Inhibiting Lung Elastase Activity Enables Lung Growth
in Mechanically Ventilated Newborn Mice
Anne Hilgendorff1, 4, Kakoli Parai1, Robert Ertsey1, Noopur Jain1, Edwin F Navarro1,
Joanna L Peterson1, Rasa Tamosiuniene2, Mark R Nicolls2, Barry C Starcher3,
Marlene Rabinovitch1 and Richard D Bland1
1Department of Pediatrics, Stanford University, Stanford, CA; 2Department of Medicine,
Stanford University, Stanford, CA; 3Department of Biochemistry, University of Texas,
Tyler, TX; 4Department of Pediatrics, University of Munich, Munich, Germany
Reprint requests: Richard D Bland MD, Stanford University School of Medicine, 269
Campus Drive, CCSR 1225, Stanford, CA 94305-5162
Corresponding author: Richard D Bland MD; e-mail: email@example.com; FAX #:
650-723-6700; telephone: 650-723-8080
Author contributions: Conception and design — AH, MR, RDB; Data acquisition,
analysis and interpretation — AH, KP, RE, NJ, EFN, JLP, RT, MRN, BCS, MR, RDB;
Writing or substantial role in revising — AH, MR, RDB
Support sources: NIH Grants HL086631 (RDB) and HL086216 (KP); DFG Grant HI
1315/3-1 (AH); NIH Grant HL082662 (MRN); and the Vera Moulton Wall Center for
Pulmonary Vascular Disease at Stanford University
Running title: Elafin prevents ventilator-induced lung injury
Descriptor number: 14.3 Neonatal Lung Disease & BPD
Word count of the body of the manuscript: 4806
Online supplement: This article has an online data supplement, which is accessible
from this issue’s table of content online at www.atsjournals.org
Page 1 of 63
AJRCCM Articles in Press. Published on May 11, 2011 as doi:10.1164/rccm.201012-2010OC
Copyright (C) 2011 by the American Thoracic Society.
AT A GLANCE COMMENTARY:
Scientific Knowledge on the Subject
Mechanical ventilation with O2-rich gas (MV-O2) offers life-saving treatment for
respiratory failure, but also promotes lung injury. In neonates MV-O2 causes increased
elastase activity and disordered elastin, leading to lung growth arrest. These features
are recapitulated in newborn mice in which MV-O2 increases lung elastase activity,
causing dysregulated elastin synthesis and assembly, TGFβ activation, apoptosis and
defective alveolarization. We hypothesized that these features are functionally related
and that inhibiting lung elastin breakdown would prevent the adverse effects of MV-O2.
What this Study Adds to the Field
This study shows for the first time that intratracheal treatment with the serine elastase
inhibitor elafin, by blocking lung elastase and MMP-9 activity, protects newborn mice
from the adverse pulmonary effects of MV-O2. Inhibiting proteolytic activity prevented
degradation and dispersion of lung elastin, as well as the increased TGFβ activation,
apoptosis and inflammation noted in the lungs of vehicle-treated controls, thereby
attenuating the emphysematous changes in alveolar structure seen in control lungs
after 24h of MV with 40% O2. This study offers new insights on the pathogenesis and
potential treatment of ventilator-induced lung injury, which is a major contributor to
chronic lung disease in newborn infants, and to ARDS morbidity and mortality in older
children and adults.
Page 2 of 63
Rationale: Mechanical ventilation with O2-rich gas (MV-O2) offers life-saving treatment
for respiratory failure, but also promotes lung injury. We previously reported that MV-O2
of newborn mice increased lung elastase activity, causing elastin degradation and
redistribution of elastic fibers from septal tips to alveolar walls. These changes were
associated with TGFβ activation and increased apoptosis leading to defective
alveolarization and lung growth arrest, as seen in neonatal chronic lung disease.
Objectives: To determine if intratracheal treatment of newborn mice with the serine
elastase inhibitor elafin would prevent MV-O2-induced lung elastin degradation and the
ensuing cascade of events causing lung growth arrest.
