Am J Physiol Lung Cell Mol Physiol
279: L562–L574, 2000.
TIMP-1, -2, -3, and -4 in idiopathic pulmonary fibrosis.
A prevailing nondegradative lung microenvironment?
MOISES SELMAN,1 VICTOR RUIZ,1 SANDRA CABRERA,2 LOURDES SEGURA,1
REMEDIOS RAMÍREZ,2 ROBERTO BARRIOS,3 AND ANNIE PARDO2
2
Facultad de Ciencias, Universidad Nacional Autonoma de Mexico, Coyocán México DF CP 04000;
1
Instituto Nacional de Enfermedades Respiratorias, México DF CP 14080, Mexico; and 3Department
of Pathology, Baylor College of Medicine, Houston, Texas 77030
Received 13 January 2000; accepted in final form 20 April 2000
tissue inhibitor of metalloproteinases; gelatinases; collagenase; fibrosis
(IPF) is a progressive and
usually lethal interstitial lung disease of unknown
etiology characterized by varying degrees of inflammation and fibrosis in the lung parenchyma (14).
The fibrotic response is generally considered an irreversible process and is characterized by a striking
increase in fibroblast population and a profound and
complex change in extracellular matrix (ECM) turnover. That brings about a progressive accumulation of
connective tissue proteins (5) and, in addition, basement membrane disruption that is observed mainly in
IDIOPATHIC PULMONARY FIBROSIS
Address for reprint requests and other correspondence: A. Pardo,
Facultad de Ciencias, U.N.A.M., Apartado Postal 21-630, Coyocán
México DF, CP 04000, Mexico (E-mail: [email protected]).
L562
the early stages of the disease (25). In this context, an
imbalance between the synthesis and degradation of
ECM molecules in the local lung microenvironment
appears to be of central importance in the pathogenesis
of the fibrotic component of IPF.
Matrix metalloproteinases (MMPs), the mediators of
matrix degradation, are a family of zinc endoproteinases that share structural domains and are collectively
capable of degrading essentially all ECM components
(2). At present, the human MMP gene family contains
17 members that can be divided by structure and
substrate specificity into several subgroups including
collagenases, gelatinases, stromelysins, and membrane-type (MT) MMPs; other MMPs do not appear to
fall into any of these subgroups (36, 37). The regulation
of these enzymes reveals similarities and differences.
They differ in cellular sources and inducibility, but
they share the property of being synthesized as inactive zymogens in which activation can occur through
intracellular, extracellular, or cell surface-mediated
proteolytic mechanisms (2, 26, 37).
A pivotal extracellular control of MMP catalytic activity is accomplished by members of a specific family
of inhibitors named tissue inhibitors of metalloproteinases (TIMPs). There are currently four members of the
TIMP family (TIMP-1 to -4) that, besides their common
MMP inhibitory action, differ in expression patterns
and other properties such as association with latent
MMPs, cell growth-promoting activity, cell survivalpromoting activity, and apoptosis (1, 7, 9).
The possible role of both MMPs and TIMPs in IPF is
at present unclear. It has been considered that TIMPs
are good candidates for tissue fibrosis, but studies in
lung fibrosis are scanty. There are two recent studies
(6, 11) exploring the localization of TIMP-2, and in one
of them (11), TIMP-1 was also evaluated. However, the
expression and localization of TIMP-3 and TIMP-4
have not been examined in any fibrotic lung disorder.
The aim of this study was to determine the localization of TIMP-1, TIMP-2, TIMP-3, and TIMP-4 in lung
tissue obtained from patients with IPF. Additionally,
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
1040-0605/00 $5.00 Copyright © 2000 the American Physiological Society
http://www.ajplung.org
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
Selman, Moises, Victor Ruiz, Sandra Cabrera,
Lourdes Segura, Remedios Ramı́rez, Roberto Barrios,
and Annie Pardo. TIMP-1, -2, -3, and -4 in idiopathic
pulmonary fibrosis. A prevailing nondegradative lung microenvironment? Am J Physiol Lung Cell Mol Physiol 279:
L562–L574, 2000.—Fibroblast proliferation and extracellular matrix accumulation characterize idiopathic pulmonary
fibrosis (IPF). We evaluated the presence of tissue inhibitor
of metalloproteinase (TIMP)-1, -2, -3, and -4; collagenase-1,
-2, and -3; gelatinases A and B; and membrane type 1 matrix
metalloproteinase (MMP) in 12 IPF and 6 control lungs.
TIMP-1 was found in interstitial macrophages and TIMP-2 in
fibroblast foci. TIMP-3 revealed an intense staining mainly
decorating the elastic lamina in vessels. TIMP-4 was expressed in IPF lungs by epithelial and plasma cells. TIMP-2
colocalized with Ki67 in fibroblasts, whereas TIMP-3 colocalized with p27 in inflammatory and epithelial cells. Collagenase-1 was localized in macrophages and alveolar epithelial
cells, collagenase-2 was localized in a few neutrophils, and
collagenase-3 was not detected. MMP-9 was found in neutrophils and subepithelial myofibroblasts. Myofibroblast expression of MMP-9 was corroborated in vitro by RT-PCR. MMP-2
was noticed in myofibroblasts, some of them close to areas of
basement membrane disruption, and membrane type 1 MMP
was noticed in interstitial macrophages. These findings suggest that in IPF there is higher expression of TIMPs compared with collagenases, supporting the hypothesis that a
nondegrading fibrillar collagen microenvironment is prevailing.
MATRIX REMODELING IN IPF
we colocalized TIMP-2 and TIMP-3, which have been
related with cell growth-promoting activity and apoptosis, respectively (1, 10), with the nuclear proteins
Ki67 (associated with proliferation) and p27 (inhibitor
of the cell cycle). To have a more integrated picture of
matrix remodeling, we also examined a variety of
MMPs including the collagenase subfamily (MMP-1,
MMP-8, and MMP-13), the gelatinase subfamily
(MMP-2 and MMP-9), and MT1-MMP.
MATERIALS AND METHODS
noglobulin followed by horseradish peroxidase-conjugated
streptavidin (BioGenex, San Ramon, CA) was used according
to the manufacturer’s instructions. 3-Amino-9-ethylcarbazole
(BioGenex) in acetate buffer containing 0.05% H2O2 was
used as the substrate. The sections were counterstained with
hematoxylin (19). The primary antibody was replaced by
nonimmune serum for negative control slides.
To identify epithelial cells, myofibroblasts, and macrophages, parallel sections were immunostained for cytokeratin 7, ␣-smooth muscle actin, and HAM56, respectively
(DAKO, Carpinteria, CA).
In the case of double-label immunostaining, either Ki67 or
p27 antibody was used first followed by a secondary biotinylated antibody and an avidin-biotin-peroxidase complex
(Vectastain Elite ABC peroxidase kit, Vector Laboratories).
