Proapoptotic Bid is required for pulmonary fibrosisG. R. Scott Budinger*†‡, Gokhan M. Mutlu*‡, James Eisenbart*, Alyson C. Fuller§, Amy A. Bellmeyer*,Christina M. Baker*, Mindy Wilson*, Karen Ridge*, Terrence A. Barrett*, Vivian Y. Lee*, and Navdeep S. Chandel*†¶

Departments of *Medicine, §Microbiology-Immunology, and †Cell and Molecular Biology, Northwestern University, Chicago, IL 60611

Edited by Philippa Marrack, National Jewish Medical and Research Center, Denver, CO, and approved January 27, 2006 (received for review August 31, 2005)

The molecular mechanisms of pulmonary fibrosis are poorly un-derstood. Previous reports indicate that activation of TGF-�1 isessential for the development of pulmonary fibrosis. Here, wereport that the proapoptotic Bcl-2 family member Bid is requiredfor the development of pulmonary fibrosis after the intratrachealinstillation of bleomycin. Mice lacking Bid exhibited significantlyless pulmonary fibrosis in response to bleomycin compared withWT mice. The attenuation in pulmonary fibrosis was observeddespite similar levels of inflammation, lung injury, and activeTGF-�1 in bronchoalveolar lavage fluid 5 days after the adminis-tration of bleomycin in mice lacking Bid and in WT controls.Bleomycin induced similar levels cell death in vitro in alveolarepithelial cells isolated from WT and bid�/� mice. By contrast,alveolar epithelial cells from bid�/� mice were resistant to TGF-�1-induced cell death. These results indicate that Bcl-2 familymembers are critical regulators for the development of pulmonaryfibrosis downstream of TGF-�1 activation.

apoptosis � Bcl-2 � TGF-�

Pulmonary fibrosis is characterized by foci of proliferatingfibroblasts that develop after lung injury (1). The outcome of

fibroblast proliferation is excessive extracellular matrix deposi-tion in the lung parenchyma. The resulting disruption of lungarchitecture impairs alveolar ventilation and gas exchange. Theunderlying initiating mechanisms that result in fibroblast prolif-eration are not fully understood. One hypothesized mechanismis that the activation of cell death pathways in the alveolarepithelium in response to injury results in activation of fibrogenicmolecules such as TGF-�1 (2–4). The intratracheal administra-tion of bleomycin in rodents results in acute lung injury, followedby fibroblast proliferation, collagen deposition, and distortion oflung architecture (5). In this model, investigators have shownthat the administration of caspase inhibitors, which preventalveolar epithelial cell death, attenuates the development ofpulmonary fibrosis (6). The intracellular pathways by which celldeath occurs and the stimuli responsible for the activation ofthese pathways, however, have not been elucidated.

Caspase activation after exposure to an apoptotic stimulususually occurs as a result of activation of the mitochondrial orreceptor-dependent cell death pathway (7). The mitochondrial-dependent cell death pathway is induced when the subset of theBcl-2 family of proteins that contain only a single BH domain(BH3 domain only) activates the proapoptotic Bcl-2 familymembers Bax or Bak (8, 9). These Bcl-2 proteins oligomerize andallow the release of cytochrome c and other proapoptoticproteins in from the mitochondrial-intermembrane space intothe cytosol. In the cytosol, cytochrome c induces the oligomer-ization of APAF-1, which activates caspase-9, resulting in theactivation of executioner caspase-3 or -7 (10–12). Activation ofthe receptor-dependent death pathway is induced by specificligands that engage death receptors, resulting in activation ofcaspase-8 or caspase-10 (13). In some cells, activation ofcaspase-8 is sufficient to directly activate executioner caspase-3or -7 whereas, in other cells, activation of caspase-8 results incleavage of the BH3 protein Bid (14). Truncated Bid activatesBax and�or Bak, resulting in cytochrome c release and activationof caspase-3 or -7 (15, 16).

