Lung organoids derived from AdSCs are attractive models for investigating epithelial stem cell potential and cellular interactions during homeostasis and disease. Both murine and human AdSC-derived organoid models have been developed from several different stem cells present along the bronchoalveolar epithelium (Figure 1) and can therefore be used to recreate specific lung environments.

Depending on the site of stem cell isolation, human basal cells can form organoids called tracheospheres (trachea), bronchospheres (large airways), or nasospheres (nasal epithelium) comprising basal, club, ciliated, and goblet cells (31–33). Tracheospheres containing basal cells and secretory and ciliated cells have also been generated from isolated murine basal cells (10). These airway models have been used to provide insights into epithelial development and signaling. For instance, SMAD and NOTCH signaling have been widely implicated in the regulation of airway stem cell proliferation and differentiation (34, 35). Distinct media supplements have been shown to modulate SMAD/BMP and NOTCH signaling during AdSC-derived organoid generation by either promoting basal cell proliferation or driving cell differentiation toward secretory and ciliated cells (30, 32, 36, 37).

Moreover, several studies have now shown that AdSC-derived organoids can model crosstalk between stem cells and their microenvironment (38–40). For instance, coculture of club cells and mesenchyme subpopulations has been used to assess mesenchymal cells’ ability to support epithelial stem cell potential. Under these conditions, cocultures gave rise to organoids called bronchiolospheres containing club and ciliated cells and revealed a distinct mesenchymal cell subset driving stem cell growth and differentiation (38, 39). Bronchiolospheres lack basal or goblet cells and are more suitable for the study of club and ciliated cell biology (39).

In 2005, BASCs were first identified at the BADJ as stem cells resistant to bronchiolar and alveolar injury in vivo (11). BASCs were found to coexpress bronchiolar and alveolar cell markers (surfactant protein C [SFTPC] and secretoglobin family 1A member 1 [SCGB1A1], respectively) and possessed self-renewal and multipotency capacities in vitro, suggesting a potential BASC contribution to club and alveolar cell maintenance (11). In line with these findings, Lee and colleagues developed a coculture system using BASCs and lung endothelial cells that gave rise to alveolar, bronchiolar, and bronchoalveolar organoids (41). In this study, the endothelium-derived BMP4/NFATc1/thrombospondin-1 (TSP1) signaling axis increased BASC proliferation and differentiation toward alveolar phenotypes. Notably, these bronchioalveolar organoids consisted of AEC2, club, ciliated, and goblet cells, but AEC1s were absent (41). Direct evidence of BASCs’ multilineage differentiation and contribution to bronchioalveolar repair after naphthalene-, bleomycin-, and influenza virus–induced injury was recently obtained using lineage tracing approaches that allowed selective labeling of SFTPC+SCGB1A1+ BASCs in the mouse lung (12, 42). Moreover, BASC coculture with resident mesenchymal cells generated complex bronchoalveolar lung organoids (BALOs) (26). These organoids were characterized by formation of tubular airway-like regions containing mature basal, secretory, and multiciliated cells, whereas distal alveolar-like areas comprised differentiated AEC1s and AEC2s capable of producing surfactant (26).

Our understanding of alveolar biology has also improved through organoids derived from murine and human AEC2s that were first described by Barkauskas et al. in 2013 (13). Using in vivo and in vitro experiments, this study revealed AEC2s as the main stem cell of the alveoli. In this organoid model, coculture of AEC2s with PDGFRα+ mesenchymal cells led to the formation of alveolospheres containing AEC2s and AEC1s (13). Since then, multiple organoid models using AEC2 subsets and mesenchymal cell subpopulations have been established to investigate alveolar stem cell niche interactions (43–46). More recently, alveolospheres have been used to investigate the effect of BMP signaling on AEC2 proliferation. In this setting, activation of BMP led to the reduction of AEC2 proliferation and increased AEC1 differentiation, while its inhibition promoted AEC2 self-renewal, suggesting that BMP modulation is crucial for alveolar niche maintenance (47).

