Cryptotanshinone is a candidate therapeutic agent for interstitial lung disease associated with a BRICHOS-domain mutation of SFTPC

Summary Interstitial lung disease (ILD) represents a large group of diseases characterized by chronic inflammation and fibrosis of the lungs, for which therapeutic options are limited. Among several causative genes of familial ILD with autosomal dominant inheritance, the mutations in the BRICHOS domain of SFTPC cause protein accumulation and endoplasmic reticulum stress by misfolding its proprotein. Through a screening system using these two phenotypes in HEK293 cells and evaluation using alveolar epithelial type 2 (AT2) cells differentiated from patient-derived induced pluripotent stem cells (iPSCs), we identified Cryptotanshinone (CPT) as a potential therapeutic agent for ILD. CPT decreased cell death induced by mutant SFTPC overexpression in A549 and HEK293 cells and ameliorated the bleomycin-induced contraction of the matrix in fibroblast-dependent alveolar organoids derived from iPSCs with SFTPC mutation. CPT and this screening strategy can apply to abnormal protein-folding-associated ILD and other protein-misfolding diseases.


INTRODUCTION
Interstitial lung disease (ILD) comprises a large group of diseases that cause lung fibrosis.Patients with progressive fibrosing ILD (PF-ILD) have poor prognoses and often experience severe respiratory failure.Its mechanism has not yet been elucidated; however, some genetic mutations associated with alveolar type 2 (AT2) cells have been reported to cause progressive pulmonary fibrosis, making it difficult to breathe and supply oxygen to the bloodstream. 1,2n recent years, nintedanib and pirfenidone were approved by the US Food and Drug Administration (FDA) for treating progressive pulmonary fibrosis, although they cannot halt and reverse fibrosis that has already occurred. 3,4The pathogenesis of pulmonary fibrosis is initiated by alveolar epithelial injury, followed by fibroblast activation.Therefore, there is an increasing demand for new therapeutic drugs targeting the alveolar epithelial cells involved in the early stages of ILD.The causes of initial alveolar epithelial injury are diverse, including the side effects of drugs such as amiodarone and environmental factors such as cigarette smoke. 5Genetic mutations have also been found to be associated with ILD.Recently, the development of next-generation sequencing has suggested that more than ten genes are responsible for familial ILD.In particular, many reports have indicated that mutations in genes critical for surfactant metabolism and functions [surfactant protein C (SFTPC), surfactant protein A1/A2 (SFTPA1/A2), and ATP binding cassette subfamily A member 3 (ABCA3)] are associated with ILD. 6 Among them, more than 70 SFTPC mutations associated with ILD have been described and found in a large and heterogeneous group of patients with sporadic and familial ILD with autosomal dominant inheritance. 7,81][12][13][14][15] This study established a compound screening system utilizing two molecular phenotypes, ER stress, and SFTPC aggregates; both are involved in the diseasecausing mechanisms invoked by BRICHOS domain mutations.Furthermore, since our group previously established methods for generating human induced pluripotent stem cells (iPSC)-derived alveolar cells in organoids, 16,17 AT2 cells differentiated from patient-specific iPSC harboring SFTPC mutations that cause ILD were used to evaluate the efficacy of the candidate compounds, and we identified Cryptotanshinone (CPT) as a potential therapeutic agent for ILD.

RESULTS
High-throughput screening and identification of compounds that reduce ER stress caused by mutant SFTPC First, we focused on ER stress among the molecular pathogeneses caused by SFTPC with BRICHOS domain mutations and developed a highthroughput screening system.To detect ER stress, we used an X-box binding protein 1 (XBP1) splicing reporter, which was constructed by fusing HiBiT protein, the split particle of NanoLuc luciferase, 18 to exon 4 of the XBP1 containing inositol requiring enzyme 1 (IRE1)-mediated splice sites (hereinafter called ''XBP1-HiBiT Reporter'') (Figure 1A).After an unfolded protein response induces ER stress, the sensor protein IRE1 is activated.Activated IRE1 mediates the splicing of 26 nucleotides of exon 4 of XBP1 in the cytoplasm, and the spliced XBP1 functions as a nuclear transcription factor.This XBP1 splicing system has been used to indicate ER stress. 19,20We examined whether the XBP1-HiBiT Reporter could quantitatively detect ER stress using tunicamycin, commonly used to induce ER stress.In HEK293 cells transfected with the XBP1-HiBiT Reporter, the relative light unit (RLU) increased in a concentration-dependent manner with tunicamycin (Figure 1B), indicating that the XBP1-HiBiT Reporter could be used to evaluate ER stress quantitatively.Next, we constructed an expression vector of myc-tagged SFTPC with the L188Q mutation (L188Q SFTPC) and the wild-type SFTPC (wild SFTPC).SFTPC encodes L188Q mutation with a mutation (c.563T>A) and substitutes glutamine for leucine at amino acid 188 in the BRICHOS domain of the SFTPC.The L188Q mutation in SFTPC was reported as a causative mutation in autosomal dominant familial ILD and an ER stress inducer. 13,21By co-transfection of L188Q SFTPC with the XBP1-HiBiT Reporter, we generated a system that could evaluate the effect of compounds on ER stress caused by the BRICHOS SFTPC mutant (Figure 1C).The Z 0 -factor, an indicator of the statistical confidence of screening, was 0.79 for this screening assay, which is considered high enough for screening 22 (Figure 1D).To confirm that L188Q SFTPC induces ER stress in this system, we compared the splicing ratio of the XBP1-HiBiT Reporter using reverse transcription-quantitative polymerase chain reaction (RT-qPCR) analysis.The splicing ratio was significantly higher in HEK293 cells co-transfected with L188Q SFTPC and the reporter than those co-transfected with wild SFTPC and the reporter (Figure 1E).We further analyzed the changes in protein and mRNA expression levels of ER stress marker genes.In RT-qPCR analysis, mRNAs levels of HSPA5 [heat shock protein family A (Hsp70) member 5] (BIP), DDIT3 (DNA damage-inducible transcript 3) (CHOP) and ATF4 (activating transcription factor 4), and endogenous XBP1 splicing ratio were significantly increased in L188Q SFTPC overexpressed cells relative to wild SFTPC overexpressed cells (Figure S1A).In western blotting (WB) analysis, BIP and CHOP protein levels were significantly increased in L188Q SFTPC overexpressed cells relative to wild SFTPC overexpressed cells (Figure S1B).To confirm the difference in transfection efficiency, we examined the expression level of neomycin phosphotransferase II (neomycin-resistance gene), which is expressed from a different promoter of the same plasmid (pcDNA3).There was no large difference in the expression levels between wild and L188Q SFTPC (Figure S1B), indicating transfection efficiency of wild and L188Q SFTPC expression vector is almost the same.These results suggest that L188Q SFTPC induces ER stress in this screening system.
We screened 2,480 compounds from the Kyoto University chemical library to identify compounds that could reduce ER stress caused by L188Q SFTPC overexpression (Figure 1F).As a positive control to reduce ER stress, we used 5 mM 4-phenyl butyric acid (4-PBA). 23,24To exclude highly toxic compounds, the cytotoxicity of the compounds at 10 mM was confirmed using a Water-Soluble Tetrazolium 8 (WST-8) assay, which measures intracellular dehydrogenase activity as an indicator of cell viability.Positive control (4-PBA): RLU of HiBiT = 51.1% G 6.4 and WST-8 = 77.2%G 3.3 [mean G SD, % of dimethyl sulfoxide (DMSO)].The threshold to determine the hit compounds was calculated based on HiBiT and WST-8 assay results for the positive control (4-PBA) because 4-PBA was reported to reduce ER stress caused by SFTPC BRICHOS mutant 23 (HiBiT: 57% and WST-8: 74%).Another threshold was used to obtain highly effective compounds with some toxicity (HiBiT: 38% and WST-8: 60%).Sixty-five hit compounds were identified (Figure 1G).

