| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Animal Model |
From the Department of Pathology and Molecular Medicine,* McMaster University, Hamilton, Canada; the Service de Pneumologie et Reanimation Respiratoire, Centre Hospitalo-Universitaire du Bocage et Universite de Bourgogne, Dijon, France; and the Developmental Biology Program, Childrens Hospital Los Angeles Research Institute, Keck School of Medicine, University of Southern California, Los Angeles, California
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
We have previously used adenovirus vectors to effect transient gene transfer of active TGF-ß1 (AdTGFß1223/225) to the lung of adult rats9 resulting in the development of progressive interstitial fibrosis. Herein we have used the same vector to transfer TGF-ß1 to the lung of 1-day-old rat pups. We show that this transient overexpression (5 to 10 days) of TGF-ß1 causes morphological and biochemical changes consistent with interstitial fibrosis along with large distorted alveoli that appear within a few weeks after birth, pathological features strikingly similar to those seen in human BPD.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
One-day-old neonatal Sprague-Dawley rats were exposed to isofluorane for 30 to 40 seconds to provide light anesthesia and were administered intranasally with 1 x 108 plaque forming units (PFUs) of either Ad dl70-3 (control vector, no transgene) or AdTGFß1223/225 (vector expressing the active form of TGF-ß1) diluted in 10 µl of sterile phosphate-buffered saline (PBS). Holding the anesthetized rat upright, the adenoviral solution was placed into the nasal passages using a P10 pipette. Using the thumb, slight pressure was placed on the bottom of the jaw to keep the mouth closed and force the animal to inhale the solution through its nose. The rats were held upright for a further 20 to 30 seconds to maximize delivery, and then allowed to recover under supervision before returning them to their mother. Preliminary studies with AdLacZ (vector expressing LacZ) showed this approach gave rise to wide distribution of the vector throughout the lung with uptake in bronchial and bronchiolar epithelial cells, similar to the distribution seen in adult rats with either intranasal or intratracheal instillation (data not shown). The control vector (empty) was chosen to avoid any confusing issues resulting from known immune responses against an expressed foreign transgene, such as LacZ.13 The TGF-ß1 gene is highly homologous and does not engender immune reactions in the host.
Detection of Transgene Expression
Because of the small and delicate nature of neonatal rat lungs, bronchoalveolar lavage could not readily be performed in a consistent manner to assess the presence of TGF-ß1 transgene. To accomplish this, three to five neonatal rats per group were sacrificed 2 days after adenovector injection. Before excising their lungs, the right ventricle of the heart was perfused with 3 ml of ice-cold PBS to remove all blood from the lungs. The lungs were then excised, weighed, and kept in ice-cold PBS. Lung weights did not vary significantly between individual rats or treated groups throughout the extent of the experiment. Shortly thereafter, lungs were transferred to 2 ml of fresh PBS and homogenized. The homogenate was pelleted by centrifugation at 12,000 rpm for 10 minutes at 4°C and the cytokine-rich supernatant was stored at -70°C until assayed. TGF-ß1 levels were determined as described below and expressed as pg of active TGF-ß1 per mg wet lung weight.
TGF-ß1 Assay
TGF-ß1 levels were determined using a human TGF-ß1 enzyme-linked immunosorbent assay kit (R&D Systems, Inc., Minneapolis, MN) which is totally cross-reactive with porcine and rodent TGF-ß1, but does not cross-react with TGF-ß2 or TGF-ß3. Total TGF-ß1 levels were measured by acid-activating samples as per the manufacturer’s protocol. Levels of active TGF-ß1 were measured by assaying samples that were not acid-activated.
Lung Hydroxyproline Determination
A modified Woessner protocol14 was used. Tissue was prepared by immediately snap-freezing the right lobe of excised lungs in liquid nitrogen, then transferring the samples to -70°C for storage until the time of assay. Samples were homogenized in 5 ml of ddH2O. One ml of the resultant homogenate was hydrolyzed in 2 ml of 6 N HCl for 16 hours at 110°C. Samples were adjusted to pH 7.0 and then 400 µl was diluted into 2.0 ml with ddH2O. The colorimetric assay was initiated with the addition of 1 ml of chloramine-T solution to the diluted sample at pH 7.0. One ml of 70% perchloric acid was added, followed by 1 ml of a dimethylbenzaldehyde solution. After 20 minutes at 60°C, the samples were returned to room temperature and the optical density at 557 nm was determined. Using hydroxyproline standards (Sigma Chemical, St. Louis, MO), results were expressed as µg hydroxyproline per mg wet lung weight.
