AJP - Lung AJP: Renal Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Lung Cell Mol Physiol 292: L1263-L1272, 2007. First published January 19, 2007; doi:10.1152/ajplung.00191.2006
1040-0605/07 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
292/5/L1263    most recent
00191.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Beaufort, N.
Right arrow Articles by Pidard, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Beaufort, N.
Right arrow Articles by Pidard, D.

The human airway trypsin-like protease modulates the urokinase receptor (uPAR, CD87) structure and functions

Nathalie Beaufort,1,2,3 Dominique Leduc,1,2 Hiroshi Eguchi,4 Karin Mengele,3 Daniela Hellmann,3 Tsukio Masegi,4 Takashi Kamimura,4 Susumu Yasuoka,5 Falko Fend,6 Michel Chignard,1,2 and Dominique Pidard1,2

1Inserm, E0336, Paris, France; 2Institute Pasteur, Unité de Défense Innée et Inflammation, Département Infection et Epidémiologie, Paris, France; 3Department of Obstetrics and Gynaecology, Technical University of Munich, Munich, Germany; 4Pharmaceutical Discovery Research Laboratories, Teijin Institute of Bio-Medical Research, Teijin Limited, Tokyo, Japan; 5Department of Nutrition and Metabolism, Graduate School of Nutrition and Bioscience, The University of Tokushima, Tokushima, Japan; and 6Institute of Pathology, Technical University of Munich, Munich, Germany

Submitted 1 June 2006 ; accepted in final form 12 January 2007


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
The human airway trypsin-like protease (HAT) is a respiratory epithelium-associated, type II transmembrane serine protease, which is also detected as an extracellular enzyme in lung fluids during airway inflammatory disorders. We have evaluated its capacity to affect the urokinase-type plasminogen activator receptor (uPAR), a membrane glycolipid-anchored, three-domain (D1D2D3) glycoprotein that plays a crucial role in innate immunity and inflammation by supporting cell migration and matrix degradation, with structure and biological properties that can be regulated via limited endoproteolysis. With the use of immunoblotting, flow immunocytometry, and ELISA analyses applied to a recombinant uPAR protein and to uPAR-expressing monocytic and human bronchial epithelial cells, it was shown that exposure of uPAR to soluble HAT in the range of 10–500 nM resulted in the proteolytic processing of the full-length (D1D2D3) into the truncated (D2D3) species, with cleavage occurring in the D1 to D2 linker sequence after arginine residues at position 83 and 89. Using immunohistochemistry, we found that both HAT and uPAR were expressed in the human bronchial epithelium. Moreover, transient cotransfection in epithelial cells showed that membrane coexpression of the two partners produced a constitutive and extensive shedding of the D1 domain, occurring for membrane-associated HAT concentrations in the nanomolar range. Because the truncated receptor was found to be unable to bind two of the major uPAR ligands, the adhesive matrix protein vitronectin and the serine protease urokinase, it thus appears that proteolytic regulation of uPAR by HAT is likely to modulate cell adherence and motility, as well as tissue remodeling during the inflammatory response in the airways.

urokinase-type plasminogen activator receptor; human bronchial epithelial cells; human monocytes; lung inflammation

AFTER LUNG TISSUE injuries, a number of trypsin- and chymotrypsin-like proteinases are involved in the (patho)physiology of inflammation, extracellular matrix (ECM) remodeling, and healing in the airways or in the lung parenchyma (13, 19, 23). Yasuoka et al. (40) reported that the overall trypsin-like proteolytic activity, which can be detected in the sputum of patients with chronic airway inflammatory diseases, was essentially associated to a single, undescribed extracellular monomeric serine (Ser) proteinase with a molecular mass of ~27 kDa, named human airway trypsin-like protease (HAT). Subsequently, HAT was characterized as a membrane-spanning, 47-kDa single-chain protein, which can be proteolytically converted into a 27-kDa soluble species corresponding to the carboxy-terminal extracellular catalytic domain (39). Although the latter shares a high homology with other trypsin-like Ser proteinases, the membrane-spanning noncatalytic domain of HAT displays a structure that relates HAT to the emerging family of type II transmembrane Ser proteinases (35, 39). Within human tissues, so far only the airways, and more recently the skin, have been shown to express HAT (14, 37). In the airways, HAT expression is restricted to the apical cilia layer of tracheobronchial epithelial cells (22, 37). With regard to potential substrates and functions, HAT has been found to target extracellular proteins, both degrading fibrinogen and activating the zymogen pro-urokinase-type plasminogen activator (pro-uPA) (22, 41), to induce IL-8 secretion by airway epithelial cells through activation of the proteinase-activated receptor (PAR)-2, as well as to stimulate mucin secretion by these cells (7, 22). These in vitro effects of HAT have been found to occur for proteinase concentrations similar to those that have been measured in the sputum of patients with various inflammatory airway diseases, i.e., in the range of 5–40 nM for HAT activity (41) and 2–100 nM for the HAT antigen measured by ELISA (21).

