Antifibrotic Role of HGF in Sarcoidosis

Martin Faehling • Martin Hetzel • Diana Anders •

Gerlinde Trischler • Max Bachem

Received: 26 June 2011 / Accepted: 4 January 2012 / Published online: 5 February 2012

� Springer Science+Business Media, LLC 2012

Abstract

Background Pulmonary sarcoidosis has a variable course

ranging from self-limiting disease to progressive fibrosis.

Activation of fibroblasts, myofibroblast transformation, and

matrix production may contribute to pulmonary damage in

sarcoidosis. These processes are influenced by pulmonary

cytokines which can be measured in bronchoalveolar

lavage fluid (BALF). In order to clarify the incompletely

understood fibrotic process in sarcoidosis, we classified

activity of sarcoidosis according to WASOG criteria,

measured TNF-a, IL-6, and HGF in BALF, and assessed

the effect of HGF and BALF on proliferation and matrix

production of human lung fibroblasts.

Results BALF was obtained from 34 consecutive patients

with sarcoidosis. BALF of active sarcoidosis contained

elevated levels of TNF-a, HGF, and IL-6 and stimulated

fibroblast proliferation. BALF of inactive sarcoidosis, but

not of active sarcoidosis, stimulated the production of

matrix proteins. HGF levels in inactive sarcoidosis were

below those of control patients. HGF suppressed TGF-b-

induced matrix expression and transformation of fibroblasts

into myofibroblasts.

Conclusion Prevention of TGF-b-induced myofibroblast

transformation may account for the inhibitory effect of

HGF on matrix production. The strong fibrogenic effect of

BALF of inactive sarcoidosis corresponds to the worse

clinical course of inactive sarcoidosis compared with active

disease and may be related to a lack of protective HGF.

Keywords Sarcoidosis � HGF � TGF � Fibrosis �Myofibroblast � Extracellular matrix

Introduction

Sarcoidosis is a multisystem granulomatous disorder of

unknown cause that affects primarily the lungs ([90%) and

lymphatic tissues [1–3]. Pulmonary sarcoidosis leads to

restrictive lung function impairment with a variable prognosis

ranging from a self-limiting course ([60%) to progressive

fibrosis (10–30%) with currently no good predictor of out-

come. However, there is a consistent pulmonary inflammatory

response characterized by a lymphocytic alveolitis with pre-

dominance of CD4 helper cells [4]. Although oral steroids

improve symptoms and chest radiograph over 6–24 months, it

is unclear whether steroid therapy has any modifying effect on

long-term disease progression [5, 6].

Immune and inflammatory cells produce a variety of

cytokines that can promote fibroblast proliferation and

deposition of matrix proteins (mainly fibronectin and

M. Faehling � D. Anders � G. Trischler

Klinik fur Innere Medizin II, Universitatsklinik Ulm, Ulm,

Baden-Wuerttemberg, Germany

Present Address:M. Faehling

Klinik fur Kardiologie und Pneumologie, Klinikum Esslingen,

Esslingen, Germany

M. Faehling (&)

Ltd. Arzt Pneumologie, Klinik fur Kardiologie und

Pneumologie, Klinikum Esslingen, Hirschlandstr. 97,

73730 Esslingen, Germany

e-mail: m.faehling@klinikum-es.de

M. Hetzel

Klinik fur Pneumologie, Krankenhaus zum Roten Kreuz,

Stuttgart, Germany

M. Bachem

Zentrale Einrichtung Klinische Chemie und Pathobiochemie,

Universitatsklinik Ulm, Ulm, Baden-Wuerttemberg, Germany

123

Lung (2012) 190:303–312

DOI 10.1007/s00408-012-9372-1

collagen) [7–9]. TGF-b induces transdifferentiation of

fibroblasts or epithelial cells present in normal lungs into a

myofibroblast-like cell type characterized by the expres-

sion of a-smooth muscle cell (a-SMC) actin [10]. Myofi-

broblasts are present in noncaseating granulomas of

sarcoidosis [11]. They are major producers of extracellular

matrix proteins and also secrete various mediators that

could influence and perpetuate the process leading to pul-

monary fibrosis [12, 13].

Bronchoalveolar lavage (BAL) is a standard diagnostic

procedure for patients with sarcoidosis [14]. In clinical

practice, the cellular content of BAL is analyzed and the

BAL fluid (BALF) is generally discarded. Although a high

lymphocyte count and a high CD4/CD8 ratio reflect the

intensity of alveolitis in sarcoidosis, they are not useful as

prognostic indicators [15].

We established an in vitro model of human lung fibro-

blasts and found marked differences in the response of adult

lung fibroblasts to various cytokines tested as single agents,

including the profibrotic cytokines TNF-a and TGF-b [16].

