Main

The current promise that stem cells hold for tissue regeneration is immense. The belief that once irreversibly damaged tissue can be restored to a normal functional capacity is providing great optimism for novel treatments in many diseases. However, this research is not without its controversies. There are ethical and moral issues surrounding the use of stem cells derived from embryos. Recent research is starting to unravel the potential for adult derived stem cells that was initially described more than 20 y ago. While the initial results from numerous disease models outside the lung show great potential for a cell replacement therapy, there is heated debate over numerous concerns within the stem cell field. The mechanism of action, the plasticity, the source, the phenotype, and the “stemness” of the stem cells used are all contentious issues. Furthermore, stem cell research has been markedly slower in lung diseases due to the complexity of the lung. The lung possesses numerous anatomical areas (upper airways, bronchioles, alveoli and their underlying capillaries) each of which contains a unique cellular population with drastically different functions.

There are two common, serious pediatric lung diseases without current treatments that stem cells hold potential promise for: bronchopulmonary dysplasia (BPD) and cystic fibrosis (CF). BPD is the chronic lung disease of prematurity that follows ventilator support. BPD is characterized by an arrest in alveolar and vascular growth (1, 2). The long-term consequences of this halted lung growth are still unknown. Therefore, the potential of a stem cell to restore or protect the alveolar epithelium is very attractive as this may reduce the morbidity and mortality associated with BPD. CF results from mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene and affects 1 in 3200 live births (3). There are five classes of gene defects in CF: 1) defective protein production; 2) aberrant intracellular protein trafficking; 3) defective regulation of chloride transport; 4) reduced chloride transport function; 5) reduced protein expression. The most common cause of CF occurs within class 2 (4). The aberrant intracellular protein trafficking is caused by the deletion of an amino acid, commonly referred to as ΔF508 (3, 4). Pulmonary complications are typically the most serious and life-threatening manifestations of CF. Major advances in the multidisciplinary management of CF have contributed to prolong the life span of affected patients to the mid 30s, but quality of life and other CF-related complications justify the search for a cure. Stem cells have the potential to act as a vehicle for corrective gene therapy, thereby restoring normal organ function. Ultimately, lung stem cell therapy may also be applicable to other devastating lung disorders such as pulmonary hypoplasia.

This review will critically analyze the controversial literature investigating the use of exogenously administered stem cells for lung diseases.

DEFINITION OF STEM CELLS

Stem cells may be artificially classified into three categories: embryonic stem (ES) cells, adult bone marrow derived stem cells (BMSC), and tissue progenitor cells (Table 1). ES cells are pluripotent cells derived from the inner cell mass of blastocyst-stage embryos (5). They are able to form tissue within each of the germ layers (the endoderm, mesoderm, and ectoderm) and have thus enormous therapeutic potential to regenerate any damaged tissue; but ES cells are also mired in great controversy due to the present need of destroying an embryo to harvest pluripotent cells (6). Conversely, progenitor cells are thought to reside within a tissue and are stimulated for repair after injury. It is these tissue progenitor cells that give the liver (the hepatic oval cell) its robust regenerative ability (7, 8). In the lung, type II alveolar epithelial cells (AEC) are the putative distal progenitor cells responsible for repair after injury (9, 10). Side population (SP) cells, identified by their ability to efflux Hoechst dye, and tracheal epithelial basal cells, are putative progenitor cells in the upper airways (11). While the pluripotency of ES and further characterization of resident lung stem cells (12, 13) hold promise for novel therapeutic approaches for lung diseases, this review will focus on the adult bone-marrow derived stem cell and their exogenous administration in various models of lung diseases.

Table 1 Terminology
Full size table

The bone marrow contains several population of primitive cells: 1) Hematopoeitic stem cells (HSCs); 2) mesenchymal stem cells (MSCs); 3) endothelial stem/progenitor cells (EPC); 4) SP cells; and 5) multipotent adult progenitor cells (MAPC). Among these, the HSCs are the best characterized of the adult stem cells (14). HSCs are multipotent and have the ability to maintain or restore the mature circulating blood cells (erythrocytes and leukocytes). Recent evidence suggest that bone-marrow derived cells have the ability to cross lineage barriers and generate differentiated tissue beyond their own tissue boundaries to form functional components of other tissues, expressing tissue-specific proteins in organs such as heart, liver, brain, skeletal muscle, and vascular endothelium (15). A true potential for adult derived stem cells lies with the multipotent MSCs and these cells have been extensively used in lung stem cell research. This potential was first proposed in the 1970s by Friedenstein et al. (16). This initial report showed that plastic adherent cells, previously believed to be committed to a specific tissue type (i.e. bone marrow), had the ability to form cells within the osteogenic, chondrogenic, and adipogenic lineages (16). There have been numerous reports characterizing the phenotype of these MSCs in both humans (17, 18) and mice (16, 19). However, it was not until over 20 y after Friedenstein's first report that the true potential of these cells were made apparent. In 2001, Krause et al. reported that a single bone-marrow derived stem cell had the ability to engraft into numerous organs as well as having a “tremendous differentiative capacity” by adopting the phenotype of epithelial cells within the liver, lung, GI tract, and skin (20).

