Recapitulating human cardio-pulmonary co-development using simultaneous multilineage differentiation of pluripotent stem cells

  1. Wai Hoe Ng
  2. Elizabeth K Johnston
  3. Jun Jie Tan
  4. Jacqueline M Bliley
  5. Adam W Feinberg
  6. Donna B Stolz
  7. Ming Sun
  8. Piyumi Wijesekara
  9. Finn Hawkins
  10. Darrell N Kotton
  11. Xi Ren  Is a corresponding author
  1. Department of Biomedical Engineering, Carnegie Mellon University, United States
  2. Advanced Medical and Dental Institute, Universiti Sains Malaysia, Malaysia
  3. Department of Materials Science and Engineering, Carnegie Mellon University, United States
  4. Center for Biologic Imaging, University of Pittsburgh, United States
  5. Center for Regenerative Medicine of Boston University and Boston Medical Center, United States

Abstract

The extensive crosstalk between the developing heart and lung is critical to their proper morphogenesis and maturation. However, there remains a lack of models that investigate the critical cardio-pulmonary mutual interaction during human embryogenesis. Here, we reported a novel stepwise strategy for directing the simultaneous induction of both mesoderm-derived cardiac and endoderm-derived lung epithelial lineages within a single differentiation of human-induced pluripotent stem cells (hiPSCs) via temporal specific tuning of WNT and nodal signaling in the absence of exogenous growth factors. Using 3D suspension culture, we established concentric cardio-pulmonary micro-Tissues (μTs), and expedited alveolar maturation in the presence of cardiac accompaniment. Upon withdrawal of WNT agonist, the cardiac and pulmonary components within each dual-lineage μT effectively segregated from each other with concurrent initiation of cardiac contraction. We expect that our multilineage differentiation model will offer an experimentally tractable system for investigating human cardio-pulmonary interaction and tissue boundary formation during embryogenesis.

Editor's evaluation

The study responded very well to the expert reviewers and offers new insights into mechanisms regulating differentiation of cardiac and pulmonary stem cells. It will stimulate further investigations into this important field.

https://doi.org/10.7554/eLife.67872.sa0

eLife digest

Organs begin developing during the first few months of pregnancy, while the baby is still an embryo. These early stages of development are known as embryogenesis – a tightly organized process, during which the embryo forms different layers of stem cells. These cells can be activated to turn into a particular type of cell, such as a heart or a lung cell.

The heart and lungs develop from different layers within the embryo, which must communicate with each other for the organs to form correctly. For example, chemical signals can be released from and travel between layers of the embryo, activating processes inside cells located in the different areas.

In mouse models, chemical signals and cells travel between developing heart and lung, which helps both organs to form into the correct structure. But it is unclear how well the observations from mouse models translate to heart and lung development in humans.

To find out more, Ng et al. developed a human model of heart and lung co-development during embryogenesis using human pluripotent stem cells. The laboratory-grown stem cells were treated with chemical signals, causing them to form different layers that developed into early forms of heart and lung cells.

The cells were then transferred into a specific growing condition, where they arranged into three-dimensional structures termed microtissues. Ng et al. found that lung cells developed faster when grown in microtissues with accompanying developing heart cells compared to microtissues containing only developing lung cells. In addition, Ng et al. revealed that the co-developing heart and lung tissues automatically separate from each other during later stage, without the need for chemical signals.

This human cell-based model of early forms of co-developing heart and lung cells may help provide researchers with new strategies to probe the underlying mechanisms of human heart and lung interaction during embryogenesis.

Introduction

Human embryogenesis is a highly orchestrated process that requires delicate coordination between organs that originate from different germ layers. As the two main organs within the chest cavity, the mesoderm-derived heart and endoderm-derived lung partake in have extensive mutual interaction that are essential for their proper morphogenesis (Peng et al., 2013; Hoffmann et al., 2009; Arora et al., 2012; Steimle, 2018). During mouse embryonic development, WNT derived from the second heart field induces specification of pulmonary endoderm, which in turn secretes SHH that signals back to the heart and regulates proper atrial septation (Steimle, 2018; Zhou, 2017; Hoffmann et al., 2009). This inter-lineage crosstalk is partly mediated by the multipotent mesodermal progenitors located between the developing heart and lung, which have the potential for lineage contribution to pulmonary endothelium, pulmonary smooth muscle and cardiomyocytes (Peng et al., 2013). However, the extent of translation of findings derived from rodent models to the understanding of developmental interplay between human cardio-pulmonary systems remains unclear. There is, therefore, a critical need for experimentally tractable systems for investigating human cardio-pulmonary co-development during organogenesis.

Much work has been done for directed differentiation of hiPSCs into either cardiomyocytes (Lian et al., 2012; Mummery et al., 2012; Burridge et al., 2014; Lian et al., 2015; Lee et al., 2017; Kattman et al., 2011), or pulmonary epithelium (Chen et al., 2017; Huang et al., 2013; Jacob et al., 2017; Dye et al., 2015; Gotoh et al., 2014; Wong et al., 2012), both of which often utilize stepwise differentiation strategies that recapitulate key developmental signaling events. To recapitulate cardiogenesis, hiPSCs were sequentially specified into mesoderm, cardiac mesoderm, and then NKX2.5+ cardiac progenitors (Lian et al., 2012; Mummery et al., 2012; Burridge et al., 2014; Lian et al., 2015; Lee et al., 2017; Kattman et al., 2011; Laflamme et al., 2007). For pulmonary induction, hiPSCs went through stages corresponding to definitive endoderm and anterior foregut endoderm, and then became NKX2.1+ lung epithelial progenitors (Chen et al., 2017; Huang et al., 2013; Jacob et al., 2017; Dye et al., 2015; Gotoh et al., 2014; Wong et al., 2012; D’Amour et al., 2005; Longmire et al., 2012). Despite the significant contributions of these models make to the mechanistic understanding of human heart and lung organogenesis, they generally focus on one organ parenchyma at a time. It remains challenging to model and investigate multi-organ co-development within a single differentiation of hiPSCs, especially when the organs of interest are derived from different germ layers, as is the case for the heart and lung.

Comparison of existing protocols for single-lineage cardiac and pulmonary differentiation from hiPSCs indicates shared regulators despite their distinct germ-layer origin. Firstly, both endodermal and mesodermal specification is facilitated by the inhibition of insulin and phosphoinositide 3-kinase signaling (Lian et al., 2012; Lian et al., 2013; Mou et al., 2012), and can be induced by a similar set of paracrine factors, including WNT, BMP, and TGF-β (Kattman et al., 2011; D’Amour et al., 2005; Loh, 2014). It is the quantitative combination of these signaling that determines endoderm versus mesoderm bifurcation (Kattman et al., 2011; Loh, 2014; Kim, 2015). This is consistent with the shared primitive streak origin of both germ layers during gastrulation (Levak-Svajger and Svajger, 1974; Lawson et al., 1991; Tam and Beddington, 1987). Secondly, WNT inhibition not only facilitates the transition from definitive endoderm to anterior foregut endoderm (Loh, 2014; Spence et al., 2011), but it also promotes cardiac mesoderm emergence (Lian et al., 2012; Willems et al., 2011; Wang et al., 2011; Ren et al., 2011; Tran, 2009). Lastly, retinoic acid (RA) signaling is required for the induction and maturation of both cardiac and pulmonary progenitors (Huang et al., 2013; Jacob et al., 2017; Mou et al., 2012; Chen, 2007; McCauley et al., 2017). These common paracrine regulation of paralleled cardiac and pulmonary specification is consistent with their close spatial coordinates within the embryonic body planning, as demonstrated by shared HOX genes expression and functional requirement (Lufkin et al., 1991; Makki and Capecchi, 2012; Chisaka and Capecchi, 1991).

In this study, we described a stepwise, growth-factor-free protocol for simultaneous induction of cardiac and pulmonary progenitors from a single culture of hiPSCs. This is accomplished by initial co-induction of mesoderm and definitive endoderm mixture, followed by their concurrent specification into cardiac (NKX2.5+) and lung (NKX2.1+) progenitors, respectively, using the same sets of small molecule cocktails modulating WNT, nodal and TGF-β signaling in a temporal specific manner. Using 3D suspension culture with continuing WNT activation, we engineered pulmonary-centered, cardio-pulmonary micro-Tissues (μTs), and demonstrated the accompanying cardiac lineage as an essential cellular niche that promoted effective alveolar maturation. Finally, following the withdrawal of WNT agonist, each concentric cardio-pulmonary μT reorganized and ultimately segregated into cardiac-only and pulmonary-only μTs. This work therefore offers an effective hiPSC-based model for investigating cardio-pulmonary co-development and tissue segregation during human embryogenesis.

Results

Simultaneous induction of cardiac and pulmonary progenitors

Building on existing protocols on cardiac (Lian et al., 2012; Mummery et al., 2012; Burridge et al., 2014; Lian et al., 2015; Lee et al., 2017; Kattman et al., 2011; Laflamme et al., 2007), and lung (Chen et al., 2017; Huang et al., 2013; Jacob et al., 2017; Dye et al., 2015; Gotoh et al., 2014; D’Amour et al., 2005; Green et al., 2011) differentiation from hiPSCs, a stepwise differentiation strategy was developed to enable simultaneous specification of both lineages within a single culture of hiPSCs. Firstly, a balanced mesodermal and endodermal induction was achieved via fine-tuning of WNT activation in the absence of insulin, activin A, and BMP4 supplementation (Stage-1). Then, a combined inhibition of WNT and TGF-β signaling initiated the specification of the co-induced mesoderm and endoderm towards cardiac and pulmonary specification, respectively (Stage-2). Lastly, reactivation of WNT signaling in the presence of retinoic acid (RA) led to concurrent emergence of NKX2.5+ cardiac and NKX2.1+ lung progenitors (Stage-3), which was our main focus in this study.

WNT signaling is required for mesodermal and endodermal specification in dose-dependent manner during embryogenesis as well as hiPSC specification (Lian et al., 2012; Kattman et al., 2011; Paige, 2010; Ng et al., 2008). Using a hiPSC line, BU3 NKX2.1GFP; SFTPCtdTomato (BU3-NGST), we examined the possibility of co-inducing mesodermal and definitive endodermal specification by exclusively modulating WNT signaling using a GSK3β inhibitor (CHIR99021, hereafter abbreviated as CHIR) and without the addition of exogenous growth factors (e.g. Activin A and BMP4). BU3-NGST were trypsinized and plated at 150,000 cells/cm2 in mTESR1 medium for 24 hr. BU3-NGST were then treated with different concentrations of CHIR at 4, 7, and 10 μM for 48 hr in mTESR1 medium, followed by incubation in growth factor-free differentiation medium (based on RPMI1640 and B-27 minus insulin) (Figure 1a). Toward the end of germ layer induction (Stage-1), we detected co-existence of both definitive endodermal (SOX17+) and mesodermal progenitors (MIXL1+ and NCAM1+), as well as the wide-spread expression of pan-mesendoderm marker (MIXL1+) (Figure 1b). In addition, definitive endoderm co-expressed both FOXA2 and SOX17 (Figure 1c). This was further confirmed by flow cytometry analysis CD13 (mesodermal marker) and SOX17 (endodermal marker) (Figure 1—figure supplement 1f-g). This observation was further confirmed by gene expression analysis of FOXA2, SOX17, and NCAM1 (Figure 1d), which together suggests that 7 μM CHIR drives balanced endoderm and mesoderm induction from hiPSCs, while further elevation of CHIR dosage selectively favors mesodermal specification. This is in line with a study by Martyn et al., 2019 demonstrating that WNT induces formation of primitive streak after 48 hr of treatment and that the newly formed primitive streak cells provide additional endogenous WNT, BMP, and Nodal signaling to further pattern the cells toward mesoderm and endoderm lineage (Martyn et al., 2019), We confirmed that upon 48 hr induction with 7 μM CHIR, the majority of cells were expressing primitive streak marker (T) and mesendoderm marker (MIXL1), but not pluripotent marker (OCT4) during early germ layer induction (Figure 1—figure supplement 1).

Figure 1 with 1 supplement see all
Mesoderm and endoderm co-induction from hiPSCs using CHIR.

(a) Diagram showing the experimental design (b) Cells following Stage-1 differentiation expressed MIXL1 (Mesendodermal lineage), SOX17 (definitive endoderm), and NCAM1 (mesoderm). Scale bar = 125 μm. (c) Majority of SOX17 cells were also FOXA2+. Scale bar = 62.5 μm. (d) Fold change of hiPSCs for FOXA2 (n = 3 each; 4 vs 7, p < 0.001; 7 vs 10, p < 0.001; 4 vs 10, p < 0.001), SOX17 (n = 3 each; 4 vs 7, p < 0.001; 7 vs 10, p < 0.001; 4 vs 10, p = 0.9978) and NCAM1 (n = 3 each; 4 vs 7, p < 0.001; 7 vs 10, p < 0.001; 4 vs 10, p < 0.001). All data are mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001. ‘n’ refers to biological replicates. Diagram created using BioRender (http://biorender.com/).

To specify the co-induced mesoderm and endoderm towards cardiac and pulmonary lineages, respectively, Day-4 cells were treated with TGF-β inhibitor (A8301) (McCauley et al., 2017; Jacob et al., 2017; Hawkins et al., 2017) and WNT inhibitor (IWP4) (Lian et al., 2012; Huang et al., 2013) for 4 days (Stage-2, Day-5 to Day-8), followed by treatment with a ventralization cocktail consisting of CHIR and RA (essential for lung progenitor specification) for 7 days to Day-15 (Stage 3) (Figure 2a and b; McCauley et al., 2017; Jacob et al., 2017; Hawkins et al., 2017). Consistent with CHIR-dependent germ layer induction (Figure 1), the efficiency of cardio-pulmonary specification was tightly regulated by CHIR dosage. We found that on Day-15, cells pre-exposed to CHIR (7 μM) during Stage-1 were able to give rise to robust co-induction of both cardiac (NKX2.5+) and pulmonary (NKX2.1+) progenitors (Figure 2c, d and e). In comparison, cells pre-treated with high-CHIR (10 μM) differentiated mainly into cardiac lineage; while low-CHIR (4 μM) failed to drive effective differentiation into either lineage (Figure 2c, d and e). FACS analysis confirmed that our protocol enabled effective induction of both NKX2.1+ and NKX2.5+ cells as compared to pulmonary-only or cardiac-only differentiation protocols (Figure 2—figure supplement 1a, b, c). Since NKX2.1 expression can also be found in neural and thyroid tissues, we performed further immunostaining analysis on Day-15 differentiated cells to inspect this possibility. Our results showed the NKX2.1+ cells did not co-express TUJ1/PAX6 (Neural) or PAX8 (Thyroid), suggesting that the specified NKX2.1+ population is of lung fate. Furthermore, no p63-expressing cells were identified, confirming the absence of airway epithelial cell population. At the same stage, NKX2.5+ cells co-expressed cardiac Troponin T (cTnT), suggesting that these cells were being specified towards cardiac lineage.COUPTFII-positive staining was also observed in some of the NKX2.5+ cells, suggesting atrial specification. However, most of the NKX2.5+ cells have yet to specific into downstream cardiac subtypes such as ventricular (MLC2v), endocardium (NFATC), and epicardium (WT1) (Figure 2—figure supplement 1d).

Figure 2 with 1 supplement see all
Stepwise cardio-pulmonary co-differentiation from hiPSCs using chemical defined, growth factor-free protocol.

