Published ahead of print on February 22, 2007, doi:10.1164/rccm.200609-1370OC
© 2007 American Thoracic Society doi: 10.1164/rccm.200609-1370OC
Cell-based Angiopoietin-1 Gene Therapy for Acute Lung Injury1 Terrence Donnelly Research Laboratories, Division of Cardiology, St. Michael's Hospital, Toronto, Ontario, Canada; 2 Institute of Medical Science, Departments of 3 Medical Biophysics and 4 Medicine and the McLaughlin Centre for Molecular Medicine, University of Toronto, Toronto, Ontario, Canada; and 5 Molecular and Cellular Biology Research, Sunnybrook Research Institute, Toronto, Ontario, Canada Correspondence and requests for reprints should be addressed to Duncan J. Stewart, M.D., F.R.C.P.C., University of Toronto, Room 6-050k, Queen Wing, Terrence Donnelly Heart Centre, St. Michael's Hospital, 30 Bond Street, Toronto, ON, M5B 1W8 Canada. E-mail: stewartd{at}smh.toronto.on.ca
Rationale: The acute respiratory distress syndrome is a significant cause of morbidity and mortality in critically ill patients. Angiopoietin-1 (Ang-1), a ligand for the endothelial Tie2 receptor, is an endothelial survival and vascular stabilization factor that reduces endothelial permeability and inhibits leukocyte–endothelium interactions. Objectives: We hypothesized that Ang-1 counteracts vascular inflammation and pulmonary vascular leak in experimental acute lung injury. Methods: We used cell-based gene therapy in a rat model of ALI. Transgenic mice overexpressing Ang-1 or deficient in the Tie2 receptor were also studied to better elucidate the mechanisms of protection. Measurements and Main Results: The present report provides data that support a strong protective role for the Ang-1/Tie2 system in two experimental models of LPS-induced acute lung injury. In a rat model, cell-based Ang-1 gene transfer improved morphological, biochemical, and molecular indices of lung injury and inflammation. These findings were confirmed in a gain-of-function conditional, targeted transgenic mouse model, in which Ang-1 reduced endothelial cell activation and the expression of adhesion molecules, associated with a marked improvement in airspace inflammation and intraalveolar septal thickening. Moreover, heterozygous Tie2-deficient mice demonstrated enhanced evidence of lung injury and increased early mortality. Conclusions: These results support a critical role for the Ang-1/Tie2 axis in modulating the pulmonary vascular response to lung injury and suggest that Ang-1 therapy may represent a potential new strategy for the treatment and/or prevention of acute respiratory distress syndrome in critically ill patients.
Key Words: acute lung injury (ALI) • acute respiratory distress syndrome (ARDS) • angiopoietin-1 • cell-based gene transfer • LPS • Tie2
Acute respiratory distress syndrome (ARDS) continues to be a common cause of morbidity and mortality in both adults and children (1). The annual incidence is estimated at 79 cases/100,000 population, and the mortality rate is estimated at 39% (2). Indeed, ARDS was the principal feature in patients with severe acute respiratory syndrome (3, 4), contributing substantially to morbidity and mortality, and emphasizing the need for new therapeutic strategies to effectively treat or prevent this pathology. ARDS is characterized by influx of polymorphonuclear leukocytes in airspaces, and massive interstitial and alveolar accumulation of protein and fluid. Although the exact pathogenic mechanisms remain poorly defined, recent observations suggest that a disruption of normal endothelial function plays an important role (5). Endothelial injury and subsequent loss of integrity of the endothelial barrier is a prerequisite for development of interstitial edema, and is associated with a marked increase in permeability in experimental models (6–9). In addition, endothelial activation and subsequent expression of molecules that mediate adhesion and signaling of leukocytes are required for the accumulation of inflammatory cells (10–12). These insights into the mechanisms of injury and inflammation during ARDS have focused attention on therapeutic strategies targeting the pulmonary endothelium. Angiopoietin (Ang)-1 is a ligand of