Comparison of Transplantation of Lung Organoid Cell Types: One Size Does Not Fit All

To the Editor:

Recent advances in identification of lung cell types and understanding their role in regeneration have given hope for potential opportunities to discover novel therapies for lung diseases. Numerous epithelial progenitor cells in the airways and alveoli repopulate the lung epithelium after injury (1–5). Concurrently, three-dimensional organoid cultures established with lung epithelial progenitor cells have provided a new way to model lung regeneration. However, additional assays are needed to document the functional capacity of lung progenitor cells in vivo; a gold standard of stem cell activity, akin to hematopoietic stem cell bone marrow transplantation assay, is still needed in the lung field.

Here we report successful transplantation using different types of lung progenitor cells, as an illustration that side-by-side comparative assays are crucial to develop optimal, robust methodology for transplantation. A limited number of studies showed cell retainment in mouse lungs after delivery of various types of lung cells, including embryonic and adult mouse progenitor epithelial cells from proximal and distal lungs (6–13). However, the experiments in published studies are conducted with varying cell types, different injury models, and disparate delivery routes, making the comparison challenging. Here, we directly compare the transplantation capabilities of different lung progenitor cells in the same orthotopic delivery assay.

We first asked whether freshly sorted lung cells enriched for epithelial cells (CD31⁻/CD45⁻) or alveolar type II cells (SCA1⁻/EPCAM⁺/CD31⁻/CD45⁻ cells; Sca1⁻ hereafter) are capable of engraftment. A cohort of 7- to 10-week-old wild-type C57BL/6 mice was preconditioned 1 day before cell transplantation with one dose of intratracheal bleomycin (Figure 1A) (3). Donor CD31⁻/45⁻ and Sca1⁻ cells from β-actin DsRed mice were flow-sorted using a CD31⁻ CD45⁻ and CD31⁻ CD45⁻ EPCAM⁺ SCA1⁻ gating strategy, respectively, as previously described (Figure 1B) (14). A total of 0.5–1 million cells were transplanted intratracheally using an orally inserted catheter. Recipient mouse lungs were harvested 1–6 weeks after transplant and analyzed by flow cytometry and histology for donor cell detection. Quantitative detection of donor cells was measured as a percentage of DsRed⁺ cells within the CD31⁻/45⁻ population of the recipient mouse lungs by flow cytometry (Figure 1E). Few to no CD31⁻/45⁻ or SCA1− cells were detected by flow cytometry (0.13 ± 0.05% vs. 0.12 ± 0.16%; P = 0.88) or immunofluorescence staining, suggesting the transplanted cells had little or no engraftment (Figures 1C and 1G).

Figure 1.

Comparison of transplantation of different cell populations. (A) Schematic of experiment design. (B) Representative flow cytometry gating from DsRed mouse in preparation of cells for transplant or organoid culture. Gating strategy is color-coded for three populations from the DAPI⁻, DsRed⁺ cells: red, CD45⁻, CD31⁻ cells; blue, CD45⁻, CD31⁻, EpCAM⁺, SCA1⁻ cells; green, CD45⁻, CD31⁻, EpCAM⁺, SCA1⁺ cells. (C) Graphical analysis of recipient mouse lungs by flow cytometry. A total of 0.5–1 million freshly sorted CD31⁻/45⁻ cells or freshly sorted SCA1⁻ cells were transplanted and analyzed 1–6 weeks later. The fraction of the CD31⁻/45⁻ population with DsRed in the recipient mouse lung is graphed. (D) Representative fluorescence images of DsRed⁺ organoids derived from SCA1⁻ cells, SCA1⁺ cells, or mouse tracheal epithelial cells (MTEC). SCA1⁻ and SCA1⁺ organoid images were captured at 20× magnification. MTEC organoids image was captured at 4× magnification. Scale bars, 50 μm. Note: SCA1⁺ cells produce alveolar and bronchiolar organoids, shown left and right, respectively, whereas SCA1⁻ cells only produce alveolar organoids. (E) Representative flow cytometry gating for analysis of recipient mouse lungs. The fraction of the live (DAPI⁻) CD31⁻/45⁻ population with DsRed in the recipient mouse lung was determined. (F) Graphical analysis of flow cytometry data as in (E) of recipient mouse lung. A total of 0.5–1 million cells from SCA1⁻ or SCA1⁺ organoid cultures were transplanted and analyzed 2–3 weeks later. (G) Graphical analysis of flow cytometry data of transplants with MTEC organoids. (H) Immunofluorescence staining of recipient mouse lungs transplanted with SCA1⁻, SCA1⁺, or MTEC organoid cultures. Lower panel shows individual channels from dashed rectangle insets seen in the upper panel. Blue: DAPI; red, anti-DsRed; green: SPC; turquoise: CCSP. Scale bars, 100 μm. (C, F, and G) Whisker box plots: The box ends with lower and upper quartiles, and within the box a horizontal line indicates the median. Lines drawn from each end of the box define the maximum and/or the minimum values. P values were determined using a Mann-Whitney rank test. N.S., P ⩾ 0.05, and *P < 0.05. CCSP = clara cell secretory protein; DsRED = fluorescent red protein; EpCAM = epithelial cell adhesion molecule; FSC = forward scatter; i.t. = intratracheal delivery; N.S. = not significant; SCA-1 = stem cell antigen-1; SPC = surfactant protein C.