In the present study, we observed that mice lacking Bid wereprotected against bleomycin-induced pulmonary fibrosis, sug-gesting that the mitochondrial-dependent cell death pathway isrequired for the development of pulmonary fibrosis in thismodel. In examining the mechanism of this protection, we foundthat mice lacking Bid had similar levels of lung injury, inflam-mation, and active TGF-�1 5 days after exposure to bleomycin,suggesting that the lack of Bid protected against TGF-�1-induced rather than bleomycin-induced cell death. In support ofthis hypothesis, alveolar epithelial cells from mice lacking Bidwere protected against TGF-�1, but not bleomycin-induced celldeath.

ResultsBid-Null Mice Are Protected from Bleomycin-Induced Pulmonary Fi-brosis. Previous studies have suggested that preventing epithelialcell death can prevent pulmonary fibrosis induced by the intra-tracheal instillation of bleomycin (17–19). However, none of theprevious strategies of preventing cell death inhibited specificintracellular death pathways. To address the role of the mito-chondrial-dependent cell death pathway in the regulation ofpulmonary fibrosis, bleomycin was instilled intratracheally intoWT and bid�/� mice. By 21 days, WT mice exposed to bleomycindisplayed profound fibrosis as assessed by Masson–Trichromestaining (Fig. 1 A and B). By contrast, bid�/� mice exposed tobleomycin did not exhibit fibrosis at day 21. Total lung collagen,as assessed by the Sircoll assay, was also significantly higher inWT mice than in bid�/� mice (Fig. 1C). Collectively, these resultsindicate that loss of Bid prevents bleomycin-induced fibrosis.

Bid Is Not Required for Bleomycin-Induced Lung Injury. The fibroticresponse due to intratracheal instillation of bleomycin is a lateresponse. The initial response to bleomycin is lung injury, whichis characterized by flooding of the alveolar space with protein-rich exudates and inflammation (20). To determine whether Bidis required for the early inflammatory response to bleomycin,histological analysis, bronchoalveolar lavage (BAL) fluid cellu-larity, and protein levels were assessed in WT and bid�/� mice5 days after exposure to bleomycin. WT and bid�/� mice bothdisplayed histological evidence of inflammation at day 5 (Fig.2A). WT and bid�/� mice demonstrated a significant increase ininflammatory cells in the BAL fluid (Fig. 2 B and C). Protein inthe BAL fluid was elevated to a similar degree in the twogenotypes (Fig. 2D). To further assess lung injury, we measuredthe increases in lung permeability to albumin 5 days aftertreatment with intratracheal bleomycin in WT mice and Bid-nullmice. WT and Bid-null mice displayed similar increased levels oflung permeability to albumin (Fig. 2E). Collectively, these

Conflict of interest statement: No conflicts declared.

This paper was submitted directly (Track II) to the PNAS office.

Freely available online through the PNAS open access option.

Abbreviations: BAL, bronchoalveolar lavage; ARDS, acute respiratory distress syndrome.

‡G.R.S.B. and G.M.M. contributed equally to this work.

¶To whom correspondence should be addressed. E-mail: nav@northwestern.edu.

© 2006 by The National Academy of Sciences of the USA

4604–4609 � PNAS � March 21, 2006 � vol. 103 � no. 12 www.pnas.org�cgi�doi�10.1073�pnas.0507604103

Dow

nloa

ded

by g

uest

on

May

17,

202

1

results indicate that loss of Bid does not alter the initial lunginjury due to bleomycin.

Active TGF-�1 Levels in BAL Fluid Are Not Affected by Bid. Thefibrotic response to bleomycin has been shown to be mediated by

the activation of TGF-�1 (21–25). To determine whether loss ofBid regulates the activation of TGF-�1, BAL fluid from WT andbid�/� mice was assessed for active TGF-�1 levels 5 days after theinstillation of bleomycin or PBS. The level of active TGF-�1 inBAL fluid of PBS treated mice in both genotypes was low (Fig.3). Mice treated with bleomycin displayed comparable levels ofactivate TGF-�1 in both genotypes (Fig. 3). Thus, the lack offibrosis in Bid-null mice is not due to diminished levels of activeTGF-�1.