This initial report from Krause has led to extensive study of these cells in the heart (21–24), liver (25–27), brain (28), kidney (29), and endothelium (30). And while these initial results show great promise in providing new therapies for numerous diseases, these studies have also generated great controversy and highlight the discrepancies that exist within adult stem cell research. These controversies include the source and phenotype of the stem cells used, the experimental model of injury, the mechanism of effect, and the methods used to determine efficacy. While stem cell research in the lung has been less prolific than in cardio-vascular or neurologic diseases, the same controversies apply and will be discussed below.

STEM CELL THERAPIES TO PREVENT LUNG INJURY

The 2001 report from Krause et al. opened the door to the potential that there may be an ethically appealing option in using stem cells for tissue replacement. Based on their previous work showing the potential of CD34+ lin− for regenerating hepatocytes (27), Krause et al. extended their observation to the lung. Myelo-ablation was induced in female mice by total body irradiation of 1050–1100 cGy (an otherwise lethal dose if the bone marrow is not reconstituted; in most cases of irradiation the lungs are also injured). CD34+ lin− cells from male donor mice were administered intravenous (IV) (20). Fluorescent in situ hybridization (FISH) to track the Y-chromosome showed 20% of AEC being derived from a donor animal. Donor derived BMSC persisted up to 11 mo post-transplant (20).

Consistent with this first observation, Kotton et al. reported that plastic adherent MSCs (1–2 × 106 injected IV) from transgenic mice overexpressing lacZ, allowing tracking of the cells via X-gal staining, were able to engraft and form type I AEC in experimental lung fibrosis induced by intratracheal (IT) bleomycin (31). Bleomycin is one of the most extensively studied and reproducible experimental models for lung fibrosis. When bleomycin is given into the airway, it produces lung epithelial injury, followed by an inflammatory response over several days, followed by lung fibrosis that eventually resolves (32). Engrafted cells had morphologic features of type I AEC (i.e. flattened with ovoid nuclei bulging into alveolar lumen and adjacent to type II AEC) and expressed type I AEC specific markers such as Lycopersicon esculentum lectin and T1α (31). The proportion of donor derived MSCs contributing to the type I AEC population was not assessed. Interestingly however, all bleomycin injured-mice (4/4) showed engraftment when compared with PBS injected controls (2/9), suggesting that prior injury amplifies stem cell engraftment, as has been observed in other tissues (7). Furthermore, there were no donor-derived type II AECs, even in lungs harvested 1 and 2.5 d postinjection. In vitro, 10% of plastic adherent donor bone marrow cells cultured for one week expressed T1α and aquaporin 5, used as markers for type I AEC, (but are not solely confined to these cells) (33–35), but not surfactant protein C (SP-C), a type II AEC specific marker (36) suggesting that cultured bone marrow cells can serve as type I AEC precursors (31).

These two reports form the prototypes for most of the experiments on BMSC in the lung that have followed. In 2002, Theise et al. reported that donor-markers for either whole bone marrow or CD34+ lin− cells were detected in 14% of type II AEC after 1200 cGy total body irradiation (37). Engraftment was still present 6 months later (for total bone marrow injection) and 8 mo later for CD34+ lin− cells, detectable by FISH. In another study, administration of MAPC to nonobese diabetic/severe combined immunodeficiency (NOD/SCID) mice following mild irradiation of 250 cGy accounted for 4% of the alveolar epithelium assessed by β-galactosidase immunofluorescence (38). In another elegant study, Abe et al. generated parabiotic mice by surgically joining green fluorescent protein (GFP) transgenic mice and wild-type littermates (39). These mice develop a common circulation (approximately 50% green cells in blood) by 2 wk after surgery. The wild-type mouse was either uninjured or irradiated or received IT elastase (a well accepted emphysema model) or the combination of radiation with IT elastase injection. Radiation or the combination of radiation with elastase significantly increased the proportion of bright green cells in the lungs of wild-type mice. These cells resembled morphologically, interstitial monocytes/macrophages, subepithelial fibroblast-like interstitial cells, and type I AEC. Approximately 5 to 20% of lung fibroblasts primary cultured from injured wild-type mice expressed GFP.

