Lung disease: Treatment by cell transplant.

Pulmonary alveolar proteinosis (PAP) is a rare lung disease characterized by the accumulation in the lung of white blood cells called alveolar macrophages that are full of surfactant — a compound of phospholipids and proteins that regulates surface tension in the lung — and of vast amounts of extracellular surfactant1. Unravelling the cause of this disease, which was first recognized in 1958, is a story that began in 1994 with the serendipitous discovery2, 3 that mice lacking the protein GM-CSF, which is important for macrophage maturation and function, had a mysterious lung disease that resembled human PAP. In this issue, Suzuki et al.4 (page 450) add a chapter to this story, reporting that transplanting macrophages that correctly respond to GM-CSF into the lungs of mice lacking the GM-CSF receptor effectively treats their disease. Studies of GM-CSF-deficient mice identified the disease-causing defect as part of the process through which surfactant is broken down by macrophages in the alveolar region of the lung5 (Fig. 1). And human studies revealed that, although alveolar macrophages from some people with PAP respond to GM-CSF stimulation in vitro6, the patients express antibodies that neutralize the protein7. PAP is now classified into three forms: autoimmune (acquired), congenital (hereditary) and secondary (linked mainly to cancers of the blood or systemic inflammatory diseases). All forms of PAP are associated with loss of GM-CSF signalling owing either to deficiencies in active GM-CSF or to GM-CSF-receptor mutations, with the exception of some congenital forms that are associated with surfactant-protein abnormalities. Further work in mouse models and human PAP samples elucidated the steps of surfactant breakdown and the roles of key proteins in alveolar-macrophage biology. These included the transcription factors PU.1, which is involved in the maturation of alveolar macrophages8, and PPARγ, which maintains lung homeostasis9. Contrary to macrophages in other parts of the body, alveolar macrophages express high levels of PPARγ9, suggesting that this protein has a special role in the lung. Indeed, it has been shown that PPARγ is a negative regulator of macrophage activation10 and that its expression is stimulated by GM-CSF11. Alveolar surfactant is 90% lipid, and its catabolism is now known to be regulated by a signalling pathway involving GM-CSF, PPARγ and the protein ABCG1 (refs 12,13). Therapeutic options for PAP were also first identified in mouse models of the disease. Initial studies showed that GM-CSF-deficient mice could be 'cured' either by the administration of exogenous GM-CSF14 or by overexpression of GM-CSF in epithelial cells of the respiratory tract15. These findings led to a trial of treating people who have autoimmune PAP with high doses of GM-CSF, administered under the skin or by inhalation, although a subset of patients failed to respond to this treatment, possibly because of high levels of anti-GM-CSF antibodies in their lungs. An alternative approach to treating the autoimmune form of PAP is to use the monoclonal antibody rituximab, which blocks production of anti-GM-CSF antibodies16 and induces increased expression of PPARγ and ABCG1 (ref. 17) in some patients. However, the only treatment for patients with hereditary PAP is whole-lung lavage (irrigation), which requires general anaesthesia. It has been proposed that hereditary PAP might be corrected by transplantation of healthy bone-marrow cells, which contain stem cells that can differentiate into normal, GM-CSF-sensitive macrophages, and this procedure has indeed been successful in mice18. But this approach requires prior myeloablation — the severe or complete depletion of existing bone-marrow cells to avoid rejection of the transplanted cells. Myeloablation is associated with a high risk of infection and death and so bone-marrow transplantation is not routinely performed in patients with PAP. Suzuki et al. sought to design a transplantation approach that circumvents the need for myeloablation. They transplanted macrophages taken from normal mice directly into the lungs of mice deficient for the β-subunit of the GM-CSF receptor (which develop a disorder identical to that in children with hereditary PAP owing to mutations in the receptor's α- or β-subunits), and found that this treatment relieved the disease symptoms, normalized the expression of disease-related proteins and extended the lifespan of these mice. The authors then repeated the experiment using macrophages taken from GM-CSF-receptor-deficient mice that had been corrected ex vivo (by a process of lentiviral transduction) such that they expressed the β-subunit again, and saw the same effect. The feasibility of translating pulmonary-macrophage transplantation into a human therapy is strongly supported by these and other recent studies in mice. A previous study18 had also demonstrated that transplantation of wild-type murine macrophage-progenitor cells into GM-CSF-receptor deficient animals effectively reduced hereditary PAP disease. Suzuki and colleagues have improved on this approach by using gene-corrected cells from genetically identical animals, thereby avoiding the need for myeloablation and immunosuppression. This bears greater resemblance to the method that would most probably be adopted in humans, which would be to take a patient's own, mutated, macrophages, correct them ex vivo, and return them to the patient. Furthermore, the authors show disease remission with lessening of symptoms in mice. Suzuki et al. also found that, although wild-type bone-marrow-derived macrophages cultured in vitro had different characteristics to alveolar macrophages, they adopted a lung-macrophage profile following transplantation into the lung. This result supported earlier work19 indicating that local microenvironments provide signals that direct macrophage development. Future studies will be needed to determine the optimal dose of transplanted cells for people, the effect of intrinsically elevated GM-CSF levels associated with hereditary PAP and whether additional GM-CSF will need to be administered to promote survival of the transplanted macrophages. The therapeutic implications of this approach reach beyond the rare disease of hereditary PAP. One can imagine transplantation of autologous gene-corrected macrophages for the treatment of other diseases, such as HIV. Macrophages serve as a reservoir for the virus and individuals lacking a certain macrophage co-receptor are HIV resistant, so transplanting macrophages without the receptor may convey immunity to the virus. Because it seems that local environments provide cues for the development of macrophages with certain characteristics, the possibilities are almost endless once the gene mutations that are relevant to certain disease states are identified. The use of whole-genome sequencing to identify aberrant genes in infants born with life-threatening conditions may further extend the options for macrophage-transplantation therapies20. Download references