Additional efforts have been made to improve the lifespan and cell differentiation of organoids derived from patients’ stem cells. Sachs et al. developed a long-term human airway organoid model from human bronchoalveolar lavage and resection material. Through modulation of TGF-β, FGF, and WNT signaling, airway organoids derived from basal cells were shown to contain basal, ciliated, and secretory/club cells that could be maintained for over a year (48). Their protocol was further optimized by Noggin (NOG) removal and DAPT/BMP4 addition into the standard airway organoid medium. Under these conditions, airway organoids generated from nasal inferior turbinate brush samples yielded significantly higher numbers of ciliated cells and could be used to model primary ciliary dyskinesia (49). In the alveoli, characterization of the WNT-responsive alveolar epithelial progenitor (AEP) subset in mice led to discovery of TM4SF1 as a surface marker for human AEP cells (14). Interestingly, TM4SF1+ AEPs isolated from human samples gave rise to functional alveolospheres when cocultured with a human fetal lung fibroblast cell line (14). In addition, Tran and colleagues created a human AEC2 immortalized cell line using SV40 large T antigen lentiviral transfection and Y-27632 (a RHO/ROCK pathway inhibitor) media supplementation. After initial 2-dimensional expansion, AEC2s cocultured with a mouse lung fibroblast cell line formed alveolospheres expressing AEC1 and AEC2 markers (50).

Innovations in organoid derivation protocols have led to the use of FSC-derived lung organoids for studies on lineage specification and cellular interactions during development and disease (Figure 2A). In 2017, Nikolić and colleagues developed organoids from human and mouse fetal lung bud tips (LBTs) (24). LBT-derived organoids were grown in medium containing seven factors known to modulate crucial signaling pathways involved in lung morphogenesis, including growth factors (EGF, FGF-7, and FGF-10), BMP signaling inhibitors (NOG and SB431542), and WNT signaling activators (RSPO1 and CHIR) (24). Through this combination, organoids retained expression of the lung-specific transcription factor NK2 homeobox 1 (NKX2-1) and coexpressed lung progenitor markers SRY-box transcription factor 2 (SOX2) and SOX9 but did not contain differentiated cell types. Commercial human airway medium was then used to drive differentiation within LBT-derived organoids toward more specialized bronchiolar lineages such as goblet, basal, and ciliated cells (24). Conversely, derivation of LBT organoids into alveolar lineages was partially achieved through coculture with freshly sorted human lung mesenchyme in medium containing FGF-7, FGF-10, CHIR, DCI (a combination of dexamethasone, cAMP, and the intracellular cAMP activator IBMX), triiodo-l-thyronine (T3), and the NOTCH inhibitor DAPT (24). These settings resulted in the formation of organoids containing AECs coexpressing SFTPC and homeodomain-only protein (HOPX) (24). Single-cell RNA sequencing (scRNA-Seq) analysis of LBT organoids also identified SMAD signaling as a major cue for airway patterning during lung development (51). In this system, SMAD activation was promoted by addition of TGF-β and BMP4 during the first days of culture followed by prolonged SMAD inhibition through A8301 and NOG addition (51). This dual signaling modulation led to the formation of proximal airway organoids composed of clonal basal cells capable of self-renewal and multilineage differentiation into basal-, goblet-, club-, and ciliated-like cells (51).

Besides FSC-derived organoids’ utility in elucidating cell fate decisions, these models can also be valuable tools to uncover novel mesenchymal-epithelial cell interactions during lung development. Transcriptional and spatial profiling of human developing lungs recently revealed a distinct mesenchymal cell population in the LBT that was enriched for the WNT agonist R-Spondin 2 (RSPO2) (52). When cocultivated with isolated LBT-derived epithelial cells in media containing FGF-7 and ATRA but not CHIR, LIFR+ (RSPO2+) mesenchyme induced organoid proliferation and alveolar lineage derivation into SFTPC+ and RAGE+ AECs (52). Alternatively, coculture with LIFR– mesenchyme increased airway marker secretoglobin family 3A member 2 (SCGB3A2) expression, implying that RSPO2+ mesenchyme provides a niche signal for distal bud tip progenitor maintenance and multipotency by supporting WNT signaling activation (52). In line with these findings, NOTUM, a WNT inhibitor, was expressed in actin alpha 2–positive (ACTA2+) myofibroblasts in the distal LBT (53). Coculture of NOTUM-overexpressing fibroblasts and alveolar-like LBT organoids led to loss of SFTPC expression, suggesting that NOTUM+ myofibroblasts control alveolar cell fate during lung development by blocking WNT signaling activation in the distal tip epithelium (53). These recent studies not only uncovered novel mesenchymal-epithelial crosstalk during lung morphogenesis but also provide insights into how these systems can be further directed to mimic developmental processes in vitro.