Ortiz et al. used plastic adherent and purified, magnetic bead immunodepleted MSCs from male bleomycin-resistant BALB/c mice expressing CD34, CD45, and CD11b in bleomycin (4 U/kg)-induced pulmonary fibrosis in female bleomycin-sensitive C57BL/6 recipients (40). The purified MSCs (5 × 105) were injected into the jugular vein. Lung engraftment quantified by real-time PCR showed that male DNA accounted for 2.21 × 10−5% of the total lung DNA in control-treated mice but prior injury with bleomycin increased engraftment by 23-fold. FISH revealed that engrafted male cells were localized to areas of bleomycin-induced injury and exhibited an epithelium-like morphology, suggesting that stem cells homed specifically to sites of injury. Furthermore, immediate administration of MSCs was able to protect the lung from the bleomycin-induced inflammation, as assessed by collagen deposition, and matrix metalloproteinase (MMP) activation (40) as compared with animals that received MSCs one week following bleomycin, indicating the importance of timing of stem cell therapy, as observed in myocardial infarct (41). The low numbers of donor-derived cells engrafting the lung did not appear sufficient to account for the therapeutic response, suggesting that donor stem cells may have other local effects.

Rojas et al. used a model in which before bleomycin injury, the animals also underwent myelosuppression via a single administration of busulfan (42). Myelosuppression increased the susceptibility to bleomycin injury, suggesting that an intact bone marrow serves to limit the extent of lung injury. After this suppression of the bone marrow, plastic adherent and immuno-depleted MSCs were given IV MSCs administration improved survival when compared with untreated animals. Furthermore, within the MSC administered group, there were a “substantial number” of donor cells within the lungs 2 wk following bleomycin insult with characteristics of AEC I and II, fibroblasts, and endothelial cells. In the bleomycin+busulfan group, 29% of cells were derived from donor MSCs. Interestingly, circulating levels of granulocyte-macrophage colony-stimulating factor (GM-CSF) and granulocyte colony stimulating factor (G-CSF) (factors known to stimulate stem cell mobilization from the bone marrow), were higher in bleomycin-injured animals receiving MSC. This suggests that, besides lung engraftment, humoral factors may also contribute to stem cell-induced tissue protection. This is consistent with the greater protection conferred by stem cell transplant in bone marrow sufficient (no busulfan) animals.

Preliminary data in oxygen-induced experimental BPD in newborn rats also suggest a protective role of MSC. Rats exposed to 95% oxygen from birth to 14 d postnatal during the alveolar period of lung development have an irreversible arrest in lung development (43). In this preliminary study, MSC were given IT, a situation that is clinically relevant since premature infants at risk are routinely given surfactant via this route and which has been used successfully with other experimental therapies (44). CFSE (a green fluorescent marker)-labeled MSC given at 4 d postnatal in hyperoxic-exposed animals prevented stunted alveolar growth, engrafted and expressed the type II AEC specific marker SP-C (unpublished data).

These studies demonstrate the potential of BMSC to home to the injured lung, engraft, and adopt the phenotype of one or many of the lung cells. These reports also illustrate the numerous methods that exist in culturing and selecting BMSC (Table 2). The differences in experimental design and the numerous techniques to determine treatment efficacy (engraftment, differentiation, improvement in lung histology) further render the interpretation of these results difficult.

Table 2 Characteristics of adult bone marrow-derived cells used in lung stem cell biology
Full size table

Furthermore, there are numerous studies that do not support the plasticity of adult stem cells in different organs. Wagers et al. attempted to determine the fate of “prospectively isolated, long-term reconstituting HSCs” by using either a chimeric animal, produced via bone marrow ablation and reconstitution with GFP HSCs, or using a parabiotic model, joining the circulation of a transgenic GFP mouse with the circulation of a wild type mouse (45). While the BM ablated group showed robust reconstitution of circulating HSCs, and the parabiotic animals had robust hematopoietic chimerism, there was little evidence for stem cell engraftment and transdifferentiation. Only one cell of 13.2 × 106 cells examined within the brain, and seven cells of 4.7 × 105 cells examined within the liver co-expressed the donor marker GFP as well as a tissue specific marker (albumin for hepatocytes) in the BM ablation group (45). There was no evidence for engraftment and transdifferentiation in the lung. Moreover, the parabiotic model had no indication of HSC engraftment and transdifferentation in any of the tissues analyzed (45).