Lung Fixation and Histological Examination
In adult mice, on lung excision, the left lobe was inflated and perfused with 10% formalin, and fixed for 24 hours before processing and paraffin embedding. In neonates, <10 days old, tissue was not perfused before fixation because of the small size of vessel for cannulation. Histological sections were stained with hematoxylin and eosin, a general nuclear stain, and elastin van Gieson, which stains collagen and elastin. Sections were also stained for collagen matrix with Sirius Red F3B (CI 35780) (Sigma Aldrich Can. Ltd., Oakville, Ontario, Canada) in saturated picric acid (Picrosirius Red stain). Select sections also underwent immunohistochemical staining for 
-smooth muscle actin (-SMA), as previously described.9
Determination of Average Alveolar Size
The size of alveolar space was determined from lung sections stained with Picrosirius Red. Mean chord length (µm) was measured at x100 magnification and is calculated from automated morphometric analysis using a Leica Q500IW high-grade image analysis system. Chord length increases with alveolar enlargement and is similar to mean linear intercept with the advantage of being independent of the thickness of septa.15
Data Analysis
Data were expressed as mean ± SEM or SD. Statistical significance was determined using Student’s two-tailed t-test, assuming unequal variances. Differences were considered statistically significant if P was <0.05. All pairwise multiple comparisons were subjected to the Tukey test. In considering comparisons between three groups of animals, a one-way analysis of variance test was performed.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
. Expressed as pg of TGF-ß1 per mg of lung tissue, the pups administered AD5E1TGFß1223/225 showed significantly higher (>30-fold) levels of total and active (10-fold) TGF-ß1 than those treated with control vector (P < 0.02). There was no significant difference in TGF-ß1 levels between control vector-treated and untreated groups.
|
and alveoli that appeared to be smaller in size than those seen at birth. The animals that were treated with the control vector, dL70-3, showed a similar hypercellular interstitium with evidence of some inflammatory cells, neutrophils, and mononuclear cells (Figure 2B)
. The animals that were treated with AdTGF-ß1 also showed a hypercellular interstitium with a number of inflammatory cells, along with areas of apparent consolidation (Figure 2C)
, seen to be positive for matrix deposition by staining with Picrosirius Red (Figure 2F)
.
|
. Alveoli were divided and subdivided into their saccular network, providing a much larger area for gas exchange than was available at birth when the alveoli had not yet fully compartmentalized. Rat pups injected with the TGF-ß1-expressing adenovirus vector developed disperse fibrotic lesions within their lungs (Figure 4)
, although not to the extent demonstrated in experiments in which the same vector was administered intratracheally to adult rats.9
There was marked deposition of matrix within the parenchyma localized to defined areas throughout the lung. These were not particularly distributed, for example, in peribronchial areas, but tended to be associated at air interfaces rather than at vascular sites. Analysis of hydroxyproline content at 29 days showed an increase in matrix accumulation within the lungs of the Ad5E1TGFß1223/225-treated pups compared to the control vector group (Figure 5)
. Using a one-way analysis of variance, the differences between the three groups showed P = 0.078. Using an independent t-test, day 28 TGF-ß1-treated animals differed from day 21 TGF-ß1-treated animals (P < 0.05) and day 28 TGF-ß1-treated animals differed from control vector-treated animals (P < 0.05). There were no significant differences seen at 21 days, consistent with the progressive nature of fibrosis initiated by TGF-ß1 overexpression in adult rats.9
The areas positive for matrix deposition were also prominent for -SMA-positive cells, most likely myofibroblasts. In addition, there were prominent -SMA-positive cells in the alveolar walls indicating an enhanced presence of myofibroblasts throughout the lung parenchyma (Figure 6)
.
|
|
|
|
. There were no significant differences in alveolar space sizes at birth between the treatment groups (Table 1)
. By 21 days, the enlarged spaces were clearly evident with 
50% increase in mean alveolar chord length from control-treated mice, and at 28 days, the increase was nearly 80% (P > 0.001 for both comparisons against untreated or control vector-treated animals) indicating the presence of markedly enlarged alveoli throughout the lung. The areas of enlarged alveoli are in many instances immediately adjacent to areas showing evidence of marked fibrosis (Figure 7)
suggestive of pathology seen in human BPD.5
In isolated cases, inflammatory cells may persist within the alveolar septa, but more often the septa resemble those of a normal lung, without inflammatory infiltrate or they are thickened from an interalveolar fibrotic process.