the endothelial-selective tyrosine-kinase receptor, Tie2, and is an essential mediator of angiogenesis, promoting vessel maturation and stabilization (13–15). Ang-1 is also an endothelial survival factor (16), and was recently shown to protect blood vessels against plasma leakage in vivo and inhibit endothelial permeability in vitro (6, 17, 18). Ang-1 inhibits leukocyte adhesion to vascular endothelium and reduces the expression of tissue factor and various adhesion molecules in endothelial cells stimulated by inflammatory cytokines (6, 19, 20). These findings strongly support an antiinflammatory role for this proangiogenic and antiapoptotic factor, with the capacity to maintain endothelial function and prevent pathologic changes in endothelial gene expression. Consistent with this role, it was recently reported that neutrophil infiltration, capillary leakage, and lung edema in experimental acute lung injury (ALI) are associated with increased production of vascular endothelial growth factor (VEGF) and down-regulation of Ang-1 and Ang-4 (21). In addition, pretreatment with an adenoviral construct encoding Ang-1 reduced mortality in mice with endotoxic shock (22); however, because transduction was limited to the liver in this experiment, it is not known whether the improved survival was due to systemic vascular effects of increased circulating levels of Ang-1 or to a direct effect on the lung, reducing vascular injury. We hypothesized that local pulmonary gene transfer of Ang-1 would protect the lungs from LPS-induced ALI. Although gene therapy conceptually remains a promising treatment option for both genetically determined and idiopathic diseases, the use of viral vectors has been problematic, ranging from host immune responses to tumorigenesis. For this reason, we used cell-based gene therapy in a rat model of ALI to achieve targeted overexpression of Ang-1 in the lung microvasculature while circumventing the problems of in vivo transfection and avoiding the confounding effects of viral proteins (23–25). Moreover, transgenic mice overexpressing Ang-1 or deficient in the Tie2 receptor were also studied to better elucidate the mechanisms of protection. We now report that Ang-1 overexpression reduced airspace inflammation and intraalveolar septal thickening after LPS-induced lung injury, associated with reduced expression of endothelial-selective adhesion molecules, and normalization of endothelial nitric oxide synthase (eNOS) and endothelin (ET)-1 expression. In contrast, deficiency in the Tie2 receptor resulted in increased susceptibility to lung injury and reduced survival after LPS exposure. These results have been previously reported in part in abstract form and in a conference proceeding (26–29).
Cell Transfection Skin fibroblast cells were isolated from syngeneic male Fisher344 rats (Charles River Co., Saint Constant, PQ, Canada) and transfected with null plasmid vector, pFLAG-CMV-1 (pFLAG), or the same vector containing the full-length cDNA for human Ang-1 (pAng-1) using Superfect (Qiagen, Valencia, CA), as previously described (23–25). After 24 hours, cells were suspended in Dulbecco's phosphate-buffered saline (PBS; Invitrogen, Carlsbad, CA) for injection into the pulmonary circulation.
Rat Model of ALI
Mouse Model of ALI
Flow Cytometry
Quantitative Reverse Transcriptase–Polymerase Chain Reaction
ELISAs Rat tumor necrosis factor (TNF)-  , IL-1 , and IL-6 ELISA kits and mouse TNF-, IL-1, IL-6, soluble ICAM-1, soluble VCAM-1, soluble E-selectin, and soluble P-selectin ELISA kits were used, following the manufacturer's instructions (R&D Systems, Minneapolis, MN). The ET-1 ELISA kit (ALPCO Diagnostics, Salem, NH) was used following the manufacturer's instructions on precipitated lavage fluid spun at 3,000 x g for 20 minutes.
Western Blots
Immunohistochemistry and Histopathology Mouse and rat lung samples were fixed in 4% paraformaldehyde, paraffin embedded, cut into 5-µm sections, and stained with H&E. Intraalveolar septal thickness was quantified by measuring all septae along a crosshair placed on each image (approximately 50 septae per animal) using ImageJ software (National Institutes of Health, Bethesda, MD).