Methods: 5d-old mice were treated via tracheotomy with recombinant human elafin or
vehicle (lactated-Ringer’s solution), followed by MV with 40%O2 for 8-24h; controls
breathed 40%O2 without MV. At study’s end, lungs were harvested to assess key
variables noted below.
Measurements and main results: MV-O2 of vehicle-treated pups increased lung
elastase and MMP-9 activity, when compared with unventilated controls, causing elastin
degradation (urine desmosine doubled), TGFβ activation (pSmad-2 tripled) and
apoptosis (cleaved-caspase-3 increased 10-fold). Quantitative lung histology showed
larger and fewer alveoli, greater inflammation, and scattered elastic fibers. Elafin
blocked these MV-O2-induced changes.
Conclusions: Intratracheal elafin, by blocking lung protease activity, prevented MV-O2
induced elastin degradation, TGFβ activation, apoptosis and dispersion of matrix
elastin, and attenuated lung structural abnormalities noted in vehicle-treated mice after
Page 3 of 63
24h of MV-O2. These findings suggest that elastin breakdown contributes to defective
lung growth in response to MV-O2 and might be targeted therapeutically to prevent MV-
O2-induced lung injury.
Word count: 248, excluding bold descriptors
Keywords: elafin; mechanical ventilation; elastin degradation; TGFβ; apoptosis; NFΚB;
alveolar septation; neonatal chronic lung disease (CLD); ventilator-induced lung injury
(VILI); bronchopulmonary dysplasia (BPD)
Page 4 of 63
Mechanical ventilation with O2-rich gas (MV-O2) offers life-saving treatment for
patients with respiratory failure. Such treatment, however, when prolonged can cause or
aggravate lung injury, leading to chronic lung disease (CLD) in neonates whose lungs
are incompletely developed, or to ventilator-induced lung injury (VILI) in older children
and adults. Neonatal CLD, a variant of what Northway et al initially described as
bronchopulmonary dysplasia (BPD) (1), is characterized by failed formation of alveoli
and lung micro-vessels, coupled with disordered lung elastin, resulting in structural and
functional abnormalities that resemble pulmonary emphysema, as seen in adults with
chronic obstructive pulmonary disease (COPD).
Elastin plays a critical role in mammalian lung development. A network of elastic
fibers within the lung helps to provide structural integrity and distensibility to conducting
airways, while enabling expansion and contraction of alveoli, pulsation of blood vessels
and elastic recoil of the surrounding matrix. Mutant mice that lack the elastin gene die
soon after birth from cardiorespiratory failure linked to smooth muscle overgrowth in
pulmonary arteries, defective airway branching and loss of alveolar septation (2, 3).
It was reported that lungs of infants who died with CLD displayed thickened, tortuous
and irregularly distributed elastic fibers in the connective tissue matrix surrounding distal
airspaces (4-6). These changes were associated with reduced secondary septation and
fewer alveoli than in lungs of infants who died without CLD. Urinary excretion of
desmosine, a biomarker of elastin degradation, increased during the first week of MV-
O2 in infants with evolving CLD (7), suggesting that scattered elastin deposition reflects
an aberrant matrix remodeling response to MV-O2. Breakdown of elastin in this disease
Page 5 of 63
has been attributed to inflammation linked to increased elastolytic activity in lung (8-10),
resulting from infection and/or hyperoxia, conditions that often complicate the course of
infants who are born very prematurely (11). Authentic animal models of CLD also exhibit
lung inflammation and increased protease activity in response to lengthy MV-O2,
resulting in lung growth arrest (12-16).