The pulmonary fibroblasts used in these experiments retain

their ability to transdifferentiate in vitro into myofibroblasts

as indicated by TGF-b-induced a-SMC actin expression.

TGF-b has been shown to be elevated in BALF of active

sarcoidosis [17]. Increased release of TNF-a by alveolar

macrophages has been shown, with TNF-a being a prog-

nostic marker for disease progression in sarcoidosis [18], and

anti-TNF-a antibodies have been used to treat refractory

sarcoidosis [19]. Increased release of IL-6, a cytokine with a

role in the regulation of inflammation and immunity [20], has

been found in sarcoidosis [21]. In vivo, these profibrotic

cytokines may be balanced by other, possibly protective

cytokines [7]. An important anti-inflammatory cytokine is

human growth factor (HGF) [22], which has been shown to

be protective and regenerative in renal fibrosis [23, 24]. More

recently, there are indications that HGF may have a protec-

tive role in pulmonary fibrosis as well [25, 26]. In order to

analyze the functional interaction of cytokines in the lung,

we measured TNF-a, IL-6, and HGF in BALF and assessed

the effect of HGF and BALF on proliferation and matrix

production of human lung fibroblasts.

Materials and Methods

Study Population

Fifty-nine nonsmoking subjects undergoing routine bron-

choscopy for diagnostic purposes were studied. Thirty-four

patients had sarcoidosis and 25 were healthy controls. No

study participant was taking oral steroids at the time of

BAL. All subjects gave written informed consent. The

study was approved by the local ethics committee.

In patients with clinically or radiologically suspected

sarcoidosis, the diagnosis was confirmed histologically by

the demonstration of noncaseating granulomas in at least one

organ, usually in transbronchial biopsy, bronchial mucosa, or

mediastinal lymph nodes. Activity of sarcoidosis was clas-

sified clinically as active disease (n = 19) or inactive disease

(n = 15) according to the recommendations of the World

Association of Sarcoidosis and Other Granulomatous Dis-

orders (WASOG) consensus conference [3]. Sarcoidosis was

regarded as active if one of the following criteria was met:

progressive respiratory symptoms (after exclusion of other

causes, e.g., cardiac or infectious diseases), fever, worsening

lung function, progressive bihilar lymphadenopathy, or

progressive interstitial infiltrates. For example, patients with

Lofgren’s syndrome fulfilled the criteria of active sarcoid-

osis. Lung function parameters of patients with sarcoidosis

were slightly restrictive with reduced VC, FEV1, and diffu-

sion capacity compared with control patients (Table 1).

Bronchoalveolar Lavage (BAL)

Patients were sedated with midazolam, and propofol if

necessary. The lower respiratory tract was anaesthetized

with 0.5% lignocaine. A flexible bronchoscope (Olympus)

was wedged into a segmental bronchus of the lingula or

middle lobe. Sterile saline (200 ml) (0.9%) was infused in

ten aliquots. Fluid was aspirated and collected on ice. The

first aliquot was discarded. The fluid was strained through

surgical gauze to remove mucus and centrifuged at

5009g for 10 min. The BAL fluid (BALF) was stored at

-80�C. For cell culture experiments, BALF was concen-

trated tenfold using centrifuge filters (NMWL 3 kDa,

Millipore).

Quantitative Determination of Albumin, TNF-a, HGF,

and IL-6 in BALF

Albumin in BALF was measured according to Lowry.

TNF-a, HGF, and IL-6 in BALF were determined using

commercially available ELISA assays (R&D). The sensi-

tivity for TNF-a was 0.18 pg/ml, for HGF 40 pg/ml, and

for IL-6 0.09 pg/ml.

Immunocytology

Human lung fibroblasts were cultured from normal lung

tissue as described previously [16]. Pulmonary fibroblasts

were cultured on glass culture slides. After growth arrest,

cells were stimulated with test substances for 36 h. Cells

were fixed in acetone and stored at 4�C. Incubation with

the primary antibody (mouse monoclonal: vimentin, des-

min, and a-smooth muscle cell actin; rabbit polyclonal:

fibronectin, collagen type I, and collagen type III) was

304 Lung (2012) 190:303–312

123

performed at working dilutions (established before in

separate experiments) at room temperature for 2 h. Non-

specific staining was controlled for by omitting primary

antibodies and including rat or mouse nonimmune serum,

as appropriate, at the same dilution as used for the specific

primary antibody. After washing, the second antibody

(biotinylated anti-mouse or biotinylated anti-rabbit, as

appropriate) was added. To stain vimentin, desmin,

a-smooth muscle cell actin, or fibronectin, streptavidin-

FITC was added for 30 min, washed, and viewed using

epifluorescence microscopy (Zeiss, Germany). Tyramide

signal amplification was used for staining collagen type I and

collagen type III. For staining collagen type I, the sequence

was primary antibody, biotinylated anti-rabbit IgG, strepta-

vidin-HRP, and biotin-TSA reagent. For collagen type III,

the sequence was primary antibody (biotinylated), strepta-

vidin-HRP, and biotin-TSA reagent. The biotinylated TSA

reagent was incubated at room temperature. After washing,

streptavidin-FITC was added for 30 min. The slides were

observed with a fluorescence microscope.