Secondary screening for compounds identified in the primary screening
As a second screening step, we tested whether the candidate compounds could reduce ER stress caused by SFTPC mutations other than L188Q.In addition, we simultaneously evaluated whether they could decrease the aggregate formation of mutant SFTPC to confirm their effect from a different molecular aspect.As a mutation other than L188Q, we chose the Y104H mutation (c.310T>C), a substitution of tyrosine for histidine at amino acid 104 in the BRICHOS domain of SFTPC, because a patient with Y104H was reported in Japan 25 and patient-derived peripheral blood mononuclear cells (PBMCs) were available for the present study.Therefore, SFTPC expression vectors with the Y104H mutation (Y104H SFTPC) were constructed.First, we investigated whether Y104H SFTPC mutation induces ER stress in the cells similar to L188Q SFTPC mutation.The splicing ratio of the XBP1-HiBiT Reporter was significantly higher in HEK293 cells co-transfected with Y104H SFTPC and the reporter compared with those co-transfected with wild SFTPC and the reporter (Figure 2A).The changes in protein and mRNA expression levels of ER stress marker genes were analyzed in HEK293 overexpressing Y104H SFTPC.In RT-qPCR analysis, mRNAs levels of BIP, CHOP, and ATF4, and endogenous XBP1 splicing ratio were significantly increased in Y104H SFTPC overexpressed cells relative to wild SFTPC-overexpressed cells (Figure S1C).In WB analysis, BIP and CHOP protein levels were significantly increased in Y104H SFTPC overexpressed cells relative to wild SFTPC-overexpressed cells (Figure S1D).Transfection efficiency was assumed to be almost the same in light of the similar expression level of the neomycin-resistance gene in the two conditions (Figure S1D).These results show that the overexpression of Y104H SFTPC induces ER stress.Next, to analyze whether the Y104H-mutant SFTPC causes SFTPC aggregates, each expression vector of the fusion genes of AcGFP and wild or Y104H SFTPC (AcGFP-wild SFTPC and AcGFP-Y104H SFTPC) was generated and transfected into HEK293 cells.Using Opera Phenix high-content screening system, we quantified the mean of the Spot Intensity of AcGFP in each cell (Spot Intensity) and the mean of the Total Spot Area of AcGFP in each cell (Total Spot Area) as an indicator of SFTPC aggregates.We found that the cells transfected with AcGFP-Y104H SFTPC showed significantly higher Spot Intensity than that of AcGFP-wild SFTPC (FC: Y104H/wild = 1.64) and significantly smaller Total Spot Area than that of AcGFP-wild SFTPC (FC: Y104H/wild = 0.79) (Figures 2B-2D).Stronger and smaller GFP spots suggested that the Y104H SFTPC forms aggregates.Because the Z 0 -factors of the assays, based on the RLT of XBP1-HiBiT Reporter and Spot Intensity of GFP, were high enough for screening [Z 0 -factor = 0.68 (reporter) and 0.72 (Spot Intensity)] (Figures 2B and 2E), the screening of 65 compounds was performed to determine whether they could reduce ER stress and aggregate formation caused by Y104H-mutant SFTPC.We obtained nine hit compounds with thresholds of 70% RLU and 80% Spot Intensity (% of DMSO) (Figure 2F).To exclude the possibility that these hit compounds affect the CMV promoter or reporter expression system, we analyzed the splicing ratio of the XBP1-HiBiT Reporter for RNA expression levels.The reporter splicing products were detected by RT-qPCR instead of the HiBiT assay and normalized to total reporter expression.Then, four compounds were selected with a threshold of 0.75 for fold change (FC) (compound/DMSO) (Figure 2G).

