAT2s were grown into alveolar organoids for up to two weeks in Matrigel domes. Single cell suspension was obtained with Dispase (Corning) and 0.25% EDTA-Trypsin as described by Jacob et al. (17). Human ILD and IMR-90 fibroblasts as well as iAT2s were counted and directly seeded either in equal 1:1 (F_low) or 1:5 (F_high) iAT2s to fibroblasts seeding densities in undiluted Matrigel domes in 8-chamber wells (20 µL Drops, Falcon), 96-well plates (50 µL Drops, Greiner) or 12-well plates (50 µL Drops, Greiner). Co-cultures used CK+DCI media that was changed every 48 h to 72 h for up to 12 days of cultivation.

3D alveolar organoids mono- and co-cultures as described in section 2.1 (mono-culture) and 2.4 (co-culture) were cultured in 8-chamber wells for immunofluorescence analysis (Nunc Lab-Tek Chamber Slide System, 8-well, Permanox slide, 0.8 cm²/well). After alveolar organoids were formed, fixation was achieved with ice cold methanol and acetone (1:1v/v) for 5 minutes at -20°C. Cells were washed with PBS and stained with the respective primary antibody in buffer containing 0.1% BSA and 0.1% Triton X-100 overnight at 4°C. The next day, cells were washed 3 times with PBS and incubated in buffer with the respective fluorescent conjugated secondary antibody at a dilution of 1:500 and DAPI diluted 1:1.000 overnight at 4°C. The following day, cells were washed gently, growth camber removed and remaining microscope slide mounted with fluorescent mounting media (Dako) and covered with a coverslip. Slides were stored at 4°C until imaging. Imaging was performed using a confocal laser scanning microscope (CLSM) Zeiss LSM 880 with Airyscan and edited afterwards using ZEN 2.5 software (Zeiss). Detailed information on the primary and secondary antibodies are given in Supplementary Table 1 .

Live imaging of all alveolar organoid mono- and co-cultures was performed using the Operetta CLS high-content analysis system (Operetta CLS, PerkinElmer) at time points 5, 8 and 12 days during the co-culture experimental set-up. Pictures were analyzed by the Harmony 3.5.2 high-content imaging and analysis software with PhenoLOGIC.

Multi-plane confocal 3D images were visualized in Napari image viewer (Python) as maximum intensity projections and the automatic measurements obtained from the “Napari organoid counter” (25) were visually checked and manually curated, resulting in output of size and numbers of formed alveolar organoids between iAT2s and human fibroblasts (ILD and control IMR-90). The Canny Edge Detection (26) algorithm is used for identifying the organoids, while pre- and post-processing steps have been included, to ensure the image matches the detector’s expected input and the number of detected organoids along with their size is returned. More specifically, the organoids are approximated to ellipses and the algorithm fits orthogonal bounding boxes around each, with the height and width of each box corresponding to the two diameters of the organoid which are then in turn used to approximate the object’s area.Quantitative real-time PCR

Co-cultures were lysed in RLT Plus Lysis Buffer (Qiagen) and RNA isolation was performed with the RNeasy Mini Kit (Qiagen) according to the manufacturer’s instructions. Cell lysis from organoids and co-culture assays was performed with peqGOLD TriFast (VWR Life Science) as recommended by the manufactures followed by RNA isolation with the RNeasy Mini Kit (Qiagen). RNA was transcribed into cDNA by reverse transcriptase using the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. 5 ng of cDNA was added to a final concentration volume of 10 µL, Random Nonamers (Metabion) and master mix (Invitrogen, Thermo Fisher Scientific) was added to each RNA sample. cDNA was diluted with ultrapure H₂O. qPCR was performed in 96-well format using the quantitative real-time PCR System (Roche 480 LightCycler). 2 µL cDNA were added to a final reaction volume of 10 µL containing H₂O, 480 SYBR Green (LightCycler, Roche Diagnostics) and the primer mix (100 µM). Gene expression was normalized to ß-Actin control for genes Vimentin (VIM), Integrin Subunit Beta 6 (ITGB6) and Cadherin 2 (CDH2), and normalized to an average of ß-Actin and HRPT control for genes Surfactant Protein C (SFTPC), Keratin 8 (KRT8), Collagen 1A1 (Col1A1), Matrix Metallopeptidase (MMP7) and Bone Morphogenic Protein 4 (BMP4), the fold change was calculated using the 2^ (-ddC) method. Sequence information of used primers are given in Supplementary Table 2 . Data obtained from qPCR are presented relative to respective control co-cultures, to demonstrate influence of disease background.

WST-1 assays were performed at day 2, 3, 5 and 7 of alveolar organoid co-cultures (human ILD or IMR-90 control fibroblasts). WST-1 reagent (Roche Diagnostics) was added to the culture medium in a 1:10 dilution. The culture medium was used as background control. After 2 h of incubation at 37°C and 5% CO₂, 100 µL of media from every sample was transferred to a microplate (Thermo Scientific; Fisher Scientific) and the absorbance of the sample against the background was measured with a TECAN reader (TECAN; infinite M200 PRO).