TGF-�1-Induced Cell Death Requires Bid. We have previously dem-onstrated that overexpression of Bcl-XL prevents bleomycin-induced cell death in rat alveolar type II cells, indicating theactivation of the mitochondrial-dependent death pathway inresponse to bleomycin (26). To decipher whether the BH3protein Bid is the upstream regulator of mitochondrial deathpathway in response to bleomycin, mouse alveolar epithelial cellsfrom WT and bid�/� mice were exposed to bleomycin (10 or 100milliunits�ml) for 48 h. Alveolar epithelial cells from WT andbid�/� mice displayed similar levels of cell death in response tobleomycin (Fig. 4A). Because Bid is not required for bleomycin-induced alveolar epithelial cell death, we decided to examine therole of Bid in TGF-�1-induced alveolar epithelial cell death.TGF-�1 has been shown to induce cell death in multipleepithelial cells, including alveolar epithelial cells (27). TGF-�1induced cell death in alveolar epithelial cells isolated from WTmice. However, TGF-�1-induced cell death was not observed inalveolar epithelial cells from bid�/� mice (Fig. 4A). To test in vivothat loss of Bid does not prevent bleomycin-induced apoptosis,we measured TUNEL-positive cells at day 5. WT and Bid-nullmice both displayed similar levels of TUNEL-positive cells (Fig.4B). These results indicate that Bid is not required for the initialcell death induced by bleomycin but is required for the laterTGF-�1-induced cell death.

Loss of Bid Does Not Affect TGF-�1-Induced Fibroblast Proliferation.The lack of fibrosis in the Bid-null mice could also be due to adefect in fibroblast proliferation in response to TGF-�1. To testwhether Bid affects fibroblast proliferation, fibroblasts from WTand bid�/� mice were subjected to TGF-�1 under low-serumconditions. TGF-�1 induced fibroblasts from WT and bid�/�

mice to proliferate at similar rates, indicating that the Bid-nullmice do not have a defect in fibroblast proliferation (Fig. 5A).We further examined whether Bid-null fibroblasts have animpaired response to death stimuli. Bid-null fibroblasts were notresistant to the classical intrinsic apoptotic stimulus staurospor-ine (Fig. 5B). As expected, the Bid-null fibroblasts were fairlyresistant to the classical extrinsic apoptotic stimulus TNF� pluscycloheximide (28). Collectively, these results indicate that lackof fibrosis observed in Bid-null mice is not due to dysregulationof fibroblast survival or proliferation.

DiscussionRecent studies have implicated apoptosis of alveolar epithelialcells as a potential initiating mechanism in the development ofpulmonary injury and fibrosis. Pharmacologic inhibition of theapoptotic pathways with caspase inhibitors has been shown toprevent bleomycin-induced fibrosis (6, 18). Activation ofcaspases by the administration of FasL, which activates thereceptor-dependent apoptotic pathway, can induce pulmonaryfibrosis (17). Furthermore, a variety of strategies that have beenshown to prevent bleomycin-induced pulmonary fibrosis, forexample the overexpression of heme oxygenase-1 (HO-1) or lossof angiotensin receptor AT1a gene, might exert their protectiveeffects by preventing alveolar epithelial cell death (29, 30). Theintracellular pathway(s) that result in the development of alve-olar epithelial cell death and pulmonary fibrosis have not beenelucidated. In the present study, we demonstrate that mice

Fig. 1. bid�/� mice are protected against bleomycin-induced pulmonaryfibrosis. Shown is histology of mouse lungs 21 days after the intratrachealinstillation of bleomycin (0.045 units) or PBS. Shown are hematoxylin�eosinstain (A) and Masson’s Trichrome stain (B) for collagen. Representative imagesare shown at �200 magnification. (C) At 21 days after the intratrachealinstillation of bleomycin (0.045 units) or PBS, collagen content in lungs fromWT or bid�/� mice was assessed by a Sircoll assay. Data are expressed as means(�SEM) of four different animals. *, P � 0.05 relative to PBS-treated mice.