Efficacy evaluation of compounds using alveolar epithelial cells differentiated from patient-derived iPSC
To examine the efficacy of the compounds identified by the developed screening system, we generated iPSC lines from PBMCs of a patient with ILD who had a heterozygous Y104H-SFTPC variant (SPC Y104H -iPSC) (Figure 3A).The expression of pluripotency markers (OCT3/4 and NANOG) was confirmed by flow cytometry to assess the pluripotency of SPC Y104H -iPSC.The differentiation potential into the three germ layers was determined by ensuring the expression of each differentiation marker [ectoderm (PAX6 and NESTIN), mesoderm (NCAM and Brachyury), and endoderm (SOX17 and FOXA2)] (Figure S2A).SPC Y104H -iPSC formed typically tightly packed round-shaped colonies (Figure S2B).Next, iPSC monoallelically expressing one wild type allele of SFTPC was generated from SPC Y104H -iPSC using CRISPR/Cas9 (hereinafter, wild type SFTPC monoallelic expressing iPSC line is called ''maeSPC Y104H -iPSC'').This is the resultant founder line that expressed no detectable mRNA of Y104H-mutated SFTPC (Figure 3B), owing to an intronic insertion of a neomycin cassette flanked by LoxP sites.SPC Y104H -iPSC and maeSPC Y104H -iPSC demonstrated a normal karyotype, as shown by G-banding analysis (Figure 3C).In addition, there were no indels at the 25 predicted off-target sites (Table S1).
To assess the therapeutic potential of the four candidate compounds (#562, #775, #2035, and #2352), SPC Y104H -, and maeSPC Y104H -iPSC were differentiated into AT2 cells in an air-liquid interface (ALI) culture (hereinafter, these differentiated alveolar epithelial cells containing AT2 cells are called ''SPC Y104H -iAT2'' and ''maeSPC Y104H -iAT2'') (Figure 3A).The expression of pro-SFTPC in SPC Y104H -and maeSPC Y104H -iAT2 was confirmed by immunostaining with an anti-proSFTPC antibody (Figure S2C), and quantitative image analysis of SFTPC accumulation was performed.As an indicator of SFTPC accumulation, fluorescence intensity/area scores were calculated by dividing the integrated values of fluorescence intensity by the fluorescence area's total value.The score of SPC Y104H -iAT2 treated with #2035 was significantly lower than that of cells treated with DMSO, although the other compounds showed no significant differences compared with DMSO (Figure 3D).In contrast, the score of maeSPC Y104H -iAT2 treated with #2035 showed no significant change (Figure 3E).The fluorescence intensity/area of SPC Y104H -iAT2 was higher than that of maeSPC Y104H -iAT2, and interestingly, the reduced fluorescence intensity/area of SPC Y104H -iAT2 treated by the #2035 was almost the same as that of maeSPC Y104H -iAT2 treated with DMSO and #2035 (Figures 3F and 3G).Compound #2035 is identical to CPT (Figures 3H and S2D).These results suggest that CPT contributed to improving SFTPC aggregates caused by the Y104H-mutant SFTPC.

CPT reduces aggregates and cell death caused by Y104H SFTPC
To evaluate the effect of CPT on mutant SFTPC in more detail, we analyzed the aggregates of mutant SFTPC for two parameters (intensity and area of spot) and concentration dependence for CPT using A549 cells, which are generally used as lung lineage cells.Spot intensity was significantly higher in A549 cells transfected with AcGFP-Y104H SFTPC than in those transfected with AcGFP-wild SFTPC (FC: Y104H/Wild = 3.6) (Figure 4A).Total Spot Area of AcGFP was significantly smaller in A549 cells transfected with AcGFP-Y104H SFTPC than in those transfected with AcGFP-wild SFTPC (FC: Y104H/Wild = 0.5) (Figure 4B).These results indicate that the GFP spot formed by the transfection of AcGFP-Y104H SFTPC was stronger and smaller than that of AcGFP-wild SFTPC.In A549 cells transfected with AcGFP-Y104H SFTPC and treated with CPT, Spot Intensity decreased, and the Total Spot Area mildly increased in a concentration-dependent manner (IC50 for Spot Intensity: 1.21 mM) (Figures 4C and 4D).These results were similar in HEK293 cells (Figure S3A) and suggested that CPT could diffuse the aggregates of mutant SFTPC in A549 cells.Given the possibility that the decrease in protein levels causes the decrease in intensity, we performed WB to quantify the level of SFTPC protein.The amount of SFTPC proteins was not reduced by CPT treatment (Figure 4E).The mutation of the SFTPC BRICHOS domain can lead to the formation of abnormal intermolecular disulfide bonds, forming aberrant oligomers, which are associated with aggregation. 21Therefore, we investigated the aberrant oligomerization caused by Y104H SFTPC and whether CPT affects it.In WB analysis using electrophoresis under non-reducing conditions for AcGFP-Y104H SFTPC, many upward-shifted bands were detected compared with that of AcGFP-wild SFTPC.In particular, large disulfide-linked oligomeric species remained at the top of the gel (Figure S3B).Transfection efficiency was assumed to be almost the same in light of the similar expression level of the neomycin-resistance gene between wild and Y104H (Figure S3C).Since the formation of abnormal oligomerization by Y104H SFTPC was confirmed, we investigated the effect of CPT on this.Concurrently, we examined the impact of CPT on quality control systems such as the ubiquitin-proteasome system and autophagy using MG132 and Bafilomycin A1.No change in the band pattern in the WB analysis of AcGFP-Y104H SFTPC was observed by CPT treatment, either by CPT treatment under MG132 or Bafilomycin A1 (Figure S3D), indicating that CPT does not affect quality control mechanisms such as protein degradation or the formation of abnormal disulfide bonds during folding.These results support that CPT affects the formation or the dissolution of aggregates.Since aggregates of mutant SFTPC are known to result in cell death, 14,26 we hypothesized that cell death is reduced by suppressing SFTPC-aggregates using CPT treatment.To verify this hypothesis, we performed an assay for detecting activated caspase-3, an effector caspase that induces apoptosis.The ratio of activated caspase-3 positive cells was higher in A549 cells overexpressing AcGFP-Y104H SFTPC compared with A549 cells overexpressing AcGFP-wild SFTPC, indicating that the expression of Y104H SFTPC induces apoptosis in A549 cells [FC (Y104H/Wild) = 1.57] (Figure 4F).Furthermore, the number of apoptotic cells expressing AcGFP-Y104H SFTPC was significantly reduced by CPT treatment [FC (CPT/DMSO) of Y104H = 0.85], and there was no significant difference in cells expressing AcGFP-wild SFTPC (Figure 4G).These results were similar in HEK293 cells [FC (Y104H/Wild) = 2.27, FC (CPT/DMSO) of Y104H = 0.69] (Figures 4H and 4I).This indicates that CPT suppressed cell death caused by overexpression of the Y104H-mutant SFTPC.
Dose-dependent cytotoxic effect of CPT was confirmed by WST-8 assay.Compared with DMSO, 1.25 mM CPT showed no obvious growth inhibitory effect within 48 h.However, the growth and viability of A549 cells treated with CPT for 24-48 h were significantly inhibited at   S3E), consistent with previous reports. 27For HEK293 cells, CPT showed no noticeable growth inhibitory effect within 48h and at concentrations <5.0 mM (Figure S3F).