(a) Schematic diagram showing the overall differentiation strategy. (b) Immunofluorescence (IF) showing staining of lung (NKX2.1+) and cardiac (NKX2.5+). (c) IF (d,e) and quantitative PCR (qPCR) analysis of the induction of lung and cardiac progenitors on Day-15 of differentiation. (c–e) The effects of different CHIR concentrations during Stage-1 of differentiation. Fold change over hiPSCs (d) NKX2.1 (n = 3 each; 4 vs 7, p < 0.001; 7 vs 10, p < 0.001; 4 vs 10, p = 0.9993) and (e) NKX2.5 (n = 3 each; 4 vs 7, p< 0.001; 7 vs 10, p = 0.0053; 4 vs 10, p < 0.001). (f–h) The effects of different exposure time of CHIR (7 µM) treatment during the first 2 days of differentiation. qPCR analysis of (g) NKX2.1 (n = 3 each; 24 vs 48, p < 0.001; 48 vs 72, p < 0.001; 24 vs 72, p < 0.001) and (h) NKX2.5 (n = 3 each; 24 vs 48, p < 0.001; 48 vs 72, p = 0.1503; 24 vs 72, p < 0.001). Scale bar = 500 μm for whole well scan; Scale bar = 125 μm for 20 X images. All data are mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001. ‘n’ refers to biological replicates. Diagram created using BioRender (http://biorender.com/).

The action of CHIR treatment on hiPSC differentiation depends not only on dosage but also the duration of exposure (Kempf et al., 2016; Zhao et al., 2019). We evaluated the efficiency of cardio-pulmonary induction following exposure to CHIR (7 µM) for different periods (24, 48, and 72 hr), and found that extended CHIR exposure for 48 or 72 hr was required to induce robust cardio-pulmonary programs (Figure 2f). Specifically, CHIR favored cardiac specification with increase in exposure time and plateaued at 48 hr of treatment (Figure 2h); while the induction of pulmonary program peaked at 48 hr of CHIR treatment (Figure 2g) and declined with further extension of the treatment. Based on these observations, for all subsequent experiments, we used 48 hr treatment of CHIR (7 µM) during Stage-1 of the co-differentiation program. Furthermore, we showed that maintaining hiPSCs in mTESR1 Plus during the initial CHIR treatment appeared to be critical for enabling effective cardio-pulmonary differentiation (Figure 3—figure supplement 1), as compared to using RPMI1640 supplemented with B-27 minus insulin as the basal medium during CHIR treatment.

Exogenous activation of nodal, TGF-β and BMP signaling during the very initial steps of hiPSC specification has been widely utilized for cardiac (Kattman et al., 2011; Laflamme et al., 2007; Ng et al., 2008) and pulmonary (Huang et al., 2013; Gotoh et al., 2014; Mou et al., 2012; Jacob et al., 2017; Dye et al., 2015) specification from hiPSCs. Here, we investigated how exogenous and endogenous nodal and BMP signaling regulates cardio-pulmonary induction during germ layer induction (Stage-1). Nodal signaling inhibition (using A8301, Day-2 to Day-4) immediately following CHIR treatment terminated both cardiac and pulmonary induction; while nodal activation through Activin A supplementation (Day-2 to Day-4) led to pulmonary-only differentiation (Figure 3a, b and c). This suggests the requirement of endogenous nodal signaling for cardio-pulmonary induction and that high-level nodal activation favors pulmonary instead of cardiac induction. In parallel, BMP inhibition (using DMH-1) during the same time period compromised cardiac induction and mildly reduced pulmonary specification; while exogenous BMP4 supplementation enhanced cardiac induction but inhibited pulmonary specification (Figure 3d, e and f). This indicates that endogenous BMP signaling is primarily required for cardiac induction and that exogenous augmentation of BMP signaling further favors the cardiac lineage at the expense of the pulmonary lineage.

Figure 3 with 3 supplements see all
The effect of Nodal and BMP signaling during Stage-1 of co-differentiation on cardio-pulmonary induction.

IF (a,d) and qPCR (b,c,e,f) analysis of the induction of lung (NKX2.1+) and cardiac (NKX2.5+) progenitors on Day-15 of differentiation (a–c) The effects of exogenous nodal activation (Activin A, 20 ng/mL) or its inhibition (A8301, 1 µM). Fold change over hiPSCs for (b) NKX2.1 (n = 3 each; Activin A /A8301 vs. Activin A+ /A8301, p = 0.1939; Activin A /A8301 vs. Activin A /A8301+, p < 0.001; Activin A+ /A8301 vs. Activin A /A8301+, p < 0.001) and (c) NKX2.5 (n = 3 each; Activin A /A8301 vs. Activin A+ /A8301, p < 0.001; Activin A /A8301 vs. Activin A /A8301+, p < 0.001; Activin A+ /A8301 vs. Activin A /A8301+, p = 0.8649). (d-f) The effects of exogenous BMP4 (20 ng/mL) or BMP inhibitor (DMH1, 2 µM). qPCR analysis of (e) NKX2.1 (n = 3 each; BMP4 /DMH1 vs. BMP4+ /DMH1, p < 0.001; BMP4 /DMH1 vs. BMP4 /DMH1+, p < 0.001; BMP4+ /DMH1 vs. BMP4 /DMH1+, p < 0.001) and (f) NKX2.5 (n = 3 each; BMP4 /DMH1 vs. BMP4+ /DMH1, p < 0.001; BMP4 /DMH1 vs. BMP4 /DMH1+, p = 0.0044; BMP4+ /DMH1 vs. BMP4 /DMH1+, p < 0.001). Scale bar = 500 μm for whole well scan; Scale bar = 125 μm for 20 X images. All data are mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001. ‘n’ refers to biological replicates. Diagram created using BioRender (http://biorender.com/).

Shared signaling for cardio-pulmonary co-differentiation from germ-layer progenitors

In previous single-lineage hiPSC differentiation studies, TGF-β and WNT inhibition is known to promote pulmonary specification from definitive endoderm (Huang et al., 2013; Gotoh et al., 2014; McCauley et al., 2017; Jacob et al., 2017; Hawkins et al., 2017; Dye et al., 2015), as well as the induction of cardiac mesoderm (Lian et al., 2012). Here, we examined how combined inhibition of both TGF-β (using A8301) and WNT (using IWP4) during Day-4 to Day-8 (Figure 3—figure supplement 2a) regulates cardio-pulmonary specification from germ-layer progenitors established in Stage-1. We found that combined TGF-β and WNT inhibition enhanced both cardiac and pulmonary specification, with TGF-β inhibition having a more profound effect on the cardiac lineage (Figure 3—figure supplement 2b,c,d). Our finding suggests shared signaling requirement for lung and heart induction from their respective germ-layer progenitors, which is consistent with their close spatial coordinates within the embryonic body planning (Lufkin et al., 1991; Makki and Capecchi, 2012; Chisaka and Capecchi, 1991).

In both mouse and human pluripotent stem cell differentiation models, exogenous BMP4 has been shown to be crucial for ventralization of the foregut endoderm to give rise to NKX2.1+ lung progenitors (Huang et al., 2013; Jacob et al., 2017; Serra et al., 2017). Here in our study, we observed effective cardio-pulmonary co-differentiation in the absence of exogenous BMP4 during ventralization (Stage 3) (Figure 2b). To address this discrepancy, we investigated how exogenous introduction of BMP4 during ventralization regulated the emergence of cardiac and pulmonary progenitors (Figure 3—figure supplement 3a). Intriguingly, there was no significant differences observed at protein and gene expression level of NKX2.1 and NKX2.5 comparing ventralization in the presence and absence of exogenous BMP4 (Figure 3—figure supplement 3b,c,d). Nonetheless, endogenous BMP4 was indeed required during this stage of differentiation, as inhibition of BMP4 using DMH1 significantly compromised the induction of both NKX2.1 and NKX2.5 (Figure 3—figure supplement 3b,c,d).

3D suspension culture platform for alveolar induction

To examine whether NKX2.1+ lung progenitors derived from the cardio-pulmonary co-induction protocol (Figure 2a) possess the ability to mature into alveolar type 2 (AT2) epithelial cells, Day-15 cells were trypsinized and re-plated into an ultra-low adhesion plate for 3D suspension culture (Figure 4a), and exposed to alveolar maturation medium containing CHIR, KGF, Dexamethasone, 8-bromoadenosine 3’, 5’-cyclic monophosphate (cAMP activator), and IBMX (CKDCI) (Jacob et al., 2017; Dye et al., 2015; de Carvalho et al., 2019). Upon transition from 2D to 3D suspension culture in CKDCI medium, the co-induced cardio-pulmonary progenitors self-assembled into pulmonary-centered, concentric, dual-lineage μTs during the overnight culture (Figure 4b). Following 3 days of 3D suspension culture in CKDCI medium, effective AT2 maturation was observed in the cardio-pulmonary μTs as indicated by robust SFTPCTdTomato fluorescence (Figure 4c) and gene expression (Figure 4—figure supplement 1c). The SFTPCTdTomato fluorescence could sustained up to Day-29 (2 weeks in alveolar maturation medium) (Figure 4—figure supplement 1a,b). As a control, we cultured Day-15 cardio-pulmonary progenitors on top of the transwell insert for air-liquid interface (ALI) culture, on 2D plastic surface for regular submerged culture or embedded in Growth Factor Reduced (GFR) Matrigel, and failed to detect obvious AT2 induction by Day-18 (Figure 4c, Figure 4—figure supplement 2, Figure 4—figure supplement 3). Consistent with the observations using fluorescence reporters, NKX2.1 and SFTPC gene expression was significantly upregulated in 3D suspension culture on Day-18 compared to the starting Day-15 cells or cells following ALI maturation (Figure 4d and e). This may in part be due to the reduction of cardiac progenitor as indicated by significant downregulation of NKX2.5 gene expression (Figure 4f, Figure 4—figure supplement 1d). Our results demonstrated 3D suspension culture as a robust platform to expedite alveolar maturation.

Figure 4 with 6 supplements see all
3D suspension culture of cardio-pulmonary μTs expedites AT2 maturation.

(a) Schematic diagram illustrating the Stage-4 maturation protocol involving replating of Day-15 cardiac and pulmonary progenitors onto ultra-low adhesion plate (for 3D suspension culture) or the transwell insert (for ALI culture). (b) Whole mount staining of cardiopulmonary μT on Day-18, scale bar 75 μm. (c) Live μT imaging of NKX2.1GFP and SFTPCTdTomato reporter signals during the first 3 days of maturation (Day-16 – Day-18). (d-f) qPCR analysis of hiPSC control, Day-15 cells and Day-18 cells (from ALI or suspension culture) for (d) NKX2.1 (n = 3 each; Day-15 vs. ALI, p = 0.9998; Day-15 vs. Suspension, p = 0.0547; ALI vs. Suspension, p = 0.0486), (e) SFTPC (n = 3 each; Day-15 vs. ALI, p = 0.1896; Day-15 vs. Suspension, p < 0.001; ALI vs. Suspension, p < 0.01). (f) NKX2.5 (n = 3 each; Day-15 vs. ALI, p < 0.001; Day-15 vs. Suspension, p = 0.8367; ALI vs. Suspension, p < 0.001). Scale bar = 125 μm. (g–n) Schematic diagram showing the differentiation procedure without (g) and with (j) Activin A during Stage-1 of differentiation (h, k) Live μT imaging of NKX2.1GFP and SFTPCTdTomato reporter signals during Day-16 to Day-18. Scale bar = 125 μm for 10 X images. (i, l) Whole mount staining of μTs on Day-18. (m–n) qPCR analysis of hiPSC control, Day-15 cells and Day-18 cells (from Activin-free or Activin) for (m) NKX2.1 (n = 3 each; Day-15 (No Activin) vs. Day-18 (No Activin), p < 0.001; Day-15 (Activin) vs Day-18 (Activin), p = 0.0147; Day-15 (No Activin) vs. Day-15 (Activin), p = 0.1316), (n) SFTPC (n = 3 each; Day-15 (No Activin) vs. Day-18 (No Activin), p < 0.001; Day-15 (Activin) vs Day-18 (Activin), p = 0.2417; Day-18 (No Activin) vs. Day-18 (Activin), p < 0.001). All data are mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001. Scale bar = 125 μm. ‘n’ refers to biological replicates. Diagram created using BioRender (http://biorender.com/).

To elucidate how the co-induced cardiac lineage modulates the alveolar maturation process, we introduced activin A (20 ng/mL) during germ-layer specification (Figure 4j), which effectively inhibited mesoderm specification and led to pulmonary-only differentiation outcome on Day-15 (Figure 3a and b). In the absence of accompanying cardiac cells, although NKX2.1GFP+ lung progenitors can be robustly induced and maintained, their alveolar maturation (as indicated by SFTPCtdTomato reporter) following 3 days of maturation in 3D suspension culture was dramatically diminished compared to the cardio-pulmonary group (Figure 4g–l). Whole mount imaging of μTs on Day-18 showed pulmonary-only differentiation mainly comprised NKX2.1+ cells, while cardio-pulmonary μTs possessed a concentric arrangement of NKX2.1+ cells, surrounded by NKX2.5+ cells (Figure 4i and j). This was further supported by gene expression analysis of NKX2.1 (Figure 4m) and SFTPC (Figure 4n). Further extension of CKDCI maturation period for 2 weeks up to Day-29 in the pulmonary-only group failed to produce AT2 induction to a level comparable to that of the cardio-pulmonary group (Figure 4—figure supplement 4). This suggests that the cardiac lineage can serve as a cellular niche in supporting alveolar maturation.

We confirmed the reproducibility of our cardio-pulmonary co-induction protocol using another independent hiPSC line (BU1), including effective induction of endodermal and mesodermal mixture on Day-4 (Figure 4—figure supplement 5a,b), induction of pulmonary (NKX2.1) and cardiac (NKX2.5) progenitors on Day-15 (Figure 4—figure supplement 5c,d), and maturation of alveolar type two epithelium (Pro-SFTPC), and cardiac structural markers (cTnT and Sarcomeric Alpha Actinin) on Day-18 μTs (Figure 4—figure supplement 5e). Further, using BU1 without built-in fluorescence reporters, we also confirmed that the GFP and TdTomato expression observed from BU3-NGST differentiation was not resulted from autofluorescence (Figure 4—figure supplement 6).

Cardio-pulmonary segregation in the dual-lineage micro-tissue (ΜT)

Spatial-temporal regulation of WNT is crucial for early cardiac differentiation (Lian et al., 2012; Zhao et al., 2019; Buikema et al., 2020), however, continuous exposure to WNT activation is known to delay contractile maturation of cardiomyocytes (Fan et al., 2018). In parallel, exogenous WNT activation using CHIR is essential for inducing AT2 maturation and its maintenance until the endogenous AT2 niche is established (Jacob et al., 2017; Abdelwahab et al., 2019; Nabhan et al., 2018; Frank et al., 2016). To investigate how CHIR removal regulates cardio-pulmonary maturation following AT2 establishment on Day-18 in 3D suspension culture (Figure 4g), we transitioned the maturation medium from CKDCI to KDCI without CHIR (Figure 5a; Jacob et al., 2017). To our surprise, upon CHIR removal, the cardiac and pulmonary components within each dual-lineage μT, which was initially arranged in the pulmonary-centered, concentric manner (Figure 5b), effectively reorganized over time and eventually segregated from each other (Figure 5a and b). When Day-15 cardio-pulmonary progenitors were transitioned to suspension culture directly in KDCI medium without CHIR, although they successfully underwent dual-lineage μT formation and segregation, there was no sign of AT2 maturation, echoing the importance of CHIR during AT2 induction (Figure 5—figure supplement 1; Huang et al., 2013; Jacob et al., 2017; Nabhan et al., 2018; Frank et al., 2016).