Statistical Analysis
Role of Ang-1 in a Rat Model of ALI Immunofluorescent staining of the lung with von Willebrand factor (green) to label the endothelium is shown in Figure 1A, a and b. At 15 minutes after injection, CMTMR-labeled fibroblasts (red) were visible in or near small arterioles (a), persisting in the lungs 48 hours later (b). CMTMR-labeled fibroblasts were not detectable in liver, spleen, or kidney (data not shown), and have previously been shown to persist in the lung for up to 6 months (24). Immunofluorescent staining of the lung for Ang-1 (green) is shown Figure 1A, c and d. The expression of the Ang-1 transgene by transplanted fibroblasts is evident by the yellow color, indicating colocalization of CMTMR and Ang-1 signals, 15 minutes after injection (c), again persisting for 48 hours after injection (d). Immunohistochemical staining confirmed that Ang-1–expressing fibroblasts persist in the lung even after LPS exposure (data not shown). Quantitative real-time RT-PCR analysis of total Ang-1 mRNA levels showed a reduction in total Ang-1 mRNA of 53% after LPS exposure compared with saline with pFLAG-transfected cells (Figure 1B). Pretreatment with pAng-1–transfected cells restored total Ang-1 mRNA to sham levels. In contrast, plasmid-derived Ang-1 mRNA was undetectable in animals pretreated with pFLAG-transfected cells, whereas plasmid Ang-1 transcript levels were similar in both groups that received injection of pAng-1–transfected cells (Figure 1C). RT-PCR analysis of the receptor tyrosine kinase, Tie2, demonstrated a reduction in Tie2 mRNA of 59% after LPS exposure compared with saline with pFLAG-transfected cells (Figure 1D); and, pretreatment with pAng-1–transfected cells partially restored Tie2 expression. Moreover, both total Tie2 and phosphorylated Tie2 protein were decreased after LPS exposure compared with saline with pFLAG-transfected cells (Figure 1E), which again was partially restored by pretreatment with pAng-1–transfected cells.
Ang-1 cell therapy attenuated intraalveolar septal thickness and airspace inflammation in rats. Administration of pAng-1–transfected cells, compared with pFLAG-transfected cells, did not significantly alter lung morphology in rats that received intratracheal instillation of normal saline (Figure 2A, a and b). Intraalveolar septal thickness and total cell count in BAL fluid (BALF) were increased two- and fourfold, respectively, after LPS exposure compared with saline (Figure 2A, c, Figures 2B and 2C). Pretreatment with pAng-1–transfected cells significantly reduced these indicators of vascular inflammation (Figure 2A, d; Figures 2B and 2C). The reduction in total BALF cells was mainly attributable to a reduction in the number of neutrophils (Table 2).
Total protein in BALF and lung wet weight–to–body weight ratio were increased after LPS exposure compared with saline (Figures 2D and 2E). There was a trend toward a reduction in these indicators of pulmonary vascular leak in animals receiving Ang-1 cell therapy (p = 0.10 for BALF protein; p = 0.09 for wet weight–to–body weight ratio).
Ang-1 cell therapy did not reduce expression of proinflammatory cytokines in rats.
Selective effect of Ang-1 cell therapy on endothelial adhesion molecule expression in rats. In animals receiving pFLAG-transfected cells, mRNA levels for ICAM-1, VCAM-1, P-selectin, and E-selectin in whole lung tissue were increased by 2- to 32-fold after exposure to LPS compared with saline (Figures 3A–3D). The increases in ICAM-1, VCAM-1, and P-selectin mRNA levels after LPS were similar in rats receiving pAng-1– compared with pFLAG-transfected cells. In contrast, pretreatment with pAng-1–transfected cells largely prevented the increase in mRNA expression of the endothelial-selective adhesion molecule, E-selectin, in LPS-treated rats (Figure 3C). Western blot analysis showed no difference in ICAM-1 or VCAM-1 protein expression between the experimental groups, whereas E-selectin protein was induced by LPS exposure, and this was substantially reduced by the administration of pAng-1–transfected cells.
Ang-1 cell therapy affects vasoactive gene expression in rats. In the control animals, the expression of heme oxygenase-1 (HO-1) mRNA in whole lung was significantly increased by pAng-1 gene transfer in the absence of LPS exposure (Table 4), whereas Ang-1 alone had no effect on the expression of the other genes studied. LPS exposure produced a marked increase in the mRNA expression of inducible NOS (55-fold; p < 0.05) in the null-transfected group, with more modest increases in mRNA levels of ET-1 (threefold; p < 0.05), HO-1 (twofold; p < 0.05) and Ang-2 (1.5-fold; p < 0.05). In LPS-treated animals, Ang-1 gene transfer reduced ET-1 and increased eNOS mRNA expression in the lung compared with null (pFLAG) gene transfer in LPS, but did not alter the expression of inducible NOS. ET-1 protein in BALF by ELISA was increased 5.6-fold after LPS compared with saline with pFLAG-transfected cells, and this was reduced (33%; p < 0.05) by pretreatment with pAng-1–transfected (pFLAG/saline, 0.63 ± 0.19; pAng-1/saline, 0.58 ± 0.24; pFLAG/LPS, 3.33 ± 0.37; pAng-1/LPS, 2.22 ± 0.16 fmol/ml). Interestingly, VEGF and Ang-2 mRNA levels were not different between the experimental groups.
