Lung cell transplantation for pulmonary fibrosis

Abstract

Idiopathic pulmonary fibrosis is a major cause of death with few treatment options. Here, we demonstrate the therapeutic efficacy for lung fibrosis of adult lung cell transplantation using a single-cell suspension of the entire lung in two distinct mouse systems: bleomycin treatment and mice lacking telomeric repeat-binding factor 1 expression in alveolar type 2 (AT2) cells (SPC-Cre TRF1^fl/fl), spontaneously developing fibrosis. In both models, the progression of fibrosis was associated with reduced levels of host lung progenitors, enabling engraftment of donor progenitors without any additional conditioning, in contrast to our previous studies. Two months after transplantation, engrafted progenitors expanded to form numerous donor-derived patches comprising AT1 and AT2 alveolar cells, as well as donor-derived mesenchymal and endothelial cells. This lung chimerism was associated with attenuation of fibrosis, as demonstrated histologically, biochemically, by computed tomography imaging, and by lung function measurements. Our study provides a strong rationale for the treatment of lung fibrosis using lung cell transplantation.

Fig. 1.. Assessment of durable BLM-induced lung fibrosis.

(A) Typical micro-CT 3D image of lung tissue from mice following treatment with phosphate-buffered saline (PBS; left) and at 8 weeks after completion of intraperitoneal BLM administration for 4 weeks (total of eight doses, of 0.035 U/g, each). Fibrotic tissue is indicated in purple. (B) Percent fibrotic tissue (tissue volume area divided by the total lung volume) was calculated on the basis of the micro-CT scans 8 weeks after completion of BLM treatment. (C) Quantitative determination of total collagen in the paraffin-embedded lung tissue measured by the hydroxyproline assay. (D to H) Representative lung staining 8 weeks after completion of BLM treatment (bottom rows) compared to treatment with PBS (top row). (D) Hematoxylin and eosin (H&E), (E) trichrome, (F) alpha–smooth muscle actin (α-SMA), (G) fibronectin, and (H) collagen IV staining. Scale bars, 500 μm (D to F, top and middle rows), 200 μm (D to F, bottom row), 200 μm (G and H, top and middle rows), and 50 μm (G and H, bottom row). (I) Lung function assessment measured by FlexiVent at different time points after completion of BLM treatments. Results were taken from one transplantation experiment. For the PBS control group and groups analyzed at 4, 6, 9, 10, 12, and 16 weeks of BLM treatment, N = 18, 5, 9, 6, 11, 9 and 10, respectively. One-way ANOVA with Dunnett’s test was used in (I), and an unpaired t test was used in (B) and (C) for statistical analysis for comparison between BLM- and PBS-treated groups; **P < 0.002, ***P < 0.0002, and ****P < 0.0001.

Fig. 2.. Patch-forming lung progenitors in donor mice treated with BLM for different time periods.

(A) TdTomato⁺ and GFP⁺ C57BL/6 male mice (donors) were treated with BLM for 1, 2, or 4 weeks, and thereafter, mice were euthanized, and a single-cell suspension derived from the lungs (1:1 mixture of both fluorescent donor types) was transplanted at different cell doses (2 × 10⁶, 4 × 10⁶, or 8 × 10⁶ cells) into recipient mice conditioned with CY and 6-Gy TBI. Donor-derived patches were determined in the lungs of recipient mice at 64 days after transplantation. (B) Typical staining at 64 days after transplantation of donor-derived lung patches (TdTomato or GFP-positive) in lungs of mice receiving lung cells from donors treated with BLM for 1, 2, or 4 weeks. Scale bars, 500 μm. (C) Quantitative analysis of the number of donor-derived patches per 2-mm² lung area in the lungs of recipient mice at 64 days after transplantation (20 to 50 regions of 2 mm² were counted, from three to four mice in each group).

Fig. 3.. Donor-derived patches after transplantation of TdTomato⁺lung cells into recipient mice treated with BLM for different periods.

(A) Experimental scheme. Recipient mice were treated with different numbers of BLM doses as indicated by the colored arrows. After completion of BLM treatment, the mice were transplanted with 8 × 10⁶ TdTomato lung cells. The lungs were harvested and assessed for the level of donor-derived patches 8 weeks after transplantation. (B) Number of donor-derived lung patches per 2-mm² lung area in recipient mice treated with BLM for different periods. Each dot represents an average number of patches per 2 mm² based on at least 10 measurements in each individual mouse. Serial sections, 12 μm in thickness, were taken along the longitudinal axis of the lobe. A fixed distance (30 to 40 μm) between the sections was maintained to allow analysis at different depths of the lung, analyzing at least 10 sections across the entire lung. Lung slices were analyzed by fluorescence microscopy. (C) Percentage of chimeric mice, defined by at least four visible donor-derived patches consisting of more than 20 donor cells each. For groups analyzed after 1, 2, 3, 4, and 5 weeks of BLM treatment, N = 5, 9, 6, 5, and 5 mice, respectively. (D) Typical examples of donor-derived (red) lung patches in low magnification views of the lung of recipient mice treated with BLM for different periods. Scale bar, 500 μm. The top and bottom panels show separate examples of two representative mice from each group. *P < 0.05, **P < 0.002.