Color was revealed with 3, 3⬘-diaminobenzidine in PBS containing 0.05% H2O2. Afterward, either TIMP-2 or TIMP-3
antibody was used, followed by a secondary biotinylated
antibody with avidin-biotin and alkaline phosphatase complexes (Vectastain ABC alkaline phosphatase kit, Vector
Laboratories). The color reaction was accomplished by incubation with Fast Red chromogen (Biomeda, Foster City, CA).
Evaluation of immunohistochemical results. The intensity
of the staining in different lung tissue components was evaluated in the IPF patients as described previously (11). A
scale from 0 to 3 was used as follows: 0, negative; 1, mild
staining; 2, moderate staining; and 3, strong staining.
In situ hybridization. Riboprobes for in situ hybridization
were generated from human cDNA TIMP-4 cloned in a
pGem-T vector provided by S. S. Apte (Cleveland Clinic
Foundation). The plasmid was linearized before translation
with Kpn I. An antisense 487-bp fragment was transcribed
with T7 and a sense 508-bp fragment with SP6 RNA polymerase, respectively. The transcription of sense and antisense transcripts was performed with a labeling mixture
containing digoxigenin-UTP (Boehringer Mannheim, Mannheim, Germany).
In situ hybridization was performed on 4-␮m sections as
previously described (19). Briefly, the sections mounted on
silanized slides were incubated in 0.001% proteinase K
(Sigma, St. Louis, MO) for 20 min at 37°C. After acetylation
with acetic anhydride, the sections were prehybridized for 1 h
at 45°C in a hybridization buffer. The sections were incubated with the digoxigenin-labeled probes at 45°C overnight.
Some sections were hybridized with a digoxigenin-labeled
sense RNA probe. The tissues were incubated with a polyclonal sheep anti-digoxigenin antibody coupled to alkaline
phosphatase (Boehringer Mannheim, Indianapolis, IN) for
1 h at room temperature. The color reaction was performed
by incubation with Fast Red chromogen (Biomeda). The sections were lightly counterstained with hematoxylin.
Fibroblast isolation and culture. Human fibroblast-like
cells were obtained from normal and IPF lungs as previously
described (20). Cells were isolated by trypsin dispersion and
grown in Ham’s F-12 medium (GIBCO BRL, Life Technologies, Grand Island, NY) supplemented with 10% fetal calf
serum. The cells were cultured at 37°C in 5% CO2-95% air in
25-cm2 Falcon flasks with F-12K medium (GIBCO BRL)
supplemented with 10% fetal bovine serum (GIBCO BRL),
100 U/ml of penicillin, and 100 ␮g/ml of streptomycin. When
the fibroblasts reached early confluence, the medium was
replaced with serum-free F-12K medium containing phorbol
12-myristate 13-acetate (PMA). Fibroblasts were collected for
RNA extraction.
Flow cytometry. In parallel experiments, fibroblasts were
fixed with ice-cold 70% ethanol and stored at ⫺20°C until
assayed. The cells were washed and incubated for 1 h at 37°C
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
Study population. Twelve nonsmoking patients with IPF
were included in this study (8 men and 4 women; mean age
58.1 ⫾ 13.2 yr). The protocol was approved by the Ethical
Committee of the Instituto Nacional de Enfermedades Respiratorias (Mexico DF, Mexico). Diagnosis of IPF was supported by clinical, radiological, computed tomography scan,
and functional findings of interstitial lung disease and was
corroborated by open lung biopsy (22, 27). The morphological
diagnosis of IPF was based on typical microscopic findings
and included patchy, nonuniform alveolar septal fibrosis and
interstitial inflammation consisting mostly of mononuclear
cells but also of neutrophils and eosinophils; a variable macrophage accumulation was observed in the alveolar spaces as
was the cuboidalization of the alveolar epithelium. Biopsies
lacked granulomas, vasculitis, microorganisms, and inorganic material as seen by polarized light microscopy. None of
the patients had been treated with steroids or other immunosuppressive drugs at the time of biopsy. As controls, histologically normal lung tissues obtained at necropsy from six
nonsmoking adult individuals (4 men and 2 women; mean
age 43.2 ⫾ 11.1 yr) who died of causes unrelated to lung
diseases were utilized.
Bronchoalveolar lavage. Bronchoalveolar lavage (BAL)
was performed with a standard technique (32). Briefly, the
fiber-optic bronchoscope was wedged in two separate segments of the right middle lobe or lingula, and 300 ml of
normal saline were instilled in 50-ml aliquots, with an average return of 70%. The recovered BAL fluid was filtered
through sterile gauze, measured, and then centrifuged at 250
g for 10 min at 4°C. The supernatants were kept frozen at
⫺70°C until used. The cell pellet was resuspended in 1 ml of
phosphate-buffered saline (PBS), and an aliquot was used to
evaluate the total number of cells. Other aliquots were fixed
in carbowax, and three slides per sample were stained with
hematoxylin and eosin, Giemsa, and toluidine blue and used
for differential cell counts. Six nonsmoking healthy volunteers were lavaged as control subjects (3 women and 3 men;
33.5 ⫾ 6 yr old).
Immunohistochemistry. Tissue sections were deparaffinized, rehydrated, and then blocked with 0.45% H2O2 in
methanol for 30 min followed by normal serum (Vector Laboratories, Burlingame, CA) diluted 1:20 in PBS for 20 min.
Before the immune reaction, antigen retrieval with 0.1 M
citrate buffer, pH 6.0, was performed. The sources of the
primary monoclonal antibodies were Fuji Chemical Industries (Toyama, Japan) for MMP-1, MMP-8, MMP-9, MMP-13,
TIMP-1, TIMP-2, and TIMP-3 and Calbiochem (San Diego,
CA) for MMP-2. TIMP-4 polyclonal antibody was kindly donated by S. S. Apte (Cleveland Clinic Foundation, Cleveland,
OH). The monoclonal antibody for mitotic inhibitor/suppressor protein p27 was from NeoMarkers (Fremont, CA), and
Ki67 rabbit polyclonal antibody was from Novocastra Laboratory (Newcastle, UK). Antibodies were applied and incubated at 4°C overnight. A secondary biotinylated anti-immu-
L563
L564
MATRIX REMODELING IN IPF
with FITC-conjugated monoclonal anti-human ␣-smooth
muscle actin antibody diluted 1:400 in 1% bovine serum
albumin in PBS, pH 7.3. The fibroblasts were washed and
resuspended in PBS for fluorescence-activated cell sorter
analysis as described elsewhere (35).
Zymography. SDS-polyacrylamide gels containing gelatin (1 mg/ml) were used to identify proteins with gelatino-
lytic activity from BAL supernatants (21). After electrophoresis, the gels were washed in a solution of 2.5% Triton
X-100 (15 ⫻ 2 min) to remove SDS, washed extensively
with water, and incubated overnight at 37oC in 100 mM
glycine, pH 7.6, containing 10 mM CaCl2 and 50 nM ZnCl2.