Figure 5 with 1 supplement see all
Cardio-pulmonary segregation in the dual-lineage μT.

(a) Schematic diagram illustrating the timeline for the investigation. Scale bar = 125 μm (b) Histological analysis of cardio-pulmonary μTs at different stages of segregation. Scale bar = 125 μm (c) Diagram showing measurement of the total perimeter of GFP+ pulmonary compartment (red color) and its overlapping perimeter with non-GFP compartment (white color) using Image J. Scale bar = 125 μm (d) Box plot showing percentage overlapping region of GFP+ with non-GFP tissues on Day- 18 (n = 30 each;; CKDCI vs. KDCI, p = 0.3979; CKDCI vs. KDCI+ IWP4; p > 0.9999; CKDCI vs. KDCI+ NSC, p = 0.4293; KDCI vs. KDCI+ IWP4, p = 0.3979; KDCI vs. KDCI+ NSC, p > 0.9999; KDCI+ IWP4 vs. KDCI+ NSC, p = 0.4293), Day-22 (n = 30 each; CKDCI vs. KDCI, p = 0.0077; CKDCI vs. KDCI+ IWP4; p = 0.0112; CKDCI vs. KDCI+ NSC, p < 0.001; KDCI vs. KDCI+ IWP4, p = 0.9994; KDCI vs. KDCI+ NSC, p = 0.8318; KDCI+ IWP4 vs. KDCI+ NSC, p = 0.7674) and Day-25 (n = 30 each; CKDCI vs. KDCI, p < 0.001; CKDCI vs. KDCI+ IWP4; p < 0.001; CKDCI vs. KDCI+ NSC, p < 0.001; KDCI vs. KDCI+ IWP4, p = 0.4271; KDCI vs. KDCI+ NSC, p = 0.7275; KDCI+ IWP4 vs. KDCI+ NSC, p = 0.9623). (e) Histological analysis of cTnT expression on the segregated cardiac and pulmonary μTs, with co-staining of NKX2.5 and NKX2.1. Scale bar = 125 μm. All data are mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001. ‘n’ refers to biological replicates. Diagram created using BioRender (http://biorender.com/).

To quantitatively assess this segregation process, we performed time-lapse single-μT tracking and determined the percentage of overlap between the cardiac and pulmonary tissues by measuring the length of the overlapping border between the GFP+ and non-GFP components and normalizing it by the total perimeter of the GFP+ pulmonary component (Figure 5c). We compared the segregation process in the presence (CKDCI) and absence (KDCI) of CHIR, and found that although cardio-pulmonary segregation took place in both medium conditions, it was significantly expedited by the withdrawal of CHIR (Figure 5d,). To investigate the requirement of endogenous WNT signaling for this segregation process, we introduced inhibitors of canonical (IWP4) and non-canonical (NSC668036, a Dishevelled inhibitor) WNT signaling (Li and Wang, 2018), and did not detect any obvious difference in the segregation process as compared to the control KDCI condition (Figure 5d). In parallel with the cardio-pulmonary segregation, cardiac contraction was observed 7 days following CHIR withdrawal (Video 1). Immunohistochemical analysis demonstrated specific co-expression of NKX2.5 and cardiac troponin T (cTnT) in the segregated cardiac μT (Figure 5e).

Video 1
Contacting cardiac μT following 7 days after withdrawal of CHIR.

Cardio-pulmonary μT maturation

NKX2.1+ has the potential to differentiate into both proximal and distal airway epithelial cells. To characterize the lung epithelial composition in cardio-pulmonary μTs, we performed whole mount staining on Day-22 μTs. The induction of AT2 cells were further confirmed by the detection of the presence of lamellar bodies by transmission electron microscopy (Figure 6a) and by the positive immunofluorescence staining of pro-SFTPC+ (Figure 6b) and pro-SFTPB+ (Figure 6—figure supplement 1a). We also observed HOPX+ cells in the μTs (Figure 6b), and upregulated HOPX gene expression (Figure 6—figure supplement 1b) suggesting the presence of AT1-like cells. S100A4 staining indicated presence of mesenchyme in the μTs (Figure 6—figure supplement 1c), which may also play an important role in promoting alveologenesis Hawkins et al., 2017. In the meantime, these μTs did not express markers for proximal airway epithelium such as ciliated cells (FOXJ1) (Figure 6—figure supplement 1d), secretory cells (MUC5AC) (Figure 6—figure supplement 1e) and basal cells (p63) (Figure 6—figure supplement 1f), which can be readily observed in airway μT engineered from bronchial epithelial cells. In parallel, cardiac elements within the μTs exhibited the striated pattern as indicated by cTnT and Sarcomeric Alpha Actinin staining (Figure 6c and d). The cardiac contractile function was confirmed via the detection of calcium influx (Video 2), and its gradual reduction with increasing concentrations of Verapamil, a calcium channel blocker (Figure 6e).

Figure 6 with 1 supplement see all
Characterization of cardio-pulmonary μT maturation.

(a) Lamellar bodies found in cardio-pulmonary μT. Scale bar = 1 μm. Cardio-pulmonary μT expressing (b) Pro-SFTPC and HOPX. Cardio-pulmonary μT also exhibited striated pattern as indicated by (c) cTnT and (d) Sarcomeric Alpha Actinin. Scale bar = 125 μm. (e) Calcium imaging of cardiac μTs following treatment with Verapamil.

Video 2
Calcium influx capability of cardiac μT loaded with Cal-520.

Discussion

Here, we described a novel strategy to model human cardio-pulmonary co-development using multi-lineage hiPSC differentiation. The current study primarily focused on co-induction of cardio-pulmonary progenitor cells, thus establishing the foundation for future investigations of how the crosstalk between these two organ lineages regulates their respective lineage maturation. We demonstrated that upon co-induction of mesoderm and endoderm, a series of shared signaling events were capable of driving simultaneous cardiac and pulmonary specification from their respective germ-layer progenitors. Upon transitioning the co-induced cardiac and pulmonary progenitors to 3D suspension culture, we observed expedited alveolar maturation within 3 days, which was supported by the accompanying cardiac lineage. In 3D suspension culture, each cardio-pulmonary μT effectively segregates into separate cardiac and pulmonary μTs, which was partially inhibited by WNT activation. This study therefore delivers an effective in vitro model for studying the mechanistic interplay between the developing heart and lung during human embryogenesis.

The extensive cardio-pulmonary mutual interaction during organogenesis has been well documented in the mouse model (Peng et al., 2013; Hoffmann et al., 2009; Steimle, 2018); however, the translatability of these findings to human embryogenesis remains elusive due to the lack of proper model systems. Human pluripotent stem cell differentiation has offered an effective means for recapitulating and investigating human organogenesis, and tremendous progress has been made toward directed cardiac or pulmonary specification (Lian et al., 2012; Mummery et al., 2012; Burridge et al., 2014; Lian et al., 2015; Lee et al., 2017; Kattman et al., 2011; Chen et al., 2017; Huang et al., 2013; Jacob et al., 2017; Dye et al., 2015; Gotoh et al., 2014; Wong et al., 2012; Laflamme et al., 2007; D’Amour et al., 2005; Longmire et al., 2012; Lian et al., 2013; Mou et al., 2012; Green et al., 2011). However, almost all existing models have been focusing on one parenchymal lineage at a time, and therefore lack the ability to support the investigation of inter-organ crosstalk. Here, building on the established understanding of signaling events necessary for cardiac and pulmonary induction (Lian et al., 2012; Mummery et al., 2012; Burridge et al., 2014; Lian et al., 2015; Lee et al., 2017; Kattman et al., 2011; Chen et al., 2017; Huang et al., 2013; Jacob et al., 2017; Dye et al., 2015; Gotoh et al., 2014; Wong et al., 2012; Laflamme et al., 2007; D’Amour et al., 2005; Longmire et al., 2012; Lian et al., 2013; Mou et al., 2012; Green et al., 2011), we have developed a robust protocol for the simultaneous co-differentiation of cardiac and pulmonary lineages from hiPSCs. Within our co-differentiation system, unrestricted interaction between cells of both lineages is enabled even before their lineage commitment.

Most current attempts on pulmonary induction from hiPSCs relies on initial nodal activation using growth factor (Activin A) supplementation, which is critical for definitive endoderm specification. Here, we showed that by fine tuning WNT signaling using a small-molecule inhibitor of GSK-3β (CHIR), robust induction of endoderm and subsequently lung progenitors can be achieved without any exogenous growth factors. This is consistent with the observation that CHIR was capable of inducing cardiac differentiation in replacement of combined effect of exogenous Activin A and BMP4 (Lian et al., 2012). In addition, Martyn et al., 2019 demonstrated that WNT is sufficient to induce primitive streak, which then creates a gradient of signals (BMP, WNT, Nodal) that could further specify fate towards mesoderm and endoderm lineages (Martyn et al., 2019). Nonetheless, Nodal and BMP signaling remains crucial in mesoderm and endoderm specification, as inhibition of these signals effectively terminated cardio-pulmonary co-induction.

Our study demonstrated the need for endogenous TGF-β signaling for effective cardio-pulmonary induction, as well as the critical role of endogenous BMP signaling in cardiogenesis. Furthermore, we found that temporal-specific action of the same set of small molecules regulating TGF-β and WNT signaling was capable of simultaneously driving mesoderm-to-cardiac and endoderm-to-pulmonary specification. Moreover, BMP4 has been shown to improve NKX2.1+ lung progenitor specification in both mouse and hiPSCs (Huang et al., 2013; Jacob et al., 2017; Serra et al., 2017). In our system, endogenous instead of exogenous BMP signaling was required during a developmental stage corresponding to foregut ventralization for effective co-emergence of cardiac and pulmonary progenitors. This is in line with the close spatial positioning of the developing heart and lung primordia within embryonic body patterning, which implies their exposure to a similar paracrine microenvironment (Steimle, 2018; Herriges and Morrisey, 2014).

To achieve alveolarization, NKX2.1+ lung progenitors are usually embedded in extracellular matrices, such as Matrigel and collagen (Jacob et al., 2017; Dye et al., 2015; de Carvalho et al., 2019). Here, we established an effective approach that enabled AT2 cell maturation within 3 days in suspension culture of 3D cell aggregates spontaneously formed from Day-15 cardiac and pulmonary progenitors. We further demonstrated that the presence of accompanying cardiac lineage is critical for robust alveolar induction. This observation is consistent with the recently reported inter-dependence between cardiac and pulmonary lineages during embryogenesis (Steimle, 2018). In the absence of cardiac lineage, NKX2.1+ lung progenitors could not achieve as effective AT2 maturation as what is observed in cardio-pulmonary μTs. This is again emphasizing that chemical cues derived from heart field also plays a role in lung development (Peng et al., 2013; Steimle, 2018). However, it should be noted that pulmonary mesenchyme could also augment distal lung differentiation, which requires further investigation with the ability to differentiate the participation and contribution of each mesodermal cell lineages (Hawkins et al., 2017; McCulley et al., 2015; Leeman et al., 2019). In addition, the presence of mesoderm-derived stromal cells have also been shown to be essential for effective alveolarization in vivo (Herriges and Morrisey, 2014; Domyan et al., 2011; Rankin et al., 2016) and in vitro (Gotoh et al., 2014; Hawkins et al., 2017). Furthermore, cells of the mesodermal lineage are known to be robust producers of extracellular matrix, which may also contribute to the effective alveolar maturation in the absence of external extracellular matrix support. The ability to enable effective alveolar induction from hiPSC-derived lung progenitors with the convenient of suspension culture also opens the door to large-scale production of alveolar cells, a critical challenge in regenerative medicine applications.

Using dual-lineage cardio-pulmonary μTs formed from the co-induced progenitors, we observed a novel process of cardio-pulmonary tissue segregation. The human body cavities are highly crowded spaces, filled with different tissues and organs that are in close contact with each other. It remains enigmatic how inter-organ boundaries are maintained to prevent undesired cell migration or tissue merging. Our cardio-pulmonary tissue segregation model suggests an intrinsic mechanism that effectively establishes a boundary between two distinct parenchymal lineages even when they are initially mingled together. Although no model of collective migration has been described in the context of cardio-pulmonary development, studies in other model systems suggest cell-cell communication and paracrine signaling (e.g. WNT) to be crucial for directed cell migration during development (Ciruna and Rossant, 2001; De Calisto et al., 2005; Carmona-Fontaine et al., 2008). Here, we found that exogenous WNT activation via GSK-3β inhibition effectively slowed down the cardio-pulmonary segregation, while inhibition of endogenous WNT (canonical and non-canonical) did not obviously affect the process. In consistence with our observation, it has been shown that the inhibition of non-canonical WNT signaling does not stop collective cell migration but distorting migration direction (Li and Wang, 2018).

Despite the presence of AT1-like cells in cardio-pulmonary μTs based on HOPX staining (Wang et al., 2018; Frank et al., 2019; Liebler et al., 2016), the organization of these cells remained unlike the thin, flat squamous cells found in the native lung (Yang et al., 2016; Williams, 2003). Although HOPX expression has been observed in cardiomyocytes (Friedman et al., 2018), we did not observe HOPX expression following establishment of cardio-pulmonary progenitors, and the induction of HOPX coincided with alveolar maturation during Day-15 to Day-18. Further study is needed to explore the possibility of obtaining AT1-like cells with a more mature phenotype using multiple markers. Our work demonstrates early emergence of SFTPC-expressing cells as early as Day-18, with CHIR and 3D suspension culture being key driving forces that promote alveologenesis (Huang et al., 2013; Jacob et al., 2017), Removal of CHIR on Day-18 led to a decreased in SFTPC expression on Day-22 (Figure 4—figure supplement 1c), suggesting that the complete alveolar niche was not fully established yet. Future investigation is needed to further clarify components of the supporting alveolar niche as well as its timeline of establishment. In parallel, cardiac contractility of the μTs was initiated 7 days following removal of CHIR, consistent with previous studies showing that GSK-3β inhibition promotes cardiomyocyte proliferation but hinders contraction by affecting myofibrillar architecture (Buikema et al., 2020; Wang et al., 2016; Tseng et al., 2006).

Heart begins to form around gestation week 3, which is the earliest organ to be developed during embryogenesis (Buckingham et al., 2005; Tan and Lewandowski, 2020). Lung development initiates soon afterwards during weeks 4–7 (Schittny, 2017). In the present study, we showed that the progenitor cells of both heart and lung can be simultaneously induced following 15 days of co-differentiation from hiPSCs. More comprehensive time-series analysis will be necessary to further delineate the fine temporal relationship between heart and lung lineage specification, which is expected to provide fundamental insights regarding cardio-pulmonary crosstalk during their paralleled organogenesis.

In conclusion, our work focuses on specification of cardio-pulmonary progenitors that have the potential to further mature into their respective descendent lineages. Further morphological and functional maturation of both cardiac and pulmonary lineages, as well as their crosstalk during this process will require future in-depth investigation. In addition, our work offers a novel model for investigating the molecular and cellular mechanisms underlying human cardio-pulmonary co-development and tissue boundary formation. We also expect this work to be of potential use for studying congenital diseases affecting both cardiovascular and pulmonary systems, such as congenital diaphragmatic hernia.

