Inducible NOS Inhibition ReversesTobacco-Smoke-Induced Emphysemaand Pulmonary Hypertension in MiceMichael Seimetz,1,5 Nirmal Parajuli,1,5 Alexandra Pichl,1 Florian Veit,1 Grazyna Kwapiszewska,1 Friederike C. Weisel,1

Katrin Milger,1 Bakytbek Egemnazarov,1 Agnieszka Turowska,4 Beate Fuchs,1 Sandeep Nikam,2 Markus Roth,1

Akylbek Sydykov,1 Thomas Medebach,1 Walter Klepetko,3 Peter Jaksch,3 Rio Dumitrascu,1 Holger Garn,4

Robert Voswinckel,2 Sawa Kostin,2 Werner Seeger,1 Ralph T. Schermuly,2 Friedrich Grimminger,1 Hossein A. Ghofrani,1

and Norbert Weissmann1,*1University of Giessen Lung Center (UGLC), Excellence Cluster Cardiopulmonary System (ECCPS), D-35392 Giessen, Germany2Max-Planck-Institute for Heart and Lung Research, D-61231 Bad Nauheim, Germany3Department of Cardiothoracic Surgery, University Hospital of Vienna, A-1090 Vienna, Austria4Biomedical Research Center (BMFZ), D-35043 Marburg, Germany5These authors contributed equally to this work

*Correspondence: norbert.weissmann@innere.med.uni-giessen.de

DOI 10.1016/j.cell.2011.08.035

SUMMARY

Chronic obstructive pulmonary disease (COPD) isone of themost common causes of death worldwide.We report in an emphysema model of mice chro-nically exposed to tobacco smoke that pulmonaryvascular dysfunction, vascular remodeling, and pul-monary hypertension (PH) precede developmentof alveolar destruction. We provide evidence for acausative role of inducible nitric oxide synthase(iNOS) and peroxynitrite in this context. Mice lackingiNOS were protected against emphysema and PH.Treatment of wild-type mice with the iNOS inhibitorN6-(1-iminoethyl)-L-lysine (L-NIL) prevented struc-tural and functional alterations of both the lungvasculature and alveoli and also reversed estab-lished disease. In chimeric mice lacking iNOS inbone marrow (BM)-derived cells, PH was dependenton iNOS from BM-derived cells, whereas emphy-sema development was dependent on iNOS fromnon-BM-derived cells. Similar regulatory and struc-tural alterations as seen in mouse lungs were foundin lung tissue from humans with end-stage COPD.

INTRODUCTION

Chronic obstructive pulmonary disease (COPD), which includes

both chronic bronchitis and emphysema, is expected to be

ranked as the third-greatest cause of death worldwide by 2020

(Murray and Lopez, 1997). One pathological concept suggests

that COPD develops through airway inflammation and remodel-

ing. The main theory behind emphysema development is the

destruction of the elastic architecture of the lung, leading to

enlargement of distal air spaces (Black et al., 2008). Moreover,

COPD/emphysema is increasingly viewed as a systemic dis-

ease, involving skeletal muscle wasting, diaphragmatic dysfunc-

tion, and systemic inflammation (Agustı et al., 2003).

An estimated 30%–70% of patients with COPD also have

pulmonary hypertension (PH); however, there is much debate

about the numbers of patients affected by PH, and many

patients with COPD have no severe PH (Minai et al., 2010).

Thus, the relevance of a vascular pathology for the pathogenesis

of COPD is still unresolved. PH was often thought to occur as

a consequence of the hypoxia associated with COPD, but there

is increasing evidence that tobacco smoke may have a direct

impact on the pulmonary vasculature (Peinado et al., 2008), indi-

cating that cor pulmonale and late-stage PH are not necessarily

secondary to hypoxia in patients with COPD. These data are

supported by studies in guinea pigs, showing that a vascular

phenotype can precede parameters of emphysema develop-

ment (Ferrer et al., 2009; Wright and Churg, 1990, 1991). How-

ever, to the best of our knowledge, no published study directly

compares the course of emphysema development and PH in

other species.

Oxidative and nitrosative stress, chronic inflammation, apo-

ptosis, and altered proliferation have been suggested as factors

in the pathogenesis of airway remodeling (Churg et al., 2008; Ric-

ciardolo et al., 2004; Stockley et al., 2009; Tsoumakidou et al.,

2005). This has led to attention being focused on the involvement

of interleukins, the vascular endothelial-derived growth factor

(VEGF) system, matrix metalloproteinases (Mmp), and reactive

oxygen species (ROS) (Taraseviciene-Stewart and Voelkel,

2008; Yoshida and Tuder, 2007) in destroying the lung architec-

ture. ROS and reactive nitrogen species have long been known

to cause protein modification and DNA damage (Wink and

Mitchell, 1998). Indeed, nitric oxide (NO) reacts with superoxide

(O2,�) to form the potent oxidant peroxynitrite (ONOO�) (Szabo

et al., 2007); this in turn can react with tyrosine residues to

form nitrotyrosine, the levels of which are increased in COPD

(Ricciardolo et al., 2004; Tsoumakidou et al., 2005).

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. 293

NO is synthesized from L-arginine by nitric oxide synthase

(NOS), which exists as three isoforms (Moncada and Erusalim-

sky, 2002). Increased NO production and nitrosative stress in

COPD may derive from enhanced expression or activity of

inducible NOS (iNOS) and endothelial NOS (eNOS) (Brindicci

et al., 2009). However, the regulation of NOS isoforms during

the course of emphysema development has not yet been

addressed.

In order to investigate the molecular mechanisms involved in

disease progression as a basis to develop new strategies to treat

COPD, we aimed to (1) decipher the role of NOS in the develop-

ment of emphysema and examine possible vascular alterations

leading to PH after tobacco-smoke exposure and (2) investigate

a possible link between vascular alterations and emphysema

development.

RESULTS

Pulmonary Hypertension Precedes Lung EmphysemaDevelopment in Wild-Type Mice Exposed to TobaccoSmokeExposure of wild-type (WT) mice to tobacco smoke for up to

8 months resulted in the development of lung emphysema after

6 months, as evident from an increase in the mean linear inter-

cept, an increase in the air space, and a decrease in the septal

wall thickness (Figures 1A–1C).

Within 3 months, tobacco-smoke exposure caused increases

in right ventricular systolic pressure and the ratio of the absolute

numbers of alveoli to the number of vessels, followed by right-

heart hypertrophy (Figures 1D–1F): i.e., development of PH

preceded the development of lung emphysema. PH was associ-

ated with an increase in the degree of muscularization in the

pulmonary arteries (diameter 20–70 mm) (Figures 1G and 1H). A

similar increase in the degree of muscularization was found in

larger pulmonary arteries (data not shown). The late onset of

lung emphysema development compared with the development

of PHwas also evident from lung-function data (Figure S1A avail-

able online). PH occurred, althoughmice did not suffer from alve-

olar hypoxia or hypoxemia (Figure S1Bi, ii), despite substantial

carbon monoxide (CO) generation during tobacco-smoke expo-

sure (Figure S1Biii, iv) previously shown to antagonize PH (Zuck-

erbraun et al., 2006). In addition, the loss of vessels seen in the

tobacco-smoke-induced emphysema model was unparalleled

in hypoxia-induced PH, although the vascular phenotype was

comparable (Figure S1C). We also observed that gene regulation

was different in hypoxia than in tobacco-smoke-induced PH

(Figure S1D).

Effects of Tobacco Smoke on iNOS and eNOSExpression in the Pulmonary Vasculatureof Wild-Type MiceWe examined eNOS and iNOS expression during the course of

tobacco-smoke exposure. Immunofluorescence staining sug-

gested an upregulation of the iNOS protein, being more promi-

nent in the pulmonary vasculature compared to alveolar septa

or bronchi in smoke-exposed mice (Figure 2A). In situ hybridiza-

tion mirrored these results and suggested some upregulation of

iNOS mRNA in bronchi; however, this was not confirmed by the

294 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.

quantitativemRNA analysis frombronchi and alveolar septa (Fig-

ure 2B). By contrast, immunofluorescence staining and in situ

hybridization suggested a downregulation of eNOS in the pulmo-

nary vasculature, with some transient upregulationwithin the first

3 months of smoke exposure (Figure 2A). These data were

confirmed by quantitative polymerase chain reaction (PCR) analy-

sis of pulmonary vessels (diameter 50–100 mm), alveolar septa,

and bronchi (diameter 140–300 mm) and by western blotting

from homogenized lung tissue (Figures 2B and 2C). Expression

of iNOS and eNOS could mostly, but not exclusively, be allocated

to cells expressing a-smooth muscle actin by in situ hybridization

(data not shown). iNOS upregulation was mirrored by an increase

in iNOS activity in the lung but not in bronchoalveolar lavage (BAL)

cells (Figure S2).

Mice Deficient in iNOS Are Protected against theDevelopment of PH, Emphysema, and FunctionalAlterations Caused by Tobacco-Smoke ExposureWe compared the development of PH and emphysema after

8 months of tobacco-smoke exposure in iNOS�/�, eNOS�/�,and WT mice. The iNOS�/� mice were protected against the

development of emphysema and PH, as evident from quantifica-

tion of (1) mean linear intercept, air space, septal wall thickness

(Figures 3A–3C); (2) right ventricular systolic pressure, the ratio of

alveoli/vessels, right-heart hypertrophy (Figures 3D–3F); and (3)

the degree of muscularization (Figure 3G). By contrast, eNOS�/�

mice developed emphysema and PH to the same degree as WT

controls (Figures 3A–3G). Similar patterns were also seen in

lung-function parameters (Figure S3A).

iNOS Inhibition Prevented Smoke-Induced FunctionalDeterioration and Reversed Deterioration of FullyEstablished EmphysemaTreatment of WT mice with the iNOS-selective inhibitor L-NIL

was started in parallel to smoke exposure and resulted in protec-

tion against the development of lung emphysema, as shown

by alveolar morphometry (Figures 3H–3J), and against PH, as

shown by hemodynamic measurements and morphometry

(Figures 3K–3N). These findings were mirrored by lung-function

parameters (Figure S3B).

Lung structure and function were restored when WT mice

were treated with L-NIL in a curative approach after full estab-

lishment of emphysema (i.e., initiation of L-NIL treatment after

8 months of chronic smoke exposure for an additional 3 month

period) (Figures 3H–3N). Lung regeneration did not occur in

placebo-treated animals. Analysis of lungs from a second, inde-

pendent set of curatively L-NIL-treated mice revealed (besides

effects on cellular components; Figures S3C and S3D) that the

number of alveoli (assessed by stereological morphometry,

which excludes lung-volume-dependent effects) was signifi-

cantly reduced after 8 months of smoke exposure and restored

by curative L-NIL treatment but not by placebo treatment (Fig-

ure 4Ai). In vivo lung-function assessment from these animals re-

vealed an increase in lung dynamic compliance after 8 months of

smoke exposure, which was reversed upon curative L-NIL treat-

ment but not with placebo (Figure 4Aii). The reversal effect of

L-NIL treatment was also evident for PH (Figure 4Aiii). Detailed

investigation and quantification of elastic-fiber structure by light,

C D

WT, 8 months of smoke exposureWT, 0 months of smoke exposure

iviii

iii

BA

F GE

H

50 µm50 µm

50 µm 50 µm

50 µm

50 µm 50 µm

50 µm 50 µm

50 µm

0 1 2 3 6 8

0

70

74

78

82

Months of smoke exposure

*

Air

sp

ac

e (

%)

0 1 2 3 6 8

0

24

26

28

30

32

Months of smoke exposure

*

Rig

ht v

en

tric

ula

r s

ys

to

lic

pre

ss

ure

(m

m H

g)

0 1 2 3 6 8

0

22

24

26

28

30

32

Months of smoke exposure

*

Me

an

lin

ea

r in

te

rc

ep

t (

µm

)

0 1 2 3 6 8

0

30

35

40

45

Months of smoke exposure

*

Alv

eo

li / v

es

se

ls

0 1 2 3 6 8

0

3.5

4.0

4.5

5.0

Months of smoke exposure

*

Se

pta

l w

all t

hic

kn

es

s (

µm

)

RV

/ (L

V +

s

ep

tu

m)

0.00

0.26

0.33

8

Months of smoke exposure

*

0.28

0.30

0.32

0.24

0 1 2 3 6

RV

/ (L

V +

s

ep

tu

m)

0.00

0.26

0.33

8

Months of smoke exposure

*

0.28

0.30

0.32

0.24

0 1 2 3 6

Pe

rc

en

t o

f to

ta

l v

es

se

l c

ou

nt

20

40

60

80

Months of smoke exposure

Full Partial None

∗

∗

∗

080 1 2 3 80 1 2 3 80 1 2 3

Pe

rc

en

t o

f to

ta

l v

es

se

l c

ou

nt

20

40

60

80

Months of smoke exposure

Full Partial None

∗

∗

∗

080 1 2 3 80 1 2 3 80 1 2 3

Months of smoke exposurePercen

to

f to

ta

l vesselco

un

t

Partial NoneFull

Months of smoke exposureMonths of smoke exposure

Figure 1. TimeCourse of the Development of Emphysema and Pulmonary Hypertension during Tobacco-Smoke Exposure inWild-TypeMice

(A–C) Alveolar morphometry given as (A) mean linear intercept, (B) air space, and (C) septal wall thickness.

(D) Right ventricular systolic pressure.

(E) Ratio of the number of alveoli to the number of vessels per area.

(F) Right-heart hypertrophy given as the ratio of the right ventricular (RV) and the left ventricular plus septum (LV+S) mass.

(G) Degree of muscularization of small pulmonary arteries. Data are given as percentages of total vessel count for fully muscularized (Full), partially muscularized

(Partial), and nonmuscularized (None) vessels.

(H) Representative histology from lung sections stained with hematoxylin and eosin (i, ii) or antibodies against a-smooth muscle actin (violet) or von Willebrand

factor (brown = endothelial cell marker; iii, iv). Themagnified histology of the alveolar structure in (i) and (ii) represents areas of emphysema formation, whereas (iii)

and (iv) depict the increase in the degree of muscularization indicated by violet color.

Data are mean ± standard error of the mean (SEM) from n = 6 lungs each in the time course of tobacco-smoke exposure for up to 8 months. *Significant

differences (p < 0.05) compared with unexposed controls (i.e., 0 months of exposure). See also Figure S1.

confocal, and electronmicroscopy showed destruction of elastic

fibers during tobacco-smoke exposure and regeneration upon

curative L-NIL treatment (Figures 4B–4D and S4).