Identical gels were incubated in the presence of 20 mM
EDTA as controls. The gels were stained with Coomassie
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
Fig. 1. Tissue inhibitor of metalloproteinase (TIMP)-1 and TIMP-2 immunolocalization in idiopathic pulmonary
fibrosis (IPF) and normal lungs. A and B: TIMP-1-immunoreactive interstitial cells in IPF lungs. Original
magnifications: ⫻10 in A; ⫻40 in B. B, inset: positive interstitial macrophages. Original magnification, ⫻100. C and
D: IPF lungs exhibiting immunolabeled subepithelial myofibroblasts stained for TIMP-2 (arrow). Original magnification, ⫻10. D, inset: positive fibroblasts. Original magnification, ⫻100. E and F: normal lungs were usually
negative for TIMP-1 and TIMP-2, respectively. Original magnification, ⫻40. G: negative control omitting the
primary antibody. Original magnification, ⫻40.
MATRIX REMODELING IN IPF
5⬘-ACAGCCAGAAGCAGTATC-3⬘ (sense) for a fragment size of
446 bp. Gelatinase B primers were 5⬘-GTCGGCCTCAAAGGTTTGGAAT-3⬘ (antisense) and 5⬘-GTGCTGGGCTGCTGCTTTGCTG-3⬘ (sense), with an amplified product of 303 bp.
To quantify the housekeeping gene GAPDH, a competitor
was constructed by cutting an internal fragment of 155 bp of
a GAPDH cDNA (originally 1,233 bp) cloned in a PBR 322
plasmid with Nco I. The modified GAPDH cDNA was subsequently religated. The competitor (cGAPDH) sequence was
obtained by PCR amplification of the modified plasmid with
the primers for GAPDH. The competitor size was 240 bp.
Fourfold serial dilutions of the standard competitor (5, 10,
15, and 20 pg) were coamplified with a constant amount of
cellular cDNA (1 ␮l). Cycling conditions were 95°C for 10 min
for 1 cycle; 95°C for 30 s, 58°C for 30 s, and 72°C for 120 s for
40 cycles; and 72°C for 7 min for the final incubation. Aliquots (5 ␮l) of the PCR product were resolved in 1.5% agarose
gel containing ethidium bromide. Band intensities were
quantitated by scanning densitometry with a Kodak digital
science electrophoresis documentation and analysis system
120 (Kodak, Rochester, NY). The logarithm of the GAPDHto-cGAPDH ratio was plotted as a function of the logarithm of
Fig. 2. TIMP-3 immunolocalization in IPF and normal lungs. A and B: immunoreactive TIMP-3 bound mainly to
elastic lamina of vessels and to extracellular matrix in IPF lungs. Original magnifications: ⫻2.5 in A; ⫻40 in B. No
TIMP-3 signal was noticed in normal lungs (C). Original magnification, ⫻40. D: interstitial macrophages with
immunoreactive TIMP-3 staining in IPF lung section. Original magnification, ⫻40. E: negative control omitting the
primary antibody. Original magnification, ⫻40.
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
brilliant blue R250 and destained in a solution of 7.5%
acetic acid and 5% methanol.
RT-PCR. Total RNA extracted from human fibroblasts as
previously described (20) was used for RT-PCR to explore
gelatinase B expression. Total RNA extracted from IPF and
control lungs was used to analyze MMP-13 and TIMP-4
expression. RNA was reverse transcribed to synthesize cDNA
according to the manufacturer’s protocol (GIBCO BRL).
PCR amplification (Perkin-Elmer 9600, Branchburg, NJ)
for glyceraldehyde-3-phosphate dehydrogenase (GAPDH),
MMP-9, MMP-13, and TIMP-4 was performed with a cDNA
working mixture in a 25-␮l reaction volume containing 20
mM Tris 䡠 HCl, pH 8.3, 50 mM KCl, 2 mM MgCl2, 200 ␮M
deoxynucleotide triphosphates, 1 ␮M specific 5⬘ and 3⬘ primers, and 1 U of Taq DNA polymerase (Perkin-Elmer).
The set of primers used for the GAPDH PCR was 5⬘-CATCCATCCCGTGACCTTAT-3⬘ and 5⬘-GCATGACTCTCACAATGCGA-3⬘, with an amplified product of 395 bp. Nucleotides
5⬘-TCATGACCTCATCTTC-3⬘ and 5⬘-GAACAGCTGCACTTAT-3⬘ were used as primers to amplify a 134-bp segment for
collagenase-3 cDNA. For TIMP-4 amplification, the primers
used were 5⬘-CCAGAGGTCAGGTGGTAA-3⬘ (antisense) and
L565
L566
MATRIX REMODELING IN IPF
the known cGAPDH amount. The point of equivalence represents the concentration of GAPDH in the cDNA sample.
Dilutions were performed to reach 5 pg/␮l of GAPDH. For
each amplification, 10 pg of GAPDH were used.
For MMP-13 and gelatinase B amplification, cycling
conditions were 95°C for 10 min for 1 cycle; 95°C for 30 s,
60°C for 30 s, and 72°C for 90 s for 35 cycles; and 72°C for
7 min for the final incubation. For TIMP-4 amplification,
the conditions were 95°C for 10 min for 1 cycle; 95°C for
30 s, 58°C for 30 s, and 72°C for 60 s for 38 cycles; and 72°C
for 7 min for the final incubation. Aliquots (5 ␮l) of the PCR
product were resolved on a 1.5% agarose gel containing
ethidium bromide.
RESULTS
TIMP-1, TIMP-2, TIMP-3, and TIMP-4 localization.
TIMP-1 was detected in 8 of 12 IPF lungs, usually
localized in isolated interstitial cells, mainly macrophages and fibroblast-like cells (Fig. 1, A and B). This
inhibitor was observed in areas of dense scar tissue.
Regarding TIMP-2, it was found in almost all IPF
lungs, primarily associated with subepithelial fibroblast/myofibroblast foci (Fig. 1C), some of them partially occupying the alveolar spaces (Fig. 1D). Myofibroblasts were identified by immunoreactive ␣-smooth
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
Fig. 3. TIMP-4 mRNA and immunoprotein localization in IPF and control lungs. A and B: in situ hybridization of
antisense TIMP-4 riboprobe showing positive epithelial cells (A) and interstitial macrophages and plasma cells (B,
arrow) in IPF lung. Original magnification, ⫻40. C: sense TIMP-4 riboprobe. Original magnification, ⫻40. D:
immunohistochemistry of TIMP-4 showing several positive plasma cells (arrows). Original magnifications: ⫻40;
⫻60 in inset. E: negative control discarding the primary antibody. Original magnification, ⫻40. F: normal lung.