The establishment of novel protocols to generate human iPSC–derived lung organoids represents a major advance in pulmonary disease modeling, drug screening, and regenerative medicine (Figure 2B). Generally, through addition of specific factors into the culture media, murine and human iPSCs are first directed toward definitive endoderm, followed by sequential generation of anterior foregut endoderm (AFE), ventral anterior foregut endoderm cells (VAFECs), and, lastly, NKX2-1+ lung progenitors (54–56). Coculture of an SFTPC-GFP reporter human iPSC cell line and human fetal lung mesenchyme showed for the first time that carboxypeptidase M–positive (CPM+) surface marker could be employed to isolate lung progenitors from VAFECs (57). Notably, organoids derived from VAFECs contained NKX2-1– and CPM-coexpressing cells, as well as AEC1s and AEC2s (57). Accordingly, Konishi and colleagues used CPM as a surface marker for NKX2-1+ VAFEC isolation. Formed organoids could be driven toward airway-like organoids comprising multiciliated cells by addition of DAPT into the medium (58).

A primary objective of the subsequent investigations with iPSCs has been to optimize media conditions to promote airway and alveolar cell maturity. In line with this goal, gene editing of human iPSC cell lines using TALENs or CRISPR/Cas9 technologies enabled more detailed characterization and isolation of NKX2-1 lung progenitors (28, 29). Airway organoids containing functional basal cells were achieved using a novel NKX2-1GFP/tumor protein P63 (TP63)tom dual fluorescent reporter iPSC line (28). Initially, NKX2-1 lung progenitors were purified based on GFP expression and cultured in medium containing FGF-2, FGF-10, DCI, and the RHO/ROCK pathway inhibitor Y-27632 (28).

After a month in culture, NKX2-1+TP63+ cells were isolated and seeded in commercially available PneumaCult-Ex Plus medium (Stemcell Technologies) along with SMAD and ROCK inhibitors. Formed organoids contained basal cells that showed clonal expansion and multilineage differentiation into basal-, club-, and ciliated-like cells when cultured in air-liquid interface conditions (28). Nonetheless, the need for reporter lines for stem cell isolation limits the system applicability and translation into clinical scenarios. In the mentioned study, the use of fluorescent reporters could be replaced by surface expression of CD47, CD26, nerve growth factor receptor (NGFR), and epithelial cell adhesion molecule (EPCAM) cell markers, facilitating prospective basal cell isolation from patient-specific iPSCs for clinical applications (28).

In a comparative study, cells solely isolated based on CPM expression contained more NKX2-1+ progenitors than cells expressing CD47 and/or CD26, indicating that further optimization of culture conditions is necessary to obtain the desired cellular differentiation (59). In this regard, similar studies developed human alveolospheres by combining NKX2-1 reporter iPSC lines and media composition optimization (60, 61). For instance, culture of sorted NKX2-1–GFP+ cells in CHIR, FGF-10, and FGF-7 media followed by addition of DCI gave rise to organoids coexpressing NKX2-1 and AEC2 markers (29). Coculture of NKX2-1+ cells with distal embryonic mesenchyme resulted in higher levels of SFTPC expression, suggesting that distal fibroblasts induce alveolar differentiation during tip specification (29). In another related study, human iPSC derivation toward organoids with AEC lineages was directed by preconditioning of NKX2-1+ VAFECs with medium containing CHIR, FGF-10, FGF-7, and DAPT. This treatment led VAFECs to express higher levels of CPM, which facilitated isolation and coculture of CPMhi cells with fetal mesenchymal cells that give rise to organoids comprising AEC2s and AEC1s (60). Notably, fibroblast-free alveolar organoids with AEC2s could be obtained by DCI, FGF-7, CHIR, Y-27632, and SB431542 media supplementation (60).