In addition, two recent reports have challenged the engraftment potential of adult-derived stem cells in the lung. These two studies take advantage of transgenic mouse that express a marker protein ubiquitously or under the control of the type II AEC specific promoter SP-C (46, 47). Chang et al. transplanted bone marrow (2.0 × 107 cells) from transgenic mice expressing the LacZ or eGFP gene ubiquitously (under the control of the actin promoter), or under the control of the human SP-C promoter into irradiated, neonatal mice. The lungs of recipients transplanted with bone marrow from transgenic mice that ubiquitously express eGFP or LacZ showed cells whose morphology and location were compatible with type II AEC (46). These green cells also co-localized with pro-SP-C. These initial results indicated that the eGFP BMSC were contributing to the alveolar epithelium. Flow cytometric analysis of SP-C immunostained cells with eGFP indicated that eGFP-fluorescent cells accounted for 50–55% of all lung cells, as well as 70% of type II AECs 2 wk after bone marrow transplantation (46). One month post-transplantation, eGFP+ cells only comprised 2-25% of total lung cells with 0% to 8% of type II AECs being donor-derived. However, upon analysis with cyto-centrifugation, the phenotype of the eGFP+ SP-C+ cells displayed a small, round cell type with a scant cytoplasm which is not congruent with the normal type II AEC phenotype of larger, cuboidal cells with an abundant cytoplasm (48). Moreover, mice that received SP-C-eGFP bone marrow transplant showed no eGFP expression, indicating that the transplanted cells are not adopting the type II AEC phenotype (46). This result led to retrospectively analyze the lungs from mice receiving actin-eGFP bone marrow transplantation using deconvolution microscopy. Deconvolution microscopy was able to generate a three dimensional image of the lung allowing to determine that the co-expression of SP-C and eGFP was a false positive: the pro-SP-C signal resided just outside the eGFP+ cells, with a difference between the two signals less than 300 nm, and this phenomenon was not apparent until the three dimensional image was viewed from multiple angles (46).

Kotton et al. also used SP-C-eGFP mice, but investigated both unfractionated BM as well as BM-derived SP cells (47). SP cells were purified from the femurs, tibias, and iliac crests and identified using a model of Hoechst dye efflux. These cells are enriched for HSC activity. Adult mice underwent myeloablation via either a single dose of 11 or 12 Gy radiation or two doses of 7 Gy (total of 14 Gy). The bone marrow was then reconstituted with either a whole bone marrow transplant or transplantation of the highly enriched HSCs SP cells from SP-C-eGFP donor mice. Three months post-transplantation, with robust blood reconstitution, the mice were challenged with a bleomycin injury (0.05 units) and engraftment was assessed 1 mo post-transplant. Using three antibody independent assays, fluorescent activated cell sorting (FACS), fluorescent microscopy, and real time PCR, the authors showed no evidence of donor cells becoming type II AECs (47).

In humans, BMSC repopulation also seems to contribute minimally to the type II AECs after cross-gender lung transplantation (49). Sequential immunohistochemistry and FISH performed on lung biopsy specimens from male recipients of transplanted lungs from female donors showed Y-chromosome–containing type II AEC in 9 of 25 biopsy specimens from five of seven gender mismatched male lung transplant recipients, that accounted for 0% to 0.553% of type II AEC (49). The number of type II AEC of male karyotype showed a statistically significant relationship to the cumulative number of episodes of acute cellular rejection. This study also suggests that BMSC contribute minimally to the type II AEC proliferation that is often present in these patients as a sequela to alveolar injury (49).

STEM CELL THERAPIES TO REPLACE DEFICIENT GENES IN THE LUNG

The use of stem cells for CF is still at the conceptual stage, therefore there are currently very few reports on the use of adult derived stem cells for CF. However, the idea that adult derived stem cells could be used as a vehicle for gene therapy is very appealing. Firstly, this method of administration would give the physician a controlled method of administering the corrected gene, circumventing some of the problems that have been seen using viral mediated gene therapy (50). Secondly, a stem cell could potentially provide a synergistic mechanism. Not only would the stem cell deliver the corrected CFTR gene to the lung of a CF patient, but these cells could also help replenish destroyed tissue.

Grove et al. first showed the potential for a marrow derived cell as a vehicle for gene delivery to the pulmonary epithelium (51). MSCs were expanded in vitro after harvest from the bone marrow by being grown in a defined basal bone marrow media supplemented with human IL-6 (IL-6), murine IL-3, murine stem cell factor, and polybrene. These cells were then transduced with a retroviral vector (MND-eGFP-SN) that has been optimized for long-term expression of eGFP in other cell lines. Following transduction, the eGFP marrow stem cells were then injected via the tail vein into mice that had received 600 cGY of radiation for 2 consecutive days. This resulted in 1% to 7% of AEC derived from donor eGFP stem cells (51).