Original magnification, ⫻10.
MATRIX REMODELING IN IPF
TIMP-3 and Ki67 was noticed (Fig. 5F), whereas
TIMP-3 was observed in alveolar epithelial cells and in
some interstitial cells showing a negative nucleus
staining for Ki67.
Collagenase immunolocalization. Immunoreactive
MMP-1 (collagenase-1) was noticed in reactive alveolar
epithelial cells and/or in clusters of alveolar macrophages in all IPF lungs (Fig. 6A). Besides, hyperplastic
type 2 pneumocytes lining honeycomb cysts in areas of
fibroconnective tissue deposition showed strong staining for MMP-1 (Fig. 6B). Cytokeratin staining corroborated the epithelial nature of these cells (data not
shown). Intriguingly, practically no interstitial cells
producing MMP-1 were noticed.
Immunoreactive MMP-8 (collagenase-2) was localized in 75% of the IPF lungs, usually in a few neutrophils inside vessels, and in two patients, the label was
also observed in the interstitium (Fig. 6C). MMP-13
(collagenase-3) was not detected either by immunohistochemistry or by RT-PCR of IPF lungs even when up
to 60 cycles were assayed. In three control lungs, scattered alveolar macrophages stained for MMP-1 were
noticed as well as some neutrophils displaying MMP-8
(Fig. 6, D and E). Control samples incubated with
nonimmune sera were negative (Fig. 6F).
Gelatinases A and B and MT-1 MMP immunolocalization. Regarding type IV collagenases, MMP-2 (gelatinase A) was found in some foci of subepithelial myofibroblasts (Fig. 7A) located close to areas of basement
membrane disruption. In 8 of 12 of the IPF lungs,
MMP-2 was also observed associated with the ECM
surrounding fibroblast foci (Fig. 7B). Normal lungs
showed no positive staining (Fig. 7C)
MT1-MMP that has been associated with the activation of pro-MMP-2 was noticed in widely spaced interstitial cells, mainly macrophages (Fig. 7D), and in
three cases in reactive alveolar epithelial cells (data
not shown). MT1-MMP was not observed in normal
lungs (Fig. 7E). IPF tissues incubated without the
primary specific antibody were negative (Fig. 7F).
Fig. 4. Analysis of TIMP-4 expression by RT-PCR. Total RNA extracted from normal and IPF lungs was reverse transcribed into
cDNA and subjected to PCR with the primers in MATERIALS AND
METHODS. After 40 cycles of amplification with TIMP-4-specific primers, the 446-bp product was clearly detected in IPF lungs (lanes 3 and
4). Normal lungs were negative even when 50 cycles were used (lanes
1 and 2).
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
muscle actin staining in parallel sections (data not
shown). In three IPF lungs, TIMP-2 was also observed
in reactive alveolar epithelial cells. Control lungs
showed virtually no signal for TIMP-1 and TIMP-2
(Fig. 1, E and F, respectively). Negative control lungs
showed no reactivity (Fig. 1G).
TIMP-3 showed the most intense extracellular staining and was mainly found coupled to the elastic lamina
of most vessels, revealing the characteristic duplication of this structure in pulmonary fibrosis (Fig. 2, A
and B). This observation was corroborated in 75% of
examined IPF lungs. Less intense staining was noticed
in thickened alveolar septa. In contrast, in normal
lungs, no positive staining was observed even in the
elastic lamina of vessels (Fig. 2C). In some areas of IPF
lungs, numerous interstitial cells, mainly macrophages
(identified by HAM56; data not shown), were strongly
stained (Fig. 2D). Additionally, in two IPF samples,
some alveolar epithelial cells were also positive (see
Fig. 5F). Control IPF samples incubated without the
primary specific antibody were negative (Fig. 2E).
By in situ hybridization, TIMP-4 transcript was localized in some alveolar epithelial cells (Fig. 3A), although a
strong positive signal was revealed in interstitial areas,
mainly in macrophages and, interestingly, plasma cells.
(Fig. 3B). The immunoreactive protein paralleled the
mRNA observations, confirming the finding of TIMP-4 in
plasma cells in almost all IPF lungs (Fig. 3D). Immunohistochemistry for TIMP-4 protein was negative in normal lungs (Fig. 3F). Tissues hybridized with the sense
probe or without the primary antibody showed no reactivity (Fig. 3, C and E).
RT-PCR for TIMP-4 mRNA. Expression of TIMP-4 in
two IPF lungs and two control lungs was analyzed by
RT-PCR. As illustrated in Fig. 4, the expected 446-bp
fragment was revealed after 40 PCR cycles only in IPF
samples (lanes 3 and 4). Control lungs did not reveal a
positive result, although an equivalent cDNA concentration as analyzed by GADPH amplification was used
(Fig. 4, lanes 1 and 2).
Ki67 and p27 immunostaining and colocalization with
TIMP-2 and TIMP-3. Ki67 and p27 were analyzed in IPF
lungs. In general, nuclear p27 staining was more abundant than Ki67 staining, although there were regional
variations in immunoreactivity. Most proliferating cells
as revealed by the presence of the Ki67 antigen were
alveolar epithelial cells and fibroblast-like cells mainly
located in areas of fibroblast foci protruding to the alveolar spaces (Fig. 5A). On the other hand, the inhibitory
cell cycle protein p27 was usually present in most inflammatory interstitial and intraluminal cells as well as in
endothelial cells (Fig. 5B).
Colocalization of Ki67 and TIMP-2 occurred in areas
of fibroblasts, mostly in the periphery of fibroblast foci
(Fig. 5C), whereas in the same areas where TIMP-2
was positive, p27 was negative (Fig. 5D).
Nuclear p27 detection colocalized with TIMP-3 in
some interstitial inflammatory cells, in endothelial
cells where elastic lamina showed extracellular
TIMP-3 accumulation, and in some epithelial cells (Fig.
5, E and F). In general terms, no colocalization of
L567
L568
MATRIX REMODELING IN IPF
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
Fig. 5. Immunohistochemical localization of p27 and Ki67 and colocalization with TIMP-2 and TIMP-3. p27 and
Ki67 were revealed with diaminobenzidine, and TIMP-2 and TIMP-3 were revealed with Fast Red. A: Ki67 in IPF
lung section. Arrow, positive nucleus in epithelial cells; arrowhead, signaling subepithelial positive nuclei. Original
magnification, ⫻10. B: IPF lung showing numerous p27-positive cells. Original magnification, ⫻10. Lung sections
in A and B were counterstained with eosin. C and D: fibroblast foci in IPF lung colocalizing cytoplasmic TIMP-2
with nuclear Ki67 and TIMP-2 with p27, respectively. Original magnification, ⫻10. E: colocalization of TIMP-3
with p27. Original magnification, ⫻10. Inset: positive cytoplasmic TIMP-3 and nuclear p27 in interstitial cells.