|
|
. The underdeveloped alveoli and localized fibrotic lesions that developed after TGF-ß1 exposure are both hallmarks of BPD in human newborns and they were not seen after treating pups with the control vector.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
In the human, at birth, nonrespiratory airways are well developed, however, the alveolar structure of the lung is not yet mature. Extensive alveolarization greatly increases the surface area available for gas exchange by dividing and subdividing the large, premature peripheral air spaces already formed. During this period of growth, the infant lung is particularly sensitive to perturbations in developmental signals. BPD is a common neonatal lung disease that may be caused by such perturbations and is particularly prevalent in premature, very low birth weight newborns. Infants suffering from BPD exhibit a dramatically reduced lung volume,17 lower levels of alveolarization,18 and much larger alveolar spaces19 compared to normal infants.6
Chronic inflammation and abnormal tissue repair can lead to pulmonary fibrosis in mature lungs, and similarly, BPD in infants may be caused by inflammation, illustrated by the positive correlation between the occurrence of BPD and high levels of the inflammatory cytokines tumor necrosis factor-, interleukin (IL)-1ß, IL-6, and IL-8 in amniotic fluid.20
Similar inflammatory mediators are also found in premature neonatal lamb lungs that have been ventilated.21
Tracheal aspirates from human premature infants that have an increased risk of developing BPD contain proinflammatory cytokines such as IL-8, while lacking anti-inflammatory cytokines such as IL-10.22
Increased levels of bioactive TGF-ß have also been detected and this can predict the need for prolonged oxygen therapy in extremely premature human infants.23
Although high O2 levels may directly contribute to BPD, it has been shown that overinflating the lung during ventilation can also lead to the production of profibrotic cytokines such as tumor necrosis factor-, basic fibroblast growth factor, and TGF-ß1 as well as extracellular matrix components.21,24,25
One might suggest that BPD possibly begins with an inflammatory insult, much like pulmonary fibrosis in adults, followed by the overexpression of TGF-ß1 and deposition of excess extracellular matrix components. The early inflammation could be initiated either by hyperventilation and/or overdistension of the lung, by increased levels of O2, as well as by exposure of the lung to inflammatory mediators within the amniotic fluid, which may be present because of invading microbes within the mother.18,20,24
To investigate the theory that overexpression of TGF-ß1 plays a key role in the pathogenesis of BPD, we administered neonatal rats intranasally (a process that delivers material to the lung epithelium) with Ad5E1TGFß1223/225, an adenovirus vector expressing the active form of TGF-ß1, and compared lung histology and collagen content with those seen in controls, either administered a control adenovirus vector or naïve neonatal lungs. Transfer of cytokine and growth factor genes to the lung by adenovirus vectors results in expression of the transgene from infected epithelial cells and lasts for 5 to 10 days, after which the gene is effectively no longer expressed from the vector. Tissue changes seen subsequent to this period are a result of this transient overexpression of transgene.9,26-28 The data shown here indicate that transient overexpression of TGF-ß1 induces pathology that closely mimics that seen in the human disease BPD and provides us with an excellent model to study the disease course.
In a study of 46 lung autopsy specimens from infants that suffered negative effects of respiratory distress syndrome, Erickson and colleagues5
defined two distinct pathological features of BPD and assigned patients into one of three distinct groups. Patients in group 1 have interstitial fibrosis, with extensive scar formation, and a marked inflammatory infiltrate. Group 1 BPD occurs in the very young infant (less than 2 months of age), and is often called early BPD. At the other end of the spectrum, seen in older infants (up to 9 months after birth) are group 3 patients. The lungs of these infants do not show evidence of interstitial fibrosis, but their lungs have reduced numbers of alveolar spaces, resulting in fewer but much larger terminal air spaces. Group 2 BPD has features from both group 1 and group 3, with distal areas of the lung experiencing both pathologies. The progression from interstitial fibrosis to large, underdeveloped airspaces with relatively normal septa is believed to be the hallmark of the BPD disease process.5
The pathology seen in our current studies of neonatal rats closely resembles the photomicrographs seen in the study by Erickson and colleagues5
with evidence of areas of fibrosis adjacent to areas of large alveoli (Figures 6 and 7)
. Consistent with the alveolar hypoplasia seen in the new BPD in low birth weight infants, TGF-ß1-treated animals exhibited numerous areas of primarily large alveoli with an average alveolar size up to twice that of untreated or control vector-treated mice.
