Rationale: Abnormal alveolar macrophages (AM) are found in chronic obstructive pulmonary disease, asthma, cystic fibrosis, and adenosine deaminase deficiency (ADA−/−). There is no specific treatment strategy to compensate for these innate immune abnormalities. Recent findings suggest AMs are of early embryonic or fetal origin. Pluripotent stem cells (PSCs) as a source of embryonic-derived AMs for therapeutic use in acute and chronic airway diseases has yet to be investigated.
Objectives: To determine if embryonic Myb−/− alveolar-like macrophages have therapeutic value on pulmonary transplantation in acute and chronic airway diseases.
Methods: Directed differentiation of murine PSCs was used in factor-defined media to produce expandable embryonic macrophages conditioned to an alveolar-like phenotype with granulocyte–macrophage colony–stimulating factor. AMs were partially depleted in mice to create an acute lung injury. To model a chronic lung disease, ADA−/− mice were used. Alveolar-like macrophages were intratracheally transplanted to the injured animals and therapeutic potential was determined.
Measurements and Main Results: The differentiation protocol is highly efficient and adaptable to human PSCs. The PSC macrophages are phenotypically like AMs both functionally and by ligand marker characterization. They engulf bacteria and apoptotic cells and are better phagocytes than bone marrow–derived macrophages. In vivo, these macrophages remain in healthy airways for at least 4 weeks, can engulf neutrophils during acute lung injury, enhance pulmonary tissue repair, and promote survival in ADA−/− mice. Animals receiving the macrophages do not develop abnormal pathology or teratomas.
Conclusions: PSCs are a reliable source to produce therapeutically active alveolar-like macrophages to treat airway disease.
Other studies have used bone marrow–derived macrophages or circulating adult macrophage progenitors to rescue alveolar proteinosis, a rare inherited pulmonary disease attributed to macrophage dysfunction. These cells differ in their etiology from tissue resident alveolar macrophages. Although some groups have made monocyte-derived macrophages from pluripotent stem cells, it is not yet known if these cells can be used for repair of airway disease.
Pluripotent stem cells were used to generate primitive yolk-sac macrophages that could be conditioned with granulocyte–macrophage colony–stimulating factor to alveolar-like macrophages, thereby being more representative of the tissue resident macrophage population within the alveolus. These functional macrophages engulf microbes and cellular material in vivo and rescue adenosine deaminase–deficient mice from lethal respiratory failure by promoting pulmonary epithelial repair. The cells are expandable in vitro and do not cause any long-term pathologic damage or teratomas. This advancement exemplifies the usefulness of pluripotent stem cell–derived alveolar-like macrophages for therapeutic applications to acute and chronic lung diseases and can be used as a tool to more efficiently promote recovery of some airway diseases.
The alveolar macrophage (AM) is the most abundant population of innate immune cells in the airways. They are environmentally adapted cells that reside on the outside of the pulmonary epithelial layers and are in constant direct contact with external stimuli (1). Studies examining AM development have suggested their origins are not from circulating adult monocytes as previously believed (2), but rather from a fetal origin (3) and that their maintenance in the lungs is from a locally expanding population much like other tissue macrophages (4, 5). This latter observation has recently been explained by identifying yolk-sac erythromyeloid progenitors as the developmental origin of many tissue-resident macrophages, including AMs (6).
A variety of chronic and acute lung diseases have been attributed to altered or compromised AM function, including pulmonary alveolar proteinosis (7), chronic obstructive pulmonary disease (8), asthma (9), bronchopulmonary dysplasia (10), cystic fibrosis (11, 12), and adenosine deaminase (ADA) deficiency (13). Furthermore, patients recovering from myeloablative radiation incur lifelong pulmonary innate immune deficiencies, often resulting in bacterial lung infections likely caused by failed replenishment of fully competent AMs following bone marrow transplantation (14–16). Recent studies suggest that intratracheal delivery of bone marrow– or blood-derived macrophages or progenitors can be used to successfully replace dysfunctional AMs in pulmonary alveolar proteinosis (17, 18). Other reports have shown that bone marrow mesenchymal stromal cells modulate pulmonary innate immune function in the context of disease; however, these cells are known to have a very short half-life in the airways and their direct function in the lungs remains unclear (19).
Given that murine AMs arise in part via embryonic yolk-sac hematopoiesis by embryonic day 8.5, proliferate independently from Myb and hematopoietic stem cells (20), and take residence in the lungs during development (not after), we reasoned that instead of adult bone marrow– or blood-derived macrophages or their progenitors, therapeutically expandable alveolar-like macrophages could best be generated through yolk-sac hematopoiesis of embryonic or induced pluripotent stem cells (PSCs).
For a detailed description of methods, consult the online supplement.
PSCs were maintained as previously described (21) in feeder-free serum-free conditions. Growth factors and cytokines were added accordingly as described in the main text and online supplement.
Flow cytometry was performed with a Beckman Coulter Gallios 10/3 flow cytometer. Cells were sorted by The SickKids-UHN Flow Cytometry Core Facility on a BD FACS ARIA FACS machine. Acquisitions were analyzed with Kaluza flow cytometry software tool (Beckman Coulter, Mississauga, Canada). Fluorescently tagged antibodies are detailed in Table E4 in the online supplement.
