Bleomycininduced lung injury in mice ... - Wiley Online Library

PRECLINICAL AND CLINICAL IMAGING Full Papers

Magnetic Resonance in Medicine 67:499–509 (2012)

Bleomycin-Induced Lung Injury in Mice Investigated by MRI: Model Assessment for Target Analysis Anna L. Babin,1,2 Catherine Cannet,1 Christelle Ge´rard,1 Pierre Saint-Mezard,3 Clive P. Page,2 Helmut Sparrer,4 Tetsuya Matsuguchi,5 and Nicolau Beckmann1* Magnetic resonance imaging (MRI) has been used to follow the course of bleomycin-induced lung injury in mice and to investigate two knockout mouse lines with the aim of providing potential therapeutic targets. Bleomycin (0.25 mg/kg) was administered intranasally six times, once a day. MRI was carried out on spontaneously breathing animals up to day 70 after bleomycin. Neither cardiac nor respiratory gating was applied during image acquisition. A long lasting response following bleomycin has been detected by MRI in the lungs of male C57BL/6 mice. Histology showed that, from day 14–70 after bleomycin, fibrosis was the predominant component of the injury. Female C57BL/6 mice displayed a smaller response than males. Bleomycin-induced injury was significantly more pronounced in C57BL/6 than in Balb/C mice. MRI and histology demonstrated a protection against bleomycin insult in female heterozygous and male homozygous cancer Osaka thyroid kinase knockout animals. In contrast, no protection was seen in cadherin-11 knockout animals. In summary, MRI can quantify, in spontaneously breathing mice, bleomycin-induced lung injury. With the ability for repetitive measurements in the same animal, the technique is attractive for in vivo target analysis and compound profiling in this murine model. Magn Reson C 2011 Wiley Periodicals, Inc. Med 67:499–509, 2012. V Key words: bleomycin; cancer Osaka thyroid kinase; fibrosis; imaging; lung; magnetic resonance imaging; mouse; pulmonary fibrosis

Pulmonary fibrosis is a progressive and lethal lung disease involving an overexuberant repair process, characterized by accumulation of inflammatory cells, excessive fibroblast proliferation, increase in collagen content, and deposition of extracellular matrix in the lungs (1). The hallmark lesions are fibroblastic foci representing focal 1 Global Imaging Group, Novartis Institutes for BioMedical Research, Basel, Switzerland. 2 Sackler Institute of Pulmonary Pharmacology, Institute of Pharmaceutical Sciences, King’s College London, London, United Kingdom. 3 Developmental and Molecular Pathways Department, Novartis Institutes for BioMedical Research, Basel, Switzerland. 4 Department of Autoimmune Diseases, Novartis Institutes for BioMedical Research, Basel, Switzerland. 5 Department of Developmental Medicine, Kagoshima University Graduate School of Medical and Dental Sciences, Kagoshima, Japan. The first two authors contributed equally to this work. Grant sponsor: 3R Research Foundation (Mu¨nsingen, Switzerland); Grant number: 82/02; Grant sponsor: Novartis Institutes for BioMedical Research. *Correspondence to: Nicolau Beckmann, Ph.D., Novartis Institutes for BioMedical Research, Global Imaging Group, Forum 1, Novartis Campus, WSJ-386.2.09, Basel CH-4056, Switzerland. E-mail: nicolau.beckmann@ novartis.com Received 13 January 2011; revised 25 March 2011; accepted 28 April 2011. DOI 10.1002/mrm.23009 Published online 7 June 2011 in Wiley Online Library (wileyonlinelibrary. com). C 2011 Wiley Periodicals, Inc. V