Normal rat lungs examined 2 to 4 days after birth demonstrate an incompletely developed lung with a hypercellular interstitium surrounding the airspaces (mean chord length 36.3 ± 6.0 µm). Evaluation of high-power fields indicates the cells in the interstitium are primarily noninflammatory cells (data not shown). Rats of an equivalent age that were administered either control adenovirus vector or Ad5E1TGFß1223/225 have similar alveolar spaces at this age (mean chord length 41.7 ± 9.0 and 39.9 ± 9.5, respectively), but the hypercellular interstitium contains evidence of inflammatory cell infiltration. The lungs of rats treated with either vector were indistinguishable at 3 days after injection (4 days old), and the reaction at this stage is primarily in response to the virus infection (data not shown). After 7 days (8 days old), the lungs treated with Ad5E1TGFß1223/225 show evidence of new matrix deposition and appear occluded, however, at this time lung hydroxyproline content is not significantly different from control animals. At this stage, these lungs resemble the lungs of infants with early BPD. In rats, the normal initial burst of alveolar formation occurs from days 3 to 8 postnatally.29 Normal untreated lungs at day 8, while still showing a hypercellular interstitium within the alveolar septa, have been undergoing septal thinning and alveolarization and their alveolar spaces have successfully divided to a much greater extent. Administration of active TGF-ß1 to the lungs during this period is likely to interfere with new alveolar formation and as the septa thin, the alveolar space may enlarge. The fact that the mean chord length of day 21 TGF-ß1-treated mice are already markedly enlarged (63.8 ± 10.3 µm) indicates that the enhanced presence of TGF-ß1 may inhibit physiological alveolarization in a similar manner to the inhibition of branching morphogenesis seen in the embryo.11,12
At 29 days of age, normal rat lung approaches maturity with the alveolar septa thinned to involve only a few cells and alveolar development has nearly reached adult levels. Rats injected with control virus had also recovered, and have apparently normally developed lungs at this time, although some lung specimens did exhibit interstitial infiltrates, which have probably persisted since vector administration. It would seem likely that the relatively low-level inflammatory insult caused by the adenovirus vector itself is enough to cause this limited pathology.30 Nonetheless, the hydroxyproline content is nearly identical between the two control groups and there is no histological evidence of fibrosis and the alveolar space size is similar in both groups. Whether the phagocytic influx is necessary for the development of fibrosis subsequent to the expression of active TGF-ß1 cannot be ascertained from these data, however, having expressed more than 20 cytokines in this manner in the lung, only two, IL-1ß and TGF-ß1, exhibit the ability to initiate progressive and widespread fibrosis.9,31
In contrast, neonatal rats at 28 days after administration of Ad5E1TGFß1223/225 (29 days of age) showed pathology closely resembling either mid-stage or late-stage human BPD. Areas with interstitial fibrosis demonstrate marked matrix deposition, leading to significantly higher hydroxyproline content of the treated lungs (150% of control levels). Stimulation of matrix production by TGF-ß includes a number of matrix glycoproteins such as fibronectin, recently shown to have a direct correlation with the extent of fibroproliferation in BPD.32
Often adjacent to fibrotic areas are large, alveolar spaces surrounded by normal-appearing septa with only limited cellularity. Although the septa have apparently undergone normal thinning, there does not appear to have been continued alveolar division, resulting in the enlarged alveolar spaces seen, characteristic of a more advanced stage of disease progression. The marked difference in mean alveolar space size seen between the TGF-ß1-treated animals and either control group (2 times higher) confirms the morphological appearances seen in Figure 4
. The appearance of -SMA-positive myofibroblasts within the interstitium of enlarged alveoli might indicate a physiological impact of these contractile cells on alveolar function (Figure 6)
. In addition, within the fibrotic areas, -SMA-positive cells would be major contributors to the deposition of matrix. Thus these cells may play different roles in different tissue sites within the lung.
At 29 days of age, while approaching lung maturity, untreated normal rats show evidence of continuing alveolar division and differentiation, as protruding alveolar septations end in elastin caps, which is a feature characteristic of budding septa. These sites seem to be equally prevalent in all groups, and show that the alveolar content of rats are not yet fully developed to adult levels at this time. The heterogeneity of the response in terms of the disease course would likely evolve throughout a longer time frame. It will be interesting to observe the final stage of the disease course at a later time point, and whether the hydroxyproline levels decrease in accordance with the resolution of interstitial fibrosis in this model, as predicted for human disease by Erickson and colleagues.5
Our results suggest that TGF-ß1 may continue to act as a negative regulator of normal alveolar differentiation in the newborn. We have shown previously in embryonic lungs that overexpression of active TGF-ß1 acts on embryonic bronchial epithelium to prevent branching morphogenesis,10 possibly by negatively regulating positive signaling through other growth factor receptors.33 Similarly, fetal stage transgenic epithelial expression of TGF-ß1 arrests lung epithelial and vascular morphogenesis in the pseudoglandular stage.34 Recently, a regulated transgenic mouse model of neonatal active TGF-ß1 expression from lung epithelium was presented that showed similar features consistent with BPD.35
The adenovirus vector delivery of constitutively active TGF-ß1 to neonatal rat lungs initiates a disease course that closely resembles BPD in human newborns. This neonatal rat model of BPD is characterized by both alveolar hypoplasia and interstitial fibrosis. Thus, the model has features of both classical and new forms of BPD. The development of an efficient intranasal delivery method makes this a promising model to further evaluate the role of other growth factors and cytokines in normal and altered neonatal lung development. Ideally, this model will be useful in the development of novel therapeutic strategies that might ameliorate or prevent the negative effects of BPD on the human neonatal lung.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by the Canadian Institutes of Health Research, Hamilton Health Sciences, St. Joseph’s Healthcare, and the National Institutes of Health (National Heart, Lung, and Blood Institute grant PO1HL60231).