CPT ameliorates the disease phenotype of alveolar organoids derived from SPC Y104H -iPSC
To investigate the efficacy of CPT for the phenotypes of alveolar organoids generated from the patient-derived iPSC, we conceived to combine the in vitro pulmonary fibrosis model using the fibroblast-dependent alveolar organoids (FD-AOs) consisting of human pluripotent stem cell-derived alveolar epithelial cells (AEC) and human fetal lung fibroblasts (Fib) 28 and the patient-derived iPSC.First, to characterize FD-AOs, we performed RNA sequencing (RNA-seq) of AEC and Fib isolated from FD-AOs generated from SPC Y104H -and maeSPC Y104H -iPSC (hereinafter, these AEC and Fib isolated from FD-AOs generated from SPC Y104H -iPSC are called ''SPC Y104H -AEC'' and ''SPC Y104H -Fib'', and those generated from maeSPC Y104H -iPSC are called ''maeSPC Y104H -AEC'' and ''maeSPC Y104H -Fib'') (Figure 5A).We found 48 upregulated and 1 downregulated differentially expressed genes (DEGs) when comparing SPC Y104H -AEC with maeSPC Y104H -AEC (padj <0.1 and log2 FC > 1.5 or < À1.5), with no apparent differences in their co-cultured Fib (Figure 5B).In addition, there were 178 upregulated genes in SPC Y104H -AEC (padj <0.1 and log2 FC > 0.5), including several genes reportedly associated with lung fibrosis (Table S2).Next, to investigate the downstream consequences of SFTPC Y104H mutation, we performed Gene Set Enrichment Analysis (GSEA) on RNA-seq transcriptome data from SPC Y104H -and maeSPC Y104H -AEC, respectively, and found that various pathways of inflammatory responses were upregulated in SPC Y104H -AEC (Figure S4A).Furthermore, lung fibrosis-and some protein folding-related gene sets were upregulated in SPC Y104H -AEC (Figures 5C and S4B).These results indicate that the expression of Y104H SFTPC caused early transcriptomic changes, including the impairment of AEC and the pathway related to protein folding.Given the no significant changes in gene expression in Fib, it is suggested that these alterations in AEC were not sufficient to activate Fib.To analyze the effect of CPT on the impairment of AEC, we selected integrin subunit beta 6 (ITGB6), which is an epithelial-specific receptor, upregulated in response to epithelial injury and highly expressed in idiopathic pulmonary fibrosis lungs, particularly in AT2 cells in the abnormal epithelial cells called ''aberrant basaloid cells''. 28,29ITGB6 was significantly upregulated in SPC Y104H -AEC than maeSPC Y104H -AEC (Figure S4C).CPT reduced the mRNA level of ITGB6 in FD-AOs generated from SPC Y104H -iPSC in each of the three independent experiments (Figure S4D), suggesting that CPT may ameliorate the impairment of AEC associated with SFTPC mutation.
Furthermore, to examine the efficacy of CPT on alveolar organoid phenotypes, we enhanced the injury of AEC by supplementing bleomycin (BLM) and performed a gel contraction assay, which allows the evaluation of the contraction of the cultivation matrix dependent on the alveolar epithelial-mesenchymal interaction. 28SPC Y104H -iPSC-derived FD-AOs were prepared and subjected to each treatment with DMSO, BLM, or BLM and CPT (Figure 5D).BLM caused significantly stronger contraction of the gels than DMSO, and CPT suppressed the BLM-induced contraction (Figures 5E and 5F).Then, SPC Y104H -AEC were isolated from FD-AOs treated by DMSO, BLM, or BLM and CPT to ask whether the ameliorative effect of CPT on BLM-induced gel contraction is attributable to AEC.RT-qPCR revealed that the expression of ITGB6 in SPC Y104H -AEC was increased by BLM treatment and decreased by CPT treatment in each of the three independent experiments (Figure S4E), suggesting that CPT can suppress BLM-enhanced injury of AEC.