Materials and methods

Materials

Detailed information regarding reagents for culture and differentiation medium was summarized in Supplementary file 1. Reagents, equipment, and probes for quantitative PCR (qPCR) analysis antibodies and reagents for immunofluorescence staining were summarized in Key Resources Table.

Maintenance of human-induced pluripotent stem cells (HiPSCs)

View detailed protocol

The BU3-NGST and BU1 hiPSC lines were obtained as kind gifts from the laboratories of Dr. Darrell Kotton and Dr. Finn Hawkins (Boston University). BU3 hiPSC line was derived from a healthy donor and carries both NKX2.1GFP (NG) and Surfactant protein C (SFTPC)tdTomato (ST) reporters (Jacob et al., 2017; Hawkins et al., 2017). BU1 hiPSC line was also derived from healthy donor but without any reporters. hiPSCs were maintained on Matrigel-coated (ESC-qualified) six-well tissue culture plate with mTESR1 Plus medium with regular medium changed every other day. hiPSCs passaging was performed every 5–7 days using ReLESR at a plating ratio of 1:10. All cells used in this study were tested negative for mycoplasma contamination using Universal Mycoplasma Detection Kit (ATCC, 30–1012 K).

Simultaneous induction of cardiac and pulmonary progenitors from hiPSCs hiPSCs maintained in mTESR Plus were dissociated into single cells using StemPro Accutase. 150,000 cells/cm2 on hESC-qualified Matrigel-coated 96-well plate, and cultured in mTESR Plus supplemented with 10 μM Y-27632 (ROCK inhibitor) for 24 hr prior to differentiation. The overall protocol for stepwise cardio-pulmonary co-differentiation was summarized in Supplementary file 1. To induce a balanced mixture of mesodermal and definitive endodermal cells, hiPSCs were first incubated in mTESR Plus medium supplemented with different concentration (4, 7, 10 μM) of CHIR99021 (GSK3β inhibitor) and 10 μM Y-27632 for 48 hr. This was followed by an additional 48 hr incubation in serum-free differentiation medium consisting of RPMI 1640 supplemented with 2% B-27 minus insulin, 1 x GlutaMAX and 10 μM Y-27632. In some experiments, Activin A (20 ng/mL), BMP4 (20 ng/mL), A8301 (Nodal or TGF-β inhibitor, 1 μM) or DMH-1 (BMP inhibitor, 2 μM) were introduced to examine how Nodal and BMP signaling regulated mesodermal and endodermal specification. Differentiation outcomes were assessed by immunostaining and qPCR analysis of mesodermal (NCAM1) and definitive endodermal (SOX17) markers.

Following Stage-1, all subsequent differentiation procedures were performed using medium recipes formulated based on RPMI 1640 medium supplemented with 2% B-27 and 1 x GlutaMAX, referred to as ‘basal medium’. To initiate simultaneous cardiac and pulmonary specification, Day-4 cells were incubated for 4 days in Stage-2 medium, containing basal medium supplemented with 1 μM A8301, 5 μM IWP4 and 10 μM Y-27632. In some experiments, co-differentiation medium without either A8301 or IWP4 was utilized to investigate the impact of inhibition of TGF-β and WNT signaling.

Following Stage-2, to induce simultaneous specification of both cardiac and lung progenitors, co-differentiating cells were incubated for 7 days in Stage-3 medium containing basal medium supplemented with 3 μM CHIR99021 and 100 nM Retinoic acid (RA). Green fluorescence of the NKX2.1GFP reporter was examined daily using EVOS FL Auto 2 Imaging System to monitor the emergence of lung progenitors. On Day-15 of co-differentiation, the expression of cardiac (NKX2.5) and lung (NKX2.1) progenitor markers was evaluated by immunofluorescence staining and qPCR.

Co-maturation of cardio-pulmonary progenitors in air-liquid interface (ALI) culture

Request a detailed protocol

On Day-15 of cardio-pulmonary co-differentiation, cells were dissociated into single cells using TrypLE Express, and re-plated at 500,000 cells/cm2 onto the apical side of each 24-well Transwell insert (pore size of 0.4 μm, pre-coated with 1% growth factor-reduced Matrigel) in 100 μL maturation medium. Basolateral side of the transwell insert was filled with 500 μL of maturation medium. The maturation medium was basal medium supplemented with 3 μM CHIR99021, 10 ng/mL Keratinocyte growth factor (KGF), 50 nM Dexamethasone, 0.1 mM 8-bromoadenosine 3’, 5’-cyclic monophosphate (cAMP, AMP-activated protein kinase activator) and 0.1 mM 3-isobutyl-1-methylxanthine (IBMX, PKA activator), which was referred to as CKDCI medium. 10 μM Y-27632 was added during the initial 24 hrs following re-plating. The next day, all medium on the apical side was removed. 200 μL of fresh CKDCI medium without Y-27632 was added to the basolateral side to establish ALI culture, and was replaced daily. Red fluorescence from the SFTPCTdTomato reporter was examined daily using EVOS Imaging System to monitor the emergence of alveolar type 2 (AT2) cells. On Day-3 of ALI maturation, Transwell membrane were excised from the insert, and analyzed by qPCR (NKX2.1, SFTPC).

Co-maturation of cardio-pulmonary μTs in 3D suspension culture

View detailed protocol

On Day-15 of cardio-pulmonary co-differentiation, cells were dissociated into single cells using TrypLE Express. A total of 250,000 cells in 500 μL CKDCI maturation medium was transferred into each well of 24-well ultra-low adherence plate and cultured with agitation at 125 rpm to form cardio-pulmonary μTs. 10 μM Y-27632 was added during the initial 24 hr following re-plating. Following 3 days of culture in CKDCI medium, CHIR99021 was removed and μT culture was continued in KDCI medium for an additional 7 days. At desired time points of 3D suspension maturation, μTs were analyzed by histology (NKX2.1, NKX2.5, cTnT) and qPCR analysis (NKX2.1, SFTPC).

Embedding cardio-pulmonary μTs in matrigel droplet

Request a detailed protocol

On Day-15 of cardio-pulmonary co-differentiation, cells were dissociated into single cells using TrypLE Express. A total of 5000 cells in 50 μL Growth Factor Reduced (GFR) Matrigel were dropped on 24-well plate and cultured in CKDCI maturation medium. 10 μM Y-27632 was added during the initial 24 hr following re-plating. Following 3 days of culture, the Matrigel was dissolved in ice-cold EDTA, and the embedded cells were then recovered for RNA extraction and qPCR analysis of NKX2.1, SFTPC, NKX2.5.

Airway μTs

Request a detailed protocol

To generate airway μTs, 96-well plate was coated with 50 μL of 40% (v/v) GFR Matrigel diluted in PneumaCult-ALI Maintenance Medium. The normal human bronchial epithelia were resuspended in 40% (v/v) GFR Matrigel in PneumaCult-ALI Maintenance Medium and added to the coated wells. A total of 100 μL PneumaCult-ALI Maintenance Medium was placed in the wells and changed every other day. Airway μTs formed were harvested following over 3 weeks of differentiation and fixed for immunostaining.

Single μT time-lapse imaging and analysis

Request a detailed protocol

To investigate the segregation of cardio-pulmonary μTs into their respective cardiac and lung μTs, following 3 days suspension culture in CKDCI medium in 24-well ultra-low adherence plate, single μT was transferred into each well in 96-well ultra-low adherence plate and cultured for an additional 7 days. The following medium recipes were examined for cardio-pulmonary segregation: KDCI medium, KDCI medium supplemented with 3 μM CHIR99021, KDCI medium with 5 μM IWP4, and KDCI medium with 50 μM NSC668036. Time-lapse imaging was performed on Day- 18, Day-22, and Day-25 following μT transfer to monitor the segregation process. The pulmonary compartment within each cardio-pulmonary μT was tracked based on the NKX2.1GFP reporter. To quantify the segregation between the two compartments within each μT. Image J was used to measure the overlapping perimeter between GFP+ (pulmonary) and non-GFP (cardiac) compartments, which was then normalized to total perimeter of GFP+ compartments and expressed as the percentage of overlapping.

Percent overlapping %=Overlapping Perimeter of GFP and nonGFP compartments μmTotal Perimeter of GFP organoids μm×100%

qPCR analysis

Request a detailed protocol

Total RNA was extracted using TRIzol, processed by chloroform extraction, precipitated using 1 volume of absolute isopropanol with 50 μg/mL of RNase-free glycoblue as carrier, washed with 75% ethanol, air-dried, solubilized in RNase-free water and quantified using NanoDrop 2000 spectrophotometer. cDNA was synthesized via reverse transcription of 1 μg total RNA with random hexamers and the High-Capacity cDNA Reverse Transcription kit according to manufacturer’s instruction. Real-time qPCR analysis was performed on CFX96 Touch Real-Time PCR Detection System using TaqMan probes. Each reaction mixture was prepared by combining 1 μL of probe, 10 μL of TaqMan Master Mix, 1 μL of cDNA (equivalent to 50 ng), and the final volume was brought up to 20 μL. The final Ct value was normalized to housekeeping gene (β-actin), using comparative Ct method. Unless otherwise specified, baseline, defined as fold change = 1, was set as undifferentiated hiPSCs, or if undetected, a cycle number of 40 was assigned to allow fold change calculations (Jacob et al., 2017). List of TaqMan probes was summarized in Key Resources Table.

Immunofluorescence staining on 2D cell samples

View detailed protocol

Cells were fixed with ice-cold methanol, air-dried, rehydrated with phosphate-buffered saline (PBS), permeabilized with 1% (v/v) Triton X-100, blocked in 1% (w/v) bovine serum albumin in PBS (blocking buffer), incubated with primary antibodies diluted in blocking buffer at 4 °C overnight, and incubated with corresponding fluorescence-conjugated secondary antibodies in blocking buffer at room temperature (RT) for 45 min. Nuclear counterstain was performed using Hoechst-33342 (1:500) in PBS. Fluorescence images were acquired using EVOS Imaging System. All antibodies used and their respective dilution were summarized in Key Resources Table.

Histology

Request a detailed protocol

The μTs were fixed with 4% paraformaldehyde, embedded in HistoGel and then in paraffin. Tissue processing and paraffin embedding was performed in Research Histology Lab of Pitt Biospecimen Core at the University of Pittsburgh Medical Center (UPMC) Shadyside Hospital. Paraffin blocks were sectioned at 5 μm thickness, transferred onto glass slides, rehydrated by sequential incubation in Histoclear, 100% ethanol, 95% ethanol and distilled water. To unmask antigen, slides were treated with Antigen Unmasking Solution at 95 °C for 20 min and cooled down to RT. Immunofluorescence staining was performed as described above for 2D cell samples. After the final wash, slides were mounted with DAPI Fluoromount-G, and imaged using EVOS Imaging System. All antibodies used and their respective dilution were summarized in Key Resources Table.

Flow cytometry

Request a detailed protocol

Cells were dissociated into single cells via incubation with TrypLE for 15 min. For NKX2.1GFP assessment, approximately 3 × 105 cells were resuspended in FACS buffer (DPBS with 1% FBS) and incubated with DAPI for 10 min on ice, followed by three washes prior to analysis. For indirect labeling of NKX2.5, cells were first trypsinized and stained with Fixable Violet Dead Cell Stain Kit (Thermo Fisher Scientific) for 10 min on ice. Cells were then fixed with 4% PFA on ice for 20 min, followed by three times washes with 1% BSA in PBS. Cells were permeabilized with 1% Triton X-100 for 20 min, followed by blocking for 30 min prior to adding primary antibody for overnight incubation. Next day, cells were washed three times in 1% BSA in PBS and incubated with fluorophore-conjugated secondary antibody for 1 hr. Following three washes with 1% BSA in PBS, cells were re-suspended in FACS buffer for flow cytometry analysis at Unified Flow Core of Department of Immunology at University of Pittsburgh Medical Center.

Contraction and calcium signal

Request a detailed protocol

To assess contraction of cardiac μT, segregated cardiac μT was stained with 5 μM of Cal-520 AM (AAT Bioquest, 21130), a calcium indicator dye. The concentration-response of cardiac μTs to calcium channel blocker (Verapamil) were assessed by treating the μTs with 0.1, 1, and 10 μM of Verapamil for 10 min. Calcium imaging (500 frames per second) was performed pre- and post- Verapamil treatment using a Prime 95B Scientific CMOS camera (Photometrics) mounted on an epifluorescent stereomicroscope (Nikon SMZ1000) with a GFP filter and an X-cite Lamp (Excelitas).

TEM

Request a detailed protocol

Cardio-pulmonary μTs were fixed in 2.5% glutaraldehyde in 0.1 M PBS (pH7.4) for at least 1 hr. After three washes in 0.1 M PBS for 10 min each, the μTs were post fixed in 1% Osmium tetroxide containing 1% potassium ferricyanide at 4 °C for 1 hr, followed by three washes in 0.1 M PBS for 10 min each. μTs were dehydrated in graded series of ethanol starting from 30%, 50%, 70%, 90% and finally 100% of ethanol for 10 min each. μTs were further dehydrated epon for 1 hr at RT. This step was repeated for another three times prior to embedding in pure epon at 37 °C for 24 hr. Finally, the μTs were cured for 48 hr at 60 °C. The presence of lamellar body in cardio-pulmonary μTs were identified using JEM 1400 Flash TEM.

Statistics

Statistical methods relevant to each figure were outlined in the accompanying figure legend. At least three biological replicates were performed for each group under comparison. Unless otherwise indicated, unpaired, two-tailed Student’s t tests were applied to comparisons between two groups. For comparisons among three or more groups, one-way ANOVA was performed followed by Tukey multiple comparison tests. Results are displayed as mean ± SD, with p < 0.05 considered statically significant. n values referred to biologically independent replicates.