Next, we asked whether epithelial cells from organoids derived from Sca1⁻ (CD31⁻/45⁻ EPCAM⁺ Sca1⁻) cells or Sca1⁺ (CD31⁻/45⁻ EPCAM⁺ Sca1⁺) cells can be successfully transplanted. Sorted Sca1⁻ and Sca1⁺ cells from β-actin DsRed mice were cultured with stromal cells in the three-dimensional organoid coculture system as previously described by Lee and colleagues. (14). Sca1⁻ cells are known to yield only alveolar organoids, whereas Sca1⁺ cells generate both alveolar and bronchiolar organoids (Figure 1D). As before, we delivered single-cell suspensions of 0.5–1 million cells from organoid cultures intratracheally 1 day after bleomycin injury and harvested recipient lungs 2–3 weeks after transplant for detection of DsRed⁺ cells by flow cytometry and histological analysis (Figures 1F–1H). Lungs from recipient mice transplanted with cells from organoids derived from Sca1⁻ cells had little to no detection of DsRed⁺ transplanted cells, similar to results from transplantation of freshly sorted Sca1⁻ cells (Figure 1F). In contrast, lungs of recipient mice transplanted with cells from organoids derived from Sca1⁺ cells had significantly higher engraftment rates as determined by flow cytometry (Sca1⁻ 0.07 ± 0.12% vs. Sca1⁺ 3.0 ± 5.2%; P = 0.001) and immunofluorescence staining (Figures 1F and 1H). Immunostaining showed that transplanted cells could express either airway (CCSP) or alveolar (SPC) epithelial cell markers, or in some cells, both (Figure 1H), suggesting that epithelial cells in Sca1⁺ organoid cultures can contribute to airway and alveolar cell engraftment. These results showed that the transplantation conditions suitable for engraftment of Sca1⁺ progenitor cells are insufficient for the transplantation of alveolar cell progenitors.

To explore if other epithelial progenitor cell types also exhibit a distinct capacity for transplantation, we tested the engraftment properties of basal cells, the progenitor cells of the murine trachea. We used mouse tracheal epithelial cells and cultured them into organoids on semipermeable supported membranes in an air–liquid interface (Figure 1D) as previously described by You and colleagues. (15). We delivered single-cell suspensions of 1 million cells from organoid cultures intratracheally 1 day after intratracheal bleomycin injury and harvested recipient mice lungs 3–6 weeks after transplant. Transplantation of mouse tracheal epithelial organoid cells showed more robust engraftment than Sca1⁺ organoid cells (3.0 ± 5.2% vs. 30.43 ± 33.11%; P = 0.015) (Figures 1G and 1H). Notably, engrafted cells were found to express the airway cell marker CCSP (Figure 1H). Thus, unlike alveolar progenitors, different epithelial progenitors with airway differentiation potential can engraft in the bleomycin-injured wild-type lung.

In conclusion, we have used a quantitative assessment to establish that the transplantation capabilities of different lung epithelial stem cell types are not identical. Cells grown in organoid cultures had more robust transplantation capacity than freshly sorted cells. Alveolar cells did not successfully engraft into the lung milieu after bleomycin treatment, whereas airway cells were robustly integrated. The utility of intravenous delivery of a combination of epithelial and hematopoietic cells for alveolar engraftment has been demonstrated (13), raising the possibility that different delivery routes may be needed for transplantation of alveolar cells. Indeed, it will be important to address the many factors that may need to be optimized for successful transplantation (e.g., delivery method, type of injury used for preconditioning, culture conditions, and host conditions) of different cell types that were not addressed in our study. Do different lung progenitor cell types sample their potential new niche differently, as did Goldilocks when trying different bears’ beds? Are the factors required for engraftment distinct, and do they need to be tempered for each cell type? These questions can now be addressed using the quantitative and qualitative system we have established. It is unlikely that any single preconditioning strategy will be fruitful for the delivery and successful transplantation of all types of pulmonary epithelial cells. Thus, for future cell therapies in the lungs, one size will not fit all.