Original magnification, ⫻40. F: double localization of Ki67 and TIMP-3 showing that most epithelial cells were
positive for TIMP-3 and negative for Ki67. Original magnification, ⫻40. Inset: epithelial cells colocalizing TIMP-3
with p27. Original magnification, ⫻100.
MATRIX REMODELING IN IPF
MMP-9 (gelatinase B) was found in almost all IPF
lungs, primarily in intravascular and interstitial neutrophils (Fig. 8A). In four patients, a weaker staining
was also noticed in clusters of alveolar macrophages.
Interestingly, positive staining was appreciated in
some subepithelial myofibroblasts and reactive alveolar epithelial cells (Fig. 8B). Similar to MMP-2, extracellular localization of MMP-9 in areas of dense scars
was also seen (Fig. 8C). The majority of the control
lungs showed scattered neutrophils positive for
MMP-9, whereas macrophages were usually negative
L569
(Fig. 8D). Control samples incubated with nonimmune
serum were negative (Fig. 8E).
A summary of the immunolocalization results of
TIMPs and MMPs in IPF lungs is depicted in Table 1.
RT-PCR for fibroblast gelatinase B mRNA. To corroborate the expression of MMP-9 by lung fibroblasts,
we examined the expression of MMP-9 in cells obtained
from normal lungs (⬃8% myofibroblasts by ␣-actin
staining and fluorescence-activated cell sorter counting) and in those derived from IPF lungs (40–80%
myofibroblasts). Total RNA from nonstimulated and
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
Fig. 6. Collagenase-1 and collagenase-2 immunolocalization in IPF and normal lungs. Immunoreactive collagenases were revealed by 3,3⬘-diaminobenzidine. A: IPF lung showing several epithelial cells (arrows) and alveolar
macrophages (arrowhead) with intracytoplasmic signal for matrix metalloproteinase (MMP)-1. Original magnification, ⫻40. B: IPF lung with MMP-1-positive cuboidal epithelial cells (arrows). Original magnification, ⫻40. C:
immunoreactive MMP-8 in several intravascular neutrophils (double arrowhead). Original magnification, ⫻40. D:
positive MMP-1 alveolar macrophage (arrowhead) in a control lung. Original magnification, ⫻40. E: MMP-8stained neutrophils were sporadically found in normal lung (double arrowhead). Original magnification, ⫻40. F:
negative control section without the primary antibody. Original magnification, ⫻40.
L570
MATRIX REMODELING IN IPF
PMA-stimulated fibroblasts was reverse transcribed,
and cDNA was amplified as described in MATERIALS AND
METHODS. Only fibroblasts derived from IPF lungs
showed a 303-bp fragment amplification that was increased approximately threefold after PMA treatment
(Fig. 9, lanes 3–6). An equivalent cDNA concentration
as measured by GADPH amplification with an internal
competitor showed no amplification even after PMA
stimulation (Fig. 9, lanes 1 and 2).
BAL. Significantly less fluid was recovered from IPF
patients (55.4 ⫾ 9.7 vs. 71.3 ⫾ 12.1% in control subjects; P ⬍ 0.01). BAL fluid obtained from IPF patients
revealed a twofold increase in total cell number and
exhibited a significant increase in the percentage of
neutrophils and eosinophils compared with those in
the control samples (Table 2).
SDS-PAGE zymography. To identify BAL fluid gelatinolytic activities, aliquots adjusted to 20 ␮l of
lavage fluid were analyzed by gelatin substrate gel
zymography. A representative gelatin zymogram
comparing IPF BAL fluid with that from normal
subjects is shown in Fig. 10. BAL normal samples
showed faint bands of 72- and 92-kDa activities
corresponding to pro-MMP-2 and pro-MMP-9, respectively (Fig. 10, lanes N). IPF patients revealed a
marked increase of BAL fluid progelatinase B activity (Fig. 10, lanes 1–4), and in most samples, the
activated form of 86 kDa was also evident. In addiDownloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
Fig. 7. MMP-2 and membrane type (MT) 1 MMP localization in IPF and normal lungs. Immunoreactive proteins
were revealed by 3,3⬘-diaminobenzidine, and the sections were lightly counterstained with hematoxylin. A: cluster
of fibroblasts showing immunoreactive MMP-2 in an IPF lung (arrow). Original magnification, ⫻10. B: higher
magnification of the same field. Original magnification, ⫻100. C: normal lung showing no positive staining for
MMP-2. Original magnification, ⫻10. D: immunohistochemical localization of MT1-MMP in several interstitial
cells in an IPF lung (arrowhead). Original magnification, ⫻10. Inset: positive macrophages. Original magnification,
⫻100. E: normal lungs showed no MT1-MMP signal. Original magnification, ⫻10. F: IPF sample with primary
antibody omitted. Original magnification, ⫻40.
L571
MATRIX REMODELING IN IPF
tion, gelatinolytic bands of higher molecular mass,
likely representing lipocalin-associated progelatinase B specific in neutrophils, were also noticed.
Densitometric analysis showed a 2.5- to 7-fold increase in progelatinase B activity compared with
that in healthy individuals (Fig. 10). Likewise, IPF
BAL fluids displayed a 5- to 15-fold increase in
progelatinase A activity compared with that in control samples. In the majority of IPF fluids, the 68kDa active form of pro-MMP-2 was also observed. All
Table 1. MMP and TIMP immunoreactivity in IPF lungs
Alveolar
epithelial cells
Alveolar
macrophages
Tissue
macrophages
Neutrophils
Plasma cells
Fibroblasts/
myofibroblasts
Extracellular
matrix
MMP-1
MMP-8
MMP-13
MMP-2
MMP-9
MT1-MMP
TIMP-1
TIMP-2
TIMP-3
TIMP-4
3
0
0
1
2
1
0
1
1–2
1–2
3
0
0
0
1–2
0
0
0
0
0
0
0
0
0
1–2
0
0
0
0
0
0
0
0
3
0
2–3
0
0
1–2
0
0
0
0
0
2–3
0
0
3
0
3
0
0
0
2
1–2
0
1
2–3
1
0–1
0
0
0
1–2
1
0
1
0
3
0
Values are staining intensity graded as 0, no staining; 1, weak or mild staining; 2, moderate staining; and 3, strong staining. MMP, matrix
metalloproteinase; MT1-MMP, membrane type 1 MMP; TIMP, tissue inhibitor of metalloproteinase.