Consistent with the above clinical and experimental findings, we discovered that
newborn mice exposed to MV with 40%O2 (MV-O2) for up to 24h exhibited increased
pulmonary elastase activity, causing lung elastin degradation and remodeling of the
extracellular matrix (17). These changes were associated with activation of TGFβ and
increased apoptosis, resulting in failed formation of alveoli and pulmonary capillaries,
and scattered deposition of lung elastin (17-19). Based on previous studies showing
that mice over-expressing the human serine elastase inhibitor elafin were protected
against several forms of lung and cardiovascular injury, including hypoxia-induced
pulmonary hypertension (20), myocardial infarction (21), viral myocarditis (22) and
models of vascular injury (23, 24), we tested the hypothesis that intra-tracheal treatment
of newborn mice with recombinant human elafin, given immediately before MV-O2 for
24h, would inhibit lung elastase activity and thereby prevent the adverse effects of MV
on matrix elastin and lung growth. Elafin treatment prevented MV-O2 induced
degradation of lung elastin, influx of inflammatory cells, TGFβ activation, apoptosis, and
maldistribution of elastic fibers, and attenuated the defective formation of alveoli seen in
the lungs of vehicle-treated controls after MV-O2 for 24h. These findings, some of which
were previously reported in the form of an abstract (25), point to a mechanistic link
Page 6 of 63
between increased elastolytic activity, inflammation, maladaptive TGFβ signaling and
lung growth arrest induced by prolonged MV with O2-rich gas.
Experimental Design. We used full-term 5d-old CD-1 mice that weighed 3.4±0.5g.
For each study, littermates were randomly assigned to one of 3 treatment groups: intra-
tracheal instillation of recombinant human elafin (Proteo-Biotech-AG, Kiel, Germany),
40ng/g-bw, in 10µl/g-bw of lactated-Ringer’s (L/R), followed by MV with 40%O2 (MV-
O2); L/R (vehicle) alone, 10µl/g-bw, followed by MV-O2; or untreated, followed by
40%O2-breathing without MV.
Mice selected for MV underwent a tracheotomy after intramuscular ketamine (~60
µg/g-bw)/xylazine (~12 µg/g-bw), followed by MV-O2 at 180-breaths/min (MicroVent
848; Harvard Apparatus, Holliston, MA) for either 8h or 24h. Tidal volumes averaged 7-
8µl/g-bw. Routine care and physiological monitoring (19) are briefly described in the
online supplement. All surgical and animal care procedures/protocols were approved by
Stanford University’s IACUC.
Serine elastase and MMP-9 activity assays. Lung tissue was stored at -80°C for
later measurement of serine elastase activity using DQ-elastin substrate (16, 17, 24)
(EnzChek®, Invitrogen, Camarillo, CA), and MMP-9 activity by gelatin zymography (24),
as detailed in the online supplement.
Postmortem processing of lungs for quantitative histology. Lungs were fixed
intra-tracheally with buffered 4%-PFA overnight @ 20cmH2O (17). Fixed lungs were
Page 7 of 63
excised, their volumes measured by fluid displacement (26), then paraffin-embedded for
isotropic uniform random (IUR) sectioning. Alveolar area was measured on random
H&E-stained 4µm-sections using the Bioquant image analysis system (R&M Biometrics,
Nashville); radial alveolar counts provided an index of alveolar number (19). Relative
amount and distribution of insoluble lung elastin was assessed by quantitative image
analysis of Hart’s-stained tissue sections. The online supplement details IUR sampling,
morphometric analysis and elastin quantification.
Assessment of TGFβ β β β activation and apoptosis. To assess nuclear localization of
pSmad-2, a marker of TGFβ activation, random tissue sections were pre-treated for
antigen retrieval (Dako, Carpinteria, CA), followed by blocking serum and application of
primary antibody (rabbit anti-pSmad-2, 1:500, Cell Signaling, Boston), with overnight
incubation at 4°C. Immune complexes were visualized with the relevant peroxidase-
coupled secondary antibody using the Vectastain Kit (Vector, Burlingame, CA).
Apoptosis was detected by TUNEL assay using the ApopTag In Situ Apoptosis
Detection kit (Chemicon, Temecula, CA) applied to PFA-fixed sections. The Bioquant
image analysis system was used to quantify stained nuclei/total nuclei in 10-15 400X-
fields in 2 random sections/animal.
Immunohistochemistry for inflammatory cells. Zinc-fixed (BD-Pharmingen, San
Jose) lung sections were incubated with blocking serum, then primary rat anti-F4/80
antibody (1:400, Abcam, Cambridge, MA) to stain monocytes, or rat anti-Ly-6G (1:200,
eBioscience, San Diego) to stain neutrophils. Sections were then incubated with
biotinylated goat anti-rat antibody (1:200, Santa Cruz Biotech), followed by streptavidin-
Page 8 of 63
HRP and diaminobenzidine (Dako). Monocytes/neutrophils were counted in 20 400X-
fields in 2 random sections/animal.