DISCUSSION
This study has shown that CPT is a potential therapeutic agent in the early stages of SFTPC-associated ILD.CPT was identified throughout the screening system based on two cellular phenotypes caused by mutant SFTPC, ER stress, and aggregate formation.The suppressive effect of CPT on mutant SFTPC aggregate formation was evaluated using AT2 cells differentiated from patient-derived iPSC and analyzed in detail using A549 cells.In addition, CPT ameliorated the global contraction of the matrix in FD-AOs with SFTPC mutation and decreased cell death of A549 cells and HEK293 cells overexpressing mutant SFTPC.
Growing evidence supports a link between ER stress and lung fibrosis. 30ER stress caused by SFTPC BRICHOS mutations has been linked to cytotoxicity and is associated with ILD pathogenesis. 9,13In transgenic mice inducibly expressing L188Q-mutant SFTPC in AT2 cells, induction of ER stress and exaggerated lung fibrosis after bleomycin treatment was observed. 13Therefore, we created a luminescent reporter to quantify ER stress and developed a high-throughput screening system using co-transfection with the reporter and SFTPC expression vector with an ILD-associated mutation, not a drug such as tunicamycin, to more closely mimic the pathological model of familial ILD.We created an ER stress reporter by utilizing cytoplasmic splicing of XBP1 but not the promoters of ER stress-inducible genes.This allowed us to normalize  S2.
the amount of spliced mRNA to the total mRNA expressed from the reporter and exclude compounds related to the expression system, such as direct promoter repression.
To narrow down the candidate chemical compounds for the specific pathogenic target, we focused on mutant SFTPC aggregates in addition to ER stress.Compounds that decrease ER stress and aggregates caused by mutant SFTPC are expected to correct the folding of unfolded mutant SFTPC or suppress the formation of aggregates, such as chemical chaperones. 31Therefore, a screening system was developed to rigorously quantify the amount of mutant SFTPC aggregates using high-content screening systems.Following these two screening systems, the compounds were evaluated using AT2 cells differentiated from patient-specific iPSC in an ALI culture.SFTPC accumulation in AT2 cells differentiated from iPSC was assessed by immunostaining.Immunostaining for SFTPC can only be used to evaluate AT2 cells because they specifically express SFTPC.These systems allowed us to identify compounds that can suppress the cytotoxicity of epithelial cells involved in the early stages of ILD.
We further evaluated the effect of CPT on the organoid phenotype.For fibrosis to develop, adult-onset mutations, such as the Y104H mutation, require a second hit, such as viral infection or bleomycin.In the mouse model, expression of L188Q SFTPC did not induce lung fibrosis without BLM stimulus.In the absence of a second hit, such mutations are likely to place AEC in a vulnerable state. 13,32,33In this study, ITGB6 and lung fibrosis-related gene sets are upregulated in SPC Y104H -AEC compared with maeSPC Y104H -AEC, suggesting that SPC Y104H -AEC are in a vulnerable state.As a second stimulus to mimic mouse models and patients, we administered BLM to FD-AOs generated from SPC Y104H -iPSC.Therefore, we confirmed the effect of CPT in the fibrosis model, which recapitulated the lung tissue of pulmonary fibrosis as the contraction of the cultivation matrix.Their feeder-Fib were derived from a distinct donor's fetal lung Fib; however, the analysis using the identical donor-derived mesenchymal cells, which we previously established, 34 might provide the model for the evaluation of drug efficacy more similar to in vivo.
As an example of the patient-derived iPSC-derived AEC harboring SFTPC mutations that cause pulmonary fibrosis, the I73T mutation has recently been reported, in which aberrant SFTPC trafficking, impaired autophagy, and metabolic reprogramming were observed. 35I73Tmutant-SFTPC is a non-BRICHOS mutation with no BRICHOS-mutant-SFTPC features, such as ER stress.The present study is the first report of drug evaluation using AEC differentiated from patient-derived iPSC harboring mutations in the BRICHOS domain of SFTPC.
Multiple inflammation-related gene sets were upregulated in SPC Y104H -AEC in GSEA, consistently with the enrichment analysis of AEC cells with I73T mutated SFTPC. 35Despite the different mechanisms implicated in cell toxicity between BRICHOS and non-BRICHOS mutations of SFTPC, similar gene expression changes in the same pathways were observed in the patient-derived AEC, suggesting that such common early epithelial transcriptomic changes may predict the future development of interstitial pneumonia and comparative analysis of alveolar organoids generated from multiple patient-derived iPSC with various SFTPC mutations would lead to elucidate the pathogenesis of ILD.Although we identified multiple genes with significant expression changes from the separated AEC and Fib in the present study, analyses at single-cell resolution may lead to a better understanding of ILD in the future.
CPT is a lipophilic compound extracted from the root of Salvia miltiorrhiza (Danshen).It has multiple pharmacological activities, including anti-tumor, anti-inflammatory, neuroprotective, and anti-metabolic effects. 36When considering CPT's mechanism of action in this study, we found that it ameliorated the formation or the dissolution of aggregates, because CPT reduced aggregation without affecting quality control mechanisms such as protein degradation and disulfide bond formation during folding.Mutations in SFTPA1/A2 and ABCA3, associated with ILD, have also been reported to cause protein misfolding. 14,37,38Accumulation of misfolded proteins is a common feature in many neurodegenerative diseases, including Parkinson's disease, Huntington's disease, and amyotrophic lateral sclerosis. 39Future research on the effects of CPT on these diseases is promising.CPT and the screening strategy used in this study can be applied to therapies targeting the initial alveolar epithelial injury involved in ILD and other protein-folding diseases.

Limitations of the study
Upregulation of protein folding-related genes in SPC Y104H -AEC suggested that the mutant SFTPC was abnormally folded; however, in AEC generated from SPC Y104H -iPSC, the expression of ER stress marker genes, including BiP and CHOP, did not increase.It is speculated that the activation of ER stress-related pathways by SFTPC mutant proteins was adapted during long-term culture. 40Viral infections such as flu may remove this adaptation necessitating us to develop the next evaluation system using patient-derived iPSC.

STAR+METHODS
Detailed methods are provided in the online version of this paper and include the following:

Western blotting (WB)
WB was performed according to the previously reported method. 47Proteins were extracted from A549 cells using a sample buffer (Nacalai Tesque), and the lysates were denatured at 95 C for 3 min.The lysates were then resolved by SDS-PAGE (SuperSep Ace gel; FUJIFILM Wako, Osaka, Japan) and transferred to a polyvinylidene difluoride membrane (Pall Corporation, Port Washington, NY, USA).Antibody reactions were performed using the Can Get Signal immunoreaction enhancer solution (Toyobo, Osaka, Japan).Immunoreactivity was visualized using Chemi-Lumi One Super (Nacalai Tesque) or ImmunoStar LD (FUJIFILM Wako) and ChemiDoc MP imaging system (Bio-Rad, Hercules, CA, USA).The antibodies used in this study are listed in the Antibody List.