Proliferation Assays

Proliferation of human lung fibroblasts was measured using

the [3

H]thymidine incorporation assay and cell counts. At

subconfluence, growth was arrested and tenfold-concen-

trated BALF was added in serum-free medium at the

dilution indicated. For [3

H]thymidine incorporation,

15 lCi/well [3

H]thymidine was added after 20 h of incu-

bation with BALF. After 4 h of incubation, [3

H]thymidine

incorporated into cellular DNA was measured as described

previously [16]. Cell counts were performed after 36 h of

incubation with BALF using an automatic cell counter

(Scharfe Systems, Germany).

Quantitative Determination of Fibronectin

Fibroblasts were cultured as for proliferation assays. Test

substances were added for 36 h. The supernatants were

collected and stored at -20�C until analysis. The cells

were counted using an automatic cell counter (Scharfe

Systems, Germany). Soluble c-fibronectin in fibroblast

supernatant or BALF was measured using time-resolved

immune fluorescence (Europium) as described previously

[27]. All measurements were obtained in duplicate. Vari-

ations of the duplicate did not exceed 8%.

Statistics

Unless otherwise stated, results are given as

mean ± standard error of the mean. Statistical significance

was calculated using the Kruskal–Wallis test (for three

groups of unpaired observations), the Mann–Whitney test

(for two groups of unpaired observations), and the Wilco-

xon rank test for paired observations, as appropriate.

Results

BAL Characteristics

BAL cytological data are given in Table 1. The data of

control patients are in line with published normal ranges

[28]. Patients with active sarcoidosis had higher cell counts

than patients with inactive disease and controls. Sarcoid-

osis patients had lymphocytic alveolitis, which was more

pronounced in active than in inactive disease. In accor-

dance with the literature, the CD4/CD8 ratio was signifi-

cantly elevated in patients with sarcoidosis [29]. However,

we found no difference in CD4/CD8 ratio between patients

with active and inactive disease.

Analysis of BAL Fluid Components

Albumin and Fibronectin in BALF

We measured albumin as a protein not produced in the lung

but which enters the lung by leakage from capillaries, and

Table 1 Clinical characteristics, lung functional data, and BAL

cytological data of patients from whom BAL fluid was studied

Controls Active

sarcoidosis

Inactive

sarcoidosis

No. of patients 25 19 15

Age (years) 46.5 ± 2.6 49.1 ± 2.9 41.5 ± 4.6

Male/female 13/12 8/11 8/7

Vital capacity (%) 102 ± 3 94 ± 4* 93 ± 4*

ITGV (%) 93 ± 6 93 ± 5 86 ± 5

FEV1 (%) 105 ± 4 92 ± 4* 91 ± 6*

DCO (%) 86 ± 5 81 ± 5* 79 ± 8*

pO2 77 ± 2 74 ± 2 77 ± 3

Cell count

(91,000/ml)

115 ± 27 288 ± 59* 147 ± 36

BAL lymphocytes

(%)

6.5 ± 1.2 40.0 ± 6.3**? 13.7 ± 3.2*

BAL neutrophils (%) 2.6 ± 0.7 1.8 ± 0.4 2.7 ± 0.5

CD4/CD8 ratio 2.0 ± 0.5 5.3 ± 1.0* 6.0 ± 1.1*

BAL-albumin (mg/l) 31.4 ± 3.8 322 ± 143* 41.0 ± 4.8

BAL-fibronectin

(lg/l)

966 ± 238 16,649 ± 8,005* 971 ± 291

ITGV intrathoracic gas volume, FEV1 forced expiratory volume in 1 s,

DCO diffusion capacity for carbon monoxide, pO2 partial pressure of

oxygen in arterialized capillary blood, CD4/CD8 ratio ratio of CD4 T

helper and CD8 T suppressor/cytotoxic lymphocytes

* p \ 0.05, ** p \ 0.01 versus control, ? p \ 0.05 versus inactive

sarcoidosis

Lung (2012) 190:303–312 305

123

fibronectin, which is produced in the lung. Compared to

control and inactive sarcoidosis, both albumin and fibro-

nectin were significantly elevated in active sarcoidosis

(Fig. 1). Albumin is significantly elevated to a lesser extent

in inactive sarcoidosis and fibronectin is not significantly

elevated in inactive sarcoidosis. Therefore, measurement of

albumin and fibronectin in BALF may serve as activity

parameters for sarcoidosis [30].