More recently, the possibility of transforming in vitro MSCs from CF patients with a viral transcript to correct the defective CFTR gene was reported (52). MSCs were first obtained from healthy volunteers as well as CF volunteers with the ΔF508 homozygous mutation. MSCs expressing GFP, co-cultured with human airway epithelium, adopted a cuboidal or columnar shape similar to epithelial cells, as well as the expression of cytokeratin-18 (CK-18) a marker of epithelial cells. When directly mixed with primary human AECs, control MSC expressed the tight junction protein occludin. Moreover, FACS was used to isolate the GFP+ cells from the co-culture system. These cells, when co-cultured with normal human AEC or with AEC from ΔF508 homozygous patients, expressed wild type CFTR, which were either not detected in MSCs before or that were not co-cultured. These experiments strongly suggest that MSCs, when exposed to AECs, are able to adopt the phenotype of AECs and show the importance of the microenvironment in inducing tissue-specific transdifferentiation. While these results are very promising and lend credence to the ability of MSC transdifferentiation into AEC, the true potential for MSC in CF are as vehicles for gene delivery. MSCs from 3 ΔF508 homozygous patients were transduced with a viral vector for expression of CFTR (vesicular stomatitis virus glycoprotein pseudotyped Moloney murine leukemia virus-CFTR-neo). In this construct, the human CFTR gene was linked by internal ribosome entry site to the neo gene to allow for selection of the corrected MSCs based on G418-resistance (a sulphate antibiotic) and drug-resistant stem cell expansion. Insertion of the gene had no effect on the multitpotency (based on adipogenic, chondrogenic, and osteogenic cell differentiation), doubling times, and numbers of colony-forming units (CFUs) compared with control MSCs. ΔF508 corrected MSCs showed improved chloride efflux in vitro. Furthermore, after co-culture with AECs, the corrected ΔF508 MSCs were able to secrete a significantly larger amount of chloride when compared with uncorrected MSCs from ΔF508 CF patients. Corrected ΔF508 MSCs responded to IBMX and forskolin (adenylate cyclase activators) to a larger degree than the uncorrected ΔF508 MSCs (52). While this in vitro proof of concept study is very exciting, definitive answers cannot be concluded until this is tested in vivo in an animal model of CF.

There is one report to date attempting to use adult BMSC for CF in vivo. In this report, Loi et al. used both plastic adherent MSCs and total bone marrow harvested from transgenic mice over-expressing GFP (53). The authors subsequently ablated the bone marrow of CFTR knockout mice using 800 rads total body irradiation and injected either the MSCs (106 cells/mouse) or total bone marrow (20 × 106 cells/mouse) via the tail vein. They also used an intraperitoneal dose of naphthalene (275 mg/kg) either three days before MSC administration or one month after bone marrow reconstitution with total bone marrow cells to induce lung injury and stimulate migration of the transplanted bone marrow cells to the lung. Stromal marrow cells were able to engraft into the alveolar epithelium and to confer CFTR mRNA to the knock-out mice, but the engraftment occurred at such a low rate (0.025% chimeric airway epithelial cells, of which 0.01% expressed CFTR protein) that it is unlikely that this method would correct the trans-epithelial chloride deficit. Naphthalene-induced airway remodeling did not significantly increase the number of chimeric airway epithelial cells expressing CFTR.

These two reports highlight the potential for BM-derived stem cell based gene therapy in CF and the need for new strategies for enhancing airway epithelial engraftment to exert a beneficial effect.

CONCLUSION

The use of stem cells in the lung is at the crossroads of understanding. Over the last decade, numerous studies have demonstrated the ability of adult BM-derived cells to differentiate into non-hematopoietic lineages, including neural, skeletal muscle, hepatic, cardiac and vascular lineages and to repair these tissues. With regard to lung repair, the literature is sparser. Initial promising reports showing adult BM-derived cell homing, engraftment, and lung phenotype adoption are now challenged. This highlights the need for more reliable techniques and assays to assess adult BM-derived cells lung engraftment and transdifferentiation. Other challenges before moving to the clinical application with this promising therapeutic strategy include: improved characterization of putative BM-derived stem cells, determining mechanisms of homing, engraftment and transdifferentiation, as well as the alternate mechanisms of action beyond pure tissue replacement, rigorous assessment of therapeutic success combining lung histology and function, and lastly and more importantly determining the long term safety of these MSCs. The potential for teratomas and other cancers is too high a risk without certain evidence that the exogenous administration of adult BM-derived stem cells is safe.