Appendix 1

Appendix 1—key resources table
Reagent type (species) or resourceDesignationSource or referenceIdentifiersAdditional information
Gene (Homo sapiens)β-actinGenBank (NM_001101.3)Hs01060665_g1
Gene (Homo sapiens)NKX2.1GenBank (NM_001079668.2)Hs00968940_m1
Gene (Homo sapiens)FOXA2GenBank (NM_02178.4)Hs00232764_m1
Gene (Homo sapiens)SOX17GenBank (NM_022454.3)Hs00751752_s1
Gene (Homo sapiens)NKX2.5GenBank (NM_004387.3)Hs00231763_m1
Gene (Homo sapiens)SFTPCGenBank (NM_001172357.1)Hs00161628_m1
Gene (Homo sapiens)NCAM1GenBank (NM_000615.6)Hs00941821_m1
Gene (Homo sapiens)HOPXGenBank (NM_001145459.1)Hs04188695_m1
Cell line (Homo sapiens)BU3-NGSTBoston University (Kotton’s Lab)RRID:CVCL_WN82
Cell line (Homo sapiens)BU1Boston University (Kotton’s Lab)-
Cell line (Homo sapiens)Normal Human Bronchial Epithelial (NHBE) cellsLonzaCat# CC-2541; RRID:CVCL_S124
Antibodyanti-NKX2.1(Rabbit monoclonal)AbcamCat# ab76013;RRID:AB_13107841:500
Antibodyanti-NKX2.5 (Goat polyclonal)R&D SystemsCat# AF2444;RRID:AB_3552691:500
Antibodyanti-Pro-SFTPB (Rabbit polyclonal)Seven HillsCat# WRAB-55522RRID:AB_1:200
Antibodyanti-Pro-SFTPC (Rabbit polyclonal)Seven HillsCat# WRAB-9937; RRID:AB_4517211:500
Antibodyanti-HOPX (Mouse monoclonal)Santa CruzCat# Sc-398703; RRID:AB_26879661:100
Antibodyanti-cTnT (Mouse monoclonal)Thermo Fisher ScientificCat# MA5-12960; RRID:AB_110007421:200
Antibodyanti-Sarcomeric Alpha Actinin (Mouse Monoclonal)Thermo Fisher ScientificCat# MA1-22863; RRID:AB_5574261:200
Antibodyanti-p63 (Mouse monoclonal)Biocare MedicalCat# CM163A; RRID:AB_105827301:100
Antibodyanti-MUC5AC (Mouse monoclonal)Thermo Fisher ScientificCat# MA5-12178;RRID:AB_109780011:100
Antibodyanti-FOXJ1 (Mouse monoclonal)Thermo Fisher ScientificCat# 14-9965-80; RRID:AB_15488361:100
Antibodyanti-PAX8 (Mouse monoclonal)ThermoFisher ScientificCat# MA1-117RRID:AB_25368281:100
Antibodyanti-β-Tubulin III (Mouse monoclonal)Sigma-AldrichCat# T8578RRID:AB_18412281:100
Antibodyanti-PAX6 (Mouse monoclonal)BioLegendsCat# 862,001RRID:AB_28012371:100
Antibodyanti-COUPTFII (Mouse monoclonal)R&D SystemsCat# PP-H7147-00RRID:AB_19642141:100
Antibodyanti-MLC2v (Rabbit Polyclonal)ProteinTech GroupCat# 10906–1-APRRID:AB_21474531:100
Antibodyanti-NFATC (Mouse monoclonal)Thermo Fisher ScientificCat# MA3-024RRID:AB_22360371:100
Antibodyanti-WT1 (Mouse monoclonal)Novus BiologicalsCat# NBP2-44606RRID:AB_not found1:100
Antibodyanti-Brachyury (Goat polyclonal)R&D SystemsCat# AF2085; RRID:AB_22002351:50
Antibodyanti-MIXL1 (Rabbit polyclonal)Thermo Fisher ScientificCat# PA5-64903; RRID:AB_26647371:50
Antibodyanti-NCAM1 (Rabbit monoclonal)Cell Signaling TechnologiesCat# 99,746T; RRID:AB_28684901:50
Antibodyanti-FOXA2 (Mouse monoclonal)Santa Cruz TechnologyCat# Sc-271103; RRID:AB_106144961:50
Antibodyanti-SOX17 (Goat polyclonal)R&D SystemsCat# AF1924; RRID:AB_3550601:200
Antibodyanti-OCT4 (Mouse monoclonal)Santa CruzCat# sc-5279RRID:AB_6280511:100
Antibodyanti-CD13 APC-conjugatedBD BiosciencesCat# 557454; RRID:AB_3986241:10
AntibodyDonkey anti-mouse IgG (H + L), Alexa Fluor 488Thermo Fisher ScientificCat# A21202; RRID:AB_1416071:500
AntibodyDonkey anti-rabbit IgG (H + L), Alexa Fluor 488Thermo Fisher ScientificCat# A21206; RRID:AB_25357921:500
AntibodyDonkey anti-rabbit IgG (H + L), Alexa Fluor 568Thermo Fisher ScientificCat# A10042; RRID:AB_27575641:500
AntibodyDonkey anti-goat IgG (H + L), Alexa Fluor 647Thermo Fisher ScientificCat# A21447; RRID:AB_1418441:500
Recombinant DNA proteinActivin AR&D Systems338-AC-010
Recombinant DNA proteinRecombinant human BMP4R&D Systems314 BP
Recombinant DNA proteinRecombinant human KGFPeproTech100–19
Commercial assay, kitHigh-Capacity cDNA Reverse Transcription kitApplied Biosystems4368814
Commercial assay, kitTaqMan Fast Advanced Master MixThermo Fisher Scientific4444556
Commercial assay, kitFixable Violet Dead Cell Stain KitThermo Fisher ScientificL34955
Chemical compound, drugshESC-qualified Matrigel Basement Membrane MatrixCorning354,234
Chemical compound, drugsGrowth Factor Reduced Basement Membrane MatrixCorning354,230
Chemical compound, drugsmTESR PlusStem Cell Technologies05825
Chemical compound, drugsDulbecco’s Phosphate-Buffered Saline (DPBS)Corning45000–430
Chemical compound, drugsReLESRStem Cell Technologies05873
Chemical compound, drugsStemPro Accutase Cell Dissociation ReagentThermo Fisher ScientificA1110501
Chemical compound, drugsRPMI1640Corning10–040-CV
Chemical compound, drugsGlutaMAXThermo Fisher Scientific35050061
Chemical compound, drugsB-27 minus insulin SupplementThermo Fisher ScientificA1895601
Chemical compound, drugsB-27 Supplement (Complete)Thermo Fisher Scientific12587–010
Chemical compound, drugsTrypLE ExpressThermo Fisher Scientific12605028
Chemical compound, drugsHyclone FetalClone 1 Serum (U.S)GE HealthcareSH30080.03
Chemical compound, drugsY-27632 dihydrochlorideCayman Chemical1000558310
Chemical compound, drugsCHIR99021Reprocell04000402
Chemical compound, drugsA8301Sigma AldrichSSML1314-1MG
Chemical compound, drugsDMH-1Tocris4126/10
Chemical compound, drugsIWP4Tocris5214/10
Chemical compound, drugsAll-trans Retinoic AcidCayman11,017
Chemical compound, drugsDexamethasoneSigma AldrichD4902
Chemical compound, drugs8-bromoadenosine 3’,5’-cyclic monophosphate sodium salt (cAMP)Sigma AldrichB7880
Chemical compound, drugs3-Isobutyl-1-methylxanthine (IBMX)Sigma AldrichI5879
Chemical compound, drugsNSC668036Tocris5813/10
Chemical compound, drugsPneumaCult-ALI Basal MediumStemcell Technologies05002
Chemical compound, drugsPneumaCult-ALI Maintenance SupplementStemcell Technologies05006
Chemical compound, drugsTRIzol ReagentThermo Fisher Scientific15596018
Chemical compound, drugsChloroformSigma-AldrichC2432
Chemical compound, drugsGlycoblueThermo Fisher ScientificAM9516
Chemical compound, drugsIsopropanolACROS Organic327272500
Chemical compound, drugsEthanol 200 ProofPharmaco-AAPLDSP-C7-18
Chemical compound, drugsMethanolFisher ChemicalBPA412-1
Chemical compound, drugsParaformaldehydeSigma AldrichP6148-500G
Chemical compound, drugsTriton X-100Sigma AldrichX100-500ML
Chemical compound, drugsBovine Serum AlbuminFisher BioReagentsBP9706-100
Chemical compound, drugsPhosphate Buffer Saline 20 XGrowcellsMRGF-695–010 L
Chemical compound, drugsHistoclearGreat LakesGL-1100–01
Chemical compound, drugsAntigen Unmasking Solution, Citric Acid BasedVector LaboratoriesH-3300; RRID:AB_2336227
Chemical compound, drugsDAPI-Fluoromount-GSouthern Biotech0100–20
Chemical compound, drugsHoechst 33,342Thermo Fisher Scientific62,249
Chemical compound, drugsTrypLE Express EnzymeThermo Fisher Scientific12605010
SoftwareImage JVersion 1.8.0.182; RRID:SCR_003070
SoftwareFlowjoVersion 7.6.1; RRID:SCR_008520
OtherTranswell insert (0.4 μm)Greiner Bio-One662,641
OtherUltra-low adherence 24-well PlateGreiner Bio-One662,970
OtherUltra-low adherence 96-well PlateGreiner Bio-One650,979
OtherNanodrop 2000 SpectrophotometerThermo Fisher ScientificND2000CLAPTOP; RRID:SCR_018042
OtherEVOS FL Auto 2 Imaging SystemThermo Fisher ScientificAMAFD2000
OtherCFX96 Touch Real-Time PCR Detection SystemBio-Rad1855196; RRID:SCR_018064
OtherImmEdge Hydrophobic Barrier PAP PenVector LaboratoriesH-4000; RRID:AB_2336517
OtherHistoGel Specimen Processing GelRichard Allen Scientific11330057

Data availability

All data supporting the findings of this study are available within the article and its supplementary files. Source data files have been provided for Figures 1 to 6.

References

    1. Levak-Svajger B
    2. Svajger A
    (1974)
    Investigation on the origin of the definitive endoderm in the rat embryo
    Journal of Embryology and Experimental Morphology 32:445–459.
    1. Tam PP
    2. Beddington RS
    (1987)
    The formation of mesodermal tissues in the mouse embryo during gastrulation and early organogenesis
    Development 99:109–126.

Decision letter

  1. Paul W Noble
    Senior and Reviewing Editor; Cedars-Sinai Medical Center, United States
  2. Sarah XL Huang
    Reviewer; The University of Texas Health Science Center at Houston, United States

Our editorial process produces two outputs: i) public reviews designed to be posted alongside the preprint for the benefit of readers; ii) feedback on the manuscript for the authors, including requests for revisions, shown below. We also include an acceptance summary that explains what the editors found interesting or important about the work.

Decision letter after peer review:

Thank you for submitting your article "Recapitulate Human Cardio-pulmonary Co-development Using Simultaneous Multilineage Differentiation of hPSCs" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Paul Noble as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Sarah XL Huang (Reviewer #3).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

The authors present a novel idea on simultaneous heart and lung induction from hPSCs. The work has a very good potential to be published in eLife. However, the manuscript in its current form is not publishable. There is consensus that the reproducibility of a given protocol must be demonstrated using multiple hPSC lines. Under COVID situation, the full protocol should be tested using at least another pluripotent cell line. In addition, the work in its current form lacks rigor and complete lineage characterization at key differentiation steps, as described in the review comments. The reviewers are confident that when such data are provided, there is a high possibility that it will support the conclusion of the paper, but the authors should not leave the readers to speculate.

The overall suggestion is that the authors should revise the manuscript to (1) perform additional thorough characterization of the differentiation culture up to the cardiac and lung progenitor specification stage of the protocol (day 15); (2) give an explicit clarification that progenitor specification is the focus of the current manuscript, and the maturation of cardiac and lung lineages will be investigated in-depth in near future studies. Otherwise, multi-lineage characterization within each of the organ and functional analyses on mature cell types are required for publication. This will take considerable time. (3) Analyses of the data that provide mechanistic insights into how each of the developmental milestones is achieved by the protocol (day 1-15), will greatly increase the impact of the paper.

Reviewer #1 (Recommendations for the authors):

1. Expression of lung (NKX2.1) and cardiac (NKX2.5) markers (Figure 1and2) is based on immunofluorescence and RT-PCR, but it remains unclear which percentage of cells expressed these markers, and which expressed neither marker. FACS analysis could have provided such data. This is relevant information for assessing the efficiency of the differentiation protocol (and comparison to those based on separate endoderm and mesoderm induction). Along the same line, it is unclear which percentage of cells ultimately expressed alveolar and cardiac markers in the 3D suspension cultures.

2. The approach to use scaffold-free 3D suspension cultures to create microtissues is not novel in the field of hiPSC and has been used e.g. in studies on cardiac myocyte and endothelial cell development. What was the reason to select this system, also because most alveolar differentiation protocols use Matrigel based organoid cultures?

3. In Figure 3, the authors use activin A to suppress development of cells of the cardiac lineage, and show that these support alveolar type 2 cell maturation. How did the authors check that activin A indeed suppressed development of cardiac cells and thus provide (indirect) evidence for the role of cardiac cells in alveolar development?

4. In Figure 3, the authors show that no alveolar type 2 cells are developed in the ALI culture system. Is there any evidence that this may be explained by loss of cells of the cardiac lineage?

5. Regarding further characterization of alveolar epithelial cells: did the authors also check for markers of type 1 alveolar epithelial cells?

6. Regarding further characterization of cardiac cells/cardiomyocytes: was any attempt made to quantify contraction, and show that this could e.g. be blocked using ca2+ channel blockers such as verapamil? And did they check other markers to show that the cells had developed into more mature heart cells, including other sarcomere markers, but also ion channels etc?

7. To achieve cardiopulmonary segregation as shown in Figure 4, the authors first used CKDCI medium from day 15-18, before switching to KDCI. What happened to alveolar type 2 marker expression upon CHIR withdrawal and segregation (was the alveolar niche fully established at day 18, and was therefore exogenous WNT activation no longer needed?), and was this segregation also observed if cultures were directly switched to KDCI on day 15?

Reviewer #2 (Recommendations for the authors):

While a potentially very interesting model, much more in depth characterization is required. Each component, as well intervening stages of development should be thoroughly characterized and benchmarked.

Reviewer #3 (Recommendations for the authors):

Here are further clarifications to some of the above-mentioned weaknesses, as well as additional suggestions on how the manuscript could be improved to increase the impact of the work.

Lines 24-30 of manuscript: the authors reasoned that WNT signaling is required for both mesodermal and endodermal specification. On this basis, the authors explained why they decided to use WNT agonist CHIR for mesoderm and DE co-induction from hiPSC line BU3. To my knowledge, the justification is not supported by lung developmental studies. In other words, the data shows convincing evidence of meso- and endo- dermal induction by the end of stage-1/day 4, however the interpretations provided by the authors are not accurate. A more plausible explanation is that, within the first 48 hours (day 1-2), high dose of WNT agonist CHIR induces primitive streak formation from part of the hiPSCs in culture. Subsequently, the newly formed primitive streak provides endogenous WNT, BMP and Nodal/Activin signaling to pattern the rest of the hiPSCs in culture to form mesoderm and DE by the end of day 4. This alternative mechanistic explanation is supported by evidence from both the existing literature and the authors' own data. Brivanlou's group elegantly demonstrated that activation of WNT signaling solely is sufficient to induce primitive streak from hPSCs. In addition, the dosage of CHIR applied in day 1-2 is way too high to be able to induce endoderm with lung potential. Furthermore, the authors maintain the cells in pluripotent medium mTESR1 for the first 48 hours (lines 48-50 of manuscript), which is critical for the success of the protocol (lines 87-90 of manuscript, suppl Figure 3). This supports my hypothesis that CHIR induces part of the hiPSCs in culture to form primitive streak during the first 48 hours, mTESR1 is required to maintain the remaining hiPS cells in culture in pluripotent state during this stage, before they start meso- and endo-dermal differentiation in day 2-4. Consistently, the data shows that Activin A or BMP inhibition during days 2-4 is detrimental for DE or mesodermal induction. Single cell analysis at 48 hr would be sufficient to prove a mixture of "primitive streak and pluripotent cells" profile, which will provide valuable mechanistic insights.

The whole manuscript is based on the differentiation of a double reporter line derived from one parental hiPSC line BU3. For scientific rigor and assurance, it is necessary to demonstrate efficient co-differentiation of lung and heart from other parental hPSC lines (both reporter and non-reporter) using the combination of time and dosage depicted in the current protocol.