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
Fig. 8. MMP-9 immunolocalization in IPF and normal lungs. A: IPF lung showing several MMP-9-positive
neutrophils (double arrowhead). Original magnification, ⫻40. B: IPF lung. Arrow, subepithelial fibroblasts
reactive to anti-MMP-9. Original magnification, ⫻10. C: higher magnification showing fibroblast and extracellular localization of MMP-9. Original magnification, ⫻40. D: MMP-9-positive neutrophil in a normal lung
(arrow). Original magnification, ⫻40. E: IPF sample with primary antibody omitted. Original magnification,
⫻40.
L572
MATRIX REMODELING IN IPF
gelatinolytic bands were fully inhibited by EDTA
(data not shown).
Table 2. Total cell number and cell profile
in BAL fluid
DISCUSSION
Fig. 9. Analysis of fibroblast gelatinase B expression by RT-PCR.
Total RNA extracted from human lung fibroblasts was reverse transcribed, and cDNA was amplified with use of the primers in MATERIALS AND METHODS. Lanes 1 and 2, normal fibroblasts (8% myofibroblasts) without and with, respectively, phorbol 12-myristate 13acetate (PMA); lanes 3 and 4, IPF fibroblasts (78% myofibroblasts)
without and with, respectively, PMA; lanes 5 and 6, IPF fibroblasts
(57% myofibroblasts) without and with, respectively, PMA; lane 7,
U-2 OS cells treated with PMA as a positive control.
Total cell count,
⫻106
Macrophages, %
Lymphocytes, %
Neutrophils, %
Eosinophils, %
IPF Patients
(n ⫽ 12)
P Value
12.7 ⫾ 3.8
80.9 ⫾ 10.5
18.1 ⫾ 11.1
1.0 ⫾ 1.0
0.0 ⫾ 0.0
23.2 ⫾ 7.1
70.8 ⫾ 12.3
19.2 ⫾ 13.3
7.5 ⫾ 7.1
2.7 ⫾ 3.6
⬍0.01
NS
NS
⬍0.05
⬍0.05
Values are means ⫾ SD; n, no. of individuals. BAL, bronchoalveolar lavage; IPF, idiopathic pulmonary fibrosis; NS, not significant.
was not found. MMP-1 was usually noticed in alveolar
epithelial cells and alveolar macrophages, whereas in
the thickened alveolar septa where increased bundles
of collagen are accumulating, no positive signal was
noticed.
Considering that the main function of collagenases is
the removal of fibrillar collagens, the absence in the
expression of this enzyme in the interstitium, at least
as detected by the methods used in this work, might
explain, in part, the presence of scars that do not
undergo resorption. Additionally, it has been proposed
that MMP-1 expression is induced in primary keratinocytes by contact with native type I collagen and that
its degradation initiates keratinocyte migration during
reepithelialization (24). In this context, it can be speculated that degradation of basement membrane by
upregulated gelatinases in IPF may induce upregulation of collagenase-1 in alveolar epithelial cells, facilitating type 2 cell migration and proliferation (15, 16).
In contrast, TIMP expression was more abundant
and usually noticed associated with the ECM or in
interstitium-located cells. Particularly TIMP-3, which
is studied here for the first time in IPF, was copiously
observed throughout the lung parenchyma. This inhibitor was found exquisitely decorating the elastic lamina in vessels and within the thickened alveolar septa.
Actually, a striking feature of TIMP-3 is that it is ECM
associated, whereas TIMP-1 and TIMP-2 are freely
diffusible extracellular proteins (7). TIMP-3 is not only
important in matrix turnover but also has been shown
to induce apoptosis by the stabilization of tumor necrosis factor-␣ receptors by an effect dependent on TIMP-3
inhibitory function (31). Moreover, TIMP-3 overexpression in vascular smooth muscle cells promoted death by
apoptosis by an effect independent of MMP inhibition
(1). Interestingly, it has been recently demonstrated in
mesangial cells that p27, a protein inhibitor of cell
proliferation, may also be involved in apoptosis (13).
Thus it seems that under certain conditions, a p27mediated increase in cyclin-dependent kinase-2 activity leads to apoptosis. In our study, TIMP-3 colocalized
with p27 in alveolar epithelial cells and interstitial
inflammatory cells. Apoptosis of type 2 pneumocytes
has been demonstrated in vivo in IPF lungs, and it has
been suggested that chronic epithelial cell death may
avoid normal reepithelialization, enhancing the fibrotic response (33). Several mechanisms may be in-
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
A dynamic balance between synthesis and degradation of ECM components is required for lung structure
and functional maintenance. The loss of regulated
turnover may result in a number of different lung
diseases including lung emphysema and pulmonary
fibrosis.
IPF is characterized by a chronic inflammatory process, with the presence of widely distributed small
aggregates of actively proliferating fibroblasts and
myofibroblasts, which, in turn, are responsible for the
excessive ECM deposition, mainly fibrillar collagens,
and alveolar structural remodeling (14). The disease is
of a patchy nature, and the progression of the fibrotic
lesion until a mature scar at any given time and place
reflects an imbalance in ECM synthesis and degradation.
The MMPs, also called matrixins, play a central role
in all processes involving ECM remodeling, and an
inappropriate regulation of their activities may have
pathological implications. Regulation is controlled at
several levels including gene transcription and activation of a latent enzyme. Locally, in the extracellular
space, MMPs are tightly regulated by TIMPs.
In IPF, the remodeling processes may involve various MMPs and their specific inhibitors, but little is
known about their possible participation during lung
fibrogenesis.
Our results showed that there are important differences in both localization and abundance of collagenases, gelatinases, and TIMPs in fibrotic tissues.
Thus the collagenases MMP-1 and MMP-8, which are
responsible for the degradation of interstitial fibrillar
collagens, showed a more localized expression compared with TIMPs, and MMP-13, another collagenase,
Control Subjects
(n ⫽ 6)
MATRIX REMODELING IN IPF
L573
Fig. 10. Identification of gelatinolytic enzymes in bronchoalveolar lavage (BAL) fluid by SDS-PAGE gelatin
zymography. BAL samples were mixed with an equal
volume of Laemmli sample buffer containing 3% SDS.
Right: lanes 1–4, IPF samples; lanes N, control samples. A, gelatinase (GEL) A; B, GEL B. Left: quantitative image analysis of the surface and intensity of the
gelatinolytic bands. F1–F4, IPF samples; N1 and N2,
control samples. Results are expressed as activity in
arbitrary units (see MATERIALS AND METHODS).
expressed the transcript and that expression of gelatinase B was closely related to the percentage of
myofibroblasts were of particular importance. Additionally, BAL fluid gelatin zymography showed increased gelatinase activities attributable to MMP-2
and MMP-9 in IPF patients.
TIMP-4 was also explored for first time in IPF lungs.