Cellular Components of PH and EmphysemaDevelopmentA detailed fluorescence-activated cell sorter analysis performed

using the BAL and the remaining lung homogenate showed

a tendency toward increased numbers of granulocytes, macro-

phages, activated macrophages, CD4+ T cells (and CD8+ T cells,

but these were generally detected in very low numbers), and acti-

vated T cells in the homogenate after 3 and 8 months of smoke

exposure. In the BAL, this trend only became evident for activated

granulocytes after 8 months of smoke exposure (Figure S3C). In-

terestingly, curative L-NIL treatment significantly downregulated

the number of granulocytes, macrophages, activated macro-

phages, and T cells (and also tended to activate T cells) in the

homogenate. In contrast, no such downregulation was detected

in theBAL (FigureS3C). Combinedquantification ofmacrophages

fromboth thealveolar andnonalveolar compartmentsby immuno-

histochemistry confirmed an upregulation of those cells and

a reduction upon curative L-NIL treatment (Figure S3D).

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. 295

Figure 2. Localization and Relative Quantification of iNOS and eNOS in Wild-Type Mouse Lungs

(A) Immunostaining (i and ii; red) and nonisotopic in situ hybridization (iii and iv; green) for iNOS and eNOS in WT mouse lung sections. V = vessel, B = bronchus.

(B) Quantitative real-time polymerase chain reaction analysis for iNOS (i) and eNOS (ii) mRNA of small pulmonary vessels (outer diameter 50–100 mm), septa, and

bronchi (outer diameter 140–300 mm). iNOS and eNOS values were related to porphobilinogen deaminase mRNA levels. (i and ii) Data are from duplicate

measurements of n = 20 vessels from n = 3 lungs each.

(C)Western blot analysis of iNOS (i) and eNOS (ii) from lung homogenate, normalized to b-actin. Values are frommeasurements of n = 6 individual lungs each. Data

are given for 3 and 8 months of tobacco-smoke exposure and for unexposed controls (0 months). A representative blot is shown on the right and densitometry is

given on the left.

Data are presented as mean ± SEM. *Significant difference (p < 0.05) compared with unexposed controls (i.e., 0 months of exposure). See also Figure S2.

To further decipher the role of iNOS in bone marrow (BM)-

derived versus non-BM-derived cells for the development of

emphysema and PH, we generated chimeric mice, where BM

from iNOS�/� mice was transplanted into WT mice (iNOS�/�/WT) and vice versa (WT/iNOS�/�). WT-to-WT BM transplanta-

tion served as a control. Quantification of alveolar numbers by

stereology, in vivo hemodynamics and lung function, and

vascular morphometry showed that WT-to-WT transplanted

mice developed emphysema and PH after 4 months of smoke

exposure. Whereas the WT/iNOS�/� mice were protected

from emphysema, iNOS�/�/WT mice were not, as seen from

the changes in the number of alveoli as well as in vivo compliance

(Figures 5A and 5B). In contrast, only those chimeric mice with

deletion of iNOS in the BM cells were protected from vascular

alterations and PH (Figures 5C and 5D).

296 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.

Molecular Pathways Explaining Lung EmphysemaDevelopment and Its Reversal upon Curative L-NILTreatmentONOO� is a possible candidate for mediating the effects of iNOS

upregulation on lung vasculature and parenchyma. As down-

stream signaling of ONOO� can be mediated via nitrotyrosine

formation, we investigated nitrotyrosine levels in WT, eNOS�/�,iNOS�/�, and L-NIL-treated mice, confirming a key role of

iNOS for nitrotyrosine formation (Figures 6A and S5A). We found

that ONOO� indeed caused nitration in primary isolated alveolar

epithelial cells, pulmonary vascular endothelial cells, and pulmo-

nary arterial smooth muscle cells from resistance vessels (data

not shown). Subsequently, apoptosis was found to be induced

in epithelial and endothelial cells. Proliferation, however, was

only reduced in epithelial cells. No significant effect was found

Figure 3. Effects of Smoke Exposure in Mice Lacking iNOS or eNOS (iNOS�/� or eNOS�/�) and in Wild-Type Mice Treated with an Inhibitor of

iNOS(A–G) Comparison of iNOS�/�, eNOS�/�, and WT mice and (H–N) comparison of WT mice treated with L-NIL or placebo. Alveolar morphometry is given as

(A and H) mean linear intercept, (B and I) air space, and (C and J) septal wall thickness. (D and K) Right ventricular systolic pressure. (E and L) Ratio of number of

alveoli to the number of vessels per area. (F and M) Right-heart hypertrophy given as the ratio of the right ventricular (RV) and the left ventricular plus septum

(LV+S) mass. (G and N) Degree of muscularization of small pulmonary arteries (diameter 20–70 mm). Data are given as percentage of total vessel count for fully

muscularized (Full), partially muscularized (Partial), and nonmuscularized (None) vessels. Data are for n = 6 lungs each presented as mean ± SEM. *Significant

difference (p < 0.05) compared with placebo-treated mice. See also Figure S3.

on apoptosis or proliferation in pulmonary arterial smooth

muscle cells (Figures 6B–6C). As a possible downstream link,

we investigated key mediators of proliferation and apoptosis,

like c-Jun N-terminal kinase (JNK), Src, and ERK phosphoryla-

tion. p-JNK was upregulated in alveolar epithelial and pulmonary

vascular endothelial cells (Figure 6D), but p-ERK and p-Src levels

were not affected by ONOO� in any cell types investigated, and

JNK, ERK, and Src inhibitors could not antagonize ONOO�

effects on proliferation and apoptosis (not shown). Investigation

of the effects of ONOO� onRtp801 and VEGF formation revealed

that the Rtp801 protein was upregulated only in epithelial cells,

and VEGF was downregulated in epithelial and pulmonary arte-

rial smooth muscle cells (Figures 6E and 6F).

In addition, we addressed mechanisms of apoptosis, pro-

liferation, extracellular matrix destruction/restoration, oxidative

stress, and inflammation in small pulmonary artery vessels

(diameter 50–100 mm), alveolar septa, and small bronchi (diam-

eter 140–300 mm; Figure S5B). This revealed that apoptosis

markers were consistently upregulated in the vascular compart-

ment by 3 months of smoke exposure, whereas no such upregu-

lation occurred in the alveolar septa. Importantly, this upregula-

tion of apoptosis markers in the vascular compartment was

strongly counter-regulated by curative L-NIL treatment, but no

such effects were seen in the alveolar or bronchial compart-

ments (Figure S5B). L-NIL treatment attenuated or reversed

the downregulation of the majority of the cell proliferation

markers (Figure S5B). This finding was supported by prolifer-

ating-cell nuclear antigen (PCNA) staining (Figure S5C). In addi-

tion, several genes found to be regulated by curative L-NIL

treatment for the categories of extracellular matrix regulation,

oxidative stress, and inflammation correlated with the restorative

effects of such treatment (Figure S5B).

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. 297

C

0 months of smoke exposure 8 months of smoke exposure

8 months of smoke exposure

with subsequent placebo

treatment

8 months of smoke exposure

with subsequent L-NIL

treatment

* **

**

**

500 nm

0 months of

smoke

exposure

8 months of

smoke

exposure

8 months of

smoke

exposure

with

subsequent

placebo

treatment

8 months of

smoke

exposure

with

subsequent

L-NIL

treatment

D

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

0

3

4

5

6

7 * *

No

. o

f alv

eo

li (m

illio

n)

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

0

3

4

5

6

7 * *

No

. o

f alv

eo

li (m

illio

n)

0

40

50

60

70

*

Co

mp

lian

ce (µ

l/cm

H2O

)

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

0

40

50

60

70

*

Co

mp

lian

ce (µ

l/cm

H2O

)

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

i iiA

0

20

22

24

26

28

30

*

Rig

ht ven

tric

ula

r systo

lic

pressu

re (m

mH

g)

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

0

20

22

24

26

28

30

*

Rig

ht ven

tric

ula

r systo

lic

pressu

re (m

mH

g)

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

0

4

6

8

10

12

14

16

*

% E

lastin

p

er m

m3

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

0

4

6

8

10

12

14

16

*

% E

lastin

p

er m

m3

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

iii

0

15

20

25

30

*

Sco

re ela

stic

fib

res/area

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

0

15

20

25

30

*

Sco

re ela

stic

fib

res/area

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

iiiii

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

0

250

300

350

*

Intact ela

stic

fib

res/area

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

Months of smoke exposure

0 8 8 8

Smoke

L-NIL

-

- -

+

-

+

+3

+

Placebo - - +3 -

0

250

300

350

*

Intact ela

stic

fib

res/area

Bi

Figure 4. Pulmonary Effects of iNOS Inhibition in Wild-Type Mice following Chronic Smoke Exposure

(A) Total number of alveoli assessed by quantitative stereology (i), in vivo dynamic lung compliance (ii), and right ventricular systolic pressure (iii); n = 5 each.

(B) Quantification of elastin by determination of the amount of intact elastic fibers (i) and elastin score (ii) via light microscopy from paraffin-embedded lungs

(n = 5 each) or elastin content in cryopreserved lungs (iii) by confocal microscopy (CM) (n = 3).

(C) Representative slides of elastin-stained lungs from CM.

(D) Elastic fibers (*) magnified by electron microscopy.

Data are presented as mean ± SEM. *Significant difference (p < 0.05) between groups as indicated. See also Figure S4.

With regard to ONOO� generation, we used lung homoge-

nates (deprived of alveolar leukocytes) versus BAL cells to

assess the activity of iNOS and O2,� generation in different

lung compartments. NO levels derived from iNOS were in-

creased in the homogenized lungs, reflecting iNOS upregula-

tion. Conversely, NO levels from BAL cells were unchanged

related to a per-cell basis (Figures S2A and S2B). The increase

in iNOS-derived NO from emphysematous lungs was reversed

by curative L-NIL treatment but not by placebo (Figure S2A).

Determination of O2,� generation from homogenized lung

tissue and from BAL cells revealed no alteration in levels

during the course of the disease or upon treatment in the

lung homogenate (when normalized to the amount of protein

or alveolar macrophages and related on a per-cell basis)

(Figure S2C).

298 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.

Comparison of Human End-Stage COPD in Smokersto a Mouse Model of Emphysema Inducedby Tobacco SmokeWhen comparing lung tissue from patients with Global Initiative

for Chronic Obstructive Lung Disease (GOLD) stage IV COPD

(Rabe et al., 2007) and a history of smoking (Table S1) to that

of healthy donors, we found a similar increase in mean linear

intercept and air space measures and a similar decrease in

septal wall thickness (Figures 7A–7C and 7F) as seen in WT

mice after 8 months of tobacco-smoke exposure. In addition,

similar to mice, an increase in the ratio of the number of alveoli

to the number of vessels and an increased degree of vessel mus-

cularization were found in the tissue samples from patients with

COPD (Figures 7D–7F) compared to healthy donors. Smokers

who had not yet developed COPD displayed similar vascular

B

C

A

D

0

2

3

4* *

No.

of a

lveo

li (m

illio

n)

Months of smoke exposure

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

0

2

3

4* *

No.

of a

lveo

li (m

illio

n)

Months of smoke exposure

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

0

20

40

100

Full Partial None

*

* *

Perc

ent o

f tot

al v

esse

l cou

nt

Months of smoke exposure

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

0

20

40

100

Full Partial None

*

* *

Perc

ent o

f tot

al v

esse

l cou

nt

Months of smoke exposure

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

0

30

40

50

60*

Com

plia

nce

(μl/c

mH

2O)

Months of smoke exposure

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

0

30

40

50

60*

Com

plia

nce

(μl/c

mH

2O)

Months of smoke exposure

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

Months of smoke exposure

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

0

24

26

28

30

32

34**

Rig

ht v

entr

icul

ar s

ysto

licpr

essu

re (m

mH

g)

Months of smoke exposure

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

0

24

26

28

30

32

34**

Rig

ht v

entr

icul

ar s

ysto

licpr

essu

re (m

mH

g)

Months of smoke exposure

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

Months of smoke exposure

0 4 0 4-

4++-

0

-

+--

--+-

+-+-

---+

+--+

WT→iNOS-/-

SmokeWT→WT

iNOS-/-→WTWT→iNOS-/-

SmokeWT→WT

iNOS-/-→WT

WT→iNOS-/-

SmokeWT→WT

iNOS-/-→WTWT→iNOS-/-

SmokeWT→WT

iNOS-/-→WT

Figure 5. Pulmonary Effects of Smoke Exposure in Bone Marrow-Transplanted, iNOS�/� Chimeric Mice

(A–C) Total number of alveoli assessed by (A) quantitative stereology (n = 8), (B) in vivo dynamic lung compliance (n R 9), and (C) hemodynamics (n R 9).

(D) Quantification of degree of muscularization of small pulmonary arteries (diameter 20–70 mm) as percent of total vessel count for fully muscularized (Full),

partially muscularized (Partial), and nonmuscularized (None) vessels (n R 9 for each).

Data are given as mean ± SEM for chimeric mice with BM transplantation fromWT to WTmice (WT/WT), WT to iNOS�/� mice (WT/iNOS�/�), and iNOS�/� to

WT (iNOS�/�/WT) mice. *Significant difference (p < 0.05) as indicated.

alterations to those observed in smokers suffering from COPD

(Figures 7A–7E). Analysis of iNOS and eNOS expression in

human lung tissue also showed a similar pattern of regulation

in the COPD lungs and in those of donor controls as in smoke-

exposed and control mice. In circumstances where iNOS was

upregulated in the pulmonary vasculature, eNOS was down-

regulated (Figures 7G and 7H). Levels of nitrotyrosine protein in

the pulmonary vasculature and alveoli were increased in

samples from patients with COPD compared with donor controls

(Figure 7I). Again, similar regulation profiles were found in

smokers without COPD.

DISCUSSION

We sought to determine whether the vascular pathology is

linked to emphysema development and identified iNOS as

a key molecular player in the underlying processes. Our data

showed that alterations in lung vascular structure and function

induced by tobacco smoke preceded emphysema in mice

and were independent of hypoxia. We also showed that emphy-

sema and PH occurred independently and are essentially asso-

ciated with iNOS in different cell types. Finally, our data indicate

that targeting iNOS by pharmacological inhibition can improve

the functional and structural destruction caused by tobacco

smoke.