Budinger et al. PNAS � March 21, 2006 � vol. 103 � no. 12 � 4605

MED

ICA

LSC

IEN

CES

Dow

nloa

ded

by g

uest

on

May

17,

202

1

lacking the proapoptotic protein Bid are protected from bleo-mycin-induced pulmonary fibrosis. Bid is a BH3 protein that canactivate the proapoptotic Bcl-2 family members Bax or Bak toinitiate the release of cytochrome c into the cytosol to triggercaspase activation (9, 15, 16). Our findings demonstrate arequirement for the mitochondrial-dependent cell death path-way in the development of pulmonary fibrosis.

Protection of bid�/� mice from pulmonary fibrosis was notexplained by a decrease in the pulmonary inflammatory re-sponse to bleomycin. In fact, Bid-null mice displayed moreinflammatory cells, particularly neutrophils, in the BAL fluid 5days after the administration of bleomycin. Similarly, BAL levelsof protein, lung permeability, and histologic evidence of lung

injury were slightly worse in bid�/� animals. The higher levels ofinflammation may result from impaired apoptosis of inflamma-tory cells in bid�/� mice. It is formally possible that the elevatedlevels of inflammation in Bid-null mice might be protectiveagainst the development of fibrosis. Future studies using bonemarrow chimeras will have to address whether the increase ininflammation observed in the bid-null mice is protective againstthe development of pulmonary fibrosis. However, our results areconsistent with previous published reports indicating that theinflammatory response can be dissociated from the fibroticresponse. For example, mice lacking the integrin �v�6 display anenhanced inflammatory response to bleomycin compared withWT controls but failed to develop pulmonary fibrosis after the

Fig. 2. Bid is not required for bleomycin-induced lung injury. (A) WT and bid–/– mice both display histologic evidence of inflammation 5 days after theintratracheal instillation of bleomycin (0.045 units). Representative images are shown at �200 magnification (hematoxylin�eosin stain). Shown are cell countswith differential (B and C), protein in BAL fluid (D), and lung permeability to FITC-labeled albumin (E) from WT and bid–/– mice at 5 days after the intratrachealinstillation of bleomycin (0.045 units) or PBS. Data are expressed as means (�SEM) from four different animals. *, P � 0.05 relative to PBS-treated mice.

4606 � www.pnas.org�cgi�doi�10.1073�pnas.0507604103 Budinger et al.

Dow

nloa

ded

by g

uest

on

May

17,

202

1

administration of bleomycin (21). These results suggest that lunginflammation is a marker for the acute lung injury rather thanthe cause of the ensuing fibrosis. Taken together, these findingsindicate that Bid is required for the late fibrosis but not the earlylung injury after the intratracheal instillation of bleomycin.

Bleomycin induced lung injury, and fibrosis is mediated by theactivation of TGF-�1. TGF-�1 in the lung is normally secreted as

an inactive complex that is bound to a latency-associated peptide(LAP), a protein that is formed by cleavage of the amino terminusof the TGF-� gene product. LAP prevents binding of matureTGF-�1 with its receptors. Binding of LAP to �v�6 integrin altersthe conformation of the complex to activate TGF-�1 (21). Accord-ingly, mice deficient in the �6 integrin, which fail to activateTGF-�1, were completely protected from bleomycin-induced lungfibrosis. The overexpression of active TGF-�1 in mice leads toprofound fibrosis, and inhibition of TGF-�1 by antibodies abrogatesbleomycin-induced fibrosis (22–24). Furthermore, blocking intra-cellular signaling pathways activated by TGF-�1, for exampleactivin receptor-like kinase 5 (ALK5) or Smad3, prevents activatedTGF-�1-induced fibrosis (25, 31). The fact that Bid-null miceexhibit similar level of active TGF-�1 compared with WT micesuggests that loss of Bid prevents TGF-�1 from inducing down-stream affects. The initial bleomycin-induced apoptosis in epithelialcells might result in the activation of TGF-�1 that is independentof Bid. Indeed, the degree of early apoptosis after bleomycinadministration was similar in WT and bid�/� animals. There yetmight be another BH3 protein that is required for the initialapoptosis elicited by intratracheal instillation of bleomycin thatresults in activation of TGF-�1.

