Patients

The study material used for IHC consists of lung tissues from 50 IPF patients who had undergone surgical lung biopsy at the Oulu University Hospital for diagnostic purposes between 1991 and 2019 as described in the previous study [16]. Lung tissue specimens taken at autopsy from 8 out of 50 patients were also studied. IPF was diagnosed according to the international guidelines [4]. The patients experiencing AE-IPF during follow-up (n = 22) were identified either based on the surgical lung biopsy and autopsy material showing DAD in parallel with UIP indicating AE-IPF (n = 9), or by applying the current criteria for AE-IPF (n = 13) [3]. Control samples (n = 10) were derived from histologically normal-looking lung tissues from non-smoking patients being operated for lung adenocarcinoma.

Date of birth, gender, age, smoking status, pharmacological treatment, and pulmonary function test results at the time of biopsy were collected from electronic patient records. Patients with less than 5 pack-years of smoking history were regarded as non-smokers. The overall survival time was calculated from biopsy date to death, transplantation, or last follow-up date (May 11, 2021). Death dates were collected from death certificates obtained from the national registry of Statistics Finland.

Immunohistochemistry

Immunohistochemical stainings were performed in serial sections for lung tissue samples from IPF and control patients. Formalin-fixed and paraffin-embedded 3.5 μm thick tissue sections were stained by Envision+ System Kit (Dako, Glostrup, Denmark) with 3,3′-diaminobenzidine chromogen as described previously [17]. Antibodies are listed in Additional file 1: Table S1. NHLRC2 expression was compared to collagen α1(IV) chain (gene name COL4A1) based on the results of our previous study on the microarray analysis of lung stromal cells [17]. In order to identify the phenotype of the cells expressing NHLRC2, few cases were also studied for alpha smooth muscle actin (α-SMA, marker for myofibroblasts, gene name ACTA2), cluster of differentiation (CD) 68 (marker for macrophages), thyroid transcription factor (TTF)-1, marker for type II pneumocytes) and CD31 (marker for endothelial cells). Rabbit isotype control (Invitrogen, Carlsbad, USA) was used as negative control.

Whole slide images were acquired with a Leica-Aperio AT2 (Leica Biosystems, Nussloch, Germany) in Biobank Borealis of Northern Finland, Oulu University Hospital or with a NanoZoom S60 scanner (Hamamatsu, Hamamatsu city, Japan) in Transgenic and Tissue Phenotyping core facility, Biocenter Oulu, University of Oulu at 40× magnification.

Digital image analysis of immunohistochemical NHLRC2 expression

Visiopharm image analysis software (Visiopharm Integrator System, Hoersholm, Denmark) provided by Transgenic and Tissue Phenotyping core facility, Biocenter Oulu, University of Oulu was used to determine the area of NHLRC2-positive staining in all types of lung cells considering all intensities (strong, middle, weak, very weak) in relation to total area of the tissue section in 50 surgical lung biopsy samples from IPF patients with a histology of UIP (n = 47) or UIP and DAD (n = 3) and 10 control lung tissue samples.

Calculation of NHLRC2-positive fibroblast foci

Digitized lung tissue specimens were examined by using Aperio Image Scope (Version 12.4.3.5008, Leica Biosystems) or NDP.view2 (Hamamatsu, Hamamatsu city, Japan). The total number of NHLRC2-positive FF were calculated from 47 samples with UIP histology. NHLRC2-positive FF was determined as consisting of more than 50% of positive stromal cells considering all intensities. In addition, the total number of FF were calculated from the 47 sections with UIP histology. The number of FF was presented in relation to the area of the tissue section.

mRNA in situ hybridization

NHLRC2 mRNA in situ hybridization was performed for surgical lung biopsy samples of 8 IPF patients with a histology of UIP (n = 6) or UIP and DAD (n = 2) and in 3 control lung tissue samples using RNAscope 2.5 HD assay—RED and probe Hs-NHLRC2 (555721) according to the manufacturer’s instructions (Advanced cell diagnostics, ACD, Newark, CA, USA). Formalin-fixed and paraffin-embedded specimens were cut into 4 µm thick sections. Target retrieval was performed by boiling the sections at 98 °C for 15 min in RNAscope target retrieval reagent using a KOS Microwave HistoSTATION (Milestone, Sorisole, Italy). Gill’s Hematoxylin (Sigma-Aldrich, St. Louis, MO, USA) was used to stain the nuclei, and coverslips were mounted with EcoMount (Biocare Medical, Pacheco, CA, USA). Positive and negative control probes (Hs-UBC 310041 and DapB 310043, ACD) were used to help qualify samples and control for background noise. Specific staining signals were identified as red dots.