Long-term exposure to tobacco smoke in our mouse model

enabled analysis of the temporal order of structural and func-

tional changes in both the pulmonary vasculature and the

airways and alveolar structures during the development of lung

emphysema. The development of lung emphysema in our model

correlated with previous reports that used mice exposed to

tobacco smoke (Churg et al., 2008). Vascular remodeling upon

tobacco-smoke exposure has been shown in humans (Peinado

et al., 2008) and in animal models (Ferrer et al., 2009; Wright

and Churg, 1991;Wright et al., 2006, 2011); however, we provide

here a detailed direct comparison of alterations in vascular struc-

ture and function and their temporal relationship to alveolar

destruction in mice. In this regard, dysregulation of iNOS and ni-

trotyrosine formation have been proposed as underlying mech-

anisms of COPD (Brindicci et al., 2009).

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. 299

Control -

ONOO0

50

100

150

VEG

F (n

g/m

l, %

)

A B

Rel

. abs

orpt

ion

(405

nm

; %)

Control

ONOO-

stau

Control

ONOO-

stau

Control

ONOO-

stau

Apo

ptos

is

Rel

. abs

orpt

ion

(405

nm

; %)

Apo

ptos

is

Rel

. abs

orpt

ion

(405

nm

; %)

Apo

ptos

is

04080100120140160

04080100120140160

04080100120140160

ns* * ** *

Control

ONOO-

CPM

(%)

Control

ONOO-

CPM

(%)

0

100

200

300

Control

ONOO-

ns

0

100

200

300

0

100

200

300

ns

Prol

ifera

tion

CPM

(%)

Prol

ifera

tion

Prol

ifera

tion

*

D

Control

ONOO-

Control

ONOO-

Control

ONOO-

p-JN

K /

JNK

(%)

0

100

200

300

400

0

100

200

300

400

* *p-

JNK

/ JN

K(%

)

p-JN

K /

JNK

(%)

*F

Control -

ONOO0

50100150200250

Rtp8

01 /β -

actin

(%)

*

Control -

ONOO0

50100150200250

Rtp8

01 /β

-act

in (%

)

Control -

ONOO0

50100150200250

Rtp

801

/β-a

ctin

(%)

AECII EC PASMC

C

E

Control -

ONOO0

50

100

150

VEG

F (n

g/m

l, %

)

Control -

ONOO0

50

100

150

VEG

F (n

g/m

l, %

)

β-actin

Nitrotyrosine

0 8 8 8 8-- + +

- +--

+- - -

- - - + -+

+

0 0

--

Months of smoke exposure

+

- - - - - +

42 kDa

0

100

200

300

400

L-NILSmokeiNOS-/-

eNOS-/-

L-NILSmokeiNOS-/-

eNOS-/-

Figure 6. Nitrotyrosine Expression in Lungs of iNOS�/� and eNOS�/� and L-NIL-TreatedWild-TypeMice andEffects of Peroxynitrite (ONOO�)on Cell Growth and Apoptosis

(A) Western blot analysis of nitrotyrosine from homogenized lung tissue in WT, iNOS�/�, eNOS�/�, and L-NIL-treated WT mice (preventive approach; n = 4 for

each).

(B and C) Assessment of (B) apoptosis and (C) proliferation in AECII, microvascular EC, and PASMC after treatment with ONOO� or KOH (control) for 20 min.

Staurosporine (stau) was used as a positive control for apoptosis. Values are normalized to controls and derived from n = 3 (AECII), n = 4 (EC) and n = 6–8 (PASMC)

independent experiments.

(D)Western blot analysis of phospho(p)-JNK fromAECII, EC, and PASMC after ONOO� incubation for 5min. Data are normalized to respective unphosphorylated

protein (n = 7 [AECII], n = 9 [EC], or n = 3 [PASMC] independent experiments).

(E)Western blot analysis of Rtp801 fromAECII, EC, and PASMCafter ONOO� and inhibitor incubation. Data are normalized to b-actin (n = 3 [AECII and EC] or n = 5

[PASMC] independent experiments).

(F) VEGF concentration in AECII, EC, and PASMC after ONOO� and inhibitor incubation by ELISA. Values are normalized to controls and derived from n = 4

(AECII), n = 3 (EC), and n = 8 (PASMC) independent experiments and are given as mean ± SEM.

*Significant difference (p < 0.05) compared with controls or between L-NIL-treated and untreated mice. See also Figure S5.

Our investigations of the cellular contribution to both emphy-

sema development and PH in chimeric mice showed that both

can occur independently. The complete dependency of pulmo-

nary vascular alterations on BM-derived iNOS-containing cells

can be explained by a derivation of a portion of pulmonary

vascular cells from BM cells. It has been suggested that BM-

derived cells contribute to pulmonary vascular remodeling, and

that pulmonary arterial smooth muscle cells may be generated

from BM-derived cells (Asosingh et al., 2008; Huertas and Pal-

ange, 2011). Indeed, the importance of BM cell-derived iNOS

300 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.

for induction of systemic vascular disease was highlighted

recently by Ponnuswamy and colleagues (Ponnuswamy et al.,

2009). Alternatively, iNOS from non-smoothmuscle, BM-derived

cells could contribute also to pulmonary vascular remodeling.

The fact that emphysema was only prevented in chimeric mice

that lacked iNOS in non-BM-derived cells excludes an essential

role for iNOS in the pathogenesis of emphysema in macro-

phages or other BM-derived cells. This supports the concept

of a prominent role of non-BM-cell-derived iNOS and thus,

e.g., vascular iNOS for emphysema development. Interestingly,

A

F G

B C D E

H I

Figure 7. Comparison of the Alveolar and Vascular Structure, eNOS and iNOS Expression, and Nitrotyrosine Formation in Lungs fromHuman

Patients with Severe COPD, Smokers without COPD, and Healthy Donor Controls

(A–C) Alveolar morphometry given as (A) mean linear intercept, (B) air space, (C) septal wall thickness.

(D) Ratio of the number of alveoli to the number of vessels.

(E) Degree of muscularization of small pulmonary arteries (diameter 20–70 mm). Data are given as percentage of total vessel count for fully muscularized (Full),

partially muscularized (Partial), and nonmuscularized (None) vessels.

(F) Representative histology from stained lung sections, representing alveolar changes (i) or antibodies against a-smooth muscle actin (violet) and vonWillebrand

factor (brown = endothelial cell marker) (ii).

(G) iNOS (i) and eNOS (ii) immunostaining. V = vessel.

(H) Expression of iNOS mRNA (i), iNOS protein (ii), eNOS mRNA (iii), and eNOS protein (iv).

(I) Densitometric data from a western blot analysis of nitrotyrosine from homogenized lung tissue.

*Significant difference (p < 0.05) compared with healthy donor controls. Data are derived from human GOLD stage IV COPD lungs (smoker+COPD), smokers

without COPD (smoker�COPD), and healthy donor control lungs (donor) and are presented as mean ± SEM. See also Table S1.

the data from the chimeric mice further suggested that emphy-

sema development can be dependent on vascular iNOS in the

absence of vascular remodeling. The fact that emphysema

development and PH in the chimeric mice were triggered by

different cell populations and, thus, can occur independently

may explain why only a portion of patients suffering from

COPDdevelop PH (if themice data are transferable to the human

situation).

The increased levels of nitrotyrosine, present as a possible

consequence of ONOO� generation (Szabo et al., 2007), in WT

mice following tobacco-smoke exposure were in accordance

with our hypothesis that ONOO� upregulation is a key step in

vascular remodeling and emphysema pathogenesis. Interest-

ingly, ONOO� both induced apoptosis and reduced proliferation

in alveolar epithelial cells but caused only apoptosis in endothe-

lial cells. This result agreed with the observed effects on alveolar

and vascular pruning in our animal studies. Such effects of

ONOO� are also concordant with previous findings of apoptosis

induction (Szabo et al., 2007), a mechanism important for the

development of lung emphysema (Yoshida and Tuder, 2007).

We have shown that ONOO� can induce JNK but not ERK

and Src phosphorylation; however, such effects could not be

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. 301

associated with proliferation or apoptosis by respective inhibitor

studies. Interestingly, ONOO� upregulated Rtp801, a protein

identified as essential for emphysema development in mice

(Yoshidaet al., 2010), in alveolar epithelial cellsbut not endothelial

or vascular smoothmuscle cells. Such a regulation has been sug-

gested to involvedownstreamVEGF inhibition, leading toalveolar

epithelial cell apoptosis (Ellisen, 2010;Yoshidaet al., 2010).VEGF

geneknockout causesemphysema inmice (Taraseviciene-Stew-

art and Voelkel, 2008), and VEGF has been shown to be essential

for lung growth and maintenance (Voelkel et al., 2006). In accor-

dance, we showed downregulation of VEGF in alveolar epithelial

type II cells (AECII) andprimarymurinepulmonary arterial smooth

muscle cells (PASMC) upon ONOO� challenge.

Analyzing gene-regulatory processes upon curative L-NIL

treatment revealed several candidates that can be linked to

lung regeneration or support regeneration by attenuation of dis-

ease progression. L-NIL reversed the upregulation of Mmp9 (a

marker of parenchyma destruction), and the mRNA of the

metalloproteinase inhibitor encoded by Timp3 was upregulated

in the vascular compartment with L-NIL treatment. It has previ-

ously been shown that Timp3 knockout mice develop emphy-

sema (Leco et al., 2001). The reversal of the mRNA downregula-

tion of pro-proliferative factors like Fgf10 and Ccna1 in alveolar

septa by curative L-NIL treatment correlates with the increase

in PCNA-positive cells. Corroborating these findings, the upre-

gulation of mRNA of proapoptotic genes Bax, Tnfsf10, Fbf1,

Traf1, and Fastk in the vascular compartment was again antago-

nized by curative L-NIL treatment.

Our detailed analysis of the elastic fiber structure of the lung

supports the concept that the increase in proliferation upon

curative L-NIL treatment is part of an active restructuring pro-

cess of the lung. Different methods of quantification and ultrafine

structural images revealed that the amount, as well as the struc-

ture, of elastin fibers is reduced and degraded, respectively,

during tobacco-smoke exposure and is substantially reversed

upon curative L-NIL treatment.

Interestingly, curative L-NIL treatment selectively reduced the

number of granulocytes, macrophages, activatedmacrophages,

and T cells in the lung-tissue compartment. Downregulation of

macrophage numbers may coincide with reduced O2,� produc-

tion, leading to decreased oxidative stress and thus reduced

ONOO� levels, but also to reduced inflammatory mediator levels

(Chung and Adcock, 2008). In addition, the observed downregu-

lation of T cells correlates with the hypothesis that autoimmune

mechanisms may contribute to COPD development (Churg

et al., 2008; Feghali-Bostwick et al., 2008; Motz et al., 2010).

Moreover, it is suggested that T cells are involved not only in

emphysema development but also in the pathological remodel-

ing of the pulmonary vasculature (Austin et al., 2010; Cuttica

et al., 2011). As our data from the chimeric mice show that

the entire vascular remodeling process is dependent on BM-

derived iNOS-expressing cells, this effect could be due either

to inflammatory cells like macrophages or to iNOS-containing

stem cells. Moreover, data from our experiments performed

in chimeric mice showed that emphysema development can

be independent on iNOS in BM-derived cells and, thus, is at least

independent from iNOS in macrophages, activated macro-

phages, granulocytes, T cells, and activated T cells.

302 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.

Examination of lung tissue from ten patients with severe COPD

(GOLD stage IV) who had undergone lung transplantation re-

vealed an upregulation of iNOS (mRNA and protein), increases

in nitrotyrosine content, and alterations in vascular and alveolar

structure and function qualitatively similar to those seen in the

smoke-exposed WT mice. In addition, we have corroborated

previous findings in lung-tissue samples from smokers who

had not developed COPD, demonstrating vascular alterations

in direct association with tobacco-smoke exposure (Peinado

et al., 2008). The similar profiles of iNOS and nitrotyrosine regu-

lation in smokers with COPD and smokers who had not devel-

oped COPD suggest that portions of the pathways deciphered

in our mouse model also may have an impact on human

COPD; however, this conclusion has to be drawn cautiously,

as our data are for a limited number of patients, are for end-stage

COPD, and are heterogeneous in nature when compared

with the mouse lung data. In addition, a confounding impact of

hypoxia or treatment including steroids should be taken into

account when applying the mouse data to the human situation.

If iNOS inhibition is to be investigated further as a clinical means

of treatment for emphysema, one has to consider possible

side effects related to disruption of iNOS function. As iNOS is

expressed in many cell types, including nonimmune cells and

immune cells (Bogdan, 2001), several cellular and systemic,

including immune, functions may be disrupted by iNOS inhibi-

tion. However, to date, no adverse effects concerning increased

susceptibility to infections have been reported in clinical studies

with selective iNOS inhibitors (Brindicci et al., 2009; Singh et al.,

2007). Owing to its effects on T cell regulation, it has been sug-

gested that abrogation of iNOS function could potentially exac-

erbate autoimmune diseases such as colitis, arthritis, or multiple

sclerosis (Niedbala et al., 2007). There are also implications for

iNOS playing a role in systemic vascular disease, and this has

been assessed in animal models including atherosclerosis (Pon-

nuswamy et al., 2009). Thus, given the potentially wide-ranging

effects of iNOS inhibition, any clinical intervention for emphy-

sema may require local application of iNOS inhibitors via inhala-

tion or cell-type-specific targeting of iNOS inhibition.

In conclusion, our study highlights that the effects of tobacco

smoke on the pulmonary circulation precede the development

of alveolar destruction and emphysema formation, and both

vascular and alveolar changes occur in an iNOS-dependent

manner in mice. Furthermore, emphysema and PH development

are not essentially linked but can occur independently. If trans-

ferable to humans, these findings could explain the hitherto

controversial discussion about the impact of PH on emphysema

development in humans, where PH is not always associated with

emphysema. Finally, we suggest that selective iNOS inhibition

offers the potential to reverse emphysema.