PSC-derived alveolar-like macrophages (PSC-AMs) were intratracheally administered (0.5–1 × 106 cells in 50 µl of phosphate-buffered saline [PBS]) to anesthetized mice. In experiments using ADA−/− pups and/or their heterozygous littermates, repeated intranasal delivery of PSC-AMs began on postnatal day (PND) 3 (1–2 × 105 cells at PND3 followed by ∼106 cells on PND 4–7). Subsequently, delivery occurred on alternating days until PND17 ± 0.5 days. Animals were monitored regularly and killed in accordance with Laboratory Animal Services at the Hospital for Sick Children as directed by Canadian Committee for Animal Care ethical regulations.
Fluorescence microscopy was performed on a Leica CTRMIC 6000 confocal microscope with a Hamamatsu C910013 spinning disc camera and a Leica DMI 3000B epifluorescent microscope with a Hamamatsu ORCA-HR camera (Leica, Richmond Hill, Canada) and analyzed with Volocity software suite (Perkin Elmer, Waltham, MA). Light microscopy was performed using a Leitz Laeorlux B light microscope with a Leica DC 200 camera.
Cells examined by electron microscopy were fixed in 2.5% (wt/vol) glutaraldehyde in 0.1M phosphate buffer at pH 7.4 followed by 1% osmium tetroxide then dehydrated and embedded in Epon Araldite resin. Ultrathin sections were prepared and stained in uranyl acetate and lead citrate before viewing. All samples were examined on a JEOL JEM 1011 transmission electron microscope (JEOL USA, Peabody, MA).
Confluent fetal rat epithelial cells, isolated as previously described (22), were grown to confluence and disrupted with a 200-μl pipette tip. During cell migrations, the denuded area was repeatedly measured to compare the linear closure rates among the conditions. All rates were normalized to the control conditions.
To determine statistical significance a t test was used to compare two groups, analysis of variance for multiple means comparison of parametric data for three or more groups, Fisher exact test for comparison of ranked data, and Kaplan-Meir curve comparison with a Mantel-Cox log-rank curve comparison test for survival. A P value less than 0.05 was considered statistically significant. Error bars on graphically presented data are ± SEM unless otherwise detailed in the figure legend.
We modified a method (20) for inducing murine PSCs into early embryonic mesoderm-derived hemangioblast (see Figures E1A and E1B) hematopoiesis that results in an efficient production of macrophages in several PSC lines (Figure 1A; see Figures E2A–E2D). The myeloid macrophages were sorted for F4/80 and took up Ac-LDL (see Figures E3A and E3B). On transfer directly to the mouse airways, these PSC-macrophages were tracked in vivo with PKH26 and shown to increase the proportion of cells expressing the AM marker CD11c (Figure 1B), in comparison with the myeloid macrophages described in Figure E2A. We recapitulated this in vitro by conditioning of the sorted PSC-macrophage cells with granulocyte–macrophage colony–stimulating factor, macrophage colony–stimulating factor, stem cell factor, and IL-3 (see Figure E4), which resulted in a sustained expression of the AM markers F4/80, CD11c, and SiglecF (Figure 1C). The in vitro differentiation protocol was highly efficient, yielding approximately 6 × 105 PSC-AMs for every 106 initial PSCs (Figure 1D).

Figure 1. Pluripotent stem cell–derived myb−/− macrophages can be conditioned to be like alveolar macrophages (AMs). (A) Schematic description of macrophage differentiation of murine pluripotent stem cells (PSCs) in serum-free, feeder-free, growth factor–defined conditions. (B) The F4/80+ macrophages that displayed only a minor CD11c population were labeled with a red fluorescent PKH26 dye and transferred to the airways of mice, where 1 week later the total bronchoalveolar lavage (BAL) and PKH26-labeled cell population were costained for F4/80 and CD11c. (C) Sorted F4/80:CD11c–coexpressing cells were compared with primary AM after in vitro expansion with GM-CSF, M-CSF, SCF, and IL-3 for expression of F4/80, CD11c, and SiglecF (gray histograms, unstained controls). (D) Efficiency of PSC-AM production from pluripotent cells (Week 0) through F4/80 macrophage sorting (Week 2) to F4/80-CD11c–positive cells (Week 3), all normalized to a starting population of 106 cells (n = 4 independent differentiations). (E) Flow cytometry showing sustained coexpression of the AM markers F4/80, CD11c, and SiglecF from cells extended in serum-free culture medium supplemented with GM-CSF. (F and G) Quantitative polymerase chain reaction for macrophage transcription factor (F) PU.1 and bone marrow myeloid transcription factor (G) Myb in PSC-AM and bone marrow cells (BMC) (n = 3 replicates, ±SEM). (H) Relative cytokine secretion of resting PSC-AMs determined with an array of 40 murine cytokines, chemokines, and other soluble factors (only detectable factors of the 40 are shown). (Inset) Representative cytokine array of supernatant from PSC-AMs. Error bars in D, F, and G are presented as mean ± SEM. Animal data in B is representative of three independent biologic replicates performed with two or more mice per replicate. BMP = bone morphogenetic protein; GM-CSF = granulocyte–macrophage colony–stimulating factor; M-CSF = macrophage colony–stimulating factor; macs = macrophages; MCP = monocyte chemoattractant protein; MIP = macrophage inflammatory protein; Mp = macrophage; mPSC = murine pluripotent stem cell; N/D = not detectable; SCF = stem cell factor; TIMP-1 = tissue inhibitor of metalloproteinase 1; TNF = tumor necrosis factor; VEGF = vascular endothelial growth factor.