TNF-a, HGF, and IL-6 in BALF

BALF from patients with active sarcoidosis contained

significantly higher concentrations of TNF-a, HGF, and

IL-6 than that from control patients (Fig. 2). TNF-a and

IL-6 were elevated to a lesser extent in inactive sarcoidosis,

whereas levels of HGF in inactive sarcoidosis were below

those of control patients. There was no correlation between

Fig. 1 Albumin and fibronectin concentration in BALF. a The

albumin concentrations of the three groups are significantly different

(p \ 0.001, Kruskal–Wallis test). Albumin in active sarcoidosis is

significantly different from that in control (**p \ 0.001) and in

inactive sarcoidosis (**p \ 0.001). Albumin in inactive sarcoidosis is

significantly different from that in control (*p = 0.037, Mann–

Whitney test). b The fibronectin concentrations of the three groups are

significantly different (p \ 0.001, Kruskal–Wallis test). Fibronectin

in active sarcoidosis is significantly different from that in control

(**p \ 0.001) and in inactive sarcoidosis (**p \ 0.001). Fibronectin

in inactive sarcoidosis is not significantly different from that in

control (p = 0.086, Mann–Whitney test)

Fig. 2 Concentrations of TNF-a, HGF, and IL-6 in BALF of patients

with sarcoidosis. a The TNF-a concentrations of the three groups are

significantly different (p = 0.046, Kruskal–Wallis test). TNF-a in

active sarcoidosis is significantly different from that in control

(*p = 0.022) but not from that in inactive sarcoidosis (p = 0.47).

TNF-a in inactive sarcoidosis is significantly different from control

(*p = 0.037, Mann–Whitney test). b The HGF concentrations of the

three groups are significantly different (p = 0.037, Kruskal–Wallis

test). HGF in active sarcoidosis is significantly different from that in

control (*p = 0.045) and from that in inactive sarcoidosis

(*p = 0.013). HGF in inactive sarcoidosis is not significantly

different from that in control (p = 0.714, Mann–Whitney test).

c Overall, the IL-6 concentrations of the three groups are not

significantly different (p = 0.053, Kruskal–Wallis test). However, IL-

6 in active sarcoidosis is significantly different from that in control

(*p = 0.014) but not from that in inactive sarcoidosis (p = 0.149).

IL-6 in inactive sarcoidosis is not significantly different from that in

control (p = 0.714, Mann–Whitney test)

306 Lung (2012) 190:303–312

123

either lymphocyte or neutrophil count in BALF and the

levels of TNF-a, HGF, or IL-6.

Effect of HGF on TGF-b-Induced Fibrogenic Effects

As assessed by immunofluorescence, collagen type I and

collagen type III are seen mainly intracellularly with a fine

granular structure, whereas fibronectin is found predomi-

nantly extracellularly with a fibrillar structure (Figs. 3, 6).

In the absence of stimulation with serum or BALF, the

fibroblasts produce some collagen type I, virtually no

collagen type III, and moderate amounts of fibronectin

(Fig. 3a, e, i). Stimulation with TGF-b results in larger

cells with increased amounts of collagen type III and

fibronectin, but only a slight increase in collagen type I

compared with control (Fig. 3c, g, k). HGF alone does not

affect basal matrix production as compared with control

(Fig. 3b, f, j), but if added together with TGF-b, it sup-

presses the TGF-b-induced expression of collagen type I

and fibronectin to control levels and reduces the expression

of collagen type III (Fig. 3d, h, l).

Fig. 3 Fluorescence micrographs of human lung fibroblasts stimu-

lated by HGF, TGF-b, or TGF-b ? HGF, showing the immunoreac-

tivity of (a–d) collagen type I, (e–h) collagen type III, and

(i–l) fibronectin. Cells were stimulated with (b, f, j) HGF

(3 nmol/l), (c, g, k) TGF-b (2.5 ng/ml), or (d, h, l) TGF-b(2.5 ng/ml) ? HGF (3 nmol/l) in serum-free medium. a, e, i Controls

without added growth factor. Original magnification 9400

Lung (2012) 190:303–312 307

123

a-SMC actin is a marker of myofibroblast transdiffer-

entiation of fibroblasts. Normal human lung fibroblasts do

not express a-SMC actin. In line with previous observa-

tions, TGF-b induces actin expression [31]. HGF com-

pletely suppresses TGF-b-induced actin expression

(Fig. 4). This indicates that HGF prevents the myofibro-

blast transformation induced by TGF-b.