Cell culture

NHLRC2 expression in stromal and epithelial cell lines were compared in vitro. Stromal cells were cultured from control lung tissue samples (n = 4) and surgical lung biopsy samples of the patients with IPF (n = 5) as described previously [18]. Briefly, the cells were cultured in medium consisting of Minimum essential medium Eagle α modification (Sigma-Aldrich) supplemented with 13% heat-inactivated fetal bovine serum (FBS-Good, Pan Biotech, Aidenbach, Germany), 2 mM l-glutamine, 100 U/ml penicillin, 0.1 g/l streptomycin, 2.5 mg/l amphotericin B and 10 mM HEPES (all from Sigma-Aldrich). These cell lines are composed of both fibroblasts and myofibroblasts as previously described in our electron microscopic analyses [18]. Cells were used for experiments in passages 3–6.

Normal human primary small airway epithelial cells (SAEC) and normal human primary bronchial/tracheal epithelial cells (PBTE) (American type culture collection, ATCC, Virginia, USA) were cultured in airway cell basal medium supplemented with bronchial epithelial growth kit (ATCC). SAEC were used for experiments in passages 6–7 and PBTE in passage 5.

In order to study the effect of TGF-β1 on NHLRC2 levels, one control and one IPF stromal cell line and SAEC were plated at a density of 9000 cells/cm2. After 24 h stromal cells were exposed to 5 ng/ml TGF-β1 (Sigma-Aldrich) in a serum-free growth medium for 24 to 72 h. SAECs were exposed to 5 ng/ml TGF-β1 in complete growth medium for 24 to 72 h. Cells cultured in similar conditions without TGF-β1 were used as controls.

RNA extraction and RT-qPCR

Total RNA was extracted from cultured cells using the RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions and the RNA concentrations were measured using the NanoDrop spectrophotometry system (Thermo Fisher Scientific, Vilnius, Lithuania). Five hundred-ng aliquots of RNA were reverse-transcribed using RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific) using oligo(dT)18 as primer according to manufacturer’s instructions. PCR amplification was performed in triplicate as previously described [17, 18] by using iQTM SYBR Green Supermix (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Primer sequences and annealing temperatures specific for each primer pair are listed in Additional file 1: Table S2. Relative gene expressions were quantified by using the 2−∆∆CT Livak method [19] and gene expression levels were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The normalized values were compared to average of the normal control stromal cell lines or non-treated samples in TGF-β1 experiments to calculate the fold changes.

Immunoblotting

The cells were lysed in 1.5% dodecyl maltoside (DDM, in phosphate buffered saline) or in radio-immunoprecipitation assay (RIPA) lysis and extraction buffer (Thermo Fisher Scientific) supplemented with a protease inhibitor cocktail tablet (Roche, Mannheim, Germany). After 45-min incubation on ice the samples were centrifuged at 20,000 g for 20 min. The protein concentration of cell lysates was determined by DC Protein Assay Kit (Bio-Rad) according to the manufacturer’s instructions. Twenty-μg aliquots of samples were loaded with Bolt LDS sample buffer (Thermo Fisher Scientific) and run on SDS–PAGE (Invitrogen Bolt Bis–Tris Mini Protein Gels, Thermo Fisher Scientific). The proteins were transferred onto 0.45 µm nitrocellulose membrane (Optitran reinforced NC, Whatman Schleicher and Schuell, Dassel, Germany). After blocking with 5% skim milk, the membranes were incubated with primary antibodies for NHLRC2, α-SMA and GAPDH followed by appropriate labelled secondary antibody incubation. Details of antibodies are listed in Additional file 1: Table S1. Protein bands were visualised with an Odyssey infrared imager (LI-COR Biosciences, Lincoln, NE, USA) and quantified with Image Studio Lite (LI-COR Biosciences). The expression levels of the target proteins were normalized to that of GAPDH.

Statistical analyses

IBM SPSS Statistics for Windows, Version 25.0 (IBM Corp, Armonk, NY) was used to perform statistical analysis. OriginPro, Version 2019b (OriginLab Corporation, Northampton, MA, USA) was used for preparing graphs. The data were presented as median values with 25 and 75% quartiles for skewed variables, or as the means with standard deviation for those with a normal distribution. Comparisons of parameters that were not normally distributed between more than two groups were performed using the Kruskal–Wallis test and post hoc analysis (Dunn’s test with Bonferroni correction) and between two groups using Mann–Whitney U test. FVC% and DLCO% were divided in two groups based on median values (75% and 53%, respectively) in IPF patients at a stable phase of the disease. Survival was evaluated using the Kaplan–Meier method and differences in survival curves were evaluated using the log-rank test. The associations of relative immunohistochemical NHLRC2 expression and the number of NHLRC2-positive FF/cm2 with survival or future AE were analysed using univariate Cox regression analysis. Median values of relative NHLRC2 expression and NHLRC2-positive FF/cm2 in IPF patients were used as cut-off values for Kaplan-Meyer and Cox regression analysis. Values of p < 0.05 were considered statistically significant.