[More] [Minimize]The conditioning strategy was effective in producing PSC-AMs from induced PSC (see Figures E5A and E5B) and is adaptable to the human embryonic CA1 stem cell-line (see Figures E6A–E6D). Furthermore, continuous expansion with granulocyte–macrophage colony–stimulating factor (and macrophage colony–stimulating factor) for 1 year maintained the expression of the three AM ligand markers (Figure 1E). Additionally, we observed coexpression of other markers commonly found on primary AMs including CD80, CD86, CD206, and SIRPα. We verified that PSC-AMs do not express MHCII or Langerin (Table 1). Using real-time polymerase chain reaction, we confirmed that the expanding PSC-AMs are dependent on the macrophage transcription factor PU.1 (Figure 1F) but, in contrast to bone marrow cells, independent of the hematopoietic transcription factor Myb (Figure 1G).
Marker | Primary AMs | PSC-AMs |
---|---|---|
CD80 | + | + |
CD86 | +/− | + |
CD206 | +/− | + |
SIRPa | + | + |
MHCII | − | − |
Langerin | − | − |
Additionally, also using quantitative polymerase chain reaction, we randomly selected 6 of the 21 tissue-specific genes previously indicated to be highly expressed in lung macrophages (23) and showed that PSC-AMs exhibit expression of all six genes with three genes highly expressed and two uniquely in comparison with bone marrow cells and RAW 264.7 cells (see Figure E7). We further examined the cytokines secreted by PSC-AMs using a protein array for 40 murine cytokines, chemokines, and other soluble factors and determined that the cells express predominantly M1 versus M2 associated cytokines (Figure 1H; see Table E5).
We further show that the PSC-AMs ultrastructurally resemble primary AMs (Figure 2A) and determined that the cells were highly efficient at phagocytosing IgG-coated beads (Figure 2B; see Figure E8A), bacteria (Figure 2C), and apoptotic neutrophils (see Figures E8B and E8C). We isolated bone marrow cells from mice and differentiated the nonadherent progenitors to bone marrow–derived macrophages (BMDMs) as previously described (17) for the purpose of comparing their expansion and phagocytic potential with PSC-AMs. Only a small proportion (<20%) of the isolated bone marrow cells and the differentiated F4/80+ BMDMs coexpressed the AM markers CD11c and SiglecF (see Figure E9A).

Figure 2. Pluripotent stem cell–derived alveolar-like macrophages (PSC-AMs) display morphologic and functional similarities to primary AM but are superior to bone marrow–derived macrophages (BMDMs) in phagocytic and expansion capacity. (A) Transmission electron microscopy comparing substructural characteristics of the PSC-AMs with primary AM. N = nucleus; PC = phagocytosed cell; p = phagosome; scale bars, 500 nm, 2 μm, and 500 nm, from left to right, respectively. (B) PSC-AMs were compared with primary AMs for their ability to phagocytose fluorescent beads (PI = phagocytic index; by t-test, P < 0.001, n = 3, independent comparisons). (C) PSC-AMs coincubated with fluorescent Staphylococcus aureus bacteria and imaged by epifluorescence (top two rows), confocal (bottom left two panels), and transmission electron microscopy (bottom right two panels, arrows indicate internalized bacteria). Scale bars: top two rows of fluorescent images, 10 μm each; bottom row confocal images left, 12 μm and right, 5 μm; transmission electron microscopy left, 2 μm and right, 500 nm. (D) Curve fitting calculated from absolute cell numbers within a constant region of interest of BMDM and PSC-AMs cultured under identical conditions (mean ± SEM of a technical replicate representative of at least three independent experiments). (E) Phagocytic index of PSC-AMs compared with BMDMs for uptake of IgG-coated beads (n = 3 replicates, ±SEM, P < 0.05 by t test). Functional assays are representative of duplicate or triplicate independent experiments, and microscope panels are representative of multiple fields of examination. BF = bright field; DAPI = 4′,6-diamidino-2-phenylindole; ROI = region of interest.
[More] [Minimize]By fluorescence-activated cell sorter analysis (see Figure E9B), we obtained the F4/80+ population and performed a side-by-side comparison with PSC-AMs, which determined that PSC-AMs expand at an exponential rate (curve equation: y = 60.744e0.7043x) (Figure 2D, right), whereas BMDMs did not expand (curve equation: y = −1.65x + 20.95) (Figure 2D, left). We also found that PSC-AMs were nearly three times more efficient at phagocytosing IgG-coated beads than were BMDMs (BMDM, 16.98 ± 2.97% vs. PSC-AMs, 43.18 ± 6.18; n = 3 independent replicates; P = 0.019). Using the DsRed-expressing ESC line (dsRed-MST) (24), we confirmed that PSC-AMs do not express the pluripotency marker Oct3/4, suggesting that the PSC-AMs do not revert back to a stage of pluripotency even after nearly 2 years of continuous expansion (see Figure E9C), thus making them highly efficient phagocytes for pulmonary therapy.
After we established that PSC-AMs are expandable in vitro without phenotypic loss, we observed that the cells bind surfactant protein D, which confirmed that they had pulmonary innate immune potential (see Figure E10). We then tracheally transferred 0.5–1 × 106 DsRed-PSC-AMs to the lungs of healthy mice. After 1 week, bronchoalveolar lavage (BAL) was performed on killed animals and their spleens were harvested. We recovered DsRed+ cells directly from the airways but not from the spleen (Figure 3A), suggesting that the cells remained mostly in the airways. To reliably distinguish the DsRed-PSC-AMs from the resident AM population, we used mice expressing green fluorescent protein (GFP) linked to histone 2B in all nucleated cells (H2B-eGFP). The BAL cells contained a population of DsRed+ cells 2 hours, 2 days, 2 weeks, and 4 weeks after tracheal transfer of the macrophages that was detected in a distinct fluorescent channel from host-derived GFP cells as measured by flow cytometry (Figure 3B, left).