Role of Ang-1 in Transgenic Models of ALI Experiments were performed using Tie2 heterozygous–deficient mice (Tie2+/–) or binary transgenic mice in which Dox-conditional Ang-1 overexpression was targeted to endothelium using the Tie1 promoter, tTA–Ang-1. WT littermates were used as control animals. Total lung Ang-1 protein expression by Western blot analysis was not different between endothelial-targeted, Ang-1–overexpressing mice and WT or Tie2 heterozygous–deficient mice (Figure 4A). This may be a result of high basal levels of pulmonary Ang-1 expression in extraendothelial cells under normal conditions (17, 18), which may overshadow the human Ang-1 transgene expression, which is restricted to the endothelium. However, the decrease in Ang-1 protein expression seen in WT mice after LPS challenge was blunted in tTA–Ang-1 mice. Paradoxically, the LPS-induced reduction in Ang-1 was also attenuated in Tie2-deficient mice, which we hypothesize might reflect a compensatory mechanism for the very low expression of Tie2 in these animals. As expected, basal Tie2 expression was reduced by about 50% in Tie2+/– compared with WT mice (Figure 4B), and Tie2 protein was markedly down-regulated by LPS exposure in WT mice, which was further reduced in Tie2+/–.
Septal thickness and airspace inflammation in transgenic mice. In the absence of LPS, lung morphology was not significantly different between groups of transgenic mice (Figure 5A). The intraalveolar septal thickness increased 2.7-fold after LPS exposure in WT mice (Figure 5B). This increase in septal thickness was significantly blunted in Ang-1–overexpressing mice, whereas Tie2-deficient mice exhibited nearly a 50% greater increase in intraalveolar septal thickness in response to LPS compared with WT animals. LPS-induced airspace inflammation mirrored these differences with the increase in total cells in BALF. This increase, attributable mainly to an increase in neutrophils, was markedly blunted in Ang-1–overexpressing mice, whereas Tie2-deficient animals consistently exhibited exaggerated airspace inflammation (Figure 5C; Table 5). Total protein in BALF was increased threefold after LPS in WT mice, and again this was suppressed in Ang-1–overexpressing mice and increased in Tie2-deficient animals (Figure 5D). In addition, mortality was higher in Tie2-haploinsufficient mice subjected to LPS challenge (60% mortality within 1 h) compared with the other experimental groups (17%; p < 0.02). Necropsy revealed that mortality was associated with massive alveolar flooding in Tie2-deficient mice.
Proinflammatory cytokines in transgenic mice. TNF- , IL-1, and IL-6 were undetectable in control mice, and were all markedly increased in lavage fluid after LPS exposure in WT mice (Figures 6A–6C). In contrast to Ang-1 gene transfer in the rat, Ang-1–transgenic mice exhibited significant reductions in all three cytokines in response to LPS. Heterozygous Tie2-deficient animals exhibited similar increases in cytokine levels compared with WT mice, with the exception of IL-6, which was higher.
Ang-1 overexpression reduces endothelial expression of adhesion molecules. To evaluate changes in the expression of adhesion molecules in endothelial cells versus other lung cell types, mouse lungs were digested in dispase and stained for the endothelial cell marker, CD31. Subsequently, flow cytometry was performed for each adhesion molecule separately using two-color flow cytometry, gated around the CD31+ and CD31– populations. The percentage of endothelial cells positive for E-selectin, P-selectin, ICAM-1, and VCAM-1 increased from 4- to 34-fold after LPS in WT mice (Figure 7A, left panels). Endothelial-targeted Ang-1 overexpression nearly completely suppressed the LPS-induced increases in endothelial expression of E-selectin, P-selectin, and VCAM-1 (Figures 7A, 7B, and 7D), but had only a marginal effect on ICAM-1 expression (Figure 7C). Surprisingly, Tie2-deficient mice showed similar, although more modest, reductions in adhesion molecule expression in response to LPS. This may be explained by greater adhesion molecule shedding in these animals (Figure 8), consistent with more advanced inflammation. In addition, similar changes, although with lower overall numbers of positive cells, were seen when the expression of these adhesion molecules was studied in all dispersed lung cells (Figure 7, right panels).