TGF-�1 is classically a mitogen for lung fibroblasts but hasrecently been shown to induce mouse alveolar epithelial celldeath through the early growth response gene Egr-1 and caspaseactivation (27). Our present data indicate that loss of Bid doesnot alter fibroblast proliferation in response to TGF-�1. How-ever, our data do demonstrate that Bid is required for TGF-�1-induced cell death in alveolar epithelial cells. Based on theseobservations, we propose a model where the presence of TGF-�1

Fig. 3. Active TGF-�1 levels in BAL fluid of WT and bid–/– mice 5 days after theintratracheal instillation of bleomycin (0.045 units) or PBS. Data are expressedas means (�SEM) from four different animals. *, P � 0.05 relative to PBS-treated mice.

Fig. 4. TGF-�1-induced alveolar epithelial cell death requires Bid. (A) Alve-olar epithelial cell isolated from WT and bid–/– mice were treated with bleo-mycin (0, 10, or 100 milliunits�ml) or TGF-�1 (10 ng�ml). Data are expressed asmeans (�SEM) from experiments done on three separate isolations. *, P � 0.05relative to untreated cells isolated from WT mice. (B) TUNEL staining wasperformed on WT and bid�/� mice 5 days after the intratracheal instillation ofbleomycin (0.045 units) or PBS. Data are expressed as means (�SEM) fromthree different animals. *, P � 0.05 relative to PBS-treated mice.

Fig. 5. Loss of Bid does not impair fibroblast proliferation. Fibroblasts fromWT and bid�/� mice were subjected to TGF-�1 (10 ng�ml) in the presence of 1%FBS (A) or were subjected to staurosporine (1 �M) or TNF� (1 ng�ml) pluscycloheximide (5 �g�ml) (B). Data are expressed as means (�SEM) fromexperiments done on four separate isolations.

Budinger et al. PNAS � March 21, 2006 � vol. 103 � no. 12 � 4607

MED

ICA

LSC

IEN

CES

Dow

nloa

ded

by g

uest

on

May

17,

202

1

sustains a low level of epithelial apoptosis to allow a permissiveenvironment for fibroblast invasion and proliferation (32). Themechanism underlying TGF-�1 activation of Bid to triggermitochondrial-dependent apoptosis in alveolar epithelial cellsremains unknown. One possibility would be the engagement ofFas by the type II TGF-�1 receptor. However, in the C57BL�6strain of mice, the Fas-dependent pathway is dispensable forbleomycin-induced fibrosis (30, 33).

In summary, our data provide two important insights into themechanisms underlying bleomycin-induced pulmonary fibrosis.First, we have identified that the mitochondrial-dependent celldeath pathway mediated by the proapoptotic Bcl-2 family mem-ber Bid is required for the development of bleomycin-inducedpulmonary fibrosis. Second, our results dissociate the earlyinflammation and lung injury in response to the administrationof bleomycin from the fibrotic response. Although the bleomy-cin-induced model of pulmonary fibrosis is used as an animalmodel of idiopathic pulmonary fibrosis, it also shares commonfeatures with the acute respiratory distress syndrome (ARDS).ARDS has an early phase that is characterized by inflammationand the accumulation of fluid and protein in the alveolar space(34). The latter phase of ARDS develops in a subset of patientsand is characterized by fibroproliferation. We have recentlyshown that active TGF-�1 is present in BAL fluid from ARDSpatients (35). The observation that Bid-null mice are protectedfrom fibrosis but not lung injury suggests that strategies to inhibitthe mitochondrial-dependent cell death pathway downstream ofthe activation of TGF-�1 might prevent fibrosis in patients withacute lung injury. This would suggest that there is a window ofopportunity to prevent ARDS-associated fibroproliferation.

Materials and MethodsMice and Bleomycin Administration. The bid�/� mice were back-crossed to C57BL�6 for 12 generations and were kindly providedby the late S. Korsmeyer (28). WT C57BL�6 mice were pur-chased from The Jackson Laboratory. Age- and sex-matched 8-to 10-week-old mice were used in all experiments. WT mice andbid�/� mice were treated with a single intratracheal injection of50-�l of PBS or bleomycin (0.045 units in 50 �l of PBS;Sigma-Aldrich). All experiments were approved by the North-western University Animal Care and Use Committee.