areas of active fibrogenesis featuring vigorous fibroblast replication and extensive extracellular matrix deposition, which may lead to obliteration of the distal air space. Currently no effective treatment exists for lung fibrosis. Animal models are important to investigate pathological mechanisms and for preclinical evaluation of novel targets and therapies. Assessment tools used in experimental models are terminal, thereby providing only a snapshot of a complex and chronic biological process (2). Besides the ethical aspects of needing less animals in the experiments, noninvasive readouts may simplify considerably the assessment of compound efficacy in such chronic models. The present work was performed to evaluate magnetic resonance imaging (MRI) as a noninvasive tool to follow the course of bleomycin-induced lung injury in mice, and in particular to validate a method for pharmacological studies. Bleomycin is an antibiotic that is useful as chemotherapy in the treatment of cancer (3). Adoption of bleomycin as an antineoplastic drug is, however, limited because it produces a dose-dependent pneumonitis, which often progresses to interstitial pulmonary fibrosis (4,5). Local instillation of bleomycin in rodents is often used to model lung fibrosis (6,7). The general events that lead to fibrosis include the initial damage, development of alveolitis and inflammation, and, subsequently, excessive accumulation of collagen in the lung. The availability of a murine model of pulmonary fibrosis provides the opportunity to investigate, in wildtype animals or using transgenic technology, novel pharmacological approaches to prevent or treat this disease. In an initial phase of the study, proton MRI was used to assess noninvasively structural changes following intranasal (i.n.) instillation of bleomycin to naive C57BL/ 6 or Balb/C mice. Measurements were performed in spontaneously breathing animals, and neither cardiac nor respiratory gating was used. Histological analysis was performed at different timepoints after bleomycin to help the characterization of signals detected by MRI. The model was then used to investigate cancer Osaka thyroid (COT, also named Tpl2 or MAP3K8) kinase deficient mice and cadherin-11 knockout mice with the aim of providing potential therapeutic targets to intervene in the disease development. The rationale for the analysis of COT kinase deficient mice is the fact that COT regulates the production of tumor necrosis factor-a (TNF-a) and interleukin-1b (IL-1b) (8–10), which have been shown to be implicated in fibrosis (11,12) and wound healing (13). Cadherin-11 knockout mice were investigated in the bleomycin model because it has been shown

499

500

Babin et al.

Table 1 Experimental Groups Analyzed Dosing scheme

Experiment

Animals

Number of mice

Repeated bleomycin (6 0.25 mg/kg i.n.) or saline Repeated bleomycin (6 0.25 mg/kg i.n.) Repeated bleomycin (6 0.25 mg/kg i.n.) Repeated bleomycin (6 0.25 mg/kg i.n.)

Time course

Male C57BL/6

Male vs. female response Male vs. female response COT kinase knockout

C57BL/6

n ¼ 10 bleomycin, n ¼ 8 vehicle n ¼ 6 male, n ¼ 6 female

Balb/C

n ¼ 8 male, n ¼ 8 female

Wildtype, HE, and HO knockout

Repeated bleomycin (6 0.25 mg/kg i.n.)

Cadherin-11 knockout

Males: n ¼ 10 wildtype, n ¼ 5 HE, n ¼ 5 HO; Females: n ¼ 6 wildtype, n ¼ 6 HE Males: n ¼ 8 wildtype, n ¼ 3 HO; Females: n ¼ 9 wildtype, n ¼ 5 HO

Wildtype and HO knockout

HO, homozygous; HE, heterozygous knockout mice.

that cadherin-11 is overexpressed in human renal tubular cells during epithelial to mesenchymal transition (14), a process that can adversely cause organ fibrosis (15), and because epithelial to mesenchymal transition has been reported to occur in bleomycin-induced pulmonary fibrosis (16). MATERIALS AND METHODS Experiments were approved by the Veterinary Authority of the City of Basel (license number 1989). Animals Seven- to nine-week-old C57BL/6 and Balb/C (male and female) mice (Charles River, L’Arbresle, France; 20 g body weight) were used. COT-kinase deficient mice were bred at Charles River Europe (Kisslegg, Germany). A partial deletion of exons 2 and 3 of COT kinase was introduced in a 129SV mouse background. This mutation deletes the ATP binding pocket. Animals were backcrossed on a C57BL/6 background over 11 generations (17). Mice were 9 weeks old at the time of the study. To generate cadherin-11 knockout mice in house, a targeting vector was constructed as follows (18): the portion of the cadherin-11 gene encoding the last 56 amino acids of the extracellular domain through most of the transmembrane domain was replaced with the PGKNeopA

gene cassette. The targeting construct was used to transfect embryonic stem cells, and one of the targeted clones produced germ-line chimeras with high efficiency. Animals heterozygous for the targeted cad11 allele were established by mating germline-transmitting chimeras with C57BL/6 mice. They were 10 weeks old at the time of the study. Experimental groups analyzed are summarized in Table 1. Bleomycin Administration Mice were lightly anesthetized with 1.5% isoflurane (Abbott, Cham, Switzerland) in a chamber and bleomycin hydrochloride (Euro Nippon Kayaku, Frankfurt am Main, Germany) in 25 mL of saline or vehicle (25 mL of saline) was administered intranasally (i.n.) via a micropipette (12.5 mL per nostril). This procedure was performed six times, once daily. An interval of 2 days existed between the 3rd and 4th administrations. Magnetic Resonance Imaging Performed at 4.7 T with a Biospec 47/40 system (Bruker Medical Systems, Ettlingen, Germany). During image acquisition, anaesthesia was maintained with 1.3% isoflurane, in a mixture of O2/N2O (1:2), administered via a nose cone. Measurements were performed on spontaneously breathing animals; neither cardiac nor respiratory