EXPERIMENTAL PROCEDURES

Animals

Adult male WT C57BL/6J, iNOS�/�, and eNOS�/� (B6.129P2-Nos2tm1Lau/J

and B6.129P2-Nos3tm1Unc/J) mice, 20–22 g, were obtained from Charles

River Laboratories, Sulzfeld, Germany. Animals were housed under con-

trolled conditions with a 12 hr light/dark cycle and food and water supply

ad libitum. Animals were randomly allocated to tobacco-smoke-exposed

and -unexposed groups of six mice each, with parallel groups for (1) alveolar

morphometry, (2) vascular morphometry including right ventricular blood-

pressure measurements, (3) protein and mRNA analysis, and (4) lung-function

tests. Selected parameters of alveolar and vascular structural and functional

measurements were assessed in a separate, independent set of experiments.

All experiments were approved by the governmental ethics committee for

animal welfare (Regierungsprasidium Giessen, Germany).

Experimental Design and Tobacco-Smoke Exposure

Wild-type, eNOS�/�, and iNOS�/� mice were exposed to mainstream smoke

of 3R4F cigarettes (Lexington, KY, USA) at 140 mg particulate matter/m3 for

6 hr/day, 5 days/week for up to 8 months. For preventive treatment, WT

mice were exposed to tobacco smoke for 8 months with parallel application

of the iNOS inhibitor L-NIL (N6-(1-Iminoethyl)-L-lysine dihydrochloride) (Bio-

tium, Hayward, CA, USA) at a concentration (600 mg/ml = 2.68 mM) known

to be highly iNOS selective in drinking water (Moore et al., 1994; Stenger

et al., 1995). Age-matched controls were kept under identical conditions to

the smoke-exposed mice but without smoke exposure. Very few of the para-

meters measured in this study were affected by the age of the control mice.

Therefore, the 8 month control values are given as control values if there

was no age effect.

In the curative approach, L-NIL treatment (2.68 mM in drinking water) was

started in WT mice after full establishment of the disease (8 months

tobacco-smoke exposure) for 3 months without further smoke exposure.

Age-matched, non-smoke-exposed mice and placebo-treated smoke-

exposed mice were used as controls. The drinking water for placebo was

adjusted to the same pH as the L-NIL solution. Each day, freshly prepared

L-NIL and placebo solutions were supplied to the animals. Tobacco-smoke

exposure was discontinued in the last 3 month treatment period. The curative

approach was performed in two independent sets of experiments.

Animal Preparation, In Vivo Hemodynamics, Alveolar and Vascular

Morphometry, Right-Heart Hypertrophy, and Lung Compliance

All animals were anesthetized with ketamine and xylazine and treated with

heparin (1000 U/kg) at the end of the experiments. Measurement of right-

ventricular systolic pressure (RVSP) was performed as described previously

(Schermuly et al., 2005). For alveolar morphometry, lungs were fixed by instil-

lation of paraformaldehyde via the trachea. For vascular morphometry and

determination of the alveoli to vessel ratio, lungs were fixed by vascular perfu-

sion with Zamboni’s fixative. The degree of muscularization was determined

from stained lung sections as described previously (Weissmann et al., 2006).

For right-heart hypertrophy, the right ventricle (RV) was separated from the

left ventricle plus septum (LV+S), and the RV to (LV+S) ratio was determined

from the dried tissue. In vivo dynamic lung compliance was assessed prior

to RVSP measurement. (For further details see Extended Experimental

Procedures online).

Isolated, Perfused Mouse Lung Experiments

For measurement of lung-function parameters, except those in Figures 4

and 5, an isolated, perfused mouse lung procedure was used as described

previously (Weissmann et al., 2006). For details, refer to Extended

Experimental Procedures online.

Isolation and Culture of Primary Murine Lung Cells, Exposure to

ONOO�, and Quantification of Proliferation and Apoptosis

PASMC, lung endothelial cells (EC), and AECII were isolated from WT mice

and cultured as described previously (Corti et al., 1996; Mittal et al., 2007;

Weissmann et al., 2006). For details, refer to Extended Experimental

Procedures online.

ONOO� was applied as described previously (Potoka et al., 2003). In brief,

the cells were washed with Dulbecco’s Phosphate-buffered Saline (DPBS,

Sigma-Aldrich, Steinheim, Germany), and ONOO� (50 mM; Alexis Biochemi-

cals, San Diego, CA, USA) or potassium hydroxide (as a control; Merck, Darm-

stadt, Germany) was mixed into the solution. For the proliferation and

apoptosis assay, cells were cultured in serum-free medium overnight prior

to ONOO� application.

Proliferation assay was performed as described previously (Mittal et al.,

2007). After incubation with ONOO�, the culture medium was replaced with

regular medium, containing serum and [3H]thymidine (Amersham, Munich,

Germany). After incubation at 37�C and 5% CO2 for 4 hr, cells were harvested

and [3H]thymidine incorporation was measured by liquid scintillation

spectrometry.

For assessment of apoptosis, the CaspACE Assay System, Colorimetric

(Promega, Mannheim, Germany), was used according to the manufacturer’s

instructions. Additionally, 1 mM staurosporine (Sigma-Aldrich, Munich,

Germany) for 4 hr was used as a positive control.

VEGF-ELISA

For detection of VEGF in cell culture medium, the RayBio Mouse VEGF ELISA

Kit (RayBiotech, Inc.) was used. Experiments were performed according to

the protocol provided by the supplier.

Generation of Bone Marrow-Transplanted Chimeric Mice

Generation of chimeric mice was performed as previously described

(Voswinckel et al., 2003) with modifications. For details, refer to Extended

Experimental Procedures online.

Alveoli Count via Design-Based Stereology

For counting of alveoli, uniform random sampling and the physical dissector

method were used as described previously (Ochs et al., 2004). For details,

refer to Extended Experimental Procedures online.

Localization of eNOS, iNOS, and Nitrotyrosine

Localization of eNOS and iNOS was investigated in lung sections from cryo-

preserved tissue by immunostaining, as described previously (Mittal et al.,

2007). Nitrotyrosine was detected in paraffin-embedded lung sections of

both mouse and human lung tissue using a rabbit anti-nitrotyrosine antibody

(Sigma-Aldrich). For details, refer to Extended Experimental Procedures

online.

PCNA Staining

For immunohistochemical localization of the proliferation marker PCNA

(Santa Cruz Biotechnology, Santa Cruz, CA, USA), the AP-fast red kit

(Zytochem, Berlin, Germany) was used on paraffin-embedded lung sections

according to the manufacturer’s instructions. All stained sections were

analyzed using digital slide scanning employing a mirax scanner and the mirax

viewer software (Carl Zeiss GmbH, Jena, Germany).

Nonisotopic In Situ Hybridization

Localization of mRNA by nonisotropic in situ hybridization (NISH) was deter-

mined in cryostat lung sections as previously described (Mittal et al., 2007).

For details, refer to Extended Experimental Procedures online.

Laser-Assisted Microdissection

Laser-assisted microdissection (LMD 6000, Leica, Nussloch, Germany) was

performed to isolate pulmonary arterial vessels, bronchi, and septa from cryo-

stat lung section as previously described (Mittal et al., 2007). For details, refer

to Extended Experimental Procedures online.

RNA Isolation, Preamplification, cDNA Synthesis,

and Real-Time PCR

RNA from laser-microdissected or homogenized mouse and human lung

tissue was isolated by RNeasy Micro and Mini kits, respectively (QIAGEN,

Hilden, Germany). The isolated RNA was converted to cDNA, and relative

quantification of the eNOS and iNOS mRNA was performed using the iQ

SYBR Green Supermix (BioRad, Munich, Germany). For details, refer to

Extended Experimental Procedures online.

Real-Time PCR-Based PCR Array

Vessels (diameter 50–100 mm), septa, and bronchi (diameter 140–300 mm)

were laser-microdissected from 8 mm sections of Tissue Tek-embedded

mouse lungs fixed after BAL. PCR-based arrays were performed using

customized 96-well plates containing primers for selected genes, according

to the manufacturer’s instructions (SA Biosciences/Biomol, Hamburg,

Germany).

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. 303

Western Blot

For the quantification of eNOS, iNOS, and nitrotyrosine in mouse and human

lung tissues, the polyclonal antibodies anti-eNOS (BD Biosciences, Heidel-

berg, Germany), or anti-iNOS (Abcam, Cambridge, UK), raised in rabbits,

and anti-nitrotyrosine (Abcam, Cambridge, UK), raised in mice, were used.

For the quantification of (phospho)-SAPK/JNK, (phospho)-ERK (both Cell

Signaling, Danvers, MA, USA), (phospho)-Src (Epitomics, Burlingame, CA,

USA), and Rtp801 (Abnova, Heidelberg, Germany) in mouse PASMC, EC,

and AECII, the respective polyclonal antibodies raised in rabbits were used.

For details, refer to Extended Experimental Procedures online.

Quantification of Lung Elastin by Image Analysis

Analysis was performed as previously published with modifications (Bigatel

et al., 1999; Black et al., 2008; Lawrence et al., 2004). For details, refer to

Extended Experimental Procedures online.

Confocal Microscopy and Quantification of Elastin Immunolabeling

Experiments were performed on cryosections of 30 mm thickness using

specific antibodies against elastin and a-smooth muscle actin. For details,

refer to Extended Experimental Procedures online.

Transmission Electron Microscopy

The tissue was fixed in 3% glutaraldehyde and embedded in Epon following

routine procedures. Ultrathin sections were double stained with uranyl acetate

and lead citrate and viewed in a Philips CM 10 or a CM 201 electron

microscope (Philips, Andover, MA, USA).

Patient Characteristics

Human lung tissues were obtained from transplanted COPD patients (GOLD

stage IV), smokers without COPD, and donor controls. The patients’ charac-

teristics are given in Table S1. The studies were approved by the Ethics

Committee of the Justus-Liebig-University School of Medicine (AZ 31/93),

Giessen, Germany.

For all other experimental procedures, please refer to the Extended

Experimental Procedures online.

Statistical Analyses

Comparison of multiple groups was performed by analysis of variance

(ANOVA) with the Student–Newman–Keuls post-test. If several groups were

compared to one control, an ANOVA with Dunnett’s test was performed. For

comparison of two groups, a Student’s t test was performed. We considered

p values below 0.05 as statistically significant for all analyses.

SUPPLEMENTAL INFORMATION

Supplemental Information includes Extended Experimental Procedures, five

figures, and one table and can be found with this article online at doi:10.

1016/j.cell.2011.08.035.

ACKNOWLEDGMENTS

The authors thank Nadja Baumgartl, Ingrid Breitenborn-Muller, Uta Eule, Lisa

Frohlich, Sabine Graf-Hochst, Carmen Homberger, Miriam Schmidt, and Karin

Quanz for technical assistance. This work was in part funded by the German

Research Foundation, Excellence Cluster Cardiopulmonary System (ECCPS),

the BMBF (ASCONET SP10), and the State of Hessen (LOEWE).

Received: August 24, 2010

Revised: April 30, 2011

Accepted: August 13, 2011

Published: October 13, 2011

304 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.

REFERENCES

Agustı, A.G., Noguera, A., Sauleda, J., Sala, E., Pons, J., and Busquets, X.

(2003). Systemic effects of chronic obstructive pulmonary disease. Eur.

Respir. J. 21, 347–360.

Asosingh, K., Aldred, M.A., Vasanji, A., Drazba, J., Sharp, J., Farver, C., Com-

hair, S.A., Xu, W., Licina, L., Huang, L., et al. (2008). Circulating angiogenic

precursors in idiopathic pulmonary arterial hypertension. Am. J. Pathol. 172,

615–627.

Austin, E.D., Rock, M.T., Mosse, C.A., Vnencak-Jones, C.L., Yoder, S.M.,

Robbins, I.M., Loyd, J.E., and Meyrick, B.O. (2010). T lymphocyte subset

abnormalities in the blood and lung in pulmonary arterial hypertension. Respir.

Med. 104, 454–462.

Bigatel, D.A., Elmore, J.R., Carey, D.J., Cizmeci-Smith, G., Franklin, D.P., and

Youkey, J.R. (1999). The matrix metalloproteinase inhibitor BB-94 limits

expansion of experimental abdominal aortic aneurysms. J. Vasc. Surg. 29,

130–138, discussion 138–139.

Black, P.N., Ching, P.S., Beaumont, B., Ranasinghe, S., Taylor, G., and Merri-

lees, M.J. (2008). Changes in elastic fibres in the small airways and alveoli in

COPD. Eur. Respir. J. 31, 998–1004.

Bogdan, C. (2001). Nitric oxide and the immune response. Nat. Immunol. 2,

907–916.

Brindicci, C., Ito, K., Torre, O., Barnes, P.J., and Kharitonov, S.A. (2009).

Effects of aminoguanidine, an inhibitor of inducible nitric oxide synthase, on

nitric oxide production and its metabolites in healthy control subjects, healthy

smokers, and COPD patients. Chest 135, 353–367.

Chung, K.F., and Adcock, I.M. (2008). Multifaceted mechanisms in COPD:

inflammation, immunity, and tissue repair and destruction. Eur. Respir. J. 31,

1334–1356.

Churg, A., Cosio, M., andWright, J.L. (2008). Mechanisms of cigarette smoke-

induced COPD: insights from animal models. Am. J. Physiol. Lung Cell. Mol.

Physiol. 294, L612–L631.

Corti, M., Brody, A.R., and Harrison, J.H. (1996). Isolation and primary culture

of murine alveolar type II cells. Am. J. Respir. Cell Mol. Biol. 14, 309–315.

Cuttica, M.J., Langenickel, T., Noguchi, A., Machado, R.F., Gladwin, M.T., and

Boehm, M. (2011). Perivascular T-cell infiltration leads to sustained pulmonary

artery remodeling after endothelial cell damage. Am. J. Respir. Cell Mol. Biol.

45, 62–71.

Ellisen, L.W. (2010). Smoking and emphysema: the stress connection. Nat.

Med. 16, 754–755.

Feghali-Bostwick, C.A., Gadgil, A.S., Otterbein, L.E., Pilewski, J.M., Stoner,

M.W., Csizmadia, E., Zhang, Y., Sciurba, F.C., and Duncan, S.R. (2008).

Autoantibodies in patients with chronic obstructive pulmonary disease.

Am. J. Respir. Crit. Care Med. 177, 156–163.

Ferrer, E., Peinado, V.I., Dıez, M., Carrasco, J.L., Musri, M.M., Martınez, A.,

Rodrıguez-Roisin, R., and Barbera, J.A. (2009). Effects of cigarette smoke

on endothelial function of pulmonary arteries in the guinea pig. Respir. Res.