RNA extraction and quantitative RT-PCR for cytokines and chemokines. RNA
was extracted from frozen lung samples, and qRT-PCR was performed using TaqMan
primer/probe sets (ABI, Foster City, CA) on a CFX384 Real Time thermal cycler (Bio-
Rad Labs, Hercules, CA) as previously described (19). ∆∆Ct analysis was used to
determine the expression level of each gene normalized to 18S using CFX384 analysis
Immunoblots for cleaved caspase-3. Lungs were frozen at -80°C for later protein
extraction and immunoblot analysis of cleaved caspase-3 (17), as detailed in the online
Nuclear extracts and immunoblots for NF-Κ Κ Κ ΚB-p65. Harvested lungs pretreated
with protease inhibitor (Pierce Biotech, Rockford, IL, cat #78410) were homogenized in
ice cold collection buffer supplied in a nuclear protein extraction kit (NE-Per Kit, Pierce
Biotech, cat #78833). Nuclear extracts, obtained according to the manufacturer’s
instructions, were incubated overnight with NF-κB-p65 primary antibody (1:700, Santa
Cruz Biotech, cat #sc-372).
ELISA for active TGFβ β β β. Lung tissue was homogenized in PBS with added protease
inhibitor (Pierce) and processed for analysis of TGFβ activity using an ELISA kit
(MB100B, R&D Systems, Minneapolis, MN) according to manufacturer’s instructions,
Additional details are in the on-line supplement.
Urinary desmosine. 24h-urine specimens were frozen for later radioimmunoassay
of desmosine/creatinine concentrations, as previously reported (27).
Page 9 of 63
Statistics. Data are mean±SD. We used 1-way ANOVA and Bonferroni’s or Dunn`s
multiple comparison test to compare results between groups. Datasets with marked
variability were compared using the Kruskal-Wallis test with Dunn’s post-hoc analysis
(28). We used Prism-5 software (GraphPad, San Diego) for statistical analysis. P<0.05
denoted significant differences.
The purpose of this study was to determine if intra-tracheal instillation of the serine
elastase inhibitor elafin would preserve matrix elastin and enable alveolar septation in
lungs of newborn mice exposed to MV-O2 for up to 24h. We applied MV with 40%O2,
rather than with air, based on earlier studies that showed a significant increase in lung
elastase activity after 8h of MV with 40%O2, but not with air (17). We did studies of 8h
duration to harvest lungs for measurement of elastase and MMP-9 activity, inflammatory
cytokine and chemokine expression, and nuclear NF-ΚB-p65 protein. Lungs harvested
at the end of 24h studies were used to assess all other endpoints. Pilot studies revealed
that pulmonary responses to MV-O2 were virtually identical in 5d-old untreated pups
exposed to MV-O2 via tracheotomy for 24h when compared to 5d-old mice treated via
tracheotomy with lactated Ringer’s solution (see online supplement, Figs E1-E6).
Elafin blocks increased lung elastase and MMP-9 activity, thereby preventing
elastin degradation and dispersion of lung elastin induced by MV-O2. MV-O2 for
8h caused a doubling of serine elastase activity in lungs of vehicle-treated mice, an
effect that was fully suppressed in pups treated with elafin (Fig 1A). Elafin treatment
Page 10 of 63
also resulted in suppression of the increased MMP-9 activity measured in lungs of
vehicle-treated mice after 8h of MV-O2 (Fig 1B). Although elastase inhibitors are not
known to suppress MMPs directly, they have been shown to block activation of the pro-
form of these enzymes and to prevent inactivation of tissue inhibitors of MMPs (29).
To determine if suppressing the increased elastase activity induced by MV-O2
prevented the breakdown of lung elastin, we assessed urinary excretion of desmosine,
a surrogate marker of elastin degradation. Elafin treatment fully suppressed the 2-fold
increase in cumulative urinary excretion of desmosine that was observed in vehicle-
treated mice after 24h of MV-O2 (Fig 1C).