Figure 3. Expanded pluripotent stem cell–derived alveolar-like macrophages (PSC-AMs) populate healthy and injured airways in vivo and remain functional. (A) DsRed cells recovered from the airways but not spleen of mice were compared by flow cytometry 1 week after airway delivery of DsRed-PSC-AMs. (B) Total bronchoalveolar lavage (BAL) cells were stained for F4/80 and CD11c and gated for DsRed and GFP cells 2 hours, 2 days, 2 weeks, and 4 weeks following intratracheal administration of DsRed-PSC-AMs to healthy GFP mice. (C) Flow cytometry staining for F4/80 and CD11c gated for DsRed and GFP for BAL cells obtained 2 hours, 2 days, and 6 days after clodronate liposome–treated GFP mice received DsRed-PSC-AMs intratracheally. (D) Confocal imaging of cytometric centrifugation of BAL cells. Host-derived GFP-positive material is indicated by arrows inside the DsRed-PSC-AMs at 2 hours, 2 days, and 6 days postinstillation (scale bars, 10 µm). Animal adoptive transfer experiments are representative of three independent biologic replicates of n = 2–3 animals per group. Injury model data are representative of two experiments of n = 3 mice per group. Data in cytometry panels are representative of the group mean. In A, DsRed fluorescence is plotted on a log10 axis, whereas forward scatter is displayed on a linear axis. FSC = forward scatter; GFP = green fluorescent protein.
[More] [Minimize]Our gating strategy (see Figure E11) revealed that most of the DsRed-PSC-AM cells transplanted to the airways continued to retain their AM expression characteristics of F4/80 and CD11c throughout the time points surveyed, with approximately 70% coexpression at 2 hours, 85% at 2 days, and 95% at both 2 and 4 weeks (Figure 3B, center column) even though resident GFP-AMs represented most BAL cells recovered from the airways at any time point surveyed (Figure 3B, right column). Using a previously published intrapulmonary dextran-fluorescein isothiocyanate recovery assay (25), we determined that the airway recovery of DsRed macrophages by BAL to be at or close to the maximal recovery potential (∼2% of total loading volume). These data demonstrate that PSC-AMs are able to attain airway residence and retain AM ligand markers while remaining present in the airways.
Next, we determined if PSC-AMs were functional during acute lung injury. Macrophages that engulf clodronate liposomes undergo apoptosis and die. This has been used as an effective model to partially deplete airway macrophages and cause injury (26, 27). In H2B-GFP mice, intranasal clodronate liposome delivery caused an airway influx of Gr-1 expressing neutrophils (see Figures E12A and E12B) within 48 hours, similar to that observed during acute lung injury. Four days later, we intratracheally delivered DsRed-PSC-AMs to reduce the risk that residual clodronate in the airways would affect the PSC-AMs. Flow cytometry of BAL cells from clodronate-treated mice showed the DsRed-PSC-AMs remained in the injured lungs for 2 hours, 2 days, and 6 days after instillation, which constituted up to 10 days after the initial clodronate injury (Figure 3C, left column). Moreover, approximately 90% of the DsRed population of cells retained their F4/80:CD11c coexpression characteristics during the lung injury, whereas the total population of GFP+ BAL cells was comprised only partially of F4/80+:CD11c+ macrophages (Figure 3C, center and right, respectively).Furthermore, 2 hours after delivery of the DsRed-PSC-AMs to the injured lungs, confocal microscopy revealed that host-derived material (containing GFP) was detectable within the DsRed-PSC-AMs that were recovered from the airways of clodronate-treated mice. This apparent phagocytosis of host-derived GFP material was observed until our endpoint 6 days after DsRed-PSC-AM delivery (Figure 3D). Differential staining of the BAL cells at each time point was consistent with lung injury resolution (see Figure E13).
In a separate and distinct experiment, DsRed-PSC-AM instillation into the lungs of healthy mice was preceded by airway delivery of fluorescent bacteria (Staphylococcus aureus). Within 2 hours, the DsRed-PSC-AMs had performed phagocytosis in vivo and engulfed the bacterial particles (see Figures E14A–E14C). Finally, a separate pathology follow-up of mice at 4 and 6 months after receiving PSC-AMs demonstrated that the mice were clear from any abnormal tissue growth or teratoma formation (see Table E6). These data illustrate that PSC-AMs persist in healthy lungs and are functionally active during and subsequent to injury resolution in vivo with no significant pathologic consequences.
We next considered the therapeutic value of PSC-AMs as an intervention for a chronic lung disease model. ADA gene (ADA) deficiency in mice leads to a variety of pulmonary abnormalities, including compromised alveologenesis, fibrosis, inflammation, alveolar proteinosis, and airway obstruction (13, 28, 29). Animals die of respiratory failure within 18–21 days of birth and their AMs are nonfunctional (13). We used this genetic disease as model test bed to evaluate the usefulness of PSC-AMs in a degenerative lung disease. Like their human counterparts, ADA−/− mice can be saved with continuous administration of polyethylene glycol–conjugated ADA (PEG-ADA). To test the effectiveness of a single intratracheal dose of PSC-AMs, we used 4-week-old ADA−/− mice that were allowed to survive by weekly administration of PEG-ADA. Before administration of PSC-AMs, PEG-ADA was discontinued and approximately 106 PSC-AMs in PBS or PBS vehicle alone was delivered directly to the airways of ADA+/− or ADA−/− mice (n ≥ 4 per group) and the survival of animals was monitored. The ADA−/− mice receiving PSC-AM appeared healthy, vital, and remained alive for at least 1 week after the single intrapulmonary dose of PSC-AMs, whereas half of the ADA−/− mice receiving only PBS were deceased by 48 hours. Within 6 days, all PBS-treated ADA−/− animals were either deceased or required killing. This difference in survival rates between the PSC-AM– and PBS-treated groups was statistically significant (log-rank, P < 0.0001) (Figure 4A).