Moreover, based on previous studies, NKX2-1GFPSFTPCtom dual reporter iPSC lines were first cultured in medium containing CHIR, BMP4, and ATRA (a combination called CBRa) to acquire NKX2-1GFP progenitors while FGF-7, CHIR, and DCI supplementation was used for subsequent NKX2-1GFPSFTPCtom alveolar organoid formation (61, 62). Nevertheless, this culture setting has substantial contamination by gastric cells (~20%), likely due to early presence of mid- and hind-gut cells (63, 64). Lastly, airway organoids derived from NKX2-1GFP cells were developed by withdrawal of CHIR from the medium, further supporting other studies indicating that WNT signaling activation promotes AEC2 proliferation while suppressing proximal lineage differentiation (65, 66). While combination of reporter iPSC lines and standardization of media components represents a valuable strategy for organoid formation and cell lineage specification, upcoming iPSC-derived lung organoid systems should be generated without the need of reporter cell lines to broaden these models’ clinical applicability.

While iPSC-derived AEC2s have the advantage of high throughput and scalability over AdSCs, the latter cells denote age-associated maturity and maintain genetic and epigenetic characteristics of the donor or patient lungs. Over the years, efforts have been made to comparatively assess the self-renewal, maturity, and differentiation capacity of iPSC-derived and primary AEC2s in ex vivo models (67, 68). In the case of iPSC-derived AEC2s, inhibition of WNT signaling using XAV939 (tankyrase inhibitor) led to AEC1 differentiation (68, 69). However, WNT withdrawal from the culture medium was not sufficient to induce AEC1 differentiation of both iPSC-derived and primary AEC2s (68, 70). Studies in mice have revealed that YAP signaling is highly enriched in AEC1s and ectopic activation of YAP maintains AEC1 cell identity and is sufficient to promote AEC2 to AEC1 differentiation (71–76). To test whether activation of YAP signaling can promote iPSC-derived AEC2 differentiation, Burgess and colleagues used both genetic and pharmacological activation approaches. Indeed, these studies revealed that activation of YAP signaling was sufficient to induce AEC1 gene expression in iPSC-derived AEC2s (77). Such studies are still required to assess FSC-derived AEC2 potential in organoid cultures. In the case of AdSCs, addition of human serum induced AEC2 to AEC1 differentiation (43). However, the specific components of the human serum capable of inducing AEC2 differentiation are not known, suggesting that identification of such molecules may have therapeutic value for both cell replacement and regenerative therapies.

Current human iPSC models contain epithelial cells found on the alveolar or airway lung compartments, but only a few systems feature proximal to distal patterning. In this regard, SOX9+SOX2+ lung bud organoids (LBOs) that resemble LBT cellular composition were generated after induction of VAFECs in the presence of FGF-10, FGF-7, and CBRa (78). Although LBOs did not exhibit mature airway cells or AEC1s, they contained goblet and club cells in the proximal structures and AEC2s in the distal tips (78). In another study, Miller and colleagues developed human lung organoids (HLOs) and bud tip organoids (BTOs) from iPSC-derived foregut spheroids (25). HLOs were created by culture in high levels of FGF-10 and included TP63+ and FOXJ1+ airway-like epithelium surrounded by a diffuse network of mesenchymal cells and epithelial cells coexpressing SFTPC and HOPX alveolar markers. Conversely, lung progenitors cultured in FGF-7, CHIR, and ATRA developed BTOs with SOX2+, mucin 5AC–positive (MUC5AC+), and SCGB1A1+ airway-like regions and SOX2+, SOX9+, SFTPC+, and ID2+ bud tip–like domains (25). In a follow-up protocol, NKX2-1+ bud tip progenitor–like cells coexpressing SOX9, SOX2, and CPM were enriched to promote organoid formation. Following digestion, cell sorting, and culture of NKX2-1GFP–expressing cells, bud tip progenitor–like cells expanded, generating spheroids with a robust NKX2-1 expression (27). Using previously published protocols, these spheroids could be efficiently directed toward alveolar or airway organoids (51, 61). Nonetheless, additional culture optimization is still necessary to establish human iPSC–derived organoids that more accurately recapitulate the lung architecture.