To see if blocking elastin breakdown helped to preserve the normal distribution of
elastic fibers at the tips of secondary septa in lungs exposed to MV-O2 for 24h, we used
quantitative image analysis to assess the amount and distribution of elastin in lung
tissue sections treated with Hart’s elastin stain. MV-O2 caused redistribution of elastin
from the tips of secondary septa, resulting in elastic fibers being scattered throughout
the walls of distal airspaces in vehicle-treated pups (Fig 2A). In contrast, lungs of elafin-
treated mice exhibited a normal distribution of elastin at the septal tips, with
considerably less dispersion of elastic fibers in alveolar walls after MV-O2 for 24h.
Quantitative image analysis confirmed that lung content of elastin, expressed as a
percent of lung tissue surface area, was similar in elafin-treated and unventilated control
mice (Fig 2B). In contrast, lungs of vehicle-treated mice showed increased amounts of
elastin that was fragmented and widely dispersed after MV-O2 for 24h.
Elafin inhibits the inflammatory response to MV-O2. MV-O2 typically induces a
leukocytic response in the lungs, which is amplified by the release of elastin degradation
Page 11 of 63
products (30). Elastic fiber fragments can recruit inflammatory cells to the lung, which in
turn can generate increased elastase and MMP-9 activity, thereby contributing to further
elastin breakdown (31). We therefore assessed the effect of elafin treatment on the
number of neutrophils and monocytes that accumulated in the lungs of newborn mice
after 24h of MV-O2. As noted in Fig 3, there was a 3-fold increase of both neutrophils
and monocytes in the lungs of vehicle-treated pups exposed to 24h of MV-O2.
Intrapulmonary elafin treatment blocked this response.
To further assess elafin’s inhibitory effects on the inflammatory response to MV-O2,
we measured lung mRNA expression of various cytokines and chemokines after 8h of
MV-O2. Elafin treatment fully suppressed the increased expression of IL-1β and IP-10
(CXCL10) noted in lungs of vehicle-treated pups after MV-O2 for 8h, and also reduced
levels of TNF-α and MCP-1 (CCL2) below levels of control and vehicle-treated pups
(on-line supplement, Fig E8).
As elafin has been reported to suppress activation of the inflammatory transcription
factor NF-ΚB in endothelial cells and macrophages exposed to endotoxin and other
atherogenic stimuli (32), we measured lung content of p65 protein, a subunit of the NF-
ΚB transcription factor complex, which plays a key role in inflammatory and immune
responses. We indeed discovered that elafin treatment markedly suppressed lung
abundance of p65 after 8h of MV-O2 when compared to unventilated controls and
vehicle-treated pups (on-line supplement, Fig E9). Lung p65 protein did not increase in
vehicle-treated pups that received MV-O2.
Elafin inhibits MV-induced upregulation of lung TGFβ β β β activity and apoptosis.
We previously found that MV, with or without associated hyperoxia, increased TGFβ
Page 12 of 63
signaling in the newborn lung (16, 18). A recent report indicates that degradation of
elastic fibers can trigger the release of active TGFβ from the extracellular matrix (ECM),
which in turn can contribute to over-production of aberrant elastic fibers in the lung
during postnatal development (33). Moreover, inflammatory cells that are recruited to
the lung in response to degradation of matrix proteins can contribute to the increased
TGFβ activity observed in the lungs after lengthy MV-O2 (34). We therefore assessed
lung expression of pSmad-2, a marker of TGFβ activity, in newborn mice exposed to
MV-O2 for 24h. The 2- to 3-fold increase in pSmad-2 expressing cells seen in lungs of
vehicle-treated pups was completely suppressed in elafin-treated mice after 24h of MV.
Consequently, lung abundance of pSmad-2 expressing cells was similar in mechanically
ventilated pups treated with elafin compared with unventilated, untreated control mice
that breathed 40%O2 for 24h (Fig 4).