Fig. 4.. Cell composition of GFP⁺ donor-derived patches in the lungs of transplanted mice following treatment with BLM for 4 weeks.

(A to D) Typical immunohistological staining of different lung cell types including (A) AQP5⁺ AT1 alveolar cells, (B) LAMP3⁺ AT2 alveolar cells, (C) CD31⁺ endothelial cells, and (D) PDGFRα⁺ mesenchymal cells. For each staining, low magnification (left column a; scale bars, 10 μm) shows a donor-derived GFP⁺ patch (green), followed by high magnification of the indicated area within each patch (right columns b to d; scale bars, 3 μm) showing typical staining with different markers. Column b: Double staining GFP expression plus indicated marker (magenta). Column c: Indicated marker (magenta) plus nuclear staining (cyan). Column d: GFP (green) and nuclei (Hoechst, cyan). Column e: Distribution of average percentages of different cell types in donor-derived lung patches. To quantitate donor-derived AT1, AT2, endothelial, and mesenchymal cells, we used double staining for GFP with different cell markers, as illustrated in (A) to (D) or in figs. S5 and S6. To rule out potential errors in tracking the boundaries of each donor-derived cell in this quantitative analysis, verification of staining along 3 mm of the z axis (at 1-mm intervals) (as illustrated in fig. S6) was performed for every counted cell. Cells exhibiting staining that was not consistent across the z axis were not included in the final percentage analysis of donor-derived lung cell subpopulations. The figure represents a minimum of 20 patches for each staining obtained from three to six chimeric mice.

Fig. 5.. Attenuation of fibrosis and functional benefit in the BLM mouse model 8 weeks after transplantation of lung cells.

(A) Experimental scheme. Mice were treated with BLM and transplanted with 4 × 10⁶ GFP⁺ and 4 × 10⁶ TdTomato⁺ lung cells. (B) Fluorescence imaging of transplanted lung tissue 8 weeks after transplantation of a 1:1 mixture of TdTomato⁺ and GFP⁺ C57BL donor-derived lung cells. Scale bar, 200 μm. (C) Chimerism level defined by the average patch number per 2-mm² lung area. Mice exhibiting less than an average of four patches per 2-mm² lung tissue (purple) were excluded from further comparisons. Data were pooled from two independent experiments. (D) Trichrome staining of lung treated with vehicle (PBS), BLM, or BLM with transplantation of 8 × 10⁶ donor-derived cells (Trans). Scale bars, 200 μm. All groups were examined for fibrosis 13 weeks after initiation of BLM treatment. (E) Ashcroft test comparing fibrosis levels based on trichrome staining. Mice were pooled from two independent transplantation experiments. For the PBS group, N = 8; for BLM alone, N = 11; and for BLM plus transplantation, N = 15 mice. (F) Quantitation of total collagen in the paraffin-embedded lung tissue by hydroxyproline assay, from one transplantation experiment. For PBS, N = 14; for BLM alone, N = 11; and for chimeric mice (Trans), N = 12. (G) Percent fibrotic lung tissue out of the total lung volume by CT. Results were pooled from two independent transplantation experiments. For PBS, N = 15; for BLM alone, N = 15; and for chimeric mice (Trans), N = 17. (H) Functional parameters measured by FlexiVent (one-way ANOVA with Dunnett’s test; **P < 0.002 and ****P < 0.0001). The lung function of mice was pooled from two independent transplantation experiments. For PBS, N = 15; BLM alone, N = 19; and for chimeric mice, N = 19.

Fig. 6.. Donor-derived patches after transplantation of TdTomato⁺ lung cells into SPC-Cre TRF1^fl/fl mice at different time points after TMX treatment.