It is convincing that most of the NKX2-1+ cells are lung fated and NKX2-5+ cells are cardiac fated, however, for scientific rigor and thoroughness, it is necessary to characterize the differentiation culture to exclude contamination of neural (EpCAM-) and thyroid (PAX8+) fated NKX2-1+ cells, and to confirm that all the NKX2-5+ cells co-express other cardiac progenitor markers such as KDR and PdgfR-α. Based on previous studies, some of the early lung progenitors co-express NKX2-5, therefore the authors must confirm whether NKX2-5+ cells at day 15 and on are NKX2-1- and EpCAM- cells.

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "Recapitulating Human Cardio-pulmonary Co-development Using Simultaneous Multilineage Differentiation of Pluripotent Stem Cells" for further consideration by eLife. Your revised article has been evaluated by Paul Noble (Senior Editor) and a Reviewing Editor.

The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

While the manuscript is improved, significant concerns remain that need to be addressed for publication. In particular, reviewer 2 has specific concerns that we would like to see addressed in a revision

Reviewer #1 (Recommendations for the authors):

The authors have made important changes and added results of new experiments in this extensively revised manuscript. This has markedly improved the quality of the manuscript, and conclusions are much better supported by the results.

As a result the revised manuscript is interesting and provides a new and relevant model to study heart and lung development in a dish, and cross-talk between these developing organs.

Reviewer #2 (Recommendations for the authors):

In this manuscript, Ng et al., report on a system where cardiac mesoderm and pulmonary endoderm co-develop from pluripotent stem cells. This is of potential interest, as it could provide an integrated model for the study of human cardiopulmonary development. I still have quite a few concerns, though, which were not fully addressed by the authors.

1. In my first review, I pointed out that the main weakness lies in the lack of thorough characterization of the resulting cells and tissues. This is still the case. The characterization relies almost entirely on reporter gene expression and PCR for the same markers. In Figure 1, representative examples are shown of a limited set of markers. It would be nice to have quantification (and for intracellular flow quantification, analysis of cells not expressing the protein of interest should be used as a control). The qPCR data do shown that these markers are expressed, but they compare something to nothing (i.e. PSCs), so that fold changes will be always be very high. One image of the ultrastructure of type 2 cells is shown now, which is a plus. Overall, this work still does begs the question why the only pulmonary markers observed in this model are basically SFTPC and NKX2.1. Furthermore, type 2 identity has not been further verified with markers such as SFTPB, ABCA3, LAMP2. Another outstanding question for the lung component is whether any pulmonary mesenchyme was generated. This was still not addressed. In this era, at scRNAseq should be performed. This would also address several of my other concerns below.

2. It is still not clear which cardiac cells are generated: ventricular, atrial, endocardium, epicardium, conducting tissue? This is the same comment as in my first review. An experiment with verapamil or some cells expressing α-actinin, as shown here, do not address this question. Atrial and ventricular cells contract. I would also add that in many PSC differentiation cultures where lineages are not sufficiently rigorously specified, contamination with beating cardiomyocytes is quite typical. This may be the case here. It is also difficult to understand why type 2 cells and unspecified cardiac cells are the only output of this model.

3. In Figure 2-S1d, the nuclear stains for NKX2.5 and NKX2.1 do not seem to align with DAPI. The same is true in Figure 3-S1a, where the higher magnifications for NKX2.1, NKX2.5 and DAPI do not seem to align. The same is true again for Figure 4-S1d, where the pattern of NKX2.1 is not the inverse of that of NK2.5 (no DAPI shown here). Furthermore, there is much more cTNT positivity than NKX2.5 expression in Figure 2-S1d. One would expect the opposite. These potential inconsistencies make the images difficult to interpret.

4. Along the same lines, in Figure 3a and 3d, as well in Figure 3-S1b and Figure 3-S3b (each time lower right panel), there seems to be co-expression of NKX2.1 and NKX2.5. Or is this superposition? Are there cells that express neither marker? This is difficult to evaluate without DAPI images. Also, could it be that in Figure 3-S1, the reduced cardiac and pulmonary differentiation is due to reduced proliferation? Again, without a DAPI image, this is difficult to evaluate.

5. This sentence is unclear: 'This may in part be due to the reduction of cardiac progenitor as indicated by significant downregulation of NKX2.5 gene expression (Figure 4f, Figure 4—figure supplement 1d).' (line 292). According to Figure 4f, there is no downregulation of NKX2.5 in suspension culture.

6. In Figure 4c, which relies entirely on reporters, it would be nice to co-stain for the actual proteins and other markers identifying type 2 cells. I note that in Figure 3c, the fluorescence patterns of the GFP and TdTomato reporters is very different, yet they should both be cytoplasmic. In the reporter studies, it would be good to co-differentiate a non-reporter expressing line as a reporter-negative control for autofluorescence in Figure 4 and its supplements. Figure 4-S1c is apparently not commented on the text.

7. The authors mention in the rebuttal evidence for type 1 cells, as shown by staining for HOPX in a non-reporter line (Figure 4-S5) and in the reporter line (Figure 6b). However, HOPX is not a marker for type 1 cells in humans, and is expressed broadly, including in cardiomyocytes (see, among others, Friedman et al., Cell Stem Cell 2018). In Figure 6b, there does not appear to be much pro-SFTPC, and that seems to be on the outside of the structure, whereas in Figures 4 and 5, the NKX2.1+ cells are on the inside.

8. The authors should not call Act. A 'TGF-β' signaling. It is a TGF-β family ligand, but signals through its own receptors. It recapitulates Nodal signaling in vitro.

9. Finally, there is still no sufficient quantification of differentiation efficiency and yield. NKX2.1, for example is also expressed in the forebrain and in the thyroid. The minimal conditions described typical yield neurectoderm. TUJ1 was examined, but other markers, such as PAX 6 should be included.

Recommendations for the authors

While a still potentially interesting model, much more in depth characterization is still required as well. In particular the nature and yield of the output should be better characterized, and figure quality and consistency should be paid attention to.

Reviewer #3 (Recommendations for the authors):

The authors addressed most of the questions and concerns raised in the 1st round of review.

https://doi.org/10.7554/eLife.67872.sa1

Author response

The authors present a novel idea on simultaneous heart and lung induction from hPSCs. The work has a very good potential to be published in eLife. However, the manuscript in its current form is not publishable. There is consensus that the reproducibility of a given protocol must be demonstrated using multiple hPSC lines. Under COVID situation, the full protocol should be tested using at least another pluripotent cell line. In addition, the work in its current form lacks rigor and complete lineage characterization at key differentiation steps, as described in the review comments. The reviewers are confident that when such data are provided, there is a high possibility that it will support the conclusion of the paper, but the authors should not leave the readers to speculate.

The overall suggestion is that the authors should revise the manuscript to (1) perform additional thorough characterization of the differentiation culture up to the cardiac and lung progenitor specification stage of the protocol (day 15); (2) give an explicit clarification that progenitor specification is the focus of the current manuscript, and the maturation of cardiac and lung lineages will be investigated in-depth in near future studies. Otherwise, multi-lineage characterization within each of the organ and functional analyses on mature cell types are required for publication. This will take considerable time. (3) Analyses of the data that provide mechanistic insights into how each of the developmental milestones is achieved by the protocol (day 1-15), will greatly increase the impact of the paper.

We appreciate the overall suggestions from the reviewers, which provide us valuable guidance to further improve this manuscript. Below we highlight key revisions in response to these overall suggestions.

(1) We have performed additional characterization of the differentiation up to Day-15. Specifically, we have included additional FACS analysis on Day-15 to provide a more comprehensive cellular characterization (Figure 2—figure supplement 1a), and have also confirmed that the NKX2.1+ progenitors are neither neural- nor thyroid-fated by co-staining with PAX8 and TUJ1 ( Figure 2—figure supplement 1d). Furthermore, we showed that NKX2.5+ progenitors are cardiac-fated as they co-express cTnT (Figure 2—figure supplement 1d).

(2) We agree with the reviewers’ comment and suggestion that the present study is primarily focused on progenitor specification. We have added explicit statements to emphasize this at discussion line 490-493; line 603-606. Nonetheless, in the revised manuscript, we have provided additional alveolar epithelial characterization in 3D cardio-pulmonary μTs to show positive immunofluorescence staining of SFTPC and HOPX (Figure 6b), as well as the presence of lamellar bodies by transmission electron microscopy (Figure 6a). In parallel, the cardiac μTs exhibited a striated pattern when stained for cTnT and sarcomeric α actinin (Figure 6c,d). Additionally, the cardiac μTs were functional and responded to calcium channel blockers (Verapamil) in a dose-dependent manner (Figure 6e).

(3) Based on the reviewers’ suggestions and to provide further mechanistic insights regarding developmental milestones, we have included staining markers of primitive streak (Brachyury) and mesendoderm (MIXL1) after 2 days of initial CHIR99021 induction, and used FACS analysis to show that the majority of resulting cells have committed to primitive streak at this stage (Figure 1—figure supplement 1).

(4) We have also validated our differentiation protocol on an additional hiPSC line, including germ layer induction, cardio-pulmonary progenitor induction, and 3D µT formation and maturation.

Reviewer #1 (Recommendations for the authors):1. Expression of lung (NKX2.1) and cardiac (NKX2.5) markers (Figure 1and2) is based on immunofluorescence and RT-PCR, but it remains unclear which percentage of cells expressed these markers, and which expressed neither marker. FACS analysis could have provided such data. This is relevant information for assessing the efficiency of the differentiation protocol (and comparison to those based on separate endoderm and mesoderm induction). Along the same line, it is unclear which percentage of cells ultimately expressed alveolar and cardiac markers in the 3D suspension cultures.

Thank you for bringing this to our attention. We have included FACS analysis of pulmonary progenitors (based on their NKX2.1-driven GFP expression) and cardiac progenitors (NKX2.5 antibody labeling) (Figure 1—figure supplement 1a). In addition, we have included the FACS analysis of endoderm-focused pulmonary protocol and mesoderm-focused cardiac protocol (Figure 1—figure supplement 1b, c) for comparison. We have also provided further characterization of alveolar and cardiac marker expression in 3D suspension culture (Figure 6).

2. The approach to use scaffold-free 3D suspension cultures to create microtissues is not novel in the field of hiPSC and has been used e.g. in studies on cardiac myocyte and endothelial cell development. What was the reason to select this system, also because most alveolar differentiation protocols use Matrigel based organoid cultures?

As the reviewer has pointed out, most current protocols for inducing alveolar differentiation utilize Matrigel embedding. 1,2 To address the reviewer’s comment, we have added a new figure (Figure 4—figure supplement 3) comparing alveolar type 2 cell induction in our 3D suspension culture versus Matrigel-embedded culture from Day-15 cardio-pulmonary progenitors, demonstrating that 3D suspension culture on ultra-low adhesion surface expedites alveolar type 2 maturation on Day-18 of differentiation (Day-3 of alveolar maturation).

3. In Figure 3, the authors use activin A to suppress development of cells of the cardiac lineage, and show that these support alveolar type 2 cell maturation. How did the authors check that activin A indeed suppressed development of cardiac cells and thus provide (indirect) evidence for the role of cardiac cells in alveolar development?

In Figure 3a-c, we showed that the presence of Activin A in our cardio-pulmonary protocol significantly downregulated NKX2.5 gene expression (by qPCR), and this was further supported by immunostaining (Figure 3a). To further confirm that there was no cardiac population in Activin A-supplemented protocol, we generated 3D μTs from Day-15 progenitors and showed that no NKX2.5 expression could be observed using confocal microscopy (Figure 4l). Therefore, we confirmed that Activin A supplementation suppressed the induction of cardiac cells, which suggested the potential positive impact of cardiac accompaniment during alveolar maturation. This is consistent with findings from rodent embryogenesis that chemical cues derived from the heart field play key roles in lung development. 3,4 We have included additional discussion in the revised manuscript (line 556-559).

4. In Figure 3, the authors show that no alveolar type 2 cells are developed in the ALI culture system. Is there any evidence that this may be explained by loss of cells of the cardiac lineage?

We would like to clarify that alveolar type 2 cells were still being developed in the ALI culture but to a very minimal extent compared to 3D suspension culture (Figure 4c). In the revised manuscript, we have added qPCR analysis to compare cardiac NKX2.5 expression in both ALI and 3D suspension culture. We found that NKX2.5 gene expression (by qPCR) was significantly downregulated in ALI culture compared to 3D suspension culture (Figure 4f). This was further supported by immunostaining (Figure 4—figure supplement 1d), which showed a reduction in the cardiac population in ALI culture.

5. Regarding further characterization of alveolar epithelial cells: did the authors also check for markers of type 1 alveolar epithelial cells?

Thank you for pointing this out. Based on the reviewer’s suggestion, in the revised manuscript, we added alveolar type 1 (AT1) cell characterization by performing HOPX staining 5-7 on Day-22 cardio-pulmonary μTs and observed the presence of AT1-like cells that did not colocalize with the AT2 cell marker SFTPC (Figure 6b). Nonetheless, the organization of these AT1-like cells in μTs remained different from that in the native lung. This suggests that further optimization of AT1 maturation is needed to overcome this limitation. We have also addressed this in the manuscript (line 585-587).

6. Regarding further characterization of cardiac cells/cardiomyocytes: was any attempt made to quantify contraction, and show that this could e.g. be blocked using ca2+ channel blockers such as verapamil? And did they check other markers to show that the cells had developed into more mature heart cells, including other sarcomere markers, but also ion channels etc?

To further investigate the function of contracting cardiac lineages, in the revised manuscript, we added an experiment to treat the cardio-pulmonary μTs with different concentrations of Verapamil (calcium channel blocker). Using Ca520 as a calcium influx indicator, we showed that the effect of Verapamil on the cardiac μTs was dose-dependent (Figure 6e). Furthermore, we have included cTnT and Sarcomeric Α Actinin staining on Day-22 μTs, which revealed a striated pattern (Figure 6c,d). Together, this provides more compelling evidence of the functionality of the induced cardiac tissues.

7. To achieve cardiopulmonary segregation as shown in Figure 4, the authors first used CKDCI medium from day 15-18, before switching to KDCI. What happened to alveolar type 2 marker expression upon CHIR withdrawal and segregation (was the alveolar niche fully established at day 18, and was therefore exogenous WNT activation no longer needed?), and was this segregation also observed if cultures were directly switched to KDCI on day 15?

In the revised manuscript, we have included a new figure showing that removal of CHIR on Day-18 led to a decrease in SFTPC expression (Figure 4—figure supplement 1c), indicating that the alveolar niche was not fully established by Day-18. Further investigation is needed to address whether the complete supporting niche for alveologenesis can be fully established and if the corresponding time point can be identified. In addition, we found that CHIR was indeed an important factor to induce alveologenesis, consistent with previous publications. 1,2 Without the addition of CHIR following Day-15 differentiation, we observed no maturation of AT2 cells on Day-18 of µT culture (Figure 5—figure supplement 1a,b,c). As for segregation, cardio-pulmonary µTs cultured directly in KDCI (without CHIR) from Day-15 were still able to undergo segregation over time (Figure 5—figure supplement 1d,e). We have included additional discussion regarding the findings and limitations of the present study (line 595-601).

Reviewer #2 (Recommendations for the authors):While a potentially very interesting model, much more in depth characterization is required. Each component, as well intervening stages of development should be thoroughly characterized and benchmarked.

Thank you for the comprehensive comments on our manuscript, we have performed an extensive revision as described above in response to the reviewer’s comments.