10, 76.

Huertas, A., and Palange, P. (2011). Circulating endothelial progenitor cells

and chronic pulmonary diseases. Eur. Respir. J. 37, 426–431.

Lawrence, D.M., Singh, R.S., Franklin, D.P., Carey, D.J., and Elmore, J.R.

(2004). Rapamycin suppresses experimental aortic aneurysm growth. J.

Vasc. Surg. 40, 334–338.

Leco, K.J., Waterhouse, P., Sanchez, O.H., Gowing, K.L., Poole, A.R., Wake-

ham, A., Mak, T.W., and Khokha, R. (2001). Spontaneous air space enlarge-

ment in the lungs of mice lacking tissue inhibitor of metalloproteinases-3

(TIMP-3). J. Clin. Invest. 108, 817–829.

Minai, O.A., Chaouat, A., and Adnot, S. (2010). Pulmonary hypertension in

COPD: epidemiology, significance, and management: pulmonary vascular

disease: the global perspective. Chest 137(6, Suppl), 39S–51S.

Mittal, M., Roth, M., Konig, P., Hofmann, S., Dony, E., Goyal, P., Selbitz, A.C.,

Schermuly, R.T., Ghofrani, H.A., Kwapiszewska, G., et al. (2007).

Hypoxia-dependent regulation of nonphagocytic NADPH oxidase subunit

NOX4 in the pulmonary vasculature. Circ. Res. 101, 258–267.

Moncada, S., and Erusalimsky, J.D. (2002). Does nitric oxide modulate mito-

chondrial energy generation and apoptosis? Nat. Rev. Mol. Cell Biol. 3,

214–220.

Moore, W.M., Webber, R.K., Jerome, G.M., Tjoeng, F.S., Misko, T.P., and

Currie, M.G. (1994). L-N6-(1-iminoethyl)lysine: a selective inhibitor of inducible

nitric oxide synthase. J. Med. Chem. 37, 3886–3888.

Motz, G.T., Eppert, B.L., Wesselkamper, S.C., Flury, J.L., and Borchers, M.T.

(2010). Chronic cigarette smoke exposure generates pathogenic T cells

capable of driving COPD-like disease in Rag2-/- mice. Am. J. Respir. Crit.

Care Med. 181, 1223–1233.

Murray, C.J., and Lopez, A.D. (1997). Alternative projections of mortality and

disability by cause 1990-2020: Global Burden of Disease Study. Lancet 349,

1498–1504.

Niedbala, W., Cai, B., Liu, H., Pitman, N., Chang, L., and Liew, F.Y. (2007).

Nitric oxide induces CD4+CD25+ Foxp3 regulatory T cells from CD4+CD25

T cells via p53, IL-2, and OX40. Proc. Natl. Acad. Sci. USA 104, 15478–15483.

Ochs, M., Nyengaard, J.R., Jung, A., Knudsen, L., Voigt, M., Wahlers, T.,

Richter, J., and Gundersen, H.J. (2004). The number of alveoli in the human

lung. Am. J. Respir. Crit. Care Med. 169, 120–124.

Peinado, V.I., Pizarro, S., and Barbera, J.A. (2008). Pulmonary vascular

involvement in COPD. Chest 134, 808–814.

Ponnuswamy, P., Ostermeier, E., Schrottle, A., Chen, J., Huang, P.L., Ertl, G.,

Nieswandt, B., and Kuhlencordt, P.J. (2009). Oxidative stress and compart-

ment of gene expression determine proatherosclerotic effects of inducible

nitric oxide synthase. Am. J. Pathol. 174, 2400–2410.

Potoka, D.A., Upperman, J.S., Zhang, X.R., Kaplan, J.R., Corey, S.J., Grishin,

A., Zamora, R., and Ford, H.R. (2003). Peroxynitrite inhibits enterocyte prolifer-

ation and modulates Src kinase activity in vitro. Am. J. Physiol. Gastrointest.

Liver Physiol. 285, G861–G869.

Rabe, K.F., Hurd, S., Anzueto, A., Barnes, P.J., Buist, S.A., Calverley, P.,

Fukuchi, Y., Jenkins, C., Rodriguez-Roisin, R., van Weel, C., and Zielinski,

J.; Global Initiative for Chronic Obstructive Lung Disease. (2007). Global

strategy for the diagnosis, management, and prevention of chronic obstructive

pulmonary disease: GOLD executive summary. Am. J. Respir. Crit. Care Med.

176, 532–555.

Ricciardolo, F.L., Sterk, P.J., Gaston, B., and Folkerts, G. (2004). Nitric oxide in

health and disease of the respiratory system. Physiol. Rev. 84, 731–765.

Schermuly, R.T., Dony, E., Ghofrani, H.A., Pullamsetti, S., Savai, R., Roth, M.,

Sydykov, A., Lai, Y.J., Weissmann, N., Seeger, W., and Grimminger, F. (2005).

Reversal of experimental pulmonary hypertension by PDGF inhibition. J. Clin.

Invest. 115, 2811–2821.

Singh, D., Richards, D., Knowles, R.G., Schwartz, S., Woodcock, A., Langley,

S., and O’Connor, B.J. (2007). Selective inducible nitric oxide synthase inhibi-

tion has no effect on allergen challenge in asthma. Am. J. Respir. Crit. Care

Med. 176, 988–993.

Stenger, S., Thuring, H., Rollinghoff, M., Manning, P., and Bogdan, C. (1995).

L-N6-(1-iminoethyl)-lysine potently inhibits inducible nitric oxide synthase and

is superior to NG-monomethyl-arginine in vitro and in vivo. Eur. J. Pharmacol.

294, 703–712.

Stockley, R.A., Mannino, D., and Barnes, P.J. (2009). Burden and pathogen-

esis of chronic obstructive pulmonary disease. Proc. Am. Thorac. Soc. 6,

524–526.

Szabo, C., Ischiropoulos, H., and Radi, R. (2007). Peroxynitrite: biochemistry,

pathophysiology and development of therapeutics. Nat. Rev. Drug Discov. 6,

662–680.

Taraseviciene-Stewart, L., and Voelkel, N.F. (2008). Molecular pathogenesis of

emphysema. J. Clin. Invest. 118, 394–402.

Tsoumakidou, M., Tzanakis, N., Chrysofakis, G., and Siafakas, N.M. (2005).

Nitrosative stress, heme oxygenase-1 expression and airway inflammation

during severe exacerbations of COPD. Chest 127, 1911–1918.

Voelkel, N.F., Vandivier, R.W., and Tuder, R.M. (2006). Vascular endothelial

growth factor in the lung. Am. J. Physiol. Lung Cell. Mol. Physiol. 290, L209–

L221.

Voswinckel, R., Ziegelhoeffer, T., Heil, M., Kostin, S., Breier, G., Mehling, T.,

Haberberger, R., Clauss, M., Gaumann, A., Schaper, W., and Seeger, W.

(2003). Circulating vascular progenitor cells do not contribute to compensatory

lung growth. Circ. Res. 93, 372–379.

Weissmann, N., Dietrich, A., Fuchs, B., Kalwa, H., Ay, M., Dumitrascu, R.,

Olschewski, A., Storch, U., Mederos y Schnitzler, M., Ghofrani, H.A., et al.

(2006). Classical transient receptor potential channel 6 (TRPC6) is essential

for hypoxic pulmonary vasoconstriction and alveolar gas exchange. Proc.

Natl. Acad. Sci. USA 103, 19093–19098.

Wink, D.A., and Mitchell, J.B. (1998). Chemical biology of nitric oxide: Insights

into regulatory, cytotoxic, and cytoprotective mechanisms of nitric oxide. Free

Radic. Biol. Med. 25, 434–456.

Wright, J.L., and Churg, A. (1990). Cigarette smoke causes physiologic and

morphologic changes of emphysema in the guinea pig. Am. Rev. Respir.

Dis. 142, 1422–1428.

Wright, J.L., and Churg, A. (1991). Effect of long-term cigarette smoke expo-

sure on pulmonary vascular structure and function in the guinea pig. Exp.

Lung Res. 17, 997–1009.

Wright, J.L., Tai, H., and Churg, A. (2006). Vasoactive mediators and pulmo-

nary hypertension after cigarette smoke exposure in the guinea pig. J. Appl.

Physiol. 100, 672–678.

Wright, J.L., Zhou, S., Preobrazhenska, O., Marshall, C., Sin, D.D., Laher, I.,

Golbidi, S., and Churg, A.M. (2011). Statin reverses smoke-induced pulmonary

hypertension and prevents emphysema but not airway remodeling. Am.

J. Respir. Crit. Care Med. 183, 50–58.

Yoshida, T., and Tuder, R.M. (2007). Pathobiology of cigarette smoke-induced

chronic obstructive pulmonary disease. Physiol. Rev. 87, 1047–1082.

Yoshida, T., Mett, I., Bhunia, A.K., Bowman, J., Perez, M., Zhang, L., Gand-

jeva, A., Zhen, L., Chukwueke, U., Mao, T., et al. (2010). Rtp801, a suppressor

of mTOR signaling, is an essential mediator of cigarette smoke-induced

pulmonary injury and emphysema. Nat. Med. 16, 767–773.

Zuckerbraun, B.S., Chin, B.Y., Wegiel, B., Billiar, T.R., Czsimadia, E., Rao, J.,

Shimoda, L., Ifedigbo, E., Kanno, S., and Otterbein, L.E. (2006). Carbon

monoxide reverses established pulmonary hypertension. J. Exp. Med. 203,

2109–2119.

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. 305

Supplemental Information

EXTENDED EXPERIMENTAL PROCEDURES

In Vivo HemodynamicsAll animals were anesthetized with ketamine (60 mg/kg body weight) and xylazine (10 mg/kg body weight) and anticoagulated with

heparin (1000 U/kg body weight) intraperitoneally. The trachea was cannulated, and the lungs were ventilated with room air at a tidal

volume of 200 ml with 150 breaths per minute. The animals were kept at physiological body temperature throughout the experiment.

Right ventricular systolic pressure was measured by inserting a PE-80 tube into the right ventricle via the right jugular vein as

described previously (Dumitrascu et al., 2006; Fink et al., 2002).

Alveolar and Vascular Morphometry (Mean Linear Intercept, Air Space, Septal-Wall Thickness, Degreeof Muscularization)For alveolar morphometry, lungs were fixed with 4.5% paraformaldehyde in phosphate-buffered saline (pH 7.0) via the trachea at

a pressure of 22 cm H2O. For vascular morphometry, Zamboni’s fixative was infused through the pulmonary artery after flushing

the lungs with saline at a vascular pressure of 22 cm H2O and a tracheal pressure of 12 cmH2O. Investigations were performed using

3 mm sections of paraffin-embedded lungs.

The mean linear intercept, mean air space, and mean septal wall thickness were measured after staining with hematoxylin and

eosin (HE). Total scans from each lung lobe were analyzed according to the procedure previously described (McGrath-Morrow

et al., 2004; Woyda et al., 2009), which was implemented into the Qwin software (Leica, Wetzlar, Germany). Horizontal lines (distance

40 mm) were placed across each lung section. The number of times the lines cross alveolar walls was calculated by multiplying the

length of the horizontal lines and the number of lines per section then dividing by the number of intercepts. Bronchi and vessels above

50 mm in diameter were excluded prior to the computerizedmeasurement. The air space was determined as the nonparenchymatous

nonstained area. The septal wall thickness was measured as the length of the line perpendicularly crossing a septum. From the

respective measurements, mean values were calculated.

The degree of muscularization of pulmonary arterial vessels was determined as described (Dumitrascu et al., 2006; Rabinovitch

et al., 1981) from lung/paraffin sections stained with a 1:900 diluted a-smooth muscle actin antibody (clone 1A4, Sigma-Aldrich, Mu-

nich, Germany) to identify a-smooth muscle actin-positive cells and a 1:900 dilution of the antihuman vonWillebrand-factor antibody

to allow identification of vessels (Dako, Hamburg, Germany). Morphometric quantification was carried out microscopically using the

Qwin software (Leica, Wetzlar, Germany). Vessels were categorized as fully muscularized (>70% vessel circumference a-smooth

muscle actin positive), partially muscularized (5% to 70% vessel circumference a-smooth muscle actin positive) and nonmuscular-

ized (<5% vessel circumference a-smooth muscle actin positive). One hundred pulmonary arteries (85 vessels of a diameter of 20–

70 mm, 10 vessels of a diameter of 71–150 mm, and 5 vessels of more than 150 mm diameter) were analyzed from each lung lobe in

a blinded fashion. The degree of muscularization is given as percentage of total vessel count.

In Vivo Lung-Function MeasurementMice were anesthetized with ketamine (60 mg/kg body weight) and xylazine (10 mg/kg body weight) intraperitoneally and anticoa-

gulated with heparin (1000 U/kg). The trachea was cannulated, and the lungs were ventilated with 10 ml/g body weight at a rate of

150 breaths per minute and a positive endexpiratory pressure of 2.0 cm H2O was used maintained at a physiological temperature

throughout the experiment. A tracheal cannula was connected to a pneumotachometer (Hugo Sachs Electronics, March-Hugstetten,

Germany) and the dynamic compliance was evaluated at the living mouse using the HSE PULMODYN software (Hugo Sachs Elec-

tronics, March-Hugstetten, Germany).

Isolated, Perfused Mouse Lung ExperimentsIsolated mouse lung perfusion was performed in a water-jacketed chamber (type 839, Hugo Sachs Elektronik, March-Hugstetten,

Germany). Deeply anesthetized and anticoagulated animals were intubated via a tracheostomy and ventilated with room air (positive

pressure ventilation 250 ml tidal volume, 90 breaths/min and 2 cm H2O positive end-expiratory pressure). A midsternal thoracotomy

was followed by an insertion of catheters into the pulmonary artery. Lungs were perfused with Krebs–Henseleit buffer (120mMNaCl,

4.3 mM KCl, 1.1 mM KH2PO4, 2.4 mM CaCl2, 1.3 mM MgCl2, and 13.32 mM glucose) as well as 5% [w/v] hydroxyethylamylopectin

(molecular weight 200,000) as an oncotic agent; NaHCO3 was adjusted to result in a constant pH of 7.37–7.40) at a flow rate of

2 ml/min using a peristaltic pump (ISM834A V2.10, Ismatec, Glattbrugg, Switzerland). In parallel to onset of perfusion, the ventilation

was changed from room air to a premixed normoxic gas (21% O2, 5.3% CO2, balanced with N2). After rinsing the lungs withR20 ml

buffer, the perfusion circuit was closed for recirculation and the left arterial pressure was set at 2.0 mmHg. Meanwhile, the flow was

slowly increased from 0.2 to 2 ml/min and the entire system was heated to 37�C. The pressure in the pulmonary artery and in the left

ventricle was registered via catheters.