Immunocytochemistry
Differentiated AT2 cells on a cell culture insert were removed with a membrane using biopsy punches (BP-50F) (Kai Corporation, Tokyo, Japan), fixed with 4% paraformaldehyde in PBS, and pre-incubated in PBS with 3% bovine serum albumin (BSA) (Sigma-Aldrich, St. Louis, MO, USA) and 0.1% Triton X-100 for 30 min at 20 C to 25 C.Primary antibodies in 3% BSA/PBS were applied overnight at 4 C, followed by 1 h at 20 C to 25 C with the secondary antibodies.The antibodies used in this study are listed in the antibody list.The nuclei were stained with Hoechst 33342 (Nacalai Tesque).The cells were observed using a fluorescence microscope (model BZX710; Keyence).Analysis was performed using three independent samples.A TCS SP8 (Leica Microsystems) was used for confocal imaging.

HiBiT assay
For the HiBiT assay, HEK293 cells (4.0 3 10^5 cells/mL) were seeded and transfected overnight with the XBP1-HiBiT Reporter and SFTPC expression vector using the FuGENE HD Transfection Reagent (Promega).Four hours after transfection, the cells were incubated with each compound of the function-known chemical library provided from Medical Research Support Center in Kyoto University (final concentration was 10 mM) or DMSO for 24 h, followed by reseeding into 96-well PureCoat amine-coated plates (Corning, NY, USA) (4.0 3 10^4 cells/mL).
To validate the quantifiability of the reporter, tunicamycin (FUJIFILM Wako) was used at 0.6, 1.25, 2.5, or 5.0 mg/mL for 6 h.Luminescence was subsequently measured using the Nano-Glo HiBiT Lytic Detection System according to the manufacturer's instructions (Promega).Relative luminescence units (RLUs) were measured using ARVO (PerkinElmer, Waltham, MA, USA).For cells treated in a similar manner, the WST-8 assay was performed by using Cell Count Reagent SF (Nacalai Tesque).
9][50] Briefly, after the consents were obtained from the patient, PBMCs were isolated using BD Vacutainer Cell Preparation tubes (BD Biosciences) and cryopreserved.Thawed PBMCs were reprogrammed into iPSC clones with episomal plasmids (pCE-hOCT3/4, pCE-hSK, pCE-hUL, pCE-mp53DD, and pCXB-EBNA1), under feeder-free culture in StemFit AK02N medium on iMatrix-511-coated plates (Nippi, Tokyo, Japan) as described previously. 49The established iPSC clones were cryopreserved in Stem Cell Banker (Takara Bio) and maintained in StemFit AK02N medium supplemented with 50 U/mL penicillin-streptomycin (Thermo Fisher Scientific).The established SPC Y104H -iPSC clone (Clone #CiRA00995-F) was evaluated using a STEMdiff Trilineage Differentiation Kit (STEMCELL Technologies, Vancouver, Canada) as described previously. 48In brief, the dissociated cells were reseeded onto culture plates at 2.0 3 10^5 cells for endoderm differentiation and 4.0 3 10^5 cells for ectoderm and mesoderm differentiation, respectively and were differentiated, according to the manufacturer's instructions.Each sample of three germ layers (1.0x10^6 cells per sample) was fixed with 4% paraformaldehyde in PBS for 20 min at 4 C and washed twice with 2%FBS/PBS.Samples were permeabilized with BD Perm/Wash buffer (BD Biosciences) for 15 min at room temperature and stained with the fluorescence-conjugated antibodies listed in the Antibody List.Then, they were washed with BD Perm/Wash buffer and subjected to flowcytometric analysis using LSR (BD Biosciences).The data was analyzed with FlowJo software (FlowJo, LLC, Ashland, OR, USA).G-banding analysis was performed at Nihon Gene Research Laboratories (Sendai, Miyagi, Japan).

Gene edition of SPC Y104H iPSC
After SPC Y104H iPSC were maintained on iMatrix-511-coated plates in StemFit AK02N medium (Ajinomoto, Tokyo, Japan), Y-27632 was supplemented at 10 mM for at least 1 h before electroporation.The cells were dissociated into single cells using Accutase (Innovative Cell Technologies, Inc., San Diego, CA, USA) at 37 C for 20 min and neutralized with DMEM/F12 (Nacalai Tesque) supplemented with 2% FBS (Sigma-Aldrich).Three plasmid vectors comprising 5 mg Cas9, 5 mg sgRNA, and 5 mg donor template with a neomycin-resistant gene cassette were simultaneously electroporated into 1.0 x 10^6 cells using a NEPA 21 electroporator (poring pulse: pulse voltage, 125 V; pulse width, 5 ms; pulse number, 2) (Nepagene, Chiba, Japan), as previously reported. 16,51The cells were then reseeded on iMatrix-511-coated plates in StemFit AK02N medium supplemented with 10 mM Y-27632 for 1-2 day.Two days after electroporation, G418 (Gibco BRL) selection was initiated at 100 mg/ml.After selection for 7-10 day, the cells were dissociated and replated at limiting dilutions of 200, 500, and 1,500 cells per iMatrix-511-coated 10-cm dish.Subclones were selected and screened using genomic PCR.When homologous recombination was suspected, Sanger sequencing was performed.Finally, the maeSPC Y104H iPSC (Clone #CiRA00995-F-Res10-2) having gene-edited SFTPC with an intronic insertion of a neomycin cassette flanked by LoxP sites.A clone with a normal karyotype was selected for subsequent G-band chromosomal analysis.The maeSPC Y104H iPSC were verified to be derived from the parental iPS cells using short tandem repeat analysis.The

QUANTIFICATION AND STATISTICAL ANALYSIS
Values are presented as mean G SD. Statistical significance was evaluated with a two-tailed Student's or Welch's t-test to analyze differences between two experimental groups (p < 0.05 was considered significant).In Figures 1B, 4C, 4E, 5F, S1, S3A, S3E, and S3F statistical significance was evaluated with one-way ANOVA with Tukey's multiple comparisons tests (statistical significance was set at p < 0.05).