Histology. A 20-gauge angiocath was sutured into the trachea, thelungs and heart were removed en bloc, and the lungs wereinflated to 25 cm of H2O with PBS and then heated to 60°C inPBS. The lungs were then inflated with 0.8 cc of 4% parafor-maldehyde. The heart and lungs were fixed in paraffin, and 5-�msections were stained with hematoxylin�eosin or Masson’sTrichrome.

BAL Analysis. BAL was performed through a 20-gauge angiocathligated into the trachea. A 1.0-ml aliquot of PBS was instilled intothe lungs and then carefully removed three times. A 200-�laliquot of the BAL fluid was placed in a cytospin and centrifugedat 500 � g for 5 min. The glass slides were Wright stained andsubjected to a blinded manual cell count and differential. Theremaining BAL fluid was centrifuged at 130 � g for 5 min, andthe supernatant was used for the measurement of BAL protein(Bradford) and cytokine analysis.

Sircoll Assay. Total collagen content was determined by harvest-ing lungs from mice 21 days after treatment with bleomycin orPBS. Exsanguination was accomplished by a left nephrectomy,followed by perfusion of the right ventricle with 2 ml of cold PBS.Both lungs were removed and homogenized (1 min with a tissuehomogenizer followed by 12 strokes in a Dounce homgenizer) in1 ml of PBS. Aliquots of lung homogenate were then assayed fortotal lung collagen levels and compared with a standard curve

prepared from rat tail collagen by using the Sircoll collagen dyebinding assay (BioColor Ltd., Newtownabbey, U.K.) accordingto the manufacturer’s directions.

Active TGF-�1 Analysis. BAL fluid (above) from animals exposedto either PBS or bleomycin was analyzed for active TGF-�1levels by using the Emax ImmunoAssay System according to themanufacturer’s directions (Promega). TGF-�1 levels were mea-sured from fresh isolated BAL fluid.

Isolation and Culture of Alveolar Epithelial Cells. Alveolar type IIcells were isolated from WT and bid�/� mice. Briefly, the lungswere perfused via the pulmonary artery, lavaged, and digestedwith elastase (1 mg�ml, Worthington). Those cells were purifiedby negative immunoselection by using magnetic beads, followedby differential adherence to CD90 pretreated dishes. Cell via-bility was assessed by trypan blue exclusion (�95%). Cells weresuspended in DMEM containing 10% FBS with 2 mM L-glutamine, 100 units�ml penicillin, and 100 �g�ml streptomycin.Cells were cultured at an air–liquid interface of 0.4-�m transwellmembranes inserted into six-well culture dishes to eliminatefibroblast contamination. Cells were incubated in a humidifiedatmosphere of 5% CO2�95% air at 37°C and used 4 days afterisolation.

Isolation and Culture of Mouse Lung Fibroblasts. Fibroblast cellswere isolated from male C57BL�6. Briefly, the lungs wereperfused via the pulmonary artery, lavaged, and digested withelastase (1 mg�ml; Worthington) and 0.25% trypsin in thepresence of 1 mM EDTA. Lung tissue was dissected from theairways, minced into 2-mm3 pieces, and placed in tissue cultureflasks for 15–20 min in a humidified incubator at 37°C under 5%CO2 atmosphere with a minimal volume of minimum essentialmedia (MEM) supplemented with 20% FBS, 50 units�ml pen-icillin, 50 �g�ml streptomycin, and 2 mM L-glutamine. Anappropriate volume of media was then added to the flasks, andcells were maintained for 7–10 days or until fibroblasts began tomigrate out from the tissue. Identification of fibroblasts wasbased on the presence of vimentin staining.

TUNEL Assay. End labeling of exposed 3�-OH ends of DNAfragments in paraffin-embedded tissue was done by using theTUNEL AP In Situ Cell Death Detection Kit (Roche Diagnos-tics) according to the manufacturer’s directions. After staining,20 fields of alveoli (�400) were randomly chosen, and TUNEL-positive nuclei were counted.