The artificial thorax was closed and the lungs were ventilated with negative pressure ranging from�2 cmH2O to�12 cmH2O. The

end-expiratory pressure was kept constant at �2 cm H2O. The tidal volume, pulmonary resistance, and dynamic lung compliance

was calculated using the HSE Pulmodyn program (Hugo Sachs Elektronik, March Hugstetten, Germany) (Held and Uhlig, 2000).

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. S1

Isolation of Murine Endothelial Cells and CultureLung EC were isolated from wild-type mice. Lung-tissue slices of �1 mm were washed in Hanks’ balanced salt solution (HBSS) to

remove any remaining blood. Minced tissues were transferred to sterile 50 ml tubes containing 20 ml HBSS without Ca2+ and Mg2+

and dispase (1 U/ml) and digested for 60min at 37�C. Residual tissuewas removed by a 70 mmmesh filter. Afterwards the suspension

was centrifuged at 290 g and cells were washed in 5 ml HBSS + 0.5% bovine serum albumin (BSA). Cells were resuspended with

800 ml HBSS/BSA and 25 ml (4 3 105 beads/ml) Dynabeads M-450 (Sheep anti-rat IgG; Invitrogen, Karlsruhe, Germany) coated

with 1.5 mg antimouse CD144 antibody produced in the rat (BD PharMingen, Heidelberg, Germany). After incubation for 30 min,

unbound cells were removed by washing five times with HBSS/BSA using a magnetic device (Dynal MPC-1; Invitrogen). Cells

were plated on fibronectin-coated cell-culture dishes, cultured in low-serum endothelial cell-growth medium (C-22020, Promocell

Heidelberg, Germany) supplemented with 15% fetal calf serum, and allowed to grow to confluency for 7–10 days. Cells were then

washed with HBSS/BSA and further purified by adding CD144-coated Dynabeads M-450 in HBSS/BSA for 60 min. After trypsiniza-

tion, unbound cells were removed as described above. The purity was confirmed by staining with an antibody against vonWillebrand

factor (#1284924; Boehringer Mannheim, Mannheim, Germany).

Isolation of Primary Murine Alveolar Type II CellsPrimary AECII cells were isolated fromwild-type mice. Lungs were lavaged twice with 1 ml sterile PBS and tissues digested with dis-

pase and minced. The suspension was sequentially filtered through 100, 20, and 10 mm nylon meshes and centrifuged at 200 g for

10 min. The pellet was resuspended in Dulbecco’s modified eagle medium (Invitrogen, Karlsruhe, Germany), and negative selection

for lymphocytes/macrophages was performed by incubation on CD16/32- and CD45-coated Petri dishes for 30 min at 37�C. Nega-tive selection for fibroblasts was performed by adherence for 45 min on cell-culture dishes. Cell purity and viability were analyzed in

freshly isolated AECII cells directly after isolation. Cell purity was assessed by epithelial cell morphology and immunofluorescence

analysis with pro-SPC (positive) as well as a-SMA andCD45 (both negative) of cytocentrifuge preparations of AECII cells. Cell viability

was measured by trypan blue exclusion. AECII cells used throughout this study demonstrated 95% ± 3% purity and greater than

97% viability. Finally, AECII cells were suspended in DMEM plus 10% FCS, 2 mM L-glutamine, 100 U/ml penicillin, and 100 g/ml

streptomycin and cultured for 24 hr to allow attachment. Phenotypic characterization was done after this time period. After media

change, cells were cultured for a maximum of 2 days in a humidified atmosphere of 5% CO2 at 37�C.

Pulmonary Arterial Smooth Muscle Cell Isolation and CulturePulmonary arterial smooth muscle cells (PASMC) were isolated from precapillary vessels modified from a previously reported

protocol (Kuzkaya et al., 2003; Marshall et al., 1996). Primary cells were cultured for 3 to 5 days without passaging. Smooth muscle

cells were identified by immunohistochemical staining with smooth muscle cell-specific a-actin and myosin antibodies and exam-

ining their morphology. The absence of endothelial cells was confirmed by staining with an antibody directed against vonWillebrand

factor.

Generation of Reciprocally Bone Marrow-Transplanted Chimeric MiceChimeric mice were generated as previously described (Voswinckel et al., 2003) with modifications. Donor mice (8 to 10 weeks old)

were anticoagulated with an intraperitoneal injection of heparin (500 U), then euthanized by isoflurane. After dissection of the femora

and tibiae, the muscles were peeled off with sterile cotton swabs and the bones were put into cold medium on ice. Under sterile

conditions, bone ends were cut off. The bone shafts were then flushedwith a 24G needle on a 5ml syringe with RPMI 1640 containing

1% fetal calf serum, 100 U/ml penicillin, and 1000 U/ml streptomycin, directly into 15 ml Falcon tube. The harvested BM was centri-

fuged for 5 min at 4003 g at 4�C. The supernatant was discarded and the pellet was resuspended in 1ml medium by repeated pipet-

ting and then transferred into a new Falcon tube through 100 mm nylon mesh. The cell strainer was rinsed with 1 ml medium. After-

wards, the suspension was centrifuged for 5 min at 400 3 g at 4�C, the supernatant was discarded and the pellet was again

resuspendedwith 1mlmediumby repeated pipetting to produce a single-cell suspension. Finally, the cells were counted using aNeu-

bauer hemocytometer. About 5 3 107 mononuclear cells were harvested from each donor.

Recipient mice (8 to 10 weeks old) were BM depleted by lethal irradiation with 60Co irradiation with 5.5 Gray two times (second

irradiation after 3 hr). Within 6�8 hr of irradiation, 2�5 3 106 donor BM cells in a 150 ml sterile medium were injected into the lateral

tail vein of the warmed recipient. BM reconstitution was allowed to occur for 6 weeks. Chimeric mice were housed individually in

sterile filter-isolator cages. Chimeras received Baytril 2.5% (enrofloxacin), 10 mg/l; Bayer, Leverkusen, Germany) in sterile drinking

water for 2 weeks.

Three different groups of chimeras were generated. A control for BM transplantation procedures where irradiated WT animals

received BM cells from WT mice (WT-to-WT [WT/WT] chimeras). In iNOS�/�-to-WT (iNOS�/�/WT) chimeras BM cells from

iNOS�/� mice were transplanted into irradiated WT mice. In WT-to- iNOS�/� (WT/iNOS�/�) chimeras, BM cells harvested from

WT mice were transferred into irradiated iNOS�/� mice. After BM reconstitution, each group was randomly allocated to controls

(= not exposed to tobacco smoke) and smoke-exposed mice.

Success of BM transplantation was verified by PCR using genomic DNA obtained from blood cells or vessel wall cells with primers

specific for WT and iNOS�/� animals. Blood (60 ml) was obtained from the heart of mice and expelled immediately into a 1.5 ml

S2 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.

microfuge tube containing 20 ml of 10 mM EDTA. Blood was stored at �20�C until processing. The genomic DNA was isolated using

DNeasy Blood and Tissue Kit (QIAGEN, Hilden, Germany).

Primers:

oIMR1216: 50-ACATGCAGAATGAGTACCGG-30

oIMR1217: 50-TCAACATCTCCTGGTGGAAC-30

oIMR1218: 50-AATATGCGAAGTGGACCTCG-30

Primer oIMR1216 is a common primer. Primer oIMR1217 amplifies 108 base pair (bp) fragment from the wild-type allele. Primer

oIMR1218 amplifies a 275 bp fragment from the disrupted allele.

PCRwas performed in a 25 ml reactionmixture using a PCR kit (Peqlab, Erlangen, Germany) containing 0.5 ml dNTPs, 18.25 ml H2O,

2.5 ml Buffer S, 0.25 ml Taq polymerase, 0.5 ml of each primer, and 2 ml DNA. The denaturing temperature was 94�C for 3 min, and the

cycling parameters were 94�C for 30 s, 59�C for 30 s, and 72�C for 30 s (35 cycles) and then 72�C for 2 min.

Volume-Related (Design-Based) Stereology for Calculation of the Total Number of AlveoliLungs were flushed blood-free with saline via the pulmonary artery and fixed in the chest by intratracheal infusion of 4.5% formal-

dehyde solution at 15 cm H2O and 20 cm H2O via the pulmonary artery for 20 min. For both stereology (right lung) and vascular

morphometry (left lung, vascular morphometry was performed from the extra set of experiments for stereological measurements

from corresponding lungs), lungs were isolated and incubated in formaldehyde overnight. Fixed lungs were transferred to 0.1 M

phosphate-buffered saline the following day. Lung volume was assessed by water-displacement method. Subsequently, the entire

right lung was cut into commensurate parallel slices (�3 mm each) then dehydrated and the pieces were embedded next to each

other in paraffin. After staining of 3 mm sections with resorcin/fuchsin and nuclear fast red (Weigert’s elastin staining), measurement

of the number of alveoli was performed on two alternate sections of 3 mm thickness, with a 3 mmdistance from each other, whichwere

placed adjacent on the same slide and examined using a light microscope (Leica, Wetzlar, Germany) equipped with newCast soft-

ware for stereology (Visiopharm, Aarhus, Denmark). The physical dissector method was used for counting as previously described

(Ochs et al., 2004). Respective counts were related to the right-lung volume.

Localization of eNOS, iNOS, and NitrotyrosineLung sections (10 mm) were fixed in ice-cold acetone and methanol (1:1), and blocked with 3% (w/v) bovine serum albumin (BSA) in

phosphate-buffered saline (PBS) for 1 hr, followed by an overnight incubation with an 1:50 dilution of anti-eNOS (BD Biosciences,

Heidelberg, Germany) or 1:100 dilution of anti-iNOS (Abcam, Cambridge, UK) antibody, diluted in PBS with 3% (w/v) BSA. Indirect

immunofluorescence was obtained by incubation for 90 min with a 1:500 dilution of Alexa Fluor 555 conjugated antirabbit antibody

(Invitrogen, Karlsruhe, Germany) in BSA. Nuclear counterstaining was performed with Hoechst-33258 (1:10000 dilution in PBS; In-

vitrogen, Karlsruhe, Germany) for 10 min.

The expression of nitrotyrosine was assessed on 3 mm, paraffin-embedded lung sections in animal and human lung-tissue

samples. After heating at 61�C, lung sections were deparaffinized in xylene and rehydrated. The endogenous peroxidase activity

was quenched with 3% (v/v) H2O2 in methanol. For staining, a 1:250 dilution of antinitrotyrosine antibody (rabbit antinitrotyrosine;

Sigma-Aldrich, Munich, Germany) was used. Subsequently the immune complexes were visualized with a peroxidase-conjugated

secondary antibody (Vector labs, LINARIS, Wertheim-Bettingen, Germany). An additional methyl green counterstaining of the

sections was performed.

Nonisotopic In Situ Hybridization Combined with Immunofluorescence on Mouse Lung SectionsThe generation of single-stranded digoxigenin (DIG)-labeled riboprobes for nonisotopic in situ hybridization was done by the in vitro

transcription method. For the generation of the probes, the template was amplified by nested PCR out of cDNA from lung homog-

enate using the following primers:

NISH iNOS_F: 50-GCCCCTGGAAGTTTCTCTTC-30

NISH iNOS_R: 50-ACCACTCGTACTTGGGATGC-30

NISH iNOS (F)_T3: 50-AATTAACCCTCACTAAAGGTTCCAGAATCCCTGGACAAG-30

NISH iNOS (R)_T7: 50-TAATACGACTCACTATAGGTGCTGAAACATTTCCTGTGC-30

NISH eNOS_F: 50-AAGTGGGCAGCATCACCTAC-30

NISH eNOS_R: 50-GTCCAGATCCATGCACACAG-30

NISH eNOS (F)_T3: 50-AATTAACCCTCACTAAAGGCTTCAGGAAGTGGAGGCTGA-30

NISH eNOS (R)_T7: 50-TAATACGACTCACTATAGGAGTAACAGGGGCAGCACATC-30

In brief, 1 mg of the purified PCR-amplified template containing T3 and T7 RNA polymerase promoter sequences was mixed with

2 ml digoxigenin-11-uridine triphosphate (Roche, Mannheim, Germany), 4 ml of 5x transcription buffer (Promega, Mannheim,

Germany), 1 ml of RNasin (Peqlab, Erlangen, Germany) and 2 ml of T3 or T7 Phage polymerase (Promega, Mannheim, Germany) in

a total reaction volume of 20 ml. The reaction mixture was incubated at 37�C for 2 hr. The RNA probes were purified with a PCR

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. S3

purification kit (QIAGEN, Hilden, Germany). The nonisotopic in situ hybridization was performed on 8 mm-thick Tissue Tek-embedded

(Sakura Finetek, Staufen, Germany) mouse lung cryostat sections as previously described (Mittal et al., 2007).

Laser-Assisted MicrodissectionCryosections (10 mm) of Tissue Tek-embedded (Sakura Finetek, Staufen, Germany) lung tissue were mounted on membrane-coated

glass slides. After hemalaun staining for 45 s,mouse lung sections were subsequently immersed in water, 70%ethanol, 96%ethanol,

and then stored in 100% ethanol until use. Human lung cryosections were stained with hematoxylin for 10 s, washed with water,

stained with eosin (1:5) for 20 s and subsequently immersed in water, 70% ethanol, 96% ethanol, and stored in 100% ethanol until

use. Intrapulmonary arteries with a diameter of 50–100 mm alveolar septa, and bronchi (diameter 140–300 mm) were selected and

microdissected under optical control using the Lasermicrodissection device LMD6000 (Leica, Wetzlar, Germany). Afterwards, the

microdissected material was collected in Eppendorf tubes filled with 25 ml RNA lysis buffer. After collection, the dissected material

was spun down in additional 275 ml of RNA lysis buffer, vortexed, and directly frozen to store in liquid nitrogen until analysis as previ-

ously described (Fink et al., 2002).