We investigated the relevance of the Ang-1/Tie2 system in modulating lung vascular injury using both gene transfer and transgenic approaches. In both the rat and mouse models, Ang-1 overexpression markedly reduced septal edema in response to LPS, as well as alveolar and endothelial inflammation, whereas Tie2-haploinsufficient mice showed an exaggerated response and had significantly greater short-term mortality. These data underscore the protective role of this pathway in maintaining pulmonary vascular homeostasis, and support the potential utility of Ang-1 in the treatment of acute lung injury.
In addition to being an essential mediator of angiogenesis (13–15), Ang-1 is an important endothelial survival factor, and has been shown to inhibit apoptosis in cultured endothelial cells induced by serum starvation, irradiation, and mannitol (16, 33). Ang-1 decreased vascular permeability in Ang-1–overexpressing mice (17), and acute administration of Ang-1 protected against VEGF-induced vascular leak (18). In a cell culture model, Ang-1 was shown to inhibit leukocyte migration across an endothelial monolayer, due to a reduction in endothelial expression of E-selectin, ICAM-1, and VCAM-1 (6, 19, 20). Karmpaliotis and colleagues (21) observed that endogenous Ang-1 expression was decreased in response to LPS in a murine model of ALI, whereas VEGF was increased. Similarly, we found a dose-dependent decrease in E-selectin, VCAM-1, and ICAM-1 by RT-PCR analysis in TNF- The endothelial monolayer plays a critical role in many aspects of the pathogenesis of ALI and ARDS (5, 34–36). Alterations in the production of vasoactive mediators by injured endothelium leads to impaired hypoxic pulmonary vasoconstriction (37, 38). Increased expression of angiogenic growth factors, such as VEGF, contributes to increased endothelial permeability and interstitial edema, increased pulmonary dead space, and vascular remodeling (6–9, 21, 39). However, perhaps the most important role of the endothelium in ALI and ARDS is in regulation of inflammation (5, 10, 40, 41). Leukocyte adhesion to the endothelium is a prerequisite for migration into the lung parenchyma, where the inflammatory cells contribute to lung injury (10–12). Indeed, previous studies have reported attenuation of lung injury in experimental ALI by blocking endothelial adhesion molecules (42, 43). Thus, strategies to selectively reduce endothelial inflammation in response to injury could be of potential benefit, not only in ALI and ARDS, but also in systemic inflammatory disorders, such as the systemic inflammatory response syndrome, sepsis, and the multiple organ dysfunction/failure syndrome. Ang-1 cell–based gene transfer also resulted in several downstream effects in the rat model of ALI. Expression of two protective enzymes, eNOS and HO-1, was increased after Ang-1 cell therapy in LPS-challenged rats. Mice overexpressing eNOS were previously shown to be protected from lung injury during endotoxic shock (44). HO-1 is known to increase during ARDS (45), and the increase in HO-1 activity is thought to be a protective mechanism (46). In addition, ET-1, which was previously shown to increase inflammation and pulmonary–vascular leak in ARDS, was reduced after Ang-1 cell therapy compared with LPS-challenged control animals (47). ET-1 receptor antagonists have been shown to be protective in an experimental model of ALI (48); thus, the mechanisms of protection of Ang-1 likely include modulation of the production of vasoactive factors. The Tie2 receptor is expressed predominantly on endothelial cells, as well as on some hematopoietic and inflammatory cells; thus, Ang-1 therapy in the lung would mainly target the vascular endothelium (49). It is likely, for this reason, that Ang-1 overexpression in the rat model only reduced the expression of adhesion molecules that are largely endothelial restricted, such as E-selectin, whereas no effect was seen for VCAM-1 and ICAM-1, which are also expressed on smooth muscle cells and type I and II epithelial cells. The transgenic mouse models overcame some of the potential limitations of the gene transfer approach. Whereas cell-based gene transfer resulted in focal regions of Ang-1 overexpression, localized mainly to the precapillary arteriole, Ang-1–tTA mice expressed Ang-1 uniformly in the lung endothelium. This may explain why the protection seen against LPS-induced microvascular injury was more complete in the transgenic model. In addition, unlike the rat model, total lung P-selectin, ICAM-1, and VCAM-1, as well as E-selectin, were all reduced to varying degrees in the Ang-1–transgenic mouse after LPS exposure. Moreover, in this model, we were able to distinguish between endothelial and nonendothelial adhesion molecule expression by flow cytometric analysis after tissue dispersion. Endothelial expression of ICAM-1, VCAM-1, P-selectin, and