Figure 4. Pluripotent stem cell–derived alveolar-like macrophages (PSC-AMs) promote survival, recovery, and mucous clearance of airways in adenosine deaminase (ADA) −/− mice. (A) Survival of ADA−/− animals after receiving one direct intratracheal instillation of PSC-AM (significance of Kaplan-Meir survival curves determined by log-rank [Mantel-Cox], P < 0.0001; n ≥ 4 per group). (B) Survival of ADA−/− mice and their healthy heterozygote littermates treated by multiple intranasal PSC-AMs or PBS (vehicle) using live/dead scoring for n ≥ 6 animals per group (***P = 0.0002, Fisher exact test with a 95% confidence interval). (C) Scoring of periodic acid–Schiff (PAS)-positive mucus-like material in the airways of PSC-AM– or vehicle (PBS)-treated heterozygote and ADA−/− mice for n = 3 mice per condition (significance determined by analysis of variance, mean ± SD). NS = not significant; PBS = phosphate-buffered saline.
[More] [Minimize]Intratracheal delivery is not possible on newborn mouse pups; thus, to test the effectiveness of PSC-AMs on ADA−/− mice that had never received PEG-ADA, we repeatedly delivered PSC-AMs intranasally to the airways of 3- to 4-day-old ADA−/− pups and continued this delivery until PND15. We intentionally killed all animals at the defined endpoint of PND17 ± 0.5 before the reported mortality limitation of PND18 of ADA−/− mice (13). All heterozygous pups survived and no treatment effect was detected; however, only 43% (3 of 7) of ADA−/− pups receiving PBS to the airways survived to the defined endpoint. All ADA−/− mice (100%; 9 of 9) treated with PSC-AM survived to the defined endpoint (Figure 4B). This was a statistically significant finding (95% confidence interval, P = 0.0002).
Interestingly, in a separate pilot experiment, we performed intranasal delivery to ADA−/− mice starting at PND12 and found that these mice died. This suggested that early delivery of PSC-AM was important. On sacrifice of surviving animals, carotid arterial blood was sampled to evaluate saturated oxygen levels. Heterozygous littermates treated with PSC-AM or PBS both displayed oxygen saturation levels of 87.7 ± 6.5% and 86.6 ± 2.2%, respectively, whereas the saturated blood oxygen levels in PBS-treated ADA−/− mice was only 65.9 ± 1.3%. Treatment of ADA−/− mice with PSC-AMs restored blood oxygen saturation levels to 90.1± 2.6%, comparable with that of their heterozygous littermates. This suggested that PSC-AM treatment of the ADA−/− mice positively affected gas exchange.
Mucus-like content has been reported to accumulate in the airways of ADA−/− mice (13); thus, we compared the proportion of periodic acid–Schiff–positive airspaces among all four groups of mice. Significantly more periodic acid–Schiff–positive airspace was detected in ADA−/− mice compared with ADA+/− mice (P = 0.001); however, the ADA−/− group treated with PSC-AM had a significantly lower percentage (P = 0.002) of periodic acid–Schiff–positive airspace (31 ± 5.3%) compared with the PBS-treated ADA−/− mice (49.3 ± 9.3%) (Figure 4C).
We further assessed the PSC-AM effect on the airways of ADA−/− mice by examining the harvested lung tissue by transmission electron microscopy. The alveoli of PBS-treated ADA−/− mice were filled with a fibrillar mucus-like substance and populated with many nonfunctional macrophages. Conversely, ADA−/− mice treated with PSC-AM displayed alveolar space with reduced mucous substance (Figure 5A, b and c vs. d). These PSC-AM–treated ADA−/− mice displayed focal areas of regeneration of alveolar tissue and cells (Figure 5B) and intact alveolar basement membranes, whereas their PBS-treated counterparts displayed degeneration and damage to their alveolar basement membranes (Figure 5B, a and b vs. d and e). Furthermore, alveolar type II pneumocytes in PBS-treated ADA−/− mice seemed to be compromised, whereas mice treated with PSC-AM had structurally normal pneumocytes (Figure 5B, c compared with f).

Figure 5. Pluripotent stem cell–derived alveolar-like macrophages (PSC-AMs) promote mucous clearance and focal repair of distal airway epithelium in adenosine deaminase (ADA) −/− mice. (A) Transmission electron microscopy of airways from PSC-AM–treated ADA−/− mice. (Aa) Alveolus of heterozygous (ADA+/−) with healthy macrophages (M) and type II pneumocytes (arrow). (Ab) Mucus (asterisk) and macrophages in alveolus of ADA−/− mouse (scale bars in Aa and Ab = 2 μm). (Ac) Magnified airspace from Ab (scale bar = 500 nm). (Ad) Alveolus of a PSC-AM–treated ADA−/− mouse (scale bar = 2 μm). (B) Transmission electron microscopy of PBS- (a–c) or PSC-AM– (d–f) treated ADA−/− mice. (Ba) Alveolar space; asterisk indicates a capillary (scale bar = 500 nm). (Bb) Basement membrane (scale bar = 500 nm); arrows in Ba and Bb indicate degeneration of basement membrane. (Bc) Pneumoctyes; asterisk indicates a dying type II cell (scale bar = 500 nm). (Bd) Alveolar space; asterisk indicates a capillary (scale bar = 2 μm). (Be) Intact basement membrane (arrows); asterisk indicates a lumen of a capillary (scale bar = 500 nm). (Bf) Live type I pneumoctyes (arrows) and type II pneumocytes (asterisk) (scale bar = 2 μm). Data are representative of animals within an experimental group from two or more repeated experiments. (C) Migration (wound healing) assay of primary fetal rat pulmonary epithelial cells with base media (BM), PSCs, or PSC-AMs (scale bar = 100 μm). (D) Comparison of epithelial cell migration normalized with BM treatment of PSC and PSC-AMs (n ≥ 3 replicates, *P < 0.05 by analysis of variance). n/s = not significant; PBS = phosphate-buffered saline.