Assessment of Lung Permeability. Five days after treatment withintratracheal bleomycin (0.045 units in 50 �l of PBS) or PBS, WTand bid�/� mice were anesthetized, and a 20-gauge angiocath wassutured into the trachea. The mice were mechanically ventilated(Minivent; Harvard Apparatus) with a respiratory rate of 100and a tidal volume of 0.2 ml. A midline laparotomy wasperformed, and the inferior vena cava was identified into which0.15 ml of a 16 mg�ml solution of FITC-labeled albumin(Sigma–Aldrich) in 5% BSA was injected. After 10 min, abronchoalveolar lavage was performed. Relative lung perme-ability was estimated from the fluorescence in the BAL fluidmeasured by using a microplate reader (excitation � 488 nm,emission � 530 nm).

Statistical Analysis. Results are expressed as mean � SEM. Datawere analyzed by using one-way analysis of variance (ANOVA).When ANOVA indicated that a significant difference waspresent, we explored individual difference with the Student t testusing Bonferroni correction for multiple comparisons. Statisticalsignificance was determined at 0.05 level. All groups werecompared with WT mice treated with PBS.

4608 � www.pnas.org�cgi�doi�10.1073�pnas.0507604103 Budinger et al.

Dow

nloa

ded

by g

uest

on

May

17,

202

1

We thank Dr. John Varga for critically evaluating the manuscript andDrs. Moises Selman and Annie Pardo for encouraging us to do thestudy. This work was supported in part by National Institutes of

Health Grants GM60472-07 and 1P01HL071643-01A10004 (toN.S.C.), 1P01HL071643-01A19001 (to K.R.), and 5K08HL067835-05(to G.R.S.B.).

1. Gross, T. J. & Hunninghake, G. W. (2001) N. Engl. J. Med. 345, 517–525.2. Selman, M., King, T. E. & Pardo, A. (2001) Ann. Intern. Med. 134, 136–151.3. Adamson, I. Y. R., Young, L. & Bowden, D. H. (1988) Am. J. Pathol. 130, 377–383.4. Gauldie, J., Kolb, M. & Sime, P. J. (2002) Respir. Res. 3, 1–3.5. Chua, F., Gauldie, J. & Laurent, G. J. (2005) Am. J. Respir. Cell Mol. Biol. 33, 9–13.6. Wang, R., Ibarra-Sunga, O., Verlinski, L., Pick, R. & Uhal, B. D. (2000) Am.

J. Physiol. 279, L143–L151.7. Danial, N. N. & Korsmeyer, S. J. (2004) Cell 116, 205–219.8. Green, D. R. & Kroemer, G. (2004) Science 305, 626–629.9. Wei, M. C., Zong, W. X., Cheng, E. H., Lindsten, T., Panoutsakopoulou, V.,

Ross, A. J., Roth, K. A., MacGregor, G. R., Thompson, C. B. & Korsmeyer,S. J. (2001) Science 292, 727–730.

10. Liu, X., Kim, C. N., Yang, J., Jemmerson, R. & Wang, X. (1996) Cell 86,147–157.

11. Zou, H., Henzel, W. J., Liu, X., Lutschg, A. & Wang, X. (1997) Cell 90, 405–413.12. Li, P., Nijhawan, D., Budihardjo, I., Srinivasula, S. M., Ahmad, M., Alnemri,

E. S. & Wang, X. (1997) Cell 91, 479–489.13. Nagata, S. (1997) Cell 88, 355–365.14. Peter, M. E. & Krammer, P. H. (1998) Curr. Opin. Immunol. 10, 545–551.15. Luo, X., Budihardjo, I., Zou, H., Slaughter, C. & Wang, X. (1998) Cell 94,

481–490.16. Li, H., Zhu, H., Xu, C. J. & Yuan, J. (1998) Cell 94, 49–501.17. Hagimoto, N., Kuwano, K., Miyazaki, H., Kunitake, R., Fujita, M., Kawasaki,

M., Kaneko, Y. & Hara, N. (1997) Am. J. Respir. Cell Mol. Biol. 17, 272–278.18. Kuwano, K., Kunitake, R., Maeyama, T., Hagimoto, N., Kawasaki, M.,

Matsuba, T., Yoshimi, M., Inoshima, I., Yoshida, K. & Hara, N. (2001) Am. J.Physiol. 280, L316–L325.