Effect of BAL Fluid on Fibroblast Proliferation

and Matrix Production

Proliferation of pulmonary fibroblasts was stimulated by

BALF from patients with active sarcoidosis in a concen-

tration-dependent manner compared with that of inactive

disease patients and healthy controls (Fig. 5). However,

Fig. 4 Fluorescence

micrographs of human lung

fibroblasts stimulated by HGF,

TGF-b, or TGF-b ? HGF,

showing immunoreactivity of

actin. Cells were stimulated

with b HGF (3 nmol/l),

c TGF-b (2.5 ng/ml), or

d TGF-b (2.5 ng/ml) and HGF

(3 nmol/l) in serum-free

medium for 36 h. a Control

without added growth factor.

Original magnification 9400

Fig. 5 Proliferation of human lung fibroblasts in response to BALF.

a [3

H]thymidine incorporation during stimulation with BALF con-

centrate. BALF concentrate (tenfold) of active sarcoidosis (n = 19,

squares), inactive sarcoidosis (n = 15, circles), or control patients

(n = 25, triangles) was added to the culture medium at various

dilutions. *p \ 0.05 versus control, ?p \ 0.05 versus inactive

sarcoidosis. b Cell count of human lung fibroblasts after 36 h of

stimulation with BALF (tenfold BALF concentrate diluted 1:10 in

serum-free culture medium). For comparison, cell count after

stimulation with 2% FCS is given. *p \ 0.05 versus control

308 Lung (2012) 190:303–312

123

BALF from patients with inactive sarcoidosis stimulated

growth as assessed by cell count, although [3

H]thymidine

incorporation was not stimulated. This difference between

[3

H]thymidine incorporation and cell counts may be due to

the longer time scale of the cell count experiments and may

be explained by secretion of proliferation-stimulating

cytokines (e.g., PDGF [32]) secreted by the fibroblasts in

response to BALF.

As shown by immunofluorescence (Fig. 6), BALF of

control patients slightly enhanced the deposition of

Fig. 6 Fluorescence micrographs of human lung fibroblasts showing

immunoreactivity of matrix proteins in response to BALF. Repre-

sentative fluorescence micrographs of human lung fibroblasts show-

ing the immunoreactivity of collagen type I (a–d), collagen type III

(e–h), or fibronectin (i–l). The fibroblasts were stimulated with BALF

from healthy controls (b, f, j), patients with inactive sarcoidosis (c, g,

k), or active sarcoidosis (d, h, l), respectively. BALF concentrate

(tenfold) was added to serum-free culture medium at a dilution of

1:10 for 36 h. Micrographs (a, e, i) were controls with serum-free

medium and no added BALF. Each figure is representative of three to

five experiments using BALF from different donors. Original

magnification 9400

Lung (2012) 190:303–312 309

123

collagen type I and collagen type III, but not deposition of

fibronectin. BALF of patients with both active and inactive

sarcoidosis enhanced the deposition of collagen type I,

collagen type III, and fibronectin. BALF of inactive sar-

coidosis stimulated the deposition of collagen type I and

fibronectin more than BALF of active sarcoidosis, whereas

for collagen type III there was no difference. Quantification

showed a higher fibronectin concentration in the superna-

tant of fibroblasts stimulated with BALF of inactive sar-

coidosis than in fibroblasts stimulated with BALF of

control patients (Fig. 7). In contrast, BALF of active sar-

coidosis did not result in an increase in fibronectin con-

centration in the supernatant.

Discussion

The elevated levels of TNF-a in sarcoidosis are in line with

its proposed profibrotic role and may thus reflect a pro-

fibrogenic alveolar milieu [33–35]. A variable response of

sarcoidosis to therapy with antibodies directed against

TNF-a has been reported [36, 37]. However, levels of

TNF-a in BALF did not predict response to anti-TNF-atherapies.

The elevation of IL-6 in our Caucasian patients with

sarcoidosis is in line with reports on African-American and

Japanese populations [38, 39]. The more pronounced ele-

vation of the proinflammatory cytokine IL-6 in active than

in inactive sarcoidosis underlines the inflammatory nature

of sarcoidosis and is in accordance with a Japanese report

which showed that elevated IL-6 levels in patients with

sarcoidosis decrease after steroid therapy [40].