[More] [Minimize]The epithelial repair we observed in ADA−/− mice treated with PSC-AMs led us to hypothesize that the PSC-AMs could actually enhance epithelial tissue repair. To test this, we performed a series of migration scratch assays on confluent primary fetal rat lung epithelial cells harvested as previously described (22), in the presence of PSCs, PSC-AMs, or no additional cells. The PSC-AMs, but not PSCs, were able to significantly increase (P < 0.05) the migration rate of the injured epithelial cells in comparison with base media and PSC alone (Figure 5C). Interestingly, to a lesser extent the conditioned media from the PSC-AMs also increased the migration rate of the damaged epithelial monolayers (see Figure E15). Taken together, the clear differences between PSC-AM–treated and untreated ADA−/− mice in alveolar function, gas exchange, and survival suggests that PSC-AM are capable of actively promoting pulmonary tissue repair in such a multifaceted degenerative chronic lung disease characterized in ADA−/− newborn mice. Furthermore, these data suggest that PSC-AMs promote direct pulmonary epithelial tissue repair and contribute positively to the recovery of respiratory physiology and function.
Direct differentiation of macrophages from PSCs through early embryonic hemangioblast-dependent hematopoiesis is simple and effectively yields expandable PSC-AMs that can take residence and promote pulmonary tissue repair in diseased airways. To our knowledge, this study is the first to describe tissue-specified myb-independent yolk-sac macrophages derived in vitro from pluripotent embryonic stem cells that are tested for their therapeutic applications in vivo and prove to promote survival and enhance tissue repair. The generation of these PSC-AMs is rapid and efficient, producing at least 6 × 105 cells per 1 million undifferentiated starting cells. The cells can be expanded for at least 2 years for scaled applications to appropriate tissue size and volume. In addition, they can be cryopreserved without notable phenotypic or functional changes or reversion to a state of pluripotency (see Figure E16). These cells more appropriately represent the embryonic origin of many myb-independent tissue macrophages that do not arise from hematopoietic stem cells but seed organs, including the lung, during embryogenesis (20), because they are made from PSCs. This advancement represents an alternative to a bone marrow cell source and produces therapeutically active hematopoietic stem cell–independent macrophages that are more efficient phagocytes without the need for laborious and invasive marrow harvest.
Pulmonary transplantation of only 5 × 105 PSC-AMs to healthy nonstrain matched mice revealed retention of the cells in the airways for several weeks with no obvious transplantation-associated injury or immune suppression requirement. In addition, 4 -and 6-month follow-up did not reveal any significant pathology. The PSC-AMs can persist in the airways following acute lung injury resulting from dead AMs (clodronate treatment) and display immediate and sustained functional phagocytosis of host-derived cellular debris and bacteria in vivo. This is in stark contrast to published experimental observations using mesenchymal stromal cells, in which cells do not persist in the airways nor do they seem to be responsible for performing phagocytosis (19, 30). Most notable was the remarkable contribution that PSC-AMs made to the recovery of the pulmonary pathology typically exhibited in ADA−/− mice.
We strategically chose ADA-deficiency as a model of severe chronic lung disease because of its extreme multifaceted presentation of inflammation, fibrosis, and impaired alveologenesis, and epithelial decay, knowing that the resident ADA−/− AM were nonfunctional (13, 28, 29, 31). Hence, we could conclude that any functional macrophage activity in the PSC-AM–treated ADA−/− would be initiated by the PSC-AMs and whose contribution (positive or negative) could be evaluated. Our data suggest that the PSC-AMs taking up residence in the airways of ADA−/− mice significantly contribute to the corrective result in pulmonary pathology, but PSC-AM contribution to the recovery of local previously nonfunctional host-derived AMs or other airway cells to a more functional status cannot be ruled out. Furthermore, we provide evidence that even a single dose (direct delivery) of PSC-AM to the airways of ADA−/− mice is sufficient to promote the survival of ADA-deficient animals for nearly 1 week. This observation underscores the potent functional and viable effect PSC-AMs have in vivo. We acknowledge that despite the effective PSC-AM treatment, animals did require killing after 1 week for humane purposes because ADA-deficiency contributes to morbidities unrelated to the lungs where airway-delivered PSC-AMs likely have no effect. Furthermore, our data in the newborn ADA−/− mice indicate that PSC-AMs provide a real nonpharmacologic therapeutic advantage by assisting in a focal regeneration of alveolar tissue that without the corrective assistance from PSC-AM would otherwise contribute to respiratory failure and death. More so, we show that the PSC-AMs can directly enhance pulmonary epithelial tissue repair in vitro. However, we recognize that a limitation of our study is that we have only tested one delivery dose range (0.5–1 × 106 cells) and did not further optimize minimum and maximum effective doses.