FIG. 1. Segmentation of MR images. Image from a C57BL/6 mouse, acquired at day 21 after last bleomycin administration (left). Segmentation of bleomycininduced signals within the border drawn in the lung (right). The corresponding volume of segmented signals was of 19.1 mL.

MRI Analysis of Bleomycin-Induced Lung Injury in Mice

501

FIG. 2. MRI of repeated bleomycin dosing in a male 8-week-old C57BL/6 mouse. a: Eight axial gradient-echo images extracted from datasets acquired in two sessions, namely at baseline and at day 7 after last administration of bleomycin. Patchy, widespread signals elicited by the antibiotic in the lungs are indicated by arrows. b: Transverse images from the same animal, corresponding to approximately the anatomical location shown in (a), acquired at several timepoints with respect to the last i.n. instillation of bleomycin. Contrast has been changed to better highlight the patchy signals in the lung (arrows).

triggering was applied (19). A gradient-echo sequence with the following parameters was applied: repetition time 5.6 ms, echo time 3.5 ms, flip angle of the excitation pulse 10 , matrix size 256 128, slice thickness 0.75 mm and field of view 3 3 cm2. A single slice image (acquisition time of 74 s) was obtained by computing the two-dimensional Fourier transform of the averaged signal from 60 individual acquisitions and interpolating the data set to 256 256 pixels. The entire lung was covered by 16 consecutive transverse slices. A birdcage resonator of 32 mm diameter was used for excitation and detection. MR Image Analysis For a given time point, the area of high intensity signals was quantified on each of the 16 images covering the whole lung, using a semiautomatic segmentation procedure implemented in the IDL (Interactive Data Language Research Systems, Boulder, CO) environment on a Linux

system. Images were first lowpass-filtered with a Gaussian profile filter and then transformed into a set of four gray level classes using adaptive Lloyd-Max histogram quantitation. The highest class in the transformed images was extracted interactively by a region grower, whose border was drawn manually to control the growing (Fig. 1). The total volume of high intensity signals was calculated by adding the areas obtained for each of the 16 images, and multiplying the summed value by the slice thickness. Differential signal volumes (parsed, baselinesubtracted data) reflecting the bleomycin effects are presented. Thus, for each mouse, the volume of high intensity signals present in baseline images was subtracted from the volumes evaluated at later time points. Histology Details are provided in (19). Animals were sacrificed R , Abbott, Baar, with an overdose of thiopental (PentothalV Switzerland; 250 mg/kg i.p., 0.2 mL) immediately after

502

Babin et al.

Statistical Analyses Raw (baseline-unsubtracted) MRI data were analyzed using analysis of variance with random effects (SYSTAT 12, Systat Software, San Jose, CA) to take into account the longitudinal structure of the data. For multiple comparisons a Bonferroni correction followed the ANOVA. Mann-Whitney analyses (SYSTAT 11) were performed on histological scores. RESULTS

FIG. 3. Quantification of responses detected by MRI in the lungs of 7–9-week-old male C57BL/6 mice. Volumes of signals (parsed, baseline-subtracted data; means 6 sem) evaluated on the MR images following repeated dosing of vehicle (saline) or bleomycin. Statistical analysis performed on raw, baseline-unsubtracted data. Levels of significance *** P < 0.001 correspond to comparisons of raw signal volumes with respect to baseline values before dosing, while # 0.01 < P < 0.05, ## 0.001 < P < 0.01, ### P < 0.001 refer to the indicated comparisons. The numbers of mice at the beginning of the experiment are furnished in parentheses.

the MRI acquisitions. The trachea was immediately ligated to avoid collapse and the intact lungs were removed and immersed in 10% neutral buffered formalin for 24 h. Slices (3 mm) including the main bronchi from each lobe were stained with (i) hematoxylin and eosin (H&E) to assess the general morphology, (ii) Masson trichrome for the demonstration of fibrosis, and (iii) Picrosirius red for the identification of collagen fibers and newly synthesized collagen.