RNA Isolation, Preamplification, cDNA Synthesis, and Real-Time PCRTotal messenger RNAwas extracted frommurine and frozen human lung tissue andmicrodissected vessels by using an RNeasyMini

orMicro Kit (QIAGEN, Hilden, Germany) according to themanufacturer’s instructions. Isolated RNA frommicrodissection was subse-

quently preamplified with a modified Quick Amp Labeling Kit (Agilent, Boblingen, Germany). PCR with reverse transcription was per-

formed with 1mg of RNA each, using the iScript cDNA Synthesis Kit (BioRad, Munich, Germany). The condition for the reverse tran-

scription was as follows: 1 cycle at 25�C for 5 min; 1 cycle at 42�C for 30 min; 1 cycle at 85�C for 5 min.

Real-time PCR was performed with the iQ SYBR Green Supermix according to the manufacturer’s instructions (BioRad, Munich,

Germany). In brief, a 25 ml mixture was used containing 12.5 ml iQ SYBR Green Supermix, 0.5 ml forward and reverse primer, 9.5 ml

sterile water, and 2 ml of the 1:5 diluted complementary DNA template. A negative control (nontemplate control) was performed in

each run. The real-time PCR was performed with a Mx3000P (Stratagene, Heidelberg, Germany) under the following conditions: 1

cycle at 95�C for 10 min, then 40 cycles at 95�C for 10 s, 59�C for 10 s, 72�C for 10 s, followed by a dissociation curve. The

intron-spanning primers were designed by using sequence information from the NCBI database. The Ct values were normalized

to the endogenous control (Porphobilinogen deaminase, PBGD).

PBGD_mouse_F: 50-GGGAACCAGCTCTCTGAGGA-30

PBGD_mouse_R: 50-GAATTCCTGCAGCTCATCCA-30

iNOS_mouse_F: 50-TGATGTGCTGCCTCTGGCT-30

iNOS_mouse_R: 50-AATCTCGGTGCCCATGTACC-30

eNOS_mouse_F: 50-ACACAAGGCTGGAGGAGCTG-30

eNOS_mouse_R: 50-TGGCATCTTCTCCCACACAG-30

PBGD_hum_F: 50-CCCACGCGAATCACTCTCAT-30

PBGD_hum_R: 50-TGTCTGGTAACGGCAATGCG-30

iNOS_hum_F: 50-ATGAGGAGCAGGTCGAGGAC-30

iNOS_hum_R: 50-CTGACATCTCCAGGCTGCTG-30

eNOS_hum_F: 50-ACCTCGTCCCTGTGGAAAGA-30

eNOS_hum_R: 50-CCTGGCCTTCTGCTCATTCT-30

Western BlotFrozen mouse and human lung-tissue samples were homogenized and cells were scraped in RIPA buffer, containing 1 mM sodium

vanadate, 0.1 mM phenylmethylsulphonyl fluoride (PMSF), 40 ml/ml protease-inhibitor mix complete (253 stock solution, Roche,

Mannheim, Germany) and 30 ml/ml b-mercaptoethanol. Subsequently the samples were centrifuged for 10 min at 8000 g. The super-

natant (containing 43 LDS loading buffer) was heated at 99�C for 10min and equal amounts of protein were loaded on an 8%or 12%

SDS polyacrylamide gel. The proteins were transferred to a polyvinylidene fluoride membrane (Pall Corporation, Dreieich, Germany)

by the semidry-blottingmethod. Themembranewaswashed for 5min with wash buffer (20mMTris, pH 7.5, 150mMNaCl, 0.1% [v/v]

Tween 20) and subsequently blocked in 6% (w/v) nonfat dry milk powder dissolved in wash buffer at room temperature. Incubation

with a diluted primary antibody (iNOS_ab3523 1:2000, Nitrotyrosine_ab7048 1:1000, both Abcam, Cambridge, UK; eNOS_610298

1:1000, BD Biosciences, Heidelberg, Germany; b-actin_A5316 1:30000, Sigma-Aldrich, Munich, Germany; phospho-SAPK/

JNK_9251 1:1000, SAPK/JNK_9252 1:1000, (phospho-)ERK 1:1000, all Cell Signaling, Danvers, MA, USA; (phospho-)Src 1:1000,

Epitomics, Burlingame, USA; Rtp801 1:250, Abnova, Heidelberg, Germany) was performed at 4�C overnight. After washing several

times with wash buffer, a horseradish peroxidase-conjugated secondary antibody (anti-rabbit_W401B and anti-mouse_W402B,

respectively; Promega, Mannheim, Germany) was applied for 1 hr at room temperature. After washing the membrane, visualization

was carried out using the enhanced chemiluminescence kit (ECL, Amersham, Braunschweig, Germany) and X-ray photo film (Kodak,

Stuttgart, Germany).

S4 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.

Quantification of Lung Elastin by Image AnalysisThe quantification of elastin (anti-elastin, Sigma-Aldrich, Munich, Germany) from paraffin sections was performed using the macro

application of Leica QWin software for cell counting at a magnification of 403. Two types of quantification were performed in parallel

(the thirdmethod used is described in the following paragraph). For each slide, 153–527 images, depending on the size of the section,

were taken. Each image covered an area of 0.07598346 mm2. For each image the number of intact elastin fibers and the amount of

intact fibers per areawas determined. Scoringwas done according to the following scale: 0 = no intact fibers; 1 = low number of intact

fibers; 2 = medium number of intact fibers; and 3 = high number of intact fibers. The number of intact fibers or the score per area was

calculated according to the equation:

intact fibers=score=

Pintact fibers=scores

Pimages3 0:07598346mm2

Confocal Microscopy and Quantification of Elastin ImmunolabelingCryosections, 30 mm thick, were air-dried and fixedwith 4%paraformaldehyde and then incubatedwith 1%bovine serum albumin for

30min to block nonspecific binding sites. After rinsing in PBS, the samples were incubated overnight with primary antibodies against

elastin (Sigma-Aldrich) and a-smooth muscle actin (Sigma-Aldrich). Tissue sections were examined by laser scanning confocal

microscopy (Leica TCS SP2). Series of confocal optical sections were taken using a Leica Planapo 340/1.00 or 363/1.32 objective

lens. Each recorded image was taken using multichannel scanning and consisted of 1024 3 1024 pixels. To improve image quality

and to obtain a high signal-to-noise ratio, each image from the series was signal-averaged and was deconvoluted using AutoQuant

X2 (Bitplane, Zurich, Switzerland) software. For three-dimensional image reconstructions, an Imaris 6.3.1 multichannel-image pro-

cessing software (Bitplane, Zurich, Switzerland) was used.

For quantification of the elastin, all tissue samples were immunolabeled simultaneously with identical conditions of fixation and

dilutions of primary and secondary antibodies. Ten random fields of vision were quantified using three-dimensional Quantification

and VoxelShop options of Imaris 6.3.1 (Bitplane, Zurich, Switzerland). The area of specific labeling for elastin was calculated as

a percentage of positive labeling per tissue area.

Arterial pO2 and COHbThe different parameters were determined from wild-type mice after 3 or 8 months of tobacco-smoke exposed and in unexposed

controls. Arterial blood was drawn from the warmed tail artery and the analysis was performed with a Bayer 348 Blood Gas Analyzer

(Siemens Healthcare Diagnostics GmbH, Eschborn, Germany). For smoke-exposed mice the values were determined during smoke

exposure at the end of each exposure period.

O2 and CO Measurements in the Smoke-Exposure ChamberGas concentration measurements were performed using a Drager Pac III (Drager, Lubeck, Germany) gas monitor equipped with

sensors for oxygen or carbon monoxide. The monitor was placed in the animal cages either under control conditions (room air) or

in the smoke-exposure chamber. Measurements were performed during the entire smoke-exposure period did not differ during

the course of smoke exposure.

Sample Preparation for NOMice were anesthetised with ketamin/xylazine and bronchoalveolar lavage (BAL) was performed by flushing the lung 10 times with

700–1000 ml 0.9% NaCl (Laboratori Diaco Biomedicali, Italy). BAL fluid was centrifuged at 320 3 g for 10 min. Supernatant was dis-

carded and the pellet was washed two times in Krebs HEPES buffer (Noxygen Science Transfer & Diagnostics GmbH, Elzach, Ger-

many). The number of cells was counted and adjusted to �20 000 cells per sample.

To obtain lung-tissue lysate for iNOS activity, lung tissue was homogenized after BAL by a Precellys 24 (Peqlab Biotechnologie

GmbH, Erlangen, Germany) homogenizer in presence of two volumes of Krebs HEPES buffer (Noxygen Science Transfer & Diagnos-

tics GmbH) containing proteinase inhibitor cocktail Complete 40 ml/ml (Roche, Mannheim, Germany). Afterwards, the samples were

centrifuged twice at 4�C, 600 g, 10 min and the supernatant was taken for analysis. Protein concentration in the tissue lysate was

determined by Bradford assay (BioRad, Munich, Germany).

iNOS Activity Measurement from Homogenized Lung TissueiNOS activity measurement was performed using the Ultrasensitive Colorimetric NOSAssay according to themanufacturer’s instruc-

tions (Oxford Biomedical Research, Oxford, MI, USA). Briefly, lung lysate (50 mg protein) was incubated at 37�C for 6 hr in presence of

NOS cofactors and absence of Ca2+. Afterwards, samples were incubated with nitrate reductase to convert nitrate to nitrite, followed

by incubation with sulfanilamide N-(1-naphthyl) ethylenediamine dihydrochloride for 5 min. The absorbance values were read at

540 nm. The nitrite concentration in experimental samples was calculated by the calibration curve. Reactions were run in parallel

in the presence and absence of L-NG-Nitroarginine (L-NNA). The L-NNA-inhibited portion of the obtained signal represents the

iNOS activity.

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. S5

Measurement of NO Release from BAL CellsMeasurement of NO was performed as previously described (MacArthur et al., 2007) from BAL cells with modifications. For NO

measurement,�20 000 cells were incubated at 37�C for 2 hr in 300 ml of Krebs HEPES buffer (Noxygen Science Transfer & Diagnos-

tics GmbH). Reactions in presence of L-NG-Nitroarginine (400 mM) were run in parallel. Afterwards, samples were centrifuged and

supernatant was collected. 150 ml of supernatant were incubated with nitrate reductase (10 mU), NAPDH (200 mM) at 22�C for

20 min. 10 ml of the product were subjected to a gas purge vessel system containing potassium iodide (1.1%), iodine (0.72%), glacial

acetic acid (77.8%) as a reducing agent. Generated gaseous NO was measured by a Sievers NOA 280 (Seeheim, Germany) NO

analyzer. Data analysis was done using the NoaWin 32 software (DeMeTec, Langgons, Germany). The L-NNA-inhibited portion of

the measurement was displayed as NO production from the cells.

Counting of Alveolar Macrophages in the Lung SectionsMacrophage counting was done in 3 mm sections in paraffin-embedded lungs. After heating at 61�C, lung sections were deparaffi-

nized in xylene and rehydrated. The endogenous peroxidase activity was quenched with 3% (v/v) H2O2 in methanol. For staining,

a 1:200 dilution of macrophage marker F4/80 antigen-specific antibody (rat monoclonal to F4/80 [ab16911], Abcam, Cambridge,

UK) was used. Immune complexes were visualized with a peroxidase-conjugated secondary antibody (Vector labs, LINARIS, Wer-

theim-Bettingen, Germany). Hematoxylin solution was used for counterstaining of the sections. The quantification of macrophages

was carried out microscopically using a Qwin macro program from Leica. For each lung lobe, the total number of macrophages was

counted. Data were calculated as a percentage of respective control groups.

FACS Analysis of Inflammatory CellsBronchoalveolar Lavage

Heparinized (1000 U/kg) animals were euthanized with ketamine and xylazine. The trachea was cannulated and the lung was lavaged

five times with 1 ml phosphate-buffered saline (PBS). The BAL was centrifuged for 10 min at 338 g, 4�C. Supernatant was discarded

and cells were fixed with 1% PFA. Cells were resuspended in 1 ml MACS Buffer and stored at 4�C until FACS analysis. The total

number of cells in BAL and remaining lung homogenate was determined using the CASY1 Cell Counting System (Scharfe Systems,

Reutlingen, Germany).

Lung Homogenate

After thoracotomy the pulmonary vasculature was flushed free of blood with saline. The left lobe was removed and homogenized.

2.5 ml of digestion medium (1 mM Na-Pyruvat, 10% FCS, 2 mM L-Glutamin, 1 mM Na-Pyruvat, 1% Pen/Strep, 30 mg/ml DNase I

[Sigma-Aldrich]; 0.7 mg/ml Collagenase A [Roche, Penzberg, Germany]) in RPMI 1640) was added and digestion was performed

by a 1 hr incubation at 37�C. The suspension was given through a 100 mm cell strainer (BD, Heidelberg, Germany) and centrifuged

for 10 min at 338 g, 4�C. Supernatant was discarded and cells were fixed with 1% PFA. Cells were resuspended in MACS Buffer and

stored at 4�C.FACS Analysis

Differential cell analysis was performed by staining with antibodies to cell-type-specific cell-surface antigens and subsequent flow

cytometric analysis as described in detail elsewhere (Garn et al., 2006).

The following antibodies with the given specificities and labels were used: FITC anti-mouse CD11c, Pacific Blue anti-mouse

CD11c, PE/Cy7 anti-mouse Ly-6G/Ly-6C (Gr-1), APC/Cy7 anti-mouse I-A/I-E, PE anti-mouse CD69 (all Biolegend, Uithoorn, The

Netherlands), V450 Rat anti-Mouse CD4 (BD Biosciences, Heidelberg, Germany), Alexa Fluor 647 anti-mouse F4/80, APC anti-

mouse F4/80 (Caltag, Burlingame, USA), FITC anti-mouse CD45 (BD PharMingen, Heidelberg, Germany).

The following isotype control antibodies were applied: FITC Armenian Hamster IgG, Pacific Blue Armenian Hamster IgG, PE Arme-

nian Hamster IgG, PE/Cy7 Rat IgG2b/k, APC Rat IgG2a/k, APC/Cy7 Rat IgG2b/k, Alexa Fluor 647 Rat IgG2a/k, Pacific Blue Rat

IgG2b/k (all Biolegend, Uithoorn, The Netherlands), FITC Rat IgG2b/k Isotype Control (e-Bioscience, Frankfurt, Germany). Flow-

cytometric analysis of stained cells was performed using a FACS Sort (BD, Heidelberg, Germany).