We consider that the usefulness of these expandable PSC-AMs is by no means limited to the rare genetic diseases like ADA-deficiency; rather, because the ADA−/− phenotype shares hallmarks of many other lung diseases, the PSC-AMs may offer sustained nonpharmacologic therapeutic alternatives or complement current treatment strategies for diseases, such as chronic obstructive pulmonary disease, bronchopulmonary dysplasia, ventilator-induced lung injury, asthma, or pneumonia. We note that PSC-AMs do not express major histocompatibility complex class II and this finding could be exploited for allogenic transplantation without the risk of graft-versus-host rejection. Moreover, because PSC-AMs expand so efficiently in vitro, they are an excellent candidate for targeted gene correction or airway delivery of therapeutic small molecules without generating a compromising immune response, a notable risk described in primary AM transfection studies (32). Finally, because the differentiation protocol is valid for human cells, future steps may include testing the efficacy of human cells in animal models of various acute and chronic lung diseases.
1. | Guth AM, Janssen WJ, Bosio CM, Crouch EC, Henson PM, Dow SW. Lung environment determines unique phenotype of alveolar macrophages. Am J Physiol Lung Cell Mol Physiol 2009;296:L936–L946. |
2. | Landsman L, Jung S. Lung macrophages serve as obligatory intermediate between blood monocytes and alveolar macrophages. J Immunol 2007;179:3488–3494. |
3. | Guilliams M, De Kleer I, Henri S, Post S, Vanhoutte L, De Prijck S, Deswarte K, Malissen B, Hammad H, Lambrecht BN. Alveolar macrophages develop from fetal monocytes that differentiate into long-lived cells in the first week of life via GM-CSF. J Exp Med 2013;210:1977–1992. |
4. | Chamoto K, Gibney BC, Ackermann M, Lee GS, Lin M, Konerding MA, Tsuda A, Mentzer SJ. Alveolar macrophage dynamics in murine lung regeneration. J Cell Physiol 2012;227:3208–3215. |
5. | Hashimoto D, Chow A, Noizat C, Teo P, Beasley MB, Leboeuf M, Becker CD, See P, Price J, Lucas D, et al. Tissue-resident macrophages self-maintain locally throughout adult life with minimal contribution from circulating monocytes. Immunity 2013;38:792–804. |
6. | Gomez Perdiguero E, Klapproth K, Schulz C, Busch K, Azzoni E, Crozet L, Garner H, Trouillet C, de Bruijn MF, Geissmann F, et al. Tissue-resident macrophages originate from yolk-sac-derived erythro-myeloid progenitors. Nature 2015;518:547–551. |
7. | Greenhill SR, Kotton DN. Pulmonary alveolar proteinosis: a bench-to-bedside story of granulocyte-macrophage colony-stimulating factor dysfunction. Chest 2009;136:571–577. |
8. | Hodge S, Hodge G, Scicchitano R, Reynolds PN, Holmes M. Alveolar macrophages from subjects with chronic obstructive pulmonary disease are deficient in their ability to phagocytose apoptotic airway epithelial cells. Immunol Cell Biol 2003;81:289–296. |
9. | Moreira AP, Hogaboam CM. Macrophages in allergic asthma: fine-tuning their pro- and anti-inflammatory actions for disease resolution. J Interferon Cytokine Res 2011;31:485–491. |
10. | Blackwell TS, Hipps AN, Yamamoto Y, Han W, Barham WJ, Ostrowski MC, Yull FE, Prince LS. NF-{kappa}B signaling in fetal lung macrophages disrupts airway morphogenesis. J Immunol 2011;187:2740–2747. |
11. | Bruscia EM, Zhang P-X, Ferreira E, Caputo C, Emerson JW, Tuck D, Krause DS, Egan ME. Macrophages directly contribute to the exaggerated inflammatory response in cystic fibrosis transmembrane conductance regulator-/- mice. Am J Respir Cell Mol Biol 2009;40:295–304. |
12. | Di A, Brown ME, Deriy LV, Li C, Szeto FL, Chen Y, Huang P, Tong J, Naren AP, Bindokas V, et al. CFTR regulates phagosome acidification in macrophages and alters bactericidal activity. Nat Cell Biol 2006;8:933–944. |
13. | Dhanju R, Min W, Ackerley C, Cimpean L, Palaniyar N, Roifman CM, Grunebaum E. Pulmonary alveolar proteinosis in adenosine deaminase-deficient mice. J Allergy Clin Immunol 2014;133:1467–1471, e1–e4. |
14. | Domingo-Gonzalez R, Wilke CA, Huang SK, Laouar Y, Brown JP, Freeman CM, Curtis JL, Yanik GA, Moore BB. Transforming growth factor-β induces microRNA-29b to promote murine alveolar macrophage dysfunction after bone marrow transplantation. Am J Physiol Lung Cell Mol Physiol 2015;308:L86–L95. |
15. | Domingo-Gonzalez R, Moore BB. Defective pulmonary innate immune responses post-stem cell transplantation; review and results from one model system. Front Immunol 2013;4:126. |
16. | Ojielo CI, Cooke K, Mancuso P, Standiford TJ, Olkiewicz KM, Clouthier S, Corrion L, Ballinger MN, Toews GB, Paine R III, et al. Defective phagocytosis and clearance of Pseudomonas aeruginosa in the lung following bone marrow transplantation. J Immunol 2003;171:4416–4424. |