(A) Generation of the SPC-Cre TRF1^fl/fl knockout (KO) mouse. TMX-induced Cre recombination leads to the deletion of TRF1 specifically in SPC⁺ AT2 alveolar cells. E1 and E2, exons 1 and 2; I1 and I2, introns 1 and 2. Red arrows indicate Lox sequences. (B) Mice were then transplanted with lung cells from TdTomato⁺ donors at the indicated time points (colored arrows) after TMX treatment and harvested 8 weeks after transplantation. (C) Representative 3D micro-CT imaging of normal lung tissue (left), and 14 weeks after initiation of TMX administration (right). Tissue volume (fibrotic area) is marked in purple. (D) Percent fibrosis (by micro-CT) at different time points after TMX without lung cell transplantation. Results were pooled from three independent experiments. Untreated mice, N = 7; 8 weeks from TMX initiation, N = 20; and 14 weeks from TMX initiation, N = 30. (E) Typical donor-derived (red) lung patches in whole-lung tissue of mice treated with donor lung cells at different time points after TMX initiation. Scale bars, 200 μm. (F) Number of donor-derived lung patches per 2-mm² lung area in transplanted mice at different time points after TMX induction. Each dot represents the average number of patches per 2 mm² based on at least 10 measurements in individual mice. Serial 12-μm sections, along the longitudinal axis of the lobe. Sections were taken every 30 to 40 μm for analysis at different lung depths. (G) Percentage of chimeric mice, with at least four visible donor-derived patches of more than 20 donor cells each. In (F) and (G), results were pooled from two independent transplantation experiments. For 8, 10, 12, and 14 weeks from initiation of TMX, N = 7, 8, 15 and 9 mice, respectively. One-way ANOVA with Dunnett’s test; *P < 0.05, **P < 0.002, ***P < 0.0002, and ****P < 0.0001.

Fig. 7.. Typical cell composition of C57BL/6-GFP⁺ donor-derived patches in the lung of SPC-Cre TRF1^fl/fl mice transplanted 14 weeks after TMX treatment initiation and harvested 8 weeks after transplantation.

(A to D) Typical immunohistological staining of different lung cell types including (A) AQP5⁺ AT1 alveolar cells, (B) LAMP3⁺ AT2 alveolar cells, (C) CD31⁺ endothelial cells, and (D) PDGFRα⁺ mesenchymal cells. For each staining, low magnification (left column a; scale bars, 10 μm) shows a donor-derived GFP⁺ patch (green), followed by high magnification of the indicated area within each patch [right columns b to d; scale bars, 4 (B) and 3 μm (A, C, and D)] showing typical staining with different markers. Column b: Double staining GFP expression plus indicated marker (magenta). Column c: Indicated marker (magenta) plus nuclear staining (cyan). Column d: GFP (green) and nuclei (Hoechst, cyan). Column e: Distribution of average percentages of different cell types in donor-derived lung patches.

Fig. 8.. Fibrosis attenuation and functional benefit in the SPC-Cre TRF1^fl/fl mouse model transplanted 14 weeks after initiation of TRF1 knockout.

(A) Induction of TRF1 KO in AT2 cells by TMX treatment. Mice were transplanted with lung cells from C57BL/6-GFP⁺ and TdTomato⁺ donors 14 weeks after initiation of TMX treatment. (B) Fluorescence imaging of transplanted lung tissue 8 weeks after transplantation of a 1:1 mixture of TdTomato⁺ and GFP⁺ C57BL donor-derived lung cells following 14 weeks of host preconditioning. Scale bar, 200 μm. (C) Chimerism level defined by the average patch number per 2-mm² lung area. Mice exhibiting less than four patches per 2-mm² lung tissue (purple) on average were excluded from further comparisons. Results show that mouse data were pooled from two independent transplantation experiments. (D) Left: Typical fibrosis determined by trichrome staining 14 weeks after initiation of TMX induction. Right: Higher magnification of the area in the square (scale bars, 500 and 100 μm, respectively). (E and F) Two examples of fibrosis determined by trichrome staining 22 weeks after initiation of TMX induction (scale bars, 200 and 100 μm, respectively). (G) Ashcroft test comparing fibrosis levels in different groups of mice based on trichrome staining. Vehicle (oil)-treated group N = 7 mice; TMX alone, N = 13 mice; and TMX plus transplantation, N = 7 mice. (H) Quantitation of total collagen in the paraffin-embedded lung tissue, measured by the hydroxyproline assay. Data were pooled from two independent transplantation experiments; in the vehicle (oil)-treated group, N = 13 mice; in TMX alone, N = 12 mice; and for TMX plus transplantation, N = 12 mice. (I) Improved lung functional parameters measured by FlexiVent. Data in mice were pooled from two independent transplantation experiments. In the vehicle (oil)-treated group, N = 21 mice; in TMX alone, N = 14 mice; and in TMX plus transplantation, N = 18 mice. One-way ANOVA with Dunnett’s test was used for statistical analysis; **P < 0.002, ***P < 0.0002, and ****P < 0.0001.