Figure 1 .
Figure 1.High-throughput screening and identification of compounds that reduce ER stress caused by mutant SFTPC (A) Diagram illustrating the construction of XBP1-HiBiT Reporter.It was spliced under ER stress and resulted in the translation of a spliced XBP1(exon 4)-HiBiT fusion protein.The STOP sign indicates the stop codon of the open reading frame of unspliced mRNA.(B) Cellular HiBiT assay using the Nano-Glo HiBiT Lytic Detection System for HEK293 cells transfected with XBP1-HiBiT Reporter and treated with tunicamycin (0.6, 1.25, 2.5 or 5.0 mg/mL) or DMSO (0.05%) for 6 h (n = 6).Data are presented as mean G SD. ***p < 0.001 [One-way analysis of variant (ANOVA) with Tukey's multiple comparisons test].(C) Diagram illustrating the screening method.(D) Validation of high-throughput screening using the cellular HiBiT assay.HEK293 cells were co-transfected with XBP1-HiBiT Reporter and empty or L188Qmutant SFTPC expression vector (L188Q SFTPC).n = 80 (L188Q SFTPC), n = 12 (empty vector).(E) RT-qPCR of the splicing ratio of XBP1-HiBiT Reporter (spliced/total expression) in HEK293 cells subjected to 24 h incubation with co-transfected XBP1-HiBiT Reporter and wild or L188Q SFTPC expression vectors.Data are presented as mean G SD (n = 3 independent experiments).**p < 0.01 (Welch's t test).(F) Scatterplot for relative luminescence units (RLU) (% of DMSO) analyzed by XBP1-HiBiT Reporter co-transfected with L188Q SFTPC expression vector in the screening of 2480 compounds.Data are presented as mean (n = 2).(G) Scatterplot showing pairwise comparisons of the result of WST-8 assay (n = 2) (% of DMSO) versus the result of HiBiT assay (from Figure 1F) in the screening of the 2480 compounds.Red and blue characters and lines indicate each threshold to identify hit compounds.See also Figure S1.

Figure 2 .
Figure 2. Secondary screening for compounds identified in the primary screening (A) RT-qPCR of the splicing ratio of XBP1-HiBiT Reporter (spliced/total expression) in HEK293 cells subjected to 24 h incubation with co-transfected XBP1-HiBiT Reporter and wild or Y104H SFTPC expression vectors.Data are presented as mean G SD (n = 3 independent experiments).***p < 0.001 (Student's t test).(B and C) The means of Spot Intensity (B) and Total Spot Area (C) for HEK293 cells transfected with the AcGFP-wild SFTPC or AcGFP-Y104H SFTPC (n = 4, each sample is averaged over 81 field-of-view 3 3 well).Data are presented as mean G SD. ***p < 0.001 (Student's t test).(D) Representative microscopic images of HEK293 cells transfected with the AcGFP-wild SFTPC or AcGFP-Y104H SFTPC.Nuclei were visualized by Hoechst 33342 (blue).Scale bar = 100 mm.(E) Validation of screening using cellular HiBiT assay.HEK293 cells were co-transfected with XBP1-HiBiT Reporter and empty or Y104H-mutated SFTPC expression vector (Y104H SFTPC).n = 80 (Y104H SFTPC), n = 12 (empty vector).(F) Scatterplot showing pairwise comparisons of the result of the means of Spot Intensity (% of DMSO) for HEK293 cells transfected with the AcGFP-Y104H SFTPC (Data are presented as the mean of n = 2, each sample is averaged over 81 field-of-view) versus the means of RLU (% of DMSO) analyzed by XBP1-HiBiT Reporter co-transfected with Y104H SFTPC in the screening of 65 compounds identified in primary screening (n = 2).Red points indicate the compounds which have passed this screening.(G) Analysis of the splicing ratio of XBP1-HiBiT Reporter (spliced/total expression) by RT-qPCR in HEK293 cells subjected to 4 h incubation with co-transfected XBP1-HiBiT Reporter and Y104H SFTPC expression vectors followed by treatment with each of 9 compounds for 24 h at the indicated concentration (n = 2).Data are presented as mean G SD. Red points indicate the compounds which have passed this screening.The Red line is FC = 0.75.See also Figure S1.

Figure 3 .
Figure 3. Efficacy evaluation of compounds using alveolar epithelial cells differentiated from the patient-derived iPSC (A) Schematic overview of generation of the patient-derived iPSC (SPC Y104H -iPSC) and their isogenic control of wild-type SFTPC monoallelic expressing iPSC (maeSPC Y104H -iPSC) and their stepwise differentiation into alveolar epithelial cells containing AT2 cells.(B) The sequence analysis of the RT-PCR product from SFTPC mRNA at 6 days after induced differentiation into AT2 cells from SPC Y104H -and maeSPC Y104H -iPSC.Red arrows indicate the position of the Y104H-mutation (c.310T>C).(C) G-banding analysis for the SPC Y104H iPSC and maeSPC Y104H -iPSC karyotypes.(D and E) Immunological fluorescence analysis for quantifying SFTPC accumulation in SPC Y104H -iAT2 (D) and maeSPC Y104H -iAT2 (E) cells differentiated from SPC Y104H iPSC and maeSPC Y104H iPSC.The iAT2 cells in each condition were treated with indicated compounds for 48 h post-7-day differentiation.The fluorescence signal intensity of the high-intensity area of SFTPC staining was quantified and normalized by the fluorescence area.Data are presented as mean G SD (n = 3 independent experiments).*p < 0.05 (Student's or Welch's t test, compared with DMSO).(F) Comparative analysis of SPC Y104H -iAT2 and maeSPC Y104H -iAT2 cells treated with CPT based on Figures 3D and 3E results.*p < 0.05, **p < 0.01 (Student's or Welch's t test, compared with SPC Y104H -iAT2 treated by DMSO).(G) Representative image of immunostaining for SFTPC (green) and nuclear counterstaining with Hoechst 33342 (blue) in SPC Y104H -iAT2 and maeSPC Y104H -iAT2 cells treated with CPT for 48 h post-7-day differentiation.Scale bar = 100 mm.(H) Structure of Cryptotanshinone (CPT), compound number #2035.See also Figure S2 and TableS1.