An overview of the experimental workflow is provided in Supplementary Figures 1A–C . iAT2s ( Figure 1A ) were successfully co-cultured with both ILD, in our study referring to cases of pulmonary fibrosis, or IMR-90 control fibroblasts, resulting in the formation of proliferative alveolar organoids ( Figure 1B ). Cell-cell contact of iAT2s and fibroblasts in (ILD) co-culture was demonstrated by partial encapsulation of alveolar organoids by α-SMA expressing fibroblasts ( Figure 1C , white arrows).

Image analysis of the 3D co-cultures revealed a reduction in organoid formation capacity in the presence of human fibroblasts ( Figure 1D ). Although alveolar organoid size was not significantly affected by fibroblast co-culture ( Figures 1B, E ), quantitative assessment of organoid size (area, µm²) and number of images obtained from alveolar organoids and co-cultures with either ILD or IMR-90 control fibroblasts in 1:1 (F_{ILD/Control low}) or 1:5 (F_{ILD/Control high}) seeding density ( Figures 1B, D ) demonstrated a negative correlation of fibroblast seeding density (ILD or IMR-90) with the alveolar organoid formation capacity.

In contrast, co-culture with ILD fibroblasts significantly increased metabolic activity of alveolar organoid in comparison to IMR-90 control co-cultures ( Figure 1F ).

In order to relate the observed changes in organoid formation capacity and metabolic activity ( Figure 1 ) to changes in gene expression, we measured critical markers of stem cell function and epithelial differentiation in co-cultured organoids.

Indicating changes in (stem) cell function and epithelial injury, we showed decreased expression of the alveolar stem cell marker SFTPC in the presence of ILD fibroblasts in both seeding ratios (F_low and F_high; Figure 2A ). In line with this, Keratin 8 (KRT8) expression levels were reduced under the impact of ILD primary fibroblasts ( Figure 2A ). Further, the distal epithelial marker Integrin Subunit Beta 6 (ITGB6) as well as Bone Morphogenetic Protein 4 (BMP4) showed increased transcription in ILD co-cultures when high seeding densities were applied ( Figure 2A ). Genes associated with regulation of extracellular matrix formation and remodeling including Collagen 1A1 (Col1A1), N-Cadherin 2 (CDH2) and Vimentin (VIM) showed increased expression in alveolar organoids co-cultured with ILD fibroblasts in high seeding ratios ( Figure 2B ). Likewise, expression levels of Matrix Metallopeptidase 7 (MMP7) were increased in ILD co-cultures compared to control co-cultures. ( Figure 2B ).

To characterize fibroblast driven communication resulting in gene expression and phenotypical changes in the alveolar epithelium in ILD, supernatants of ILD and non-CLD fibroblasts were subjected to mass spectrometry (MSMS).

MS analysis detected an overall of 2625 expressed proteins, of which 47 were significantly more and 55 significantly less abundant when comparing ILD-derived fibroblast to non-CLD controls ( Supplementary Table 3, 4 , Figure 3A ). The top 15 differentially expressed proteins (increased and decreased abundance) are listed in Supplementary Tables 1 , 2 .

Proteins with increased abundance predominantly belonged to the C-X-C motif chemokine family (CXCL1, CXCL3, and CXCL8), as well as to the interleukin family (IL13RA, IL11) and included gap junction proteins (connexin 43, GJA1). Further, Pregnancy Specific Beta-1-Glycoprotein 4 (PSG4) as well as WNT signaling modulator SFRP4 were found amongst the top 15 proteins.

Accordingly, pathway enrichment analysis of proteins with increased abundance classified the responses as cytokine activity, chemokine-mediated signaling pathway and TNF-signaling pathway. Furthermore, cellular/response to chemokines, IL17 signaling pathways and neutrophil chemotaxis were identified, indicative of a strong inflammatory response.

ClueGo, a Cytoscape plug-in for network analysis, highlighted proteins associated with rheumatoid arthritis, a disease often complicated by the development of lung fibrosis and characterized by the presence of inflammatory chemokines ( Figure 3B ).

Proteins with decreased abundance in the in ILD secretome were dominated by candidates involved in ECM production, ECM assembly or ECM reorganization as well as coordination of myofibroblast differentiation (PDGFRL) together with a downregulation of proteins involved in complement and coagulation cascade pathways ( Figure 3C ).

Based on the MSMS secretome analysis, IL11 emerged as a top player in the mesenchymal-to-epithelial disease crosstalk. In consequence, we exposed alveolar organoids ( Figure 4A ) to IL11 in order to investigate its functional relevance. Experiments (Dose_low = 0.5 ng/mL, Dose_high = 5 ng/mL) ( Figures 4B–D ) recapitulated the results observed in epithelial-fibroblasts co-cultures (section 3.1), i.e., we demonstrated reduced organoid formation capacity ( Figures 4Biii, C ) and increased metabolic activity (WST-1) in IL11 treated alveolar organoids (treatment from day 7 - 14 of culture) ( Figure 4D ). In addition, treatment of growing alveolar organoid monocultures with IL11 (20 ng/mL) led to an increase in alveolar organoid size followed by apoptosis within 5 days of culture.