Histological Analysis Sections were analyzed by an observer blinded to the treatment and the strain of the animals. The extent and the severity of parenchyma infiltration with inflammatory cells and fibrosis were analyzed using a scoring system adapted from the literature (20). Scores were assigned to edema (perivascular, peribronchial, alveolar), infiltration of inflammatory cells (perivascular, peribronchial, parenchymal) and fibrosis, following the classification: none (score 0), slight (score 1), moderate (score 2), marked (score 3), and severe (score 4). The extent of the lesions was classified based on their percentage distribution as focal (51%). Abnormal histological features were recorded according to their extent and severity. Collagen was quantified using ‘‘Histolab’’ (Microvision Instruments, Evry, France). Picrosirius-stained slides were examined with a light microscope (Eclipse E600, Nikon, Egg, Switzerland) connected to a 3CCD video color camera (DXC970MD, Sony, Tokyo, Japan). The whole surface of each lung section was captured at 10 magnification. The color corresponding to picrosirius was extracted by threshold setting and the area corresponding to picrosirius staining calculated.

In a first step, the course of bleomycin-induced lung injury in C57BL/6 or Balb/C mice was followed by MRI and the corresponding signals detected noninvasively in vivo were characterized using histology. MRI was performed at baseline and at several time points following repeated bleomycin administration. To highlight the responses induced by bleomycin, differential volumes of MRI signals, from which the respective baseline values for the individual animals have been subtracted, are presented. However, statistical analyses of MRI signals have been performed on the raw, baseline-unsubtracted values. Repeated Bleomycin Administration to C57BL/6 or Balb/C Mice Following repetitive administration of a low dose of bleomycin (6 0.25 mg/kg) to male C57BL/6 mice, MRI revealed the presence of patchy, diffuse signals throughout the lung, up to day 70 (Fig. 2). No animal died during the experimental period, and a maximum weight loss of 10% in the first two weeks after the last bleomycin administration was followed by weight recovery. The course of differential MRI signal volumes (means 6 sem; parsed, baseline-subtracted data) is presented in Fig. 3. No effect was seen for vehicle (saline) treatment, while in contrast bleomycin induced a sustained signal. Histology revealed (i) slight to moderate infiltration of inflammatory cells and alveolar edema, and moderate to marked multifocal fibrosis at day 7; (ii) slight multifocal infiltration of inflammatory cells and moderate multifocal fibrosis at day 14; (iii) almost no infiltration of inflammatory cells but a marked to severe multifocal fibrosis at days 21, 35, and 70 after bleomycin treatment (Table 2). Figure 4 presents images from three different animals, acquired at different time points after bleomycin. For each image acquired in vivo, the corresponding picrosirius-stained histological section at approximately the same anatomical location is presented. Fibrotic regions indicated by arrows in the histological sections corresponded to areas of significant signal in the in vivo images. Picrosirius-stained slides reflecting collagen content were examined with a light microscope connected to a 3CCD video color camera. The color corresponding to picrosirius was extracted by threshold setting and the area corresponding to picrosirius staining calculated (Fig. 5a). The mean collagen content assessed in picrosirius-stained slices at different times after repeated bleomycin is summarized in Fig. 5b. Two histological slices were analyzed for each animal.


503

Table 2 Summary of Histological Scores (Means 6 sem) of Inflammation on Samples from C57BL/6 Mice Subjected to Repeated Bleomycin (6 0.25 mg/kg i.n.) or Saline Dosing Edema Treatment Saline (n ¼ 2) Day 14 Saline (n ¼ 4) Day 70 Bleomycin (n ¼ 5) Day Bleomycin (n ¼ 3) Day Bleomycin (n ¼ 3) Day Bleomycin (n ¼ 4) Day Bleomycin (n ¼ 8) Day

Cellular infiltration

Perivascular

Peribronchial

Alveolar

Perivascular

Peribronchial

Parenchymal

0 0 0 0 0 0 0

0 0 0 0 0 0 0

0 0 1.6 6 0.5 (MF) 0.3 6 0.3 (F) 0.3 6 0.3 (MF) 0 0

0 0 1.4 6 0.4 0.3 6 0.2 0 0 0

0 0 1.2 6 0.4 0 0 0 0

0 0 1.8 6 0.7 (MF) 0.8 6 0.2 (MF) P ¼ 0.002 0.2 6 0.2 (MF) 0.9 6 0.1 (MF) 0.4 6 0.1 (MF)

7 14 21 35 70

Mean scores (6sem) for the number of animals assessed are provided. The levels of significance P refer to Mann-Whitney comparisons of scores for saline and bleomycin treatments, at days 14 and 70 after last dosing. F focal, MF multifocal. Original scores were none (score 0), slight (score 1), moderate (score 2), marked (score 3) and severe (score 4).