This inhibitor can effectively inhibit human MMPs,
and it has been suggested that as TIMP-2, it is more
specific for MMP-2. Moreover, it binds to the COOHterminal domain of MMP-2 in a manner similar to
TIMP-2 (12). In normal tissues, TIMP-4 mRNA is confined primarily to the adult heart, and no transcript
has been detected in normal lungs (8). In the present
study, we demonstrated by RT-PCR, in situ hybridization, and immunohistochemistry that TIMP-4 is highly
expressed in IPF lungs, primarily by interstitial macrophages, clusters of plasma cells, and alveolar epithelial cells. Although its role in IPF is largely unknown,
the widespread expression found in this study strongly
suggests that it may contribute to a profibrotic microenvironment in the IPF lungs.
MT1-MMP has been mainly implicated in the activation of progelatinase A through a trimolecular complex comprising MT1-MMP, MMP-2, and TIMP-2 (3).
Intriguingly, by immunohistochemistry, we found that
although gelatinase A and TIMP-2 were expressed by
the same type of cells, primarily myofibroblasts, MT1MMP was mainly observed in interstitial cells. Studies
are ongoing in our laboratory to analyze the distribution of other MT-MMPs in this disease.
In summary, our findings indicate that during lung
fibrogenesis, 1) there is a wider distribution of TIMPs
compared with the collagenases MMP-1 and MMP-8 in
the lung parenchyma. These findings, together with
previous results from our laboratory (18, 23, 28–30) in
which we have consistently found a decrease in collagenolytic activity associated with the development of
fibrosis, suggest that in IPF a nondegrading fibrillar
collagen microenvironment is prevailing; and 2) excessive MMP-2 and MMP-9 production might play a role
in basement membrane disruption, enhancing fibroblast invasion to the alveolar spaces.
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
volved in alveolar epithelial cell apoptosis, including
the Fas-Fas ligand pathway (17) and fibroblast secretion of apoptotic factors (34), and the results of the
present study suggest the possible participation of
TIMP-3.
TIMP-2 was found almost exclusively associated with
fibroblast foci. A fibroblast focus is characterized by a
distinct cluster of fibroblasts and/or myofibroblasts
within the alveolar wall. The fibroblast foci are distributed throughout the lung parenchyma, representing a
characteristic morphological feature in IPF and indicating that fibrosis is actively ongoing (14). These foci may
occur in the interstitium as well as in the alveolar spaces
(14–16). In the latter, fibroblasts migrate through gaps in
the alveolar epithelial basement membranes and proliferate within the alveolar spaces, producing intraluminal
fibrosis. This process might be at least partially associated with the secretion of the gelatinases MMP-2 and
MMP-9. In this context, both MMP-2 and MMP-9 were
observed in subepithelial myofibroblasts and occasionally
in areas of denuded alveolar basement membranes, suggesting that these MMPs may play a role in the migration of these cells to the alveolar spaces. As mentioned,
subepithelial myofibroblasts were also positive for
TIMP-2. Although MMP inhibition is the main function
of TIMPs, paradoxically, TIMP-2 might be influencing an
enhanced activation of this enzyme through its binding to
pro-MMP-2. This feature may be of important physiological significance in modulating the cell surface activation
of pro-MMP-2. Such a mechanism should depend on the
molar equilibrium between the enzyme and the inhibitor
(4). In addition, growing evidence is supporting a wide
variety of other functions for TIMPs, including cell
growth-promoting activity. In this context, we found that
areas of TIMP-2-positive myofibroblasts close to the epithelium showed a positive signal for Ki67, a protein
expressed during the cell division cycle in the G1, S, and
G2 phases as well as in mitosis.
As previously reported (6, 11), subepithelial myofibroblasts in IPF lungs exhibited immunoreactive
MMP-9. Considering that lung fibroblasts do not
express MMP-9 in vitro, as we also showed by RTPCR in cells obtained from human normal lungs, the
findings that fibroblasts obtained from IPF lungs
L574
MATRIX REMODELING IN IPF
We are most grateful to Juan Carlos Valencia for technical photographic support.
This work was supported by Consejo Nacional de Ciencia y Tecnologia (Mexico) Grant CONACYT 27518M and by Programa Universitario de Investigacion en Salud (Universidad Nacional Autónoma de México, Mexico).
19.
20.
REFERENCES
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
Downloaded from http://ajplung.physiology.org/ by 10.220.32.247 on June 15, 2017
1. Baker AH, Zaltsman AB, George SJ, and Newby AC. Divergent effects of tissue inhibitor of metalloproteinase-1, -2, or -3
overexpression on rat vascular smooth muscle cell invasion,
proliferation, and death in vitro. TIMP-3 promotes apoptosis.
J Clin Invest 101: 1478–1487, 1998.
2. Birkedal-Hansen H, Moore WG, Bodden MK, Windsor LJ,
Birkedal-Hansen B, DeCarlo A, and Engler JA. Matrix metalloproteinases: a review. Crit Rev Oral Biol Med 4: 197–250, 1993.
3. Cao J, Sato H, Takino T, and Seiki M. The C-terminal region
of membrane type matrix metalloproteinase is a functional
transmembrane domain required for pro-gelatinase A activation.
J Biol Chem 270: 801–805, 1995.
4. Corcoran ML, Hewitt RE, Kleiner DE, and Stetler-Stevenson WG. MMP-2: expression, activation, and inhibition. Enzyme
Protein 49: 7–19, 1996.
5. Crouch E. Pathobiology of pulmonary fibrosis. Am J Physiol
Lung Cell Mol Physiol 259: L159–L184, 1990.
6. Fukuda Y, Ishizaki M, Kudoh S, Kitaichi M, and Yamanaka N. Localization of matrix metalloproteinases-1, -2, and
-9, and tissue inhibitor of metalloproteinase-2 in interstitial lung
diseases. Lab Invest 78: 687–698, 1998.
7. Gómez D, Alonso D, Yoshiji U, and Thorgeirsson P. Tissue
inhibitors of metalloproteinases: structure, regulation and biological functions. Eur J Cell Biol 74: 111–122, 1997.
8. Greene J, Wang M, Liu YE, Raymond LA, Rosen C, and Shi
YE. Molecular cloning and characterization of human tissue inhibitor of metalloproteinase-4. J Biol Chem 271: 30375–30380, 1996.
9. Guedez L, Stetler-Stevenson WG, Wolff L, Wang L, Fukushima P, Mansoor A, and Stetler-Stevenson M. In vitro
suppression of programmed cell death of B cells by tissue inhibitor of metalloproteinases-1. J Clin Invest 102: 2002–2010, 1998.
10. Hayakawa T, Yamashita K, Ohuchi E, and Shinagawa A.
Cell growth-promoting activity of tissue inhibitor of metalloproteinases-2 (TIMP-2). J Cell Sci 107: 2373–2379, 1994.
11. Hayashi T, Stetler-Stevenson WG, Fleming MV, Fishback N,
Koss MN, Liotta LA, Ferrans VJ, and Travis WD. Immunohistochemical study of metalloproteinases and their tissue inhibitors
in the lungs of patients with diffuse alveolar damage and idiopathic
pulmonary fibrosis. Am J Pathol 149: 1241–1256, 1996.