17. | Suzuki T, Arumugam P, Sakagami T, Lachmann N, Chalk C, Sallese A, Abe S, Trapnell C, Carey B, Moritz T, et al. Pulmonary macrophage transplantation therapy. Nature 2014;514:450–454. |
18. | Happle C, Lachmann N, Škuljec J, Wetzke M, Ackermann M, Brennig S, Mucci A, Jirmo AC, Groos S, Mirenska A, et al. Pulmonary transplantation of macrophage progenitors as effective and long-lasting therapy for hereditary pulmonary alveolar proteinosis. Sci Transl Med 2014;6:250ra113. |
19. | Curley GF, Scott JA, Laffey JG. Therapeutic potential and mechanisms of action of mesenchymal stromal cells for acute respiratory distress syndrome. Curr Stem Cell Res Ther 2014;9:319–329. |
20. | Schulz C, Gomez Perdiguero E, Chorro L, Szabo-Rogers H, Cagnard N, Kierdorf K, Prinz M, Wu B, Jacobsen SE, Pollard JW, et al. A lineage of myeloid cells independent of Myb and hematopoietic stem cells. Science 2012;336:86–90. |
21. | Shojaie S, Ermini L, Ackerley C, Wang J, Chin S, Yeganeh B, Bilodeau M, Sambi M, Rogers I, Rossant J, et al. Acellular lung scaffolds direct differentiation of endoderm to functional airway epithelial cells: requirement of matrix-bound HS proteoglycans. Stem Cell Rep 2015;4:419–430. |
22. | Kroon AA, Delriccio V, Tseu I, Kavanagh BP, Post M. Mechanical ventilation-induced apoptosis in newborn rat lung is mediated via FasL/Fas pathway. Am J Physiol Lung Cell Mol Physiol 2013;305:L795–L804. |
23. | Gautier EL, Shay T, Miller J, Greter M, Jakubzick C, Ivanov S, Helft J, Chow A, Elpek KG, Gordonov S, et al.; Immunological Genome Consortium. Gene-expression profiles and transcriptional regulatory pathways that underlie the identity and diversity of mouse tissue macrophages. Nat Immunol 2012;13:1118–1128. |
24. | Vintersten K, Monetti C, Gertsenstein M, Zhang P, Laszlo L, Biechele S, Nagy A. Mouse in red: red fluorescent protein expression in mouse ES cells, embryos, and adult animals. Genesis 2004;40:241–246. |
25. | Ridsdale R, Roth-Kleiner M, D’Ovidio F, Unger S, Yi M, Keshavjee S, Tanswell AK, Post M. Surfactant palmitoylmyristoylphosphatidylcholine is a marker for alveolar size during disease. Am J Respir Crit Care Med 2005;172:225–232. |
26. | Nakamura T, Abu-Dahab R, Menger MD, Schäfer U, Vollmar B, Wada H, Lehr CM, Schäfers HJ. Depletion of alveolar macrophages by clodronate-liposomes aggravates ischemia-reperfusion injury of the lung. J Heart Lung Transplant 2005;24:38–45. |
27. | Traeger T, Kessler W, Hilpert A, Mikulcak M, Entleutner M, Koerner P, Westerholt A, Cziupka K, van Rooijen N, Heidecke CD, et al. Selective depletion of alveolar macrophages in polymicrobial sepsis increases lung injury, bacterial load and mortality but does not affect cytokine release. Respiration 2009;77:203–213. |
28. | Chunn JL, Molina JG, Mi T, Xia Y, Kellems RE, Blackburn MR. Adenosine-dependent pulmonary fibrosis in adenosine deaminase-deficient mice. J Immunol 2005;175:1937–1946. |
29. | Blackburn MR, Volmer JB, Thrasher JL, Zhong H, Crosby JR, Lee JJ, Kellems RE. Metabolic consequences of adenosine deaminase deficiency in mice are associated with defects in alveogenesis, pulmonary inflammation, and airway obstruction. J Exp Med 2000;192:159–170. |
30. | Devaney J, Horie S, Masterson C, Elliman S, Barry F, O’Brien T, Curley GF, O’Toole D, Laffey JG. Human mesenchymal stromal cells decrease the severity of acute lung injury induced by E. coli in the rat. Thorax 2015;70:625–635. |
31. | Grunebaum E, Cutz E, Roifman CM. Pulmonary alveolar proteinosis in patients with adenosine deaminase deficiency. J Allergy Clin Immunol 2012;129:1588–1593. |
32. | Wilson AA, Murphy GJ, Hamakawa H, Kwok LW, Srinivasan S, Hovav AH, Mulligan RC, Amar S, Suki B, Kotton DN. Amelioration of emphysema in mice through lentiviral transduction of long-lived pulmonary alveolar macrophages. J Clin Invest 2010;120:379–389. |
Supported by operating grants (MOP-77751 and RMF-92088) from Canadian Institutes of Health Research. E.G. is supported by the Donald and Audrey Campbell Chair for Immunology. M.P. is the holder of a tier 1 Canadian Research Chair in Fetal, Neonatal, and Maternal Health.
Author Contributions: M.L.L. and M.P. conceived the project, designed the experiments, and wrote the manuscript. M.L.L., J.L., J.W., and T.J.W. conducted experiments and performed analysis. C.A. performed electron microscopy and pathology analysis. E.G. provided expertise in adenosine deaminase deficiency and advised on adenosine deaminase deficiency animal management and experimentation.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1164/rccm.201509-1838OC on January 5, 2016
Author disclosures are available with the text of this article at www.atsjournals.org.