19. Kuwano, K., Hagimoto, N., Kawasaki, M., Yatomi, T., Nakamura, N., Nagata,S., Suda, T., Kunitake, R., Maeyama, T., Miyazaki, H., et al. (1999) J. Clin.Invest. 104, 13–19.

20. Pittet, J. F., Griffiths, M. J., Geiser, T., Kaminski, N., Dalton, S. L., Huang, X.,Brown, L. A., Gotwals, P. J., Koteliansky, V. E., Matthay, M. A., et al. (2001)J. Clin. Invest. 107, 1537–1544.

21. Munger, J. S., Huang, X., Kawakatsu, H., Griffiths, M. J., Dalton, S. L., Wu,J., Pittet, J. F., Kaminski, N., Garat, C., Matthay, M. A., et al. (1999) Cell 96,319–328.

22. Giri, S. N., Hyde, D. M. & Hollinger, M. A. (1993) Thorax 48, 959–966.23. Sime, P. J., Xing, Z., Graham, F. L., Csaky, K. G. & Gauldie, J. (1997) J. Clin.

Invest. 100, 768–776.24. Wang, Q., Wang, Y., Hyde, D. M., Gotwals, P. J., Koteliansky, V. E., Ryan, S. T.

& Giri, S. N. (1999) Thorax 54, 805–812.25. Bonniaud, P., Margetts, P. J., Kolb, M., Schroeder, J. A., Kapoun, A. M.,

Damm, D., Murphy, A., Chakravarty, S., Dugar, S., Higgins, L., et al. (2005) Am.J. Respir. Crit. Care Med. 171, 889–898.

26. Lee, V. Y., Schroedl, C., Brunelle, J. K., Buccellato, L. J., Akinci, O. I., Kaneto,H., Snyder, C., Eisenbart, J., Budinger, G. R. & Chandel, N. S. (2005) Am. J.Physiol. Lung Cell Mol. Physiol. 289, L521–L528.

27. Lee, C. G., Cho, S. J., Kang, M. J., Chapoval, S. P., Lee, P. J., Noble, P. W.,Yehualaeshet, T., Lu, B., Flavell, R. A., Milbrandt, J., et al. (2004) J. Exp. Med.200, 377–389.

28. Yin, X. M., Wang, K., Gross, A., Zhao, Y., Zinkel, S., Klocke, B., Roth, K. A.& Korsmeyer, S. J. (1999) Nature 400, 886–891.

29. Li, X., Rayford, H. & Uhal, B. D. (2003) Am. J. Pathol. 163, 2523–2530.30. Tsuburai, T., Suzuki, M., Nagashima, Y., Suzuki, S., Inoue, S., Hasiba, T., Ueda,

A., Ikehara, K., Matsuse, T. & Ishigatsubo, Y. (2002) Hum. Gene Ther. 13,1945–1960.

31. Bonniaud, P., Kolb, M., Galt, T., Robertson, J., Robbins, C., Stampfli, M.,Lavery, C., Margetts, P. J., Roberts, A. B. & Gauldie, J. (2004) J. Immunol. 173,2099–2108.

32. Sheppard, D. (2001) J. Clin. Invest. 107, 1501–1502.33. Aoshiba, K., Yasui, S., Tamaoki, J. & Nagai, A. (2000) Am. J. Respir. Crit. Care

Med. 162, 695–700.34. Ware, L. & Matthay, M. (2000) N. Engl. J. Med. 342, 1334–1349.35. Budinger, G. R., Chandel, N. S., Donnelly, H. K., Eisenbart, J., Oberoi, M. &

Jain, M. (2004) Intensive Care Med. 31, 121–128.

Budinger et al. PNAS � March 21, 2006 � vol. 103 � no. 12 � 4609

MED

ICA

LSC

IEN

CES

Dow

nloa

ded

by g

uest

on

May

17,

202

1