12. Heather F, Bigg Y, Shi E, Yiliang L, Steffensen B, and
Overall CM. Specific, high-affinity binding of tissue inhibitor of
metalloproteinases-4 (TIMP-4) to the COOH-terminal hemopexin-like domain of human gelatinase A. TIMP-4 binds
progelatinase A and the COOH-terminal domain in a similar
manner to TIMP-2. J Biol Chem 272: 15496–15500, 1997.
13. Hiromura K, Pippin JW, Fero ML, Roberts JM, and Shankland SJ. Modulation of apoptosis by the cyclin-dependent kinase
inhibitor p27(Kip1). J Clin Invest 103: 597–604, 1999.
14. Katzenstein ALA and Myers JL. Idiopathic pulmonary fibrosis. Clinical relevance of pathologic classification. Am J Respir
Crit Care Med 157: 1301–1315, 1998.
15. Kuhn C, Boldt J, King TE, Crouch E, Vartio T, and McDonald JA. An immunohistochemical study of architectural
remodeling and connective tissue synthesis in pulmonary fibrosis. Am Rev Respir Dis 140: 1693–1703, 1989.
16. Kuhn C and McDonald JA. The roles of the myofibroblast in
idiopathic pulmonary fibrosis. Ultrastructural and immunohistochemical features of sites of active extracellular matrix synthesis. Am J Pathol 138: 1257–1265, 1991.
17. Kuwano K, Miyazaki H, Hagimoto N, Kawasaki M, Fujita
M, Kunitake R, Kaneko Y, and Hara N. The involvement of
Fas-Fas ligand pathway in fibrosing lung diseases. Am J Respir
Cell Mol Biol 20: 53–60, 1999.
18. Montaño M, Ramos C, González G, Vadillo F, Pardo A, and
Selman M. Lung collagenase inhibitors and spontaneous and
latent collagenase activity in idiopathic pulmonary fibrosis and
hypersensitivity pneumonitis. Chest 96: 1115–1119, 1989.
Pardo A, Barrios R, Maldonado V, Meléndez J, Pérez J,
Ruiz V, Segura L, Sznajder JI, and Selman M. Gelatinases A
and B are upregulated in lung rats by subacute hyperoxia.
Pathogenetic implications. Am J Pathol 153: 833–844, 1998.
Pardo A, Selman M, Ramı́rez R, Ramos C, Montaño M,
Stricklin G, and Raghu G. Production of collagenase and
tissue inhibitor of metalloproteinases by fibroblasts derived from
normal and fibrotic human lungs. Chest 102: 1085–1089, 1992.
Pardo A, Selman M, Ridge K, Barrios R, and Sznajder JI.
Increased expression of gelatinases and collagenase in rat lungs
exposed to 100% oxygen. Am J Respir Crit Care Med 154:
1067–1075, 1996.
Pérez-Padilla R, Salas J, Chapela R, Carrillo G, Sansores R,
Gaxiola M, and Selman M. Mortality in Mexican patients with
chronic pigeon breeders lung compared with those with usual
interstitial pneumonia. Am Rev Respir Dis 148: 49–53, 1993.
Pérez-Ramos J, Segura L, Ramı́rez R, Vanda B, Selman M,
and Pardo A. Matrix metalloproteinases 2, 9, and 13 and tissue
inhibitor of metalloproteinases 1 and 2 in early and late lesions
of experimental lung silicosis. Am J Respir Crit Care Med 160:
1274–1282, 1999.
Pilcher BK, Dumin JA, Sudbeck BD, Krane SM, Welgus
HG, and Parks WC. The activity of collagenase-1 is required for
keratinocyte migration on a type I collagen matrix. J Cell Biol
137: 1445–1457, 1997.
Raghu G, Striker L, Hudson LD, and Striker GE. Extracellular matrix in normal and fibrotic lungs. Am Rev Respir Dis 131:
281–289, 1985.
Seiki M. Membrane-type matrix metalloproteinases. APMIS
107: 137–143, 1999.
Selman M, Carrillo G, Salas J, Pérez R, Sansores R, and
Chapela R. Colchicine, D-penicillamine, and prednisone in the
treatment of idiopathic pulmonary fibrosis: a controlled clinical
trial. Chest 114: 507–512, 1998.
Selman M, Montaño M, Ramos C, Barrios R, and PérezTamayo R. Experimental pulmonary fibrosis induced by paraquat plus oxygen in rats: a morphological and biochemical sequential study. Exp Mol Pathol 50: 147–166, 1989.
Selman M, Montaño M, Ramos C, and Chapela R. Concentration, biosynthesis and degradation of collagen in idiopathic
pulmonary fibrosis. Thorax 41: 355–359, 1986.
Selman M, Montaño M, Ramos C, Chapela R, González G,
and Vadillo F. Lung collagen metabolism and the clinical course of
hypersensitivity pneumonitis. Chest 94: 347–353, 1988.
Smith MR, Kung H, Durum SK, Colburn NH, and Sun Y.
TIMP-3 induces cell death by stabilizing TNF-alpha receptors on
the surface of human colon carcinoma cells. Cytokine 9: 770–780,
1997.
The BAL Cooperative Group Steering Committee. Bronchoalveolar lavage constituents in healthy individuals, idiopathic pulmonary fibrosis, and selected comparison groups. Am
Rev Respir Dis 141, Suppl: 188–192, 1990.
Uhal BD, Iravati J, Ramos C, Pardo A, and Selman M.
Colocalization of alveolar epithelial cell death and underlying
myofibroblasts in advanced fibrotic human lung. Am J Physiol
Lung Cell Mol Physiol 275: L1192–L1199, 1998.
Uhal BD, Joshi I, Mundle S, Raza A, Pardo A, and Selman
M. Fibroblasts isolated after fibrotic lung injury induce apoptosis
of alveolar epithelial cells in vitro. Am J Physiol Lung Cell Mol
Physiol 269: L819–L828, 1995.
Uhal BD, Ramos C, Joshi I, Bifero A, Pardo A, and Selman
M. Cell size, cell cycle, and ␣-smooth muscle actin expression by
primary human lung fibroblasts. Am J Physiol Lung Cell Mol
Physiol 275: L998–L1005, 1998.
Velasco G, Pendás AM, Fueyo A, Knauper V, Murphy G,
and López-Otı́n C. Cloning and characterization of human
MMP-23, a new matrix metalloproteinase predominantly expressed in reproductive tissues and lacking conserved domains
in other family members. J Biol Chem 274: 4570–4576, 1999.
Woessner JF Jr. The family of matrix metalloproteinase family. Ann NY Acad Sci 732: 11–21, 1994.