Table 3 shows the comparison between the volumes of responses determined in vivo by MRI and the collagen content in histological sections. For each animal, the mean value of picrosirius analyses on two histological sections was taken. A significant correlation was found when including data from all time points measured. According to the histological analysis summarized in Table 2, some alveolar edema was present at day 7 after the last bleomycin administration and could have contributed to the MRI signals detected at this time point. Therefore, we compared the MRI signals with the histology of collagen at later time points, for which no alveolar edema had been detected histologically. The correlations became significantly stronger when data corresponding to day 7 after the last bleomycin administration were excluded from the comparison, suggesting that the MRI signals at time points later than day 7 reflected more accurately the development of pulmonary fibrosis in the model.

FIG. 4. Histology of the lungs of male bleomycin-challenged 7–9week-old C57BL/6 mice. Picrosirius-stained histological sections corresponding to approximately the anatomical location illustrated by the MR images. A marked to severe infiltration with inflammatory cells was found at day 7. Fibrotic foci (arrows) were noted within inflammatory areas. Slight multifocal infiltration of inflammatory cells and multifocal fibrosis (arrows) was observed at day 28. Almost no infiltration with inflammatory cells but a multifocal fibrosis (arrows) was seen on day 70.

In another experiment, we compared the responses induced by repeated bleomycin in the lungs of male and female C57BL/6 mice. The course of MRI signal volumes (means 6 sem), for which the volumes of baseline signals have been parsed (subtracted) at each timepoint, is presented in Fig. 6. The response in female C57BL/6 mice as detected by MRI was less pronounced and resolved more quickly than in male animals. Multiple bleomycin instillation also led to a smaller response in the lungs of male Balb/C compared to male C57BL/6 mice as revealed by MRI (Fig. 7). The signals evaluated in male Balb/C animals were above baseline at all three time points measured. In accordance with the MRI results, histology at day 21 after last bleomycin administration on male Balb/C mice showed slight multifocal fibrosis (mean collagen area 0.28 6 0.04 mm2), slight perivascular and peribronchial infiltration of inflammatory cells (mean score 1), and slight to moderate multifocal parenchymal infiltration of inflammatory

504

Babin et al.

FIG. 5. Quantification of picrosirius-stained histological slices of the lungs of male bleomycinchallenged 7–9-week-old C57 BL/6 mice. a: Picrosirius-stained slices at day 70 after last administration of saline or bleomycin. Areas corresponding to collagen accumulation are automatically recognized (green zones, right panels). In these examples, the total collagen areas were of 0.11 and 0.51 mm2, respectively for the saline and the bleomycintreated mouse. b: Quantification of collagen content in picrosirius-stained slices. Values are given as means (6sem) for 3–6 animals per time point. Levels of significance *** P < 0.001, ** P ¼ 007, and * 0.01 < P < 0.05 correspond to Anova comparisons with respect to the control group (saline challenge). Control mice were analyzed at days 14 and 70 after saline.

cells (mean score 1.5 6 0.5). In contrast, at all three time points measured after bleomycin, the responses in female Balb/C mice were not significantly different from baseline and, albeit tendentially smaller than the values obtained for male Balb/C animals, no significance was attained at any time point. This was consistent with the smaller response observed for female as compared to male C57BL/6 mice reported earlier. Following this initial verification, MRI has been applied to study the effects of repeated bleomycin dosing (6 0.25 mg/kg i.n.) in COT kinase and cadherin-11 knockout mice to investigate the role these targets may play in pulmonary fibrosis. COT Kinase Knockout Mice Figure 8a summarizes the MRI signals obtained in heterozygous (HE) and homozygous (HO) COT kinase knockout mice and in wildtype animals following bleomycin administration. Since we had observed before that the effects of bleomycin are both mouse strain- and genderspecific, we analyzed the response of the COT KO mice