Measurement of Reactive Oxygen Species from Lung-Tissue Lysate and Isolated MacrophagesSuperoxide release from homogenized lung tissue and BAL cells was measured as described (Kuzkaya et al., 2003) with modifica-

tions. Briefly, 50 mg of lung-tissue lysate was incubatedwith 0.5mM1-hydroxy-3-carboxy-2,2,5,5-tetramethylpyrrolidine (CPH). After

incubation for 30 min at 37�C, samples were shock frozen. For the analysis of BAL cells a similar protocol was applied. Electronspin-

resonance (ESR) measurements were performed at 77�K using anMS 100 ESR spectrometer (Magnettech, Berlin, Germany). Values

obtained in parallel reactions with SOD (100 U/ml) were subtracted and represent the amount of superoxide.

Human Lung-Tissue Extraction and FixationThe human lung tissue was snap-frozen directly after explantation for mRNA and protein extraction or fixed in 4.5% paraformalde-

hyde or Tissue Tek (Sakura Finetek, Staufen, Germany), respectively for histology and laser-assisted microdissection.

S6 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.

SUPPLEMENTAL REFERENCES

Dumitrascu, R., Weissmann, N., Ghofrani, H.A., Dony, E., Beuerlein, K., Schmidt, H., Stasch, J.P., Gnoth, M.J., Seeger, W., Grimminger, F., and Schermuly, R.T.

(2006). Activation of soluble guanylate cyclase reverses experimental pulmonary hypertension and vascular remodeling. Circulation 113, 286–295.

Fink, L., Kohlhoff, S., Stein, M.M., Hanze, J., Weissmann, N., Rose, F., Akkayagil, E., Manz, D., Grimminger, F., Seeger, W., and Bohle, R.M. (2002). cDNA array

hybridization after laser-assisted microdissection from nonneoplastic tissue. Am. J. Pathol. 160, 81–90.

Garn, H., Siese, A., Stumpf, S., Wensing, A., Renz, H., and Gemsa, D. (2006). Phenotypical and functional characterization of alveolar macrophage subpopula-

tions in the lungs of NO2-exposed rats. Respir. Res. 7, 4.

Held, H.D., and Uhlig, S. (2000). Mechanisms of endotoxin-induced airway and pulmonary vascular hyperreactivity in mice. Am. J. Respir. Crit. Care Med. 162,

1547–1552.

Kuzkaya, N., Weissmann, N., Harrison, D.G., and Dikalov, S. (2003). Interactions of peroxynitrite, tetrahydrobiopterin, ascorbic acid, and thiols: implications for

uncoupling endothelial nitric-oxide synthase. J. Biol. Chem. 278, 22546–22554.

MacArthur, P.H., Shiva, S., and Gladwin, M.T. (2007). Measurement of circulating nitrite and S-nitrosothiols by reductive chemiluminescence. J. Chromatogr. B

Analyt. Technol. Biomed. Life Sci. 851, 93–105.

Marshall, C., Mamary, A.J., Verhoeven, A.J., and Marshall, B.E. (1996). Pulmonary artery NADPH-oxidase is activated in hypoxic pulmonary vasoconstriction.

Am. J. Respir. Cell Mol. Biol. 15, 633–644.

McGrath-Morrow, S.A., Cho, C., Soutiere, S., Mitzner, W., and Tuder, R. (2004). The effect of neonatal hyperoxia on the lung of p21Waf1/Cip1/Sdi1-deficient

mice. Am. J. Respir. Cell Mol. Biol. 30, 635–640.

Mittal, M., Roth, M., Konig, P., Hofmann, S., Dony, E., Goyal, P., Selbitz, A.C., Schermuly, R.T., Ghofrani, H.A., Kwapiszewska, G., et al. (2007). Hypoxia-depen-

dent regulation of nonphagocytic NADPH oxidase subunit NOX4 in the pulmonary vasculature. Circ. Res. 101, 258–267.

Ochs, M., Nyengaard, J.R., Jung, A., Knudsen, L., Voigt, M., Wahlers, T., Richter, J., and Gundersen, H.J. (2004). The number of alveoli in the human lung. Am. J.

Respir. Crit. Care Med. 169, 120–124.

Rabinovitch, M., Gamble, W.J., Miettinen, O.S., and Reid, L. (1981). Age and sex influence on pulmonary hypertension of chronic hypoxia and on recovery. Am. J.

Physiol. 240, H62–H72.

Voswinckel, R., Ziegelhoeffer, T., Heil, M., Kostin, S., Breier, G., Mehling, T., Haberberger, R., Clauss, M., Gaumann, A., Schaper, W., and Seeger, W. (2003).

Circulating vascular progenitor cells do not contribute to compensatory lung growth. Circ. Res. 93, 372–379.

Woyda, K., Koebrich, S., Reiss, I., Rudloff, S., Pullamsetti, S.S., Ruhlmann, A., Weissmann, N., Ghofrani, H.A., Gunther, A., Seeger, W., et al. (2009). Inhibition of

phosphodiesterase 4 enhances lung alveolarisation in neonatal mice exposed to hyperoxia. Eur. Respir. J. 33, 861–870.

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. S7

Chronic hypoxiaSmoke exposure

Apoptosis

down

no

up

Reg

ulat

ion

ProliferationExtracellular

matrix Oxidative stress Inflammation

Vegfa

Vegfc

Flt1

Fgf10

Ccna1

Vegfb

Kdr

Cdk2

Control SmokeHypoxia020406080100120 *

Alv

eoli/

vess

els

(%)

C Alveoli:vessels ratio

A

B viiiiii i

Bcl2

Fastk

Fbf1

Traf1

Tnfsf10

Bax

Pecam1

Mmp9

Mmp13

Vcam1

Timp3

Mmp12

Nox4

Noxo1

Cybb

Prdx2

Mpo

Gpx1

Noxa1

Sod2

Sod1

Ccl5

Ccl9

Ccl6

Ccl25

Cd68

Csf2

Ccl2

Il1b

Il8rb

D

0 3 80

5

10

15 **

**

Months of smoke exposure

CO

Hb

(%)

0

100

200

300

400

500 *

No smokeexposure

Smoke exposurechamber

CO

con

cent

ratio

n(p

pm)

0 3 80

20406080

100120

Months of smoke exposure

(pO

2 m

mH

g)

05

1015

20

25

No smoke exposure

Smoke exposurechamber

O2

conc

entr

atio

n(%

)

Figure S1. Characterization of Wild-Type Mice Exposed to Tobacco Smoke and Comparison with Wild-Type Mice Exposed to Chronic

Hypoxia, Related to Figure 1

(A) Lung compliance, tidal volume, and airway resistance quantified in isolated perfused lungs from WT mice after smoke exposure compared with unexposed

controls.

(B) (i) O2 concentration and (iii) carbon monoxide (CO) concentration in the smoke-exposure chamber and in the ambient air of non-smoke-exposed mice. (ii)

Arterial blood O2 partial pressure and (iv) concentration of COHb in the blood of smoke-exposed and non-smoke-exposed mice. Data are derived from at least

n = 5 mice or measurements each.

(C) Ratio of the number of alveoli to the number of vessels per area frommice exposed to chronic hypoxia (n = 5) or tobacco smoke (n = 6), related to the ambient

air controls.

(D) Changes in mRNA expression of small pulmonary vessels (diameter 50–100 mm) from mice exposed to chronic hypoxia or tobacco smoke (n = 3 each). Data

are categorized as ‘‘significant upregulation (up),’’ ‘‘significant downregulation (down),’’ or ‘‘no significant regulation (no).’’ The degree of pulmonary hypertension

induced by chronic hypoxia exposure was similar to that noted after 3 months of smoke exposure (data not given).

Data are presented as mean ± SEM. *Significant difference (p < 0.05) compared with controls.

S8 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.

BA

V

C

iii

1500

2000

2500

e (A

U)

1500

2000

2500

e (A

U)

0

500

1000

1500

0

smoke - + + ++

3 8 88

Su

pero

xid

e

0

500

1000

1500

0

smoke - + + ++

3 8 88

Su

pero

xid

e

s o e

Placebo

L-NIL curative -

-

+

-

-

+

-

-

+

+3

-

+

-

+3

Months of smoke exposure

s o e

Placebo

L-NIL curative -

-

+

-

-

+

-

-

+

+3

-

+

-

+3

Months of smoke exposure

Figure S2. Effects of Tobacco Smoke on Pulmonary Nitric Oxide Synthase Activity and Pulmonary Superoxide Generation inWild-TypeMice,

Related to Figure 2

(A) iNOS activity in homogenized lungs.

(B) NO release from BAL cells. Data were normalized to amount of protein (homogenized lungs) or number of cells (BAL) and are given for n = 4 lungs each.#Significant effect of smoke exposure compared to non-smoke-exposed controls (p < 0.05). *Significant difference between placebo and curative L-NIL

treatment (p < 0.05).

(C) Superoxide release from (i) homogenized lung tissue and (ii) BAL cells. Data were normalized to amount of protein (homogenized lungs) or number of cells

(BAL) and are given for n = 4 lungs each.

Data are presented as mean ± SEM.

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. S9

C

D

BA

i

iii

ii

iv

0 8 8 0 8 8

Months of exposure

Smoke

-L-NIL preventive

L-NIL curative +3

Placebo

-

-- -

- - -

--

-- -

+3

8

88

0 8 8 0 8 8

Months of exposure

Smoke

-L-NIL preventive

L-NIL curative +3

Placebo

-

-- -

- - -

--

-- -

+3

8

88

Co

mp

lia

nc

e (µ

l/c

m H

2O

)

preventive curative preventive curative preventive curative

Tid

al vo

lu

me

(µ

l)

Re

sis

ta

nc

e

(c

m H

2O

/µ

l/s

x

1

0-4)

Co

mp

lia

nc

e (µ

l/c

m H

2O

)

Tid

al vo

lu

me

(µ

l)

Re

sis

ta

nc

e

(c

m H

2O

/µ

l/s

x

1

0-4)

0 8 0 8

WT iNOS-/-eNOS

-/-

0 8

Months of smoke exposure Months of smoke exposure

0 8 0 8

WT iNOS-/-eNOS

-/-

0 80 8 8 0 8 8

Months of exposure

Smoke

-L-NIL preventive

L-NIL curative +3

Placebo

-

-- -

- - -

--

-- -

+3

8

88

200

300

400

∗

∗

0

*

*

*

*

0 8 0 8

WT iNOS-/-eNOS

-/-

0 8

Months of smoke exposure

Figure S3. iNOS Deficiency or Inhibition Protects against Smoke-Induced Functional Deterioration, and iNOS Inhibition Reverses

Established Emphysema, Related to Figure 3

Lung compliance, tidal volume, and airway resistance quantified in isolated perfused lungs from from (A) iNOS�/�, eNOS�/�, and wild-type (WT) mice with and

without smoke exposure, and (B) smoke-exposedWTmicewith andwithout L-NIL treatment comparedwith placebo controls. L-NIL treatment was either given in

parallel to the smoke exposure (preventive, left panels) or for 3 months after cessation of smoke exposure (curative, right panels).

(C) Number of hematopoetic cells in (i) lung homogenate after BAL and (iii) BAL from smoke-exposed mice compared with nonexposed controls, and in (ii) lung

homogenate (after BAL) and (iv) BAL from curatively L-NIL-treated mice relative to placebo-treated controls (n = 3–4 each). CD45+ cells were differentiated by

FACS by the following markers: GR1+ (granulocytes), CD11c+ (macrophages), CD11c+/MHCII+ (activated macrophages), CD4+ (T cells), CD4+/CD69+ (activated

T cells).

(D) Number of macrophages in nonlavaged lungs of smoke-exposed WT mice with and without curative L-NIL treatment compared with nonexposed controls

(n = 4 each). T cells were only occasionally detected in BAL.

Data are presented as mean ± SEM. *Significant difference compared with controls (p < 0.05).

S10 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.

A iNOS-/-

B

erusopxe ekoms fo shtnom 8erusopxe ekoms fo shtnom 0 iNOS-/-

with 8 months of

smoke exposure

8 months of smoke exposure

with curative L-NIL treatment

3 months of smoke exposure

C

* *

500 nm

8 months of smoke exposure

Smoke

iNOS-/-

Figure S4. Smoke-Induced Destruction of Elastic Fibers in the Lung Is Prevented by iNOS Deficiency and Reversed by iNOS Inhibition,

Related to Figure 4

(A) Elastin content in cryopreserved lungs (n = 3). Data are presented as mean ± SEM. *Significant difference between smoke-exposed mice and non-smoke-

exposed controls.

(B) Representative slides of elastin (green) and a-smooth muscle actin (a-SMA) (red) stained lungs. Nuclei are displayed in blue. Images of elastin staining from

control, 8 months of smoke exposure, and curatively L-NIL treated mice are identical to those of Figure 4C and are repeated for comparison with a-SMA staining.

(C) Representative electron microscopy image of elastic fibers from an iNOS�/� mouse exposed to smoke for 8 months. *Indicates elastic fibers.

Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc. S11

Figure S5. Effect of iNOS Deficiency or Inhibition on Nitrotyrosine Expression, mRNA Regulation, and Cell Proliferation in the Lungs ofSmoke-Exposed Mice, Related to Figure 6

(A) Nitrotyrosine staining (violet color) in smoke-exposed WT, iNOS�/�, eNOS�/�, and L-NIL-treated WT mice (preventive and curative approach) and nonex-

posed controls. Representative vessels (i and iii) and alveolar structures (ii and iv) are depicted. L-NIL treatment was either given in parallel to the smoke exposure

(preventive) or for 3 months after cessation of smoke exposure (curative).

(B) Comparison of mRNA expression levels in laser-microdissected vessels, septa, and bronchi from lungs of smoke-exposed, L-NIL-treated WT mice (curative

approach) and nonexposed controls (n = 3 each). Data are given as mean ± SEM. *Significant difference (p < 0.05) compared with controls.

(C) Immunostaining for the proliferation marker PCNA (red) in lung sections from smoke-exposed, L-NIL-treated WT mice (curative approach) and nonexposed

controls. Arrows indicate positive staining. Data are given for different magnifications, showing that PCNA-positive staining occurs also in the vicinity of small as

well as larger vessels.

S12 Cell 147, 293–305, October 14, 2011 ª2011 Elsevier Inc.