Reviewer #3 (Recommendations for the authors):Here are further clarifications to some of the above-mentioned weaknesses, as well as additional suggestions on how the manuscript could be improved to increase the impact of the work.

Lines 24-30 of manuscript: the authors reasoned that WNT signaling is required for both mesodermal and endodermal specification. On this basis, the authors explained why they decided to use WNT agonist CHIR for mesoderm and DE co-induction from hiPSC line BU3. To my knowledge, the justification is not supported by lung developmental studies. In other words, the data shows convincing evidence of meso- and endo- dermal induction by the end of stage-1/day 4, however the interpretations provided by the authors are not accurate. A more plausible explanation is that, within the first 48 hours (day 1-2), high dose of WNT agonist CHIR induces primitive streak formation from part of the hiPSCs in culture. Subsequently, the newly formed primitive streak provides endogenous WNT, BMP and Nodal/Activin signaling to pattern the rest of the hiPSCs in culture to form mesoderm and DE by the end of day 4. This alternative mechanistic explanation is supported by evidence from both the existing literature and the authors' own data. Brivanlou's group elegantly demonstrated that activation of WNT signaling solely is sufficient to induce primitive streak from hPSCs. In addition, the dosage of CHIR applied in day 1-2 is way too high to be able to induce endoderm with lung potential. Furthermore, the authors maintain the cells in pluripotent medium mTESR1 for the first 48 hours (lines 48-50 of manuscript), which is critical for the success of the protocol (lines 87-90 of manuscript, suppl Figure 3). This supports my hypothesis that CHIR induces part of the hiPSCs in culture to form primitive streak during the first 48 hours, mTESR1 is required to maintain the remaining hiPS cells in culture in pluripotent state during this stage, before they start meso- and endo-dermal differentiation in day 2-4. Consistently, the data shows that Activin A or BMP inhibition during days 2-4 is detrimental for DE or mesodermal induction. Single cell analysis at 48 hr would be sufficient to prove a mixture of "primitive streak and pluripotent cells" profile, which will provide valuable mechanistic insights.

We thank the reviewer for sharing the comprehensive opinion. We agree that endogenous signaling is the key to promote differentiation of both mesoderm and endoderm in our culture. In the revised manuscript, we have included new FACS data for T (Brachyury) and MIXL1 after 2 days of CHIR treatment and found that the majority of cells were T and MIXL1 positive (primitive streak/mesendoderm) (Figure 1—figure supplement 1d,e). Furthermore, we performed analysis of pluripotent cell marker (OCT4) and did not observe any remaining pluripotent cells after exposure to CHIR for 2 days (Figure 1—figure supplement 1c). Lastly, we appreciate the reviewer for bringing the related literature to our attention and have included the references as well as the related discussion to the revised manuscript (line 522-524).

The whole manuscript is based on the differentiation of a double reporter line derived from one parental hiPSC line BU3. For scientific rigor and assurance, it is necessary to demonstrate efficient co-differentiation of lung and heart from other parental hPSC lines (both reporter and non-reporter) using the combination of time and dosage depicted in the current protocol.

Thank you for pointing this out. In the revised manuscript, we have further showed that the presented protocol can be reproduced in another independent hiPSC line (BU1), as indicated by differentiation into cardiopulmonary progenitors on Day-15. Furthermore, we have included additional data for BU1 cell line regarding mesoderm and endoderm induction during Stage-1 (Figure 4—figure supplement 5b) and cardio-pulmonary μT formation from Day-15 progenitor cells. On Day-18, the BU1-derived dual-lineage μTs were positive for NKX2.1 and NKX2.5, similar to what we observed in BU3. Furthermore, the μTs were stained positive for SFTPC and HOPX, indicative of their potential for further alveolar maturation. Additionally, the BU1-derived NKX2.5 cardiac lineages co-expressed cTnT and Sarcomeric Α Actinin (Figure 4—figure supplement 5e).

It is convincing that most of the NKX2-1+ cells are lung fated and NKX2-5+ cells are cardiac fated, however, for scientific rigor and thoroughness, it is necessary to characterize the differentiation culture to exclude contamination of neural (EpCAM-) and thyroid (PAX8+) fated NKX2-1+ cells, and to confirm that all the NKX2-5+ cells co-express other cardiac progenitor markers such as KDR and PdgfR-α. Based on previous studies, some of the early lung progenitors co-express NKX2-5, therefore the authors must confirm whether NKX2-5+ cells at day 15 and on are NKX2-1- and EpCAM- cells.

To exclude the possibility of neural- and thyroid-fate of NKX2.1+ cells, in the revised manuscript, we have included the staining of TUJ1 (neural) and PAX8 (thyroid) on Day-15 differentiated cells. We did not observe any co-localization between either TUJ1 or PAX8 with NKX2.1 (Figure 2—figure supplement 2d). To confirm the identity of NKX2.5+ cells, we performed immunostaining and demonstrated their colocalization with cTnT, further confirming their cardiac fate. Regarding possible co-expression of NKX2.1 and NKX2.5, please refer to whole mount images in Figure 4b and Figure 4i. We have carefully checked each confocal section and did not observe colocalization between the two markers.

References:

1. Jacob, A. et al. Differentiation of Human Pluripotent Stem Cells into Functional Lung Alveolar Epithelial Cells. Cell stem cell 21, 472-488, doi:10.1016/j.stem.2017.08.014 (2017).

2. Huang, S. X. L. et al. Efficient generation of lung and airway epithelial cells from human pluripotent stem cells. Nat Biotech 32, 84-91, doi:10.1038/nbt.2754 (2014).

3. Steimle, J. D. et al. Evolutionarily conserved Tbx5-Wnt2/2b pathway orchestrates cardiopulmonary development. Proceedings of the National Academy of Sciences 115, E10615-E10624, doi:10.1073/pnas.1811624115 (2018).

4. Peng, T. et al. Coordination of heart and lung co-development by a multipotent cardiopulmonary progenitor. Nature 500, 589-592, doi:10.1038/nature12358 (2013).

5. Wang, Y. et al. Pulmonary alveolar type I cell population consists of two distinct subtypes that differ in cell fate. Proceedings of the National Academy of Sciences of the United States of America 115, 2407-2412, doi:10.1073/pnas.1719474115 (2018).

6. Frank, D. B. et al. Early lineage specification defines alveolar epithelial ontogeny in the murine lung. Proceedings of the National Academy of Sciences 116, 4362, doi:10.1073/pnas.1813952116 (2019).

7. Liebler, J. M. et al. Combinations of differentiation markers distinguish subpopulations of alveolar epithelial cells in adult lung. American journal of physiology. Lung cellular and molecular physiology 310, L114-120, doi:10.1152/ajplung.00337.2015 (2016).

8. Wang, S. et al. GSK-3beta Inhibitor CHIR-99021 Promotes Proliferation Through Upregulating β-Catenin in Neonatal Atrial Human Cardiomyocytes. Journal of cardiovascular pharmacology 68, 425-432, doi:10.1097/fjc.0000000000000429 (2016).

9. Tseng, A. S., Engel, F. B. and Keating, M. T. The GSK-3 inhibitor BIO promotes proliferation in mammalian cardiomyocytes. Chemistry and biology 13, 957-963, doi:10.1016/j.chembiol.2006.08.004 (2006).

10. Buikema, J. W. et al. Wnt Activation and Reduced Cell-Cell Contact Synergistically Induce Massive Expansion of Functional Human iPSC-Derived Cardiomyocytes. Cell stem cell 27, 50-63.e55, doi:10.1016/j.stem.2020.06.001 (2020).

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

While the manuscript is improved, significant concerns remain that need to be addressed for publication. In particular, reviewer 2 has specific concerns that we would like to see addressed in a revision

Reviewer #1 (Recommendations for the authors):

The authors have made important changes and added results of new experiments in this extensively revised manuscript. This has markedly improved the quality of the manuscript, and conclusions are much better supported by the results.

As a result the revised manuscript is interesting and provides a new and relevant model to study heart and lung development in a dish, and cross-talk between these developing organs.

Thank you for the comment. In the revised manuscript (line 473-475), we have provided additional discussion regarding HOPX expression and highlighted the need of future investigations of alveolar maturation (in particular towards AT1 fate) in a more comprehensive manner. We have followed the overall suggestion from the reviewers and editor and have provided an explicit clarification that progenitor specification is the focus of the current manuscript.

Reviewer #2 (Recommendations for the authors):In this manuscript, Ng et al., report on a system where cardiac mesoderm and pulmonary endoderm co-develop from pluripotent stem cells. This is of potential interest, as it could provide an integrated model for the study of human cardiopulmonary development. I still have quite a few concerns, though, which were not fully addressed by the authors.

1. In my first review, I pointed out that the main weakness lies in the lack of thorough characterization of the resulting cells and tissues. This is still the case. The characterization relies almost entirely on reporter gene expression and PCR for the same markers. In Figure 1, representative examples are shown of a limited set of markers. It would be nice to have quantification (and for intracellular flow quantification, analysis of cells not expressing the protein of interest should be used as a control). The qPCR data do shown that these markers are expressed, but they compare something to nothing (i.e. PSCs), so that fold changes will be always be very high. One image of the ultrastructure of type 2 cells is shown now, which is a plus. Overall, this work still does begs the question why the only pulmonary markers observed in this model are basically SFTPC and NKX2.1. Furthermore, type 2 identity has not been further verified with markers such as SFTPB, ABCA3, LAMP2. Another outstanding question for the lung component is whether any pulmonary mesenchyme was generated. This was still not addressed. In this era, at scRNAseq should be performed. This would also address several of my other concerns below.

We thank the reviewer for the suggestions and have organized our response in three subsections as described below.

(a) It would be nice to have quantification (and for intracellular flow quantification), analysis of cells not expressing the protein of interest should be used as a control.

We have performed additional flow cytometry experiments to characterize and quantitate mesoderm and endoderm induction during Stage-1 of differentiation. Our data showed 47.0% of CD13-expressing mesodermal cells and 45.2% of SOX17-expressing endodermal cells (Figure 1—figure supplement 1f-g). We have also included hiPSCs as our control, which expressed neither of these markers.

(b) The qPCR data do shown that these markers are expressed, but they compare something to nothing (i.e. PSCs), so that fold changes will be always be very high.

We thank the reviewer to point this out. In figures such as Figure 1d, we agree that normalization based on hiPSCs that did not express the target gene led to fold changes that appeared to be very high. However, the main comparisons that we intend to deliver in these experiments were those between different differentiation conditions, such as different CHIR concentrations, where in all conditions there were measurable expression of the target gene and the fold change between groups is much more realistic than the comparison versus hiPSCs. The high fold change values relative to hiPSC control did not affect us to perform accurate comparison regarding gene expression levels between different differentiation conditions.

(c) Overall, this work still does begs the question why the only pulmonary markers observed in this model are basically SFTPC and NKX2.1. furthermore, type 2 identity has not been further verified with markers such as SFTPB, ABCA3, LAMP2. Another outstanding question for the lung component is whether any pulmonary mesenchyme was generated.

In this manuscript, we used CKDCI (CHIR99021, KGF, dexamethasone, cAMP, and IBMX) maturation medium that have been shown to selectively promote differentiation of alveolar type 2 (AT2) cells of the distal lung [1, 2]. Further, based on the reviewer’s suggestion, we have included characterization of an additional marker (pro-SFTPB) to verify AT2 identity (Figure 6—figure supplement 1a). Staining of S100A4, a marker for mesenchyme, has been performed in the 3D cardio-pulmonary μT (Figure 6—figure supplement 1c).

2. It is still not clear which cardiac cells are generated: ventricular, atrial, endocardium, epicardium, conducting tissue? This is the same comment as in my first review. An experiment with verapamil or some cells expressing α-actinin, as shown here, do not address this question. Atrial and ventricular cells contract. I would also add that in many PSC differentiation cultures where lineages are not sufficiently rigorously specified, contamination with beating cardiomyocytes is quite typical. This may be the case here. It is also difficult to understand why type 2 cells and unspecified cardiac cells are the only output of this model.

(a) It is still not clear which cardiac cells are generated.

In the revised manuscript, we have performed additional staining to characterize the identity of the induced cardiac cells, including ventricular myocytes (MLC2v), atrial myocytes (COUPTFII), endocardium (NFATC) and epicardium (WT1). In Figure 2—figure supplement 1d, we observed presence of NKX2.5+COUPTFII+ cardiac cells but negative for the rest markers such as MLC2v, NFATC and WT1, suggesting that the some of the cardiac cells have been specified into atrial myocytes. The observation of atrial induction is consistent with the presence of retinoic acid in our co-differentiation medium, which has been shown to be critical for atrial specification [3-5]. Further induction of other cardiac cell lineages is expected to require specifically formulated maturation medium and extended maturation time.

(b) Atrial and ventricular cells contract. I would also add that in many PSC differentiation cultures where lineages are not sufficiently rigorously specified, contamination with beating cardiomyocytes is quite typical. This may be the case here.

We agree to the reviewer’s point that minor cardiac population could emerge during differentiation where mesodermal cells are permitted. However, based on our flow cytometry analysis (Figure 2—figure supplement 1a), a balanced induction of pulmonary (32.8%) and cardiac (39.3%) cells were observed. In Figure 2—figure supplement 1b, we further confirmed that emergence of cardiac cells could be modulated by Activin A. Thus, the emergence of cardiac lineage in the presented co-differentiation system was a controlled process.

(c) It is also difficult to understand why type 2 cells and unspecified cardiac cells are the only output of this model.

In this manuscript, we used CKDCI (CHIR99021, KGF, dexamethasone, cAMP, and IBMX) maturation medium that have been shown to selectively promote differentiation of alveolar type 2 (AT2) cells of the distal lung [1, 2, 6]. The same goes for cardiac subtype specification. The observation of atrial induction is consistent with the presence of retinoic acid in our co-differentiation medium, which has been shown to be critical for atrial specification [3-5]. Further induction of other cardiac cell lineages is expected to require specifically formulated maturation medium and extended maturation time.

3. In Figure 2-S1d, the nuclear stains for NKX2.5 and NKX2.1 do not seem to align with DAPI. The same is true in Figure 3-S1a, where the higher magnifications for NKX2.1, NKX2.5 and DAPI do not seem to align. The same is true again for Figure 4-S1d, where the pattern of NKX2.1 is not the inverse of that of NK2.5 (no DAPI shown here). Furthermore, there is much more cTNT positivity than NKX2.5 expression in Figure 2-S1d. One would expect the opposite. These potential inconsistencies make the images difficult to interpret.

Due to high confluency after long term culture (15 days), the cells were arranged in multiple layers, resulting in superimposition, and thus it was difficult to appreciate DAPI staining with the presence of different cell lineages. Although confocal imaging would have solved this issue, but the co-differentiation was performed in 96 well plate, rendering difficulty to image with confocal microscopy. We have included respective DAPI images in Figure 2—figure supplement 1d, Figure 3—figure supplement 1a, and Figure 4—figure supplement 1d. Figure 4-S1d represented the cells cultured on Transwell Membrane. While one may expect the inverse relationship between NKX2.1 and NKX2.5 staining, within Day-15 differentiated cells, there were still about 20% of cell that belong to neither cardiac or pulmonary lineages. In Figure 2—figure supplement 1d, we observed that majority of the NKX2.5 cells were cTnT positive.