(backcrossed to C57BL/6 background) separately for male and female mice as well as for the influence of gene dosage (homozygous versus heterozygous). The signals detected in male mice were attenuated in COT HO compared to wildtype mice. Statistical analyses of unparsed data showed a significant increase of signal volumes with respect to baseline values until day 21 after bleomycin for male HO COT kinase knockout animals, whilst signals were significantly increased throughout the experiment (until day 42) in male wildtype and male HE COT kinase deficient mice (Fig. 8a). In female animals, the bleomycin-induced response was smaller and of shorter duration in HE COT kinase compared to wildtype mice (Fig. 8a). These data suggest protection against bleomycin treatment for male HO and female HE knockout mice, as further evidenced by comparisons of the areas under the curves (AUCs) (Fig. 8a). Histology at day 42 after bleomycin demonstrated (i) slight parenchymal infiltration with inflammatory cells (score 1) as well as moderate multifocal fibrosis (score 2) for the HE male mice; (ii) minimal perivascular and peribronchic infiltration, minimal parenchymal infiltration with


505

FIG. 6. Repeated bleomycin dosing: male vs. female 7–9-weekold C57BL/6 mice. Volumes of signals (parsed, baseline-subtracted data; means 6 sem) evaluated on MR images acquired after repeated dosing of bleomycin (6 0.25 mg/kg). Statistical analysis performed on raw, baseline-unsubtracted data. Levels of significance ** P ¼ 0.09 and *** P < 0.001 correspond to comparisons of raw signal volumes with respect to baseline values before dosing, while # P ¼ 0.03, ## P ¼ 0.007, ### P < 0.001 refer to the indicated comparisons. The numbers of mice at the beginning of the experiment are indicated in parentheses.

FIG. 7. Repeated bleomycin dosing: male vs. female 7–9-weekold Balb/C mice. Volumes of signals (parsed, baseline-subtracted data; means 6 sem) evaluated on MR images following repeated dosing of bleomycin (6 0.25 mg/kg). Levels of significance *0.01 < P < 0.05 and **P ¼ 0.001 correspond to comparisons of raw signal volumes with respect to baseline values before dosing. The numbers of mice at the beginning of the experiment are indicated in parentheses.

inflammatory cells, as well as minimal fibrosis for male HO mice; and (iii) slight fibrosis (score 1) for the female HE mice. The collagen areas in histological samples from male HO and female HE COT kinase knockout mice challenged with bleomycin (Fig. 8b) were comparable to those detected in naive mice treated with saline (Fig. 5b). Taken together, it was shown that besides the dose of bleomycin, gender-specific effects and gene dosage effects can affect the course of the fibrosis.

no effective treatment exists. Murine models of lung fibrosis are important to investigate pathological mechanisms and for the preclinical evaluation of novel therapies. The present study was performed with the aim of validating MRI as a noninvasive tool to follow the course of lung injury in bleomycin-induced fibrosis in mice, as a model for future pharmacological investigations. Measurements were performed on spontaneously breathing animals, and neither cardiac nor respiratory gating was applied. As no intubation or mechanical ventilation was necessary, this procedure allowed repetitive measurements in the same animal and avoided potential lung damage caused by mechanical ventilation (22). A lasting response was obtained following multiple bleomycin administration, translated by MRI signals being detectable in male C57BL/6 mice up to day 70, when measurements were interrupted. Histology revealed early fibrotic foci within inflammatory areas at day 7 after repeated bleomycin. This observation agrees with the literature suggesting that the major fibro-proliferative phase induced by bleomycin in mice occurs within the first week following its administration and that this process coexists with inflammation (23). In our study, histology also showed that from day 14 onwards fibrosis was the predominant component of the response elicited in the lungs of male C57BL/6 mice by repeated bleomycin. Overall, the data obtained here demonstrate that repeated bleomycin administration at a low dose led to consistent and sustained fibrosis formation with moderate initial inflammation, a result that is in agreement with a recent publication showing that repetitive intratracheal bleomycin in mice models elicits several features of idiopathic pulmonary fibrosis (24). The fact that the MRI signals slowly decreased over time is consistent with the fact that bleomycin may induce fibrosis that

Cadherin-11 Knockout Mice Repeated bleomycin dosing (6 0.25 mg/kg) elicited a comparable response in wildtype and in homozygous cadherin-11 deficient mice as detected by MRI (Fig. 9). In accordance with previous results, the signal volumes were smaller in female than in male animals. These observations suggest that the cadherin-11 knockout mice were not protected against bleomycin-induced lung injury. DISCUSSION Although considerable progress has been made in understanding the pathophysiology of pulmonary fibrosis (21), Table 3 Comparisons Between Volumes of MRI Signals and the Collagen Content Assessed by Histology Time points Days Days Days Days

7–70 14–70 21–70 28–70

Correlation coefficient

P

Number of mice

0.5 0.78 0.75 0.8

0.0025