To complement immunohistochemical assessment of pSmad-2 in peripheral lung as
an index of TGFβ activation, which elafin inhibited during MV-O2, we used an ELISA
assay to measure active TGFβ in whole lung homogenates obtained from the three
groups of mice. While the results were consistent with a suppressive effect of elafin on
TGFβ signaling (on-line supplement, Fig E10), apparent differences in TGFβ activity
between elafin-treated and control groups did not reach statistical significance (p=.08),
perhaps related to small sample size (n=4/group), considerable variability in the control
and vehicle-treated groups, and regional differences in TGFβ activity that may be
obscured by measurements made on protein extracts of whole lung tissue.
Page 13 of 63
As activation of TGFβ signaling can induce apoptosis of both alveolar epithelial cells
(35) and microvascular endothelial cells (36), we assessed apoptosis in lungs of mice
that had received 24h of MV-O2 with or without prior elafin treatment. Measurement of
cleaved caspase-3 protein, a marker of apoptosis, showed nearly a 10-fold increase in
lungs of vehicle-treated mice, whereas MV-O2 yielded no significant change of cleaved
caspase-3 in lungs of elafin-treated pups (Fig 5A). Elafin inhibition of apoptosis was
confirmed by quantitative IHC for terminal uridine deoxynucleotidyl transferase dUTP
terminal nick end labeling (TUNEL assay) in lung sections from vehicle-treated mice
compared with elafin-treated mice exposed to MV-O2 for 24h (Fig 5B & 5C).
Elafin enables lung septation during MV-O2. Previous studies showed that
prolonged MV of newborn mice with either air or 40%O2 inhibits alveolar septation
leading to lung growth arrest (18, 19). To determine if blocking lung elastase activity
would help to preserve lung growth during MV-O2, we used quantitative morphometry to
assess lung structure in elafin-treated mice compared to vehicle-treated mice after MV-
O2 for 24h. Lung morphometry showed that distal airspace size was greater and radial
alveolar counts were reduced after 24h of MV-O2 in vehicle-treated mice when
compared with unventilated controls (Fig 6). Elafin treatment attenuated these changes
in lung structure.
This study shows for the first time that intra-tracheal instillation of the serine elastase
inhibitor elafin, by blocking lung protease activity, can protect newborn mice from the
Page 14 of 63
adverse pulmonary effects of MV with O2-rich gas. Elafin, delivered directly into the Download full-text
lungs via the airways, fully suppressed the increased elastase and MMP-9 activity
measured in the lungs of untreated pups after 8h of MV-O2. Inhibition of proteolytic
activity prevented degradation and dispersion of lung elastin, as well as the increased
TGFβ activation, apoptosis and influx of neutrophils and monocytes noted in the lungs
of vehicle-treated mice after 24h of MV-O2. Elafin treatment also attenuated the MV-
induced emphysematous changes in alveolar structure seen in lungs of untreated mice
after 24h of MV-O2. These findings offer new insights on the pathogenesis and potential
treatment of CLD in neonates.
Elafin inhibition of lung inflammation and injury. Elafin is a low molecular weight
(~6kD) inhibitor of human neutrophil elastase (HNE) and proteinase-3. It is expressed in
several human tissues, including skin and lung (37), where it is secreted by pulmonary
epithelial cells and detected in bronchial secretions (38, 39). Elafin is secreted as a pro-
form, termed pre-elafin (also known as trappin-2), which contains a transglutaminase
domain that facilitates its binding to extracellular matrix proteins and helps to preserve
its anti-proteolytic function (40). Lung epithelial cell secretion of elafin increases in
response to HNE (41). Elafin is not expressed in rodents (42), but transgenic mice
engineered to express elafin in the lung were protected against injury induced by
intratracheal injection of Pseudomonas aeruginosa (43). In another study, mice that
were pretreated via the airways with recombinant pre-elafin exhibited less lung
inflammation than did vehicle-treated controls 6h after intranasal delivery of LPS (44). A
more recent report by the same group of investigators described a beneficial effect of
intranasal treatment with recombinant pre-elafin in reducing the acute lung inflammatory
Page 15 of 63