4. Along the same lines, in Figure 3a and 3d, as well in Figure 3-S1b and Figure 3-S3b (each time lower right panel), there seems to be co-expression of NKX2.1 and NKX2.5. Or is this superposition? Are there cells that express neither marker? This is difficult to evaluate without DAPI images. Also, could it be that in Figure 3-S1, the reduced cardiac and pulmonary differentiation is due to reduced proliferation? Again, without a DAPI image, this is difficult to evaluate.

We thank the reviewer for the comment. In Figure 3a, 3d, Figure 3—figure supplement 1b, Figure 3—figure supplement 3b, due to high cell confluency, the resulting cells from different lineages were superimposed. As the reviewer suggested, we have included DAPI images. By referring to Figure 4b, using confocal imaging of cardio-pulmonary μT, we did not observe obvious co-localization between NKX2.1 and NKX2.5 expression. Indeed, there were cells that do not express either cardiac or pulmonary marker as shown by flow cytometry analysis, where there were about 20% cells that expressed neither NKX2.1 nor NKX2.5 (Figure 2—figure supplement 1a). In Figure 3—figure supplement 1, it is still possible that the reduced cardio-pulmonary differentiation was due to reduced proliferation since initial differentiation medium in B27 minus insulin was harsh compared to mTESR1. However, the well was full of cells at the end of differentiation.

5. This sentence is unclear: 'This may in part be due to the reduction of cardiac progenitor as indicated by significant downregulation of NKX2.5 gene expression (Figure 4f, Figure 4—figure supplement 1d).' (line 292). According to Figure 4f, there is no downregulation of NKX2.5 in suspension culture.

Thank you for the comment. The sentence “This may in part be due to the reduction of cardiac progenitor as indicated by significant downregulation of NKX2.5 gene expression (Figure 4f, Figure 4—figure supplement 1d).” In this context, we were comparing 3D suspension culture with ALI culture. We found 3D suspension is a more robust system for alveolar maturation as indicated by SFTPC gene induction. We anticipated that the lower alveolar specification efficiency in ALI culture was due to reduction of cardiac progenitor as pointed out in Figure 4f (ALI vs. Day-15 cells) and Figure 4—figure supplement 1d (Immunostaining on ALI) (line 295).

6. In Figure 4c, which relies entirely on reporters, it would be nice to co-stain for the actual proteins and other markers identifying type 2 cells. I note that in Figure 3c, the fluorescence patterns of the GFP and TdTomato reporters is very different, yet they should both be cytoplasmic. In the reporter studies, it would be good to co-differentiate a non-reporter expressing line as a reporter-negative control for autofluorescence in Figure 4 and its supplements. Figure 4-S1c is apparently not commented on the text.

We thank the reviewer for pointing this out. We have included staining for both pro-SFTPC and pro-SFTPB to further verify type 2 identify as demonstrated in Figure 6b and Figure 6—figure supplement 1a. The fluorescence reporter TdTomato is generally a subset of GFP, as only a portion of the GFP+ lung progenitors have matured into type 2 cells (labelled by TdTomato). Thus, not all GFP cells would express SFTPC. To address the possibility of autofluorescence, as the reviewer suggested, we have added representative images of non-reporter BU1 hiPSC line as a control (Figure 4—figure supplement 6). We have also commented Figure 4—figure supplement 1c in the text (line 286-287).

7. The authors mention in the rebuttal evidence for type 1 cells, as shown by staining for HOPX in a non-reporter line (Figure 4-S5) and in the reporter line (Figure 6b). However, HOPX is not a marker for type 1 cells in humans, and is expressed broadly, including in cardiomyocytes (see, among others, Friedman et al., Cell Stem Cell 2018). In Figure 6b, there does not appear to be much pro-SFTPC, and that seems to be on the outside of the structure, whereas in Figures 4 and 5, the NKX2.1+ cells are on the inside.

(a) HOPX is not a marker for type 1 cells in humans, and is expressed broadly, including in cardiomyocytes.

Thank you for the comment. We agree that HOPX alone does not fully define AT1 cells. In the revised manuscript (line 473-475), we have provided additional discussion regarding HOPX expression and highlighted the need of future investigations of alveolar maturation (in particular towards AT1 fate) in a more comprehensive manner. Although HOPX expression has been observed in cardiomyocytes, [7] we did not observe HOPX expression following establishment of cardio-pulmonary progenitors, and the induction of HOPX coincided with alveolar maturation during Day-15 to Day-18.

(b) In Figure 6b, there does not appear to be much pro-SFTPC, and that seems to be on the outside of the structure, whereas in Figures 4 and 5, the NKX2.1+ cells are on the inside.

In Figure 6b, the μTs have been cultured in maturation medium without CHIR for 4 days. With CHIR withdrawal after Day-18, tissues in the μTs would have reorganized themselves and the pulmonary tissue started to undergo segregation from the cardiac tissue, leading to the pulmonary lineages to move towards the edge (Figure 5a). While in Figure 4, and Figure 5, the μTs were of Day-18 (before cardio-pulmonary segregation).

8. The authors should not call Act. A 'TGF-β' signaling. It is a TGF-β family ligand, but signals through its own receptors. It recapitulates Nodal signaling in vitro.

Thanks for the recommendation. We have changed the term “TGF-β signaling” to “nodal signaling” for activin A.

9. Finally, there is still no sufficient quantification of differentiation efficiency and yield. NKX2.1, for example is also expressed in the forebrain and in the thyroid. The minimal conditions described typical yield neurectoderm. TUJ1 was examined, but other markers, such as PAX 6 should be included.

The yield of NKX2.1- and NKX2.5-expressing cells on Day-15 cells were as shown in Figure 2—figure supplement 1a. We have also demonstrated that NKX2.1+ cells on Day-15 do not co-express another forebrain marker such as PAX6. The data has been included in Figure 2—figure supplement 1d.

Recommendations for the authors

While a still potentially interesting model, much more in depth characterization is still required as well. In particular the nature and yield of the output should be better characterized, and figure quality and consistency should be paid attention to.

Thank you for the comprehensive suggestions, and we have revised the manuscript to address comments as described above.

References:

1. Huang, S.X.L., et al., Efficient generation of lung and airway epithelial cells from human pluripotent stem cells. Nat Biotech, 2014. 32(1): p. 84-91.

2. Jacob, A., et al., Differentiation of Human Pluripotent Stem Cells into Functional Lung Alveolar Epithelial Cells. Cell Stem Cell, 2017. 21(4): p. 472-488.e10.

3. Zhang, Q., et al., Direct differentiation of atrial and ventricular myocytes from human embryonic stem cells by alternating retinoid signals. Cell Research, 2011. 21(4): p. 579-587.

4. Miao, S., et al., Retinoic acid promotes metabolic maturation of human Embryonic Stem Cell-derived Cardiomyocytes. Theranostics, 2020. 10(21): p. 9686-9701.

5. Gassanov, N., et al., Retinoid acid-induced effects on atrial and pacemaker cell differentiation and expression of cardiac ion channels. Differentiation, 2008. 76(9): p. 971-80.

6. Hawkins, F.J., et al., Derivation of Airway Basal Stem Cells from Human Pluripotent Stem Cells. Cell Stem Cell, 2021. 28(1): p. 79-95.e8.

7. Friedman, C.E., et al., Single-Cell Transcriptomic Analysis of Cardiac Differentiation from Human PSCs Reveals HOPX-Dependent Cardiomyocyte Maturation. Cell stem cell, 2018. 23(4): p. 586-598.e8.

8. Hogan, B.L., Bone morphogenetic proteins in development. Curr Opin Genet Dev, 1996. 6(4): p. 432-8.

9. Schier, A.F., Nodal signaling in vertebrate development. Annu Rev Cell Dev Biol, 2003. 19: p. 589-621.

10. Yamaguchi, T.P., Heads or tails: Wnts and anterior-posterior patterning. Curr Biol, 2001. 11(17): p. R713-24.

https://doi.org/10.7554/eLife.67872.sa2

Article and author information

Author details

  1. Wai Hoe Ng

    Department of Biomedical Engineering, Carnegie Mellon University, Pittsburgh, United States
    Contribution
    Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review and editing
    Competing interests
    This author is a co-inventor of a related provisional patent application (No. 63/124422) entitled 'Methods for simultaneous cardio-pulmonary differentiation and alveolar maturation from human pluripotent stem cells'
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-6876-6542
  2. Elizabeth K Johnston

    Department of Biomedical Engineering, Carnegie Mellon University, Pittsburgh, United States
    Contribution
    Investigation, Methodology, Writing – review and editing
    Competing interests
    This author is a co-inventor of a related provisional patent application (No. 63/124422) entitled 'Methods for simultaneous cardio-pulmonary differentiation and alveolar maturation from human pluripotent stem cells'
  3. Jun Jie Tan

    Advanced Medical and Dental Institute, Universiti Sains Malaysia, Penang, Malaysia
    Contribution
    Supervision, Writing – review and editing
    Competing interests
    No competing interests declared
  4. Jacqueline M Bliley

    1. Department of Biomedical Engineering, Carnegie Mellon University, Pittsburgh, United States
    2. Department of Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, United States
    Contribution
    Investigation, Writing – review and editing
    Competing interests
    No competing interests declared
  5. Adam W Feinberg

    1. Department of Biomedical Engineering, Carnegie Mellon University, Pittsburgh, United States
    2. Department of Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, United States
    Contribution
    Resources, Supervision, Writing – review and editing
    Competing interests
    No competing interests declared
  6. Donna B Stolz

    Center for Biologic Imaging, University of Pittsburgh, Pittsburgh, United States
    Contribution
    Investigation
    Competing interests
    No competing interests declared
  7. Ming Sun

    Center for Biologic Imaging, University of Pittsburgh, Pittsburgh, United States
    Contribution
    Investigation
    Competing interests
    No competing interests declared
  8. Piyumi Wijesekara

    Department of Biomedical Engineering, Carnegie Mellon University, Pittsburgh, United States
    Contribution
    Investigation, Visualization
    Competing interests
    No competing interests declared
  9. Finn Hawkins

    Center for Regenerative Medicine of Boston University and Boston Medical Center, Boston, United States
    Contribution
    Resources, Supervision, Writing – review and editing
    Competing interests
    No competing interests declared
  10. Darrell N Kotton

    Center for Regenerative Medicine of Boston University and Boston Medical Center, Boston, United States
    Contribution
    Resources, Supervision
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-9604-8476
  11. Xi Ren

    Department of Biomedical Engineering, Carnegie Mellon University, Pittsburgh, United States
    Contribution
    Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Project administration, Resources, Supervision, Writing – original draft, Writing – review and editing
    For correspondence
    xiren@cmu.edu
    Competing interests
    This author is a co-inventor of a related provisional patent application (No. 63/124422) entitled 'Methods for simultaneous cardio-pulmonary differentiation and alveolar maturation from human pluripotent stem cells'
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-3187-1311

Funding

Samuel & Emma Winters Foundation (A025662)

  • Xi Ren

Carnegie Mellon University

  • Xi Ren

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

This work was supported by Samuel & Emma Winters Foundation A025662 (to XR) and the Department of Biomedical Engineering and College of Engineering at Carnegie Mellon University. We are grateful to Drs. Yu-li Wang and David Li for advice on collective cell migration, to Misti West for laboratory management, and to Anthony Green and the Pitt Biospecimen Core at the University of Pittsburgh for assistance with histology. We also thank Barbie Varghese for proofreading the manuscript.

Senior and Reviewing Editor

  1. Paul W Noble, Cedars-Sinai Medical Center, United States

Reviewer

  1. Sarah XL Huang, The University of Texas Health Science Center at Houston, United States

Publication history

  1. Received: February 25, 2021
  2. Preprint posted: March 3, 2021 (view preprint)
  3. Accepted: January 7, 2022
  4. Accepted Manuscript published: January 12, 2022 (version 1)
  5. Version of Record published: February 15, 2022 (version 2)

Copyright

© 2022, Ng et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 1,710
    Page views
  • 319
    Downloads
  • 4
    Citations

Article citation count generated by polling the highest count across the following sources: PubMed Central, Crossref, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Wai Hoe Ng
  2. Elizabeth K Johnston
  3. Jun Jie Tan
  4. Jacqueline M Bliley
  5. Adam W Feinberg
  6. Donna B Stolz
  7. Ming Sun
  8. Piyumi Wijesekara
  9. Finn Hawkins
  10. Darrell N Kotton
  11. Xi Ren
(2022)
Recapitulating human cardio-pulmonary co-development using simultaneous multilineage differentiation of pluripotent stem cells
eLife 11:e67872.
https://doi.org/10.7554/eLife.67872
  1. Further reading

Further reading

    1. Developmental Biology
    Yanling Xin, Qinghai He ... Shuyi Chen
    Research Article

    N 6-methyladenosine (m6A) is the most prevalent mRNA internal modification and has been shown to regulate the development, physiology, and pathology of various tissues. However, the functions of the m6A epitranscriptome in the visual system remain unclear. In this study, using a retina-specific conditional knockout mouse model, we show that retinas deficient in Mettl3, the core component of the m6A methyltransferase complex, exhibit structural and functional abnormalities beginning at the end of retinogenesis. Immunohistological and single-cell RNA sequencing (scRNA-seq) analyses of retinogenesis processes reveal that retinal progenitor cells (RPCs) and Müller glial cells are the two cell types primarily affected by Mettl3 deficiency. Integrative analyses of scRNA-seq and MeRIP-seq data suggest that m6A fine-tunes the transcriptomic transition from RPCs to Müller cells by promoting the degradation of RPC transcripts, the disruption of which leads to abnormalities in late retinogenesis and likely compromises the glial functions of Müller cells. Overexpression of m6A-regulated RPC transcripts in late RPCs partially recapitulates the Mettl3-deficient retinal phenotype. Collectively, our study reveals an epitranscriptomic mechanism governing progenitor-to-glial cell transition during late retinogenesis, which is essential for the homeostasis of the mature retina. The mechanism revealed in this study might also apply to other nervous systems.

    1. Developmental Biology
    2. Genetics and Genomics
    Xiaodong Li, Patrick J Gordon ... Edward M Levine
    Research Article

    An important question in organogenesis is how tissue-specific transcription factors interact with signaling pathways. In some cases, transcription factors define the context for how signaling pathways elicit tissue- or cell-specific responses, and in others, they influence signaling through transcriptional regulation of signaling components or accessory factors. We previously showed that during optic vesicle patterning, the Lim-homeodomain transcription factor Lhx2 has a contextual role by linking the Sonic Hedgehog (Shh) pathway to downstream targets without regulating the pathway itself. Here, we show that during early retinal neurogenesis in mice, Lhx2 is a multilevel regulator of Shh signaling. Specifically, Lhx2 acts cell autonomously to control the expression of pathway genes required for efficient activation and maintenance of signaling in retinal progenitor cells. The Shh co-receptors Cdon and Gas1 are candidate direct targets of Lhx2 that mediate pathway activation, whereas Lhx2 directly or indirectly promotes the expression of other pathway components important for activation and sustained signaling. We also provide genetic evidence suggesting that Lhx2 has a contextual role by linking the Shh pathway to downstream targets. Through these interactions, Lhx2 establishes the competence for Shh signaling in retinal progenitors and the context for the pathway to promote early retinal neurogenesis. The temporally distinct interactions between Lhx2 and the Shh pathway in retinal development illustrate how transcription factors and signaling pathways adapt to meet stage-dependent requirements of tissue formation.