Accepted for publication August 27, 2993.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
, interleukin-1ß, and interleukin-8) and the risk for the development of bronchopulmonary dysplasia. Am J Obstet Gynecol 1997, 177:825-830[Medline]
to rat lung induces severe pulmonary inflammation and patchy interstitial fibrogenesis with induction of TGF-ß1 and myofibroblasts. Am J Pathol 1998, 153:825-832This article has been cited by other articles:
 |
|
 |
|
  N. Ambalavanan, T. Nicola, J. Hagood, A. Bulger, R. Serra, J. Murphy-Ullrich, S. Oparil, and Y.-F. Chen Transforming growth factor-{beta} signaling mediates hypoxia-induced pulmonary arterial remodeling and inhibition of alveolar development in newborn mouse lung Am J Physiol Lung Cell Mol Physiol, July 1, 2008; 295(1): L86 - L95. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Buckley, W. Shi, L. Barsky, and D. Warburton TGF-{beta} signaling promotes survival and repair in rat alveolar epithelial type 2 cells during recovery after hyperoxic injury Am J Physiol Lung Cell Mol Physiol, April 1, 2008; 294(4): L739 - L748. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Nakanishi, T. Sugiura, J. B. Streisand, S. M. Lonning, and J. D. Roberts Jr TGF-beta-neutralizing antibodies improve pulmonary alveologenesis and vasculogenesis in the injured newborn lung Am J Physiol Lung Cell Mol Physiol, July 1, 2007; 293(1): L151 - L161. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. D. Bland, L. Xu, R. Ertsey, M. Rabinovitch, K. H. Albertine, K. A. Wynn, V. H. Kumar, R. M. Ryan, D. D. Swartz, K. Csiszar, et al. Dysregulation of pulmonary elastin synthesis and assembly in preterm lambs with chronic lung disease Am J Physiol Lung Cell Mol Physiol, June 1, 2007; 292(6): L1370 - L1384. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Alejandre-Alcazar, G. Kwapiszewska, I. Reiss, O. V. Amarie, L. M. Marsh, J. Sevilla-Perez, M. Wygrecka, B. Eul, S. Kobrich, M. Hesse, et al. Hyperoxia modulates TGF-beta/BMP signaling in a mouse model of bronchopulmonary dysplasia Am J Physiol Lung Cell Mol Physiol, February 1, 2007; 292(2): L537 - L549. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kunzmann, C. P. Speer, A. H. Jobe, and B. W. Kramer Antenatal inflammation induced TGF-beta1 but suppressed CTGF in preterm lungs Am J Physiol Lung Cell Mol Physiol, January 1, 2007; 292(1): L223 - L231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Gauldie, M. Kolb, K. Ask, G. Martin, P. Bonniaud, and D. Warburton Smad3 Signaling Involved in Pulmonary Fibrosis and Emphysema Proceedings of the ATS, November 1, 2006; 3(8): 696 - 702. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Chen, J. Sun, S. Buckley, C. Chen, D. Warburton, X.-F. Wang, and W. Shi Abnormal mouse lung alveolarization caused by Smad3 deficiency is a developmental antecedent of centrilobular emphysema Am J Physiol Lung Cell Mol Physiol, April 1, 2005; 288(4): L683 - L691. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. G. Vicencio, C. G. Lee, S. J. Cho, O. Eickelberg, Y. Chuu, G. G. Haddad, and J. A. Elias Conditional Overexpression of Bioactive Transforming Growth Factor-{beta}1 in Neonatal Mouse Lung: A New Model for Bronchopulmonary Dysplasia? Am. J. Respir. Cell Mol. Biol., December 1, 2004; 31(6): 650 - 656. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