E-selectin were attenuated in Ang-1–overexpressing mice compared with WT mice; there were no significant differences in nonendothelial adhesion molecule expression between the experimental groups. In contrast, Tie2-deficient mice exhibited increased susceptibility to lung injury after LPS, with higher levels of inflammatory cells and protein in BALF, increased intraalveolar septal thickening, and higher early mortality than in WT mice. The only exception was the unexpected reduction in adhesion molecule expression in dispersed lung in Tie2-deficient mice compared with WT mice. However, these animals also exhibited increased circulating levels of soluble adhesion molecules, and it is possible that the lower cellular expression of adhesion molecules may reflect more extensive and rapid shedding from the endothelial cells, as well as other lung cell types, in the Tie2-haploinsufficient mice. It is important to note that circulating soluble adhesion molecule expression has been correlated with severity of ARDS, as well as mortality rates (19, 50, 51). Soluble adhesion molecule expression may increase when adhesion molecules are shed or proteolytically cleaved after endothelial–leukocyte interaction, or by up-regulation of a soluble variant of these adhesion molecules (52–54). Ang-1 cell therapy represents a possible therapeutic strategy to prevent development of clinical ARDS (55). Cell-based gene therapy could be advantageous over intravenous administration of recombinant Ang-1 protein, as it allows for more targeted expression of the transgene and overcomes issues of short protein half-life after injection. A number of clinical trials are currently underway assessing the safety and efficacy of cell therapy in several different pathologies. In addition, endothelial progenitor cells and mesenchymal stem cells may also provide protection above and beyond that provided by increased Ang-1 expression, as these cells may release many additional mediators that could be beneficial (56, 57). In this report, we demonstrate a strong protective role for the Ang-1/Tie2 system in experimental ALI. The benefits of Ang-1 overexpression were seen in two experimental models, showing reduced vascular endothelial inflammation and leakage. Overexpression of Ang-1 blunted endothelial adhesion molecule expression, increased HO-1 and eNOS expression, and decreased ET-1 expression, all of which likely contributed to reduced airspace inflammation, intraalveolar septal thickening, and early mortality. Therefore, these results suggest that Ang-1 therapy may represent a potential new treatment strategy to reduce the vascular consequences of lung injury, which are a major determinant of morbidity and mortality in critically ill patients.
The authors thank Ms. Lakshmi Kugathasan for maintaining the transgenic mouse colony and for doing the initial investigations with the transgenic mice, as well as Dr. Gerald A. Proteau and Mr. Malcolm Robb for their technical assistance.
Supported by the Canadian Institutes of Health Research (CIHR; grant MOP-74752) and Northern Therapeutics, Inc. S.D.M. is supported by the Heart & Stroke Foundation of Canada/Pfizer Canada Fellowship Award and the TACTICS Strategic Training Program in Cardiovascular Research. S.H.J.M holds an NSERC Canada Graduate Ph.D. Scholarship. C.H.P. was supported by a University of Toronto Institute of Medical Science Summer Student Scholarship. R.D.H. was supported by a University of Toronto R&D Health Research Foundation Summer Studentship in Medicine. R.S.S. holds a studentship and travel award from CIHR/Canadian Hypertension Society/Pfizer. Originally Published in Press as DOI: 10.1164/rccm.200609-1370OC on February 22, 2007 Conflict of Interest Statement: S.D.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. S.H.J.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. P.F.H.L. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Q.W.Z. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. C.H.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. R.S.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. R.D.H. received $4,500 in 2004 in the form of an R&D Health Research Foundation Summer Studentship in medicine. Y.D.Z. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Y.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. R.N.N.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. D.J.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. D.J.S. has been provided research funding from Northern Therapeutics for the Lung Injury Project. He is a founding Scientist and Chief Scientific Officer of Northern Therapeutics. Received in original form September 26, 2006; accepted in final form February 16, 2007
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
CT analysis was used to calculate expression in comparison to 18S RNA. Primers for genes of interest are listed in 