Figure 4 .
Figure 4. CPT reduces aggregates and cell death caused by Y104H SFTPC (A and B) The means of Spot Intensity of AcGFP in each cell (A) and Total Spot Area of AcGFP in each cell (B) were analyzed for A549 cells transfected with the AcGFP-wild SFTPC or AcGFP-Y104H SFTPC for 48 h (n = 3, each sample is averaged over 81 fields of view).Data are presented as the mean G SD. ***p < 0.001 (Student's t test).(C) Analysis of Spot Intensity and Total Spot Area of A549 cells transfected with the AcGFP-Y104H SFTPC for 4 h and subsequently treated with CPT (1.25, 2.5, or 5 mM) or DMSO (0.1%) for 48 h (n = 3, each sample is averaged over 81 fields of view).Data are presented as the mean G SD. *p < 0.05, **p < 0.01, ***p < 0.001 (One-way ANOVA with Tukey's multiple comparisons test, compared with DMSO control).(D) Representative image of A549 cells transfected with AcGFP-Y104H SFTPC for 4 h and subsequently treated with CPT (2.5 mM) or DMSO (0.1%) for 48 h.Nuclei were visualized by Hoechst 33342 (blue).Scale bar = 50 mm.(E) Representative results of WB using anti-SFTPC antibody on A549 cells transfected with the AcGFP-wild SFTPC or AcGFP-Y104H SFTPC for 4 h and subsequently treated with CPT (1.25, 2.5, or 5 mM) or DMSO (0.1%) for 48 h.a-tubulin was used as a loading control (Left panel).The signal intensities were

Figure 4 .
Figure 4. Continued quantified using densitometry (Right panel).Three independent experiments were conducted, and the data are presented as the mean G SD. *p < 0.05 (Student's t test, compared with DMSO control).(F) The number of double-positive cells of activated caspase-3 and SFTPC-GFP in A549 cells transfected with the AcGFP-wild SFTPC or AcGFP-Y104H SFTPC for 24 h and calculated by normalizing the number of GFP-positive cells (n = 3, each sample is averaged over 169 fields of view).Data are presented as the mean G SD. **p < 0.01 (Student's t test).(G) The number of double-positive cells of activated caspase-3 and SFTPC-GFP in A549 cells transfected with the AcGFP-Y104H SFTPC (Left graph) or AcGFPwild SFTPC (Right graph) for 4 h and subsequently treated with 1.25 mM CPT or DMSO (0.1%) for 24 h, and calculated by normalizing the total cell number (n = 3, each sample is averaged over 169 fields of view).Data are presented as the mean G SD. *p < 0.05 (Student's t test).(H) The number of double-positive cells of activated caspase-3 and SFTPC-GFP in HEK293 cells transfected with the AcGFP-wild SFTPC or AcGFP-Y104H SFTPC for 24 h and calculated by normalizing the number of GFP-positive cells (n = 3, each sample is averaged over 169 fields of view).Data are presented as the mean G SD. ***p < 0.001 (Student's t test).(I) The number of double-positive cells of activated caspase-3 and SFTPC-GFP in HEK293 cells transfected with the AcGFP-Y104H SFTPC (Left graph) or AcGFPwild SFTPC (Right graph) for 4 h and subsequently treated with 2.5 mM CPT or DMSO (0.1%) for 24 h, and calculated by normalizing the total cell number (n = 3, each sample is averaged over 169 fields of view).Data are presented as the mean G SD. ***p < 0.001 (Student's t test).See also Figure S3.

Figure 5 .
Figure 5. Characterization of FD-AOs generated from SPC Y104H iPSC and its application to BLM-induced contraction assay for validating the efficacy of CPT (A) Schematic overview of generation of the fibroblast-dependent alveolar organoids (FD-AOs) and RNA-seq analysis.(B) Volcano plot generated from the results of DESeq2 analysis.Left panel: SPC Y104H -AEC vs. maeSPC Y104H -AEC, Right panel: SPC Y104H -Fib vs. maeSPC Y104H -Fib (n = 3 independent experiments, respectively).AEC and Fib were separated from FD-AOs derived from iPSC, respectively.Red character indicates the number of significantly upregulated genes and blue character indicates the number of significantly downregulated genes.(C) GSEA enrichment plot of the ranked gene expression data by ''WP LUNG FIBROSIS'' gene set in SPC Y104H -and maeSPC Y104H -AEC.NES, normalized enrichment score; NOM, nominal; FDR, false discovery rate.(D) Schematic overview of the generation of FD-AOs and their application to the bleomycin (BLM)-induced pulmonary fibrosis model to analyze the efficacy of CPT.(E) Representative whole-well imaging of the cultivation matrices at day 17.Each well was treated with BLM, BLM, and 10 mM CPT or DMSO from day 11 to day 14 and with 10 mM CPT or DMSO from day 14 to day 17.Scale bars, 2 mm.(F) Quantifying matrix area changes on Day 17 (% of Day 14, n = 3 independent experiments, each sample is averaged over three wells).Data are presented as the mean G SD. **p < 0.01 (One-way ANOVA with Tukey's multiple comparisons test).See also Figure S4 and TableS2.

TABLE
d RESOURCE AVAILABILITY B Lead contact B Materials availability B Data and code availability d EXPERIMENTAL MODEL AND STUDY PARTICIPANT DETAILS B Ethics B Cell culture and transfection B Plasmids (Continued on next page)