The regenerative factor HGF is elevated in active but

not in inactive sarcoidosis. Using the pulmonary fibroblast

model, we show that HGF prevents both TGF-b-induced

matrix deposition and TGF-b-induced transdifferentiation

of fibroblasts into myofibroblasts. This is in line with the

demonstration that HGF prevents myofibroblast transdif-

ferentiation of rat epithelial cells [40]. The relevant role of

fibroblast transdifferentiation in pulmonary fibrosis is

supported by a study that combined in situ hybridization

analysis for procollagen I mRNA and immunochemistry

for a-SMC actin expression and showed that the first cells

to become a-SMC actin-positive and expressing procolla-

gen mRNA are fibroblasts in areas with incipient fibrosis

[41]. Our findings are in accordance an antifibrotic effect of

HGF, which was demonstrated in bleomycin-treated mice

[25, 42, 43].

The effect of BALF on proliferation and matrix produc-

tion of human lung fibroblasts reflects the activity of sar-

coidosis. BALF of patients with inactive sarcoidosis induces

little stimulation of fibroblast proliferation but pronounced

increases in the deposition of matrix proteins. This increase

is not found in active sarcoidosis, which in turn strongly

induces proliferation, indicating that proliferation and

matrix production by human lung fibroblasts are regulated

independently by factors contained in BALF. The low rate of

matrix synthesis by lung fibroblasts stimulated with BALF

from active sarcoidosis in vitro may imply that the rate of

matrix deposition and thus the risk of irreversible fibrotic

changes in vivo may be low, whereas this risk may be higher

in patients with clinically inactive sarcoidosis. This corre-

sponds to the good prognosis of patients with active sar-

coidosis, e.g., Lofgren’s syndrome [3]. The discrepancy

between clinical disease activity and fibrogenic activity in

sarcoidosis is in line with the lack of reliable clinical pre-

dictors of the outcome of sarcoidosis and underlines the need

for better markers of the fibrosing propensity of sarcoidosis

in an individual patient (ref editorial WASOG). Measure-

ment of the fibrogenic effect of BALF on human adult lung

fibroblasts may provide such markers.

Activation of the epidermal growth factor receptor (EGFR)

results in collateral activation of the HGF receptor (cMET)

[44]. Gefitinib, a tyrosine kinase inhibitor of EGFR, inhibits

the activation of cMET and may cause interstitial lung disease.

With our findings, it is tempting to speculate that interstitial

lung disease in gefitinib therapy might be due to inhibition of

Fig. 7 Fibronectin production of fibroblasts in response to BALF.

BALF concentrate (tenfold) from patients with the given diagnosis

was added to the culture medium of growth-arrested normal human

lung fibroblasts at 1:10 dilution. Cellular fibronectin was measured in

the cell supernatants after 36 h. The fibronectin production of the

three groups is significantly different (p = 0.015, Kruskal–Wallis

test). The fibronectin production of active sarcoidosis is not signif-

icantly different from control (p = 0.804). However, the fibronectin

production of inactive sarcoidosis is different from that of control

(**p = 0.006) and from that of active sarcoidosis (?p = 0.028,

Mann–Whitney test)

310 Lung (2012) 190:303–312

123

the antifibrotic activity of HGF. With both small molecules

(ARQ197) and antibodies (METMAB) directed against the

HGF receptor (cMET) currently being studied in various fields

of oncology, including lung cancer [45], it will be interesting

to see whether interstitial lung disease occurs at an increased

rate with the use of cMET inhibitors.

Taken together, our findings suggest a protective role of

HGF in pulmonary fibrogenesis in sarcoidosis. The strong

fibrogenic effect of BALF of inactive sarcoidosis may be

related to low levels of HGF which are insufficient to balance

the profibrotic alveolar milieu which is reflected in BALF.

Future clinical studies should test whether a low level of HGF

in BALF identifies sarcoidosis patients at risk of developing

progressive fibrosis and whether restoration of normal pul-

monary HGF levels in these patients prevents fibrosis.

Conflict of interest The authors have no conflicts of interest to

disclose.

References

1. Baughman RP, Culver DA, Judson MA (2011) A concise review

of pulmonary sarcoidosis. Am J Respir Crit Care Med 183:

573–581

2. Iannuzzi MC, Rybicki BA, Teirstein AS (2007) Sarcoidosis.

N Engl J Med 357:2153–2165

3. ATS/ERS/WASOG Committee (1999) Statement on sarcoidosis.

Am J Respir Crit Care Med 160:736–755

4. Grunewald J, Eklund A (2007) Role of CD4? T-cells in sar-

coidosis. Proc Am Thorac Soc 4:461–464

5. Paramothayan NS, lasserson TJ, Jones PW (2005) Corticosteroids

for pulmonary sarcoidosis. Cochrane Database Syst Rev

CD001114

6. Baughman RP, Costabel U, du Bois RM (2008) Treatment of

sarcoidosis. Clin Chest Med 29:533–548

7. Gurrieri C, Bortoli M, Brunetta E, Piazza F, Agostini C (2005)

Cytokines, chemokines and other biomolecular markers in sar-

coidosis. Sarcoidosis Vasc Diffus Lung Dis 22:S9–S14

8. Zhang K, Phan SH (1996) Cytokines and pulmonary fibrosis. Biol

Signals 5:232–239

9. Thickett DR, Poole AR, Millar AB (2001) The balance between

collagen synthesis and degradation in diffuse lung disease. Sar-

coidosis Vasc Diffus Lung Dis 18:27–33

10. Willis BC, Borok Z (2007) TGF-beta-induced EMT: mechanisms

and implications for fibrotic lung disease. Am J Physiol Lung

Cell Mol Physiol 293:L525–L534

11. Kaarteenaho-Wiik R, Sademies O, Paakko P, Ristele J, Soini Y

(2007) Extracellular matrix proteins and myofibroblasts in gran-

ulomas of sarcoidosis, atypical mycobacteriosis, and tuberculosis

of the lung. Hum Pathol 38:147–153

12. Serini G, Gabbiani G (1999) Mechanism of myofibroblast activity

and phenotype modulation. Exp Cell Res 250:273–283

13. Hinz B, Phan SH, Thannickal VJ, Galli A, Bochaton-Piallat ML,

Gabbiani G (2007) The myofibroblast: one function, multiple

origins. Am J Pathol 170:1807–1816

14. Costabel U, Guzman J (2001) Bronchoalveolar lavage in inter-

stitial lung disease. Curr Opin Pulm Med 7:255–261

15. Verstraeten A, Demedts M, Verwilghen J, van den Eeckhout A,

Marien G, Lacquet LM, Ceuppens JL (1990) Predictive value of

bronchoalveolar lavage in pulmonary sarcoidosis. Chest 98:

560–567

16. Hetzel M, Bachem M, Anders D, Trischler G, Faehling M (2005)

Different effects of growth factors on proliferation and matrix

production of normal and fibrotic human lung fibroblasts. Lung

183:225–237

17. Salez F, Gosset P, Copin MC, Lamblin Degros C, Tonnel AB,

Wallaert B (1998) Transforming growth factor-beta1 in sarcoid-

osis. Eur Respir J 12:913–919

18. Armstrong L, Foley NM, Millar AB (1999) Inter-relationship

between tumour necrosis factor-alpha (TNF-alpha) and TNF

soluble receptors in pulmonary sarcoidosis. Thorax 54:524–530

19. Baughman RP, Lower EE, Drent M (2008) Inhibitors of tumor

necrosis factor (TNF) in sarcoidosis: who, what, and how to use

them. Sarcoidosis Vasc Diffus Lung Dis 25:76–89

20. Kishimoto T (2010) IL-6: from its discovery to clinical applica-

tions. Int Immunol 22:347–352

21. Ziegenhagen MW, Benner G, Zissel P, Zabel P, Schlaak M,

Mueller-Quernheim J (1997) Sarcoidosis: TNF-alpha release

from alveolar macrophages and serum level of sIL-2R are

prognostic markers. Am J Respir Crit Care Med 156:1586–1592

22. Mizuno S, Matsumoto K, Kurosawa T, Mizuno-Horikawa Y,

Nakamura T (2000) Reciprocal balance of hepatocyte growth

factor and transforming growth factor-beta 1 in renal fibrosis in

mice. Kidney Int 57:937–948

23. Funakoshi H, Nakamura T (2003) Hepatocyte growth factor:

from diagnosis to clinical applications. Clin Chim Acta 327:1–23

24. Iekushi K, Taniyama Y, Azuma J, Sanada F, Kusunoki H, Yokoi

T, Koibuchi N, Okayama K, Rakugi H, Morishita R (2010)

Hepatocyte growth factor attenuates renal fibrosis through TGF-

b1 suppression of apoptosis of myofibroblasts. J Hypertens

28:2454–2461

25. Gazdhar A, Fachinger P, van Leer C, Pierog J, Gugger M, Friis R,

Schmid RA, Geiser T (2007) Gene transfer of hepatocyte growth

factor by electroporation reduces bleomycin-induced lung fibro-

sis. Am J Physiol Lung Cell Mol Physiol 292:L529–L536

26. Panganiban RA, Day RM (2011) Hepatocyte growth factor in

lung repair and pulmonary fibrosis. Acta Pharmacol Sin 32:12–20

27. Bachem MG, Schneider E, Groß H, Weidenbach H, Schmid RM,

Menke A, Siech M, Beger H, Grunert A, Adler G (1998) Iden-

tification, culture, and characterization of pancreatic stellate cells

in rats and humans. Gastroenterology 115:421–432

28. Lensmar C, Elmberger G, Sandgren P, Skold CM, Eklund A

(1998) Leukocyte counts and macrophage phenotypes in induced

sputum and BALF from normal subjects. Eur Respir J 12:

595–600

29. Costabel U (1997) CD4/CD8 ratios in bronchoalveolar lavage

fluid: of value for diagnosing sarcoidosis? Eur Respir J 10:

2699–2700

30. Kriegova E, Melle C, Kolek V, Hutyrova B, Mrazek F, Bleul A,

du Bois RM, von Eggeling F, Petrek M (2006) Protein profiles of

bronchoalveolar lavage fluid from patients with pulmonary sar-

coidosis. Am J Respir Crit Care Med 173:1145–1154

31. Desmouliere A, Geinoz A, Gabbiani F, Gabbiani G (1993) TGF-

b1 induces aSMC actin expression in granulation tissue and in

quiescent and growing fibroblasts. J Cell Biol 122:103–111

32. Bonner JC (2010) Mesenchymal cell survival in airway and

interstitial pulmonary fibrosis. Fibrogenesis Tissue Repair 3:15

33. Piguet PF, Ribaux C, Karpuz V, Grau GE, Kapanci Y (1993)

Expression and localization of tumor necrosis factor-alpha and its

mRNA in idiopathic pulmonary fibrosis. Am J Pathol 143:

651–655

34. Hasegawa M, Fujimoto M, Kikuchi K, Takehara K (1997) Ele-

vated serum tumor necrosis factor-alpha levels in patients with

systemic sclerosis: association with pulmonary fibrosis. J Rheu-

matol 24:663–665

Lung (2012) 190:303–312 311

123

35. Oikonomou N, Harokopos V, Zalevsky J, Valavanis C, Kotani-

dou A, Szymkowski DE, Kollias G, Aidinis V (2006) Soluble

TNF mediates the transition from pulmonary inflammation to

fibrosis. PLoS One 1:e108

36. Doty JD, Mazur JE, Judson MA (2005) Treatment of sarcoidosis

with infliximab. Chest 127:1064–1071

37. Utz JP, Limper AH, Kalra S, Specks U, Scott JP, Vuk-Pavlovic Z,

Schroeder DR (2003) Etanercept for the treatment of stage II and

III progressive pulmonary sarcoidosis. Chest 124:177–185

38. Girgis RE, Basha MA, Maliarik M, Popovich J Jr, Iannuzzi MG

(1995) Cytokines in the bronchoalveolar lavage fluid of patients

with active pulmonary sarcoidosis. Am J Respir Crit Care Med

152:71–75

39. Takizawa H, Satoh M, Akazaki H, Matsusaki G, Suzuki N, Ishii

A, Suko M, Okudaira H (1997) Increased IL-6 and IL-8 in

bronchoalveolar lavage fluid (BALF) from patients with sar-

coidosis: correlation with clinical parameters. Clin Exp Immunol

107:175–181

40. Shukly MN, Rose JL, Lathrop KL, Ray A, Ray P (2009) Hepa-

tocyte growth factor inhibits epithelial to myofibroblast transition

in lung cells via Smad7. Am J Respir Cell Mol Biol 40:643–653

41. Zhang K, Rekhter MD, Gordon D, Phan SH (1994) Myofibro-

blasts and their role in lung collagen gene expression during

pulmonary fibrosis. Am J Pathol 145:114–125

42. Watanabe M, Ebina M, Orson FM, Nakamura A, Kubota K,

Koinuma D, Akiyama K, Maemondo M, Okouchi S, Tahara M,

Matsumoto K, Nukiwa T (2005) Hepatocyte growth factor gene

transfer to alveolar septa for effective suppression of lung fibro-

sis. Mol Ther 12:58–67

43. Mizuno S, Matsumoto K, Li MY, Nakamura T (2005) HGF

reduces advancing lung fibrosis in mice: a potential role for

MMP-dependent myofibroblast apoptosis. FASEB J 19:580–582

44. Dulak AM, Gubish CT, Stabile LP, Henry C, Siegfried JM (2011)

HGF-independent potentiation of EGFR action by c-Met. Onco-

gene 30:3625–3635

45. Sattler M, Reddy MM (2011) The role of the c-Met pathway in

lung cancer and the potential for targeted therapy. Ther Adv Med

Oncol 4:171–184

312 Lung (2012) 190:303–312

123