The urokinase-type plasminogen activator receptor (uPAR, CD87) is a highly glycosylated, glycosylphosphatidylinositol (GPI)-anchored cell membrane protein. It is made of three structurally homologous globular domains, named D1, D2, and D3 from the amino to the carboxy terminus of the molecule, each containing a conserved arrangement of disulfide bonds and separated by short interdomain linker sequences (27). uPAR is present on the surface of many cell types, including monocytes and macrophages, polymorphonuclear neutrophils, vascular endothelial, smooth muscle, and epithelial cells, and is markedly overexpressed during inflammatory processes (4, 8, 33). uPAR binds pro-uPA with high affinity, and the uPAR-bound Ser proteinase uPA efficiently catalyzes the activation of plasminogen into plasmin, which in turn activates a proteolytic cascade [including matrix metalloproteinases (MMP)], thus conferring to uPAR-expressing cells a high potential for pericellular proteolysis, ECM processing, and cell motility (4, 27). uPAR participates in cell adherence and migration through other routes that include 1) its capacity to bind the ECM adhesive protein vitronectin (28, 30), 2) its physical and functional interaction with various integrins (4, 15), and 3) its capacity to expose an intrinsic, proteinase-regulated chemotactic epitope localized within the D1-D2 linker sequence (4). In agreement with these properties, uPAR and the downstream plasminogen activation system are involved in the migration of both inflammatory leukocytes to sites of injury (11, 24) and epithelial cells, including those of the respiratory tract, during repair (6, 13, 17).

Interestingly, uPAR is highly susceptible to endoproteolysis within the D1-D2 linker sequence, which contains cleavage sites for proteinases of the plasminogen activation cascade itself, i.e., uPA, plasmin, and various MMP (1, 3, 12), but also for the inflammatory leukocyte Ser proteinases elastase and cathepsin G (2). Because the D1 domain of uPAR plays a major role in the interactions of the receptor with its various ligands (27, 28), its proteolytic shedding is thus a likely pathway for controlling pericellular proteolysis and cell adherence and migration (4, 28). Along with this assumption, soluble uPAR species (suPAR), including the free D1 domain and truncated D2D3 forms, can be detected in the biological fluids of healthy individuals (31), whereas suPAR concentrations are markedly increased in infectious and inflammatory disorders and can be of prognostic value for a negative outcome, including in lung diseases (25).

In the scope of a growing interest for the (patho)physiological processes of the regulation of membrane receptor activity through proteolytic ectodomain shedding (10), we provide new insights on the putative functions of HAT, by establishing that this airway proteinase regulates human uPAR (h-uPAR), mostly through cleavage of the D1-D2 linker sequence and generation of truncated uPAR species of potential importance for the cell behavior under inflammatory conditions.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Reagents and antibodies. The glycosylated recombinant h-uPAR (Leu1-Arg281), devoid of a carboxy-terminal GPI anchor, which is replaced by a polyhistidine tag (rhuPAR[His]6), was from R&D Systems (Minneapolis, MN). Bacterial phosphatidylinositol-specific phospholipase C (PI-PLC; 5,100 U/mg) and benzamidine were from Sigma-Aldrich (St. Louis, MO). FITC-conjugated human recombinant uPA was from American Diagnostica (Greenwich, CT), and purified human plasma vitronectin was from BioSource International (Camarillo, CA). The recombinant 27-kDa soluble, catalytically active HAT species was produced and purified as previously reported (22, 39), with a specific activity of 30.6 U/mg, and was stored at –80°C in a 20 mM sodium phosphate buffer.

Antibodies used in this study were mouse MAbs 3931 (also referred to as IIIF10, against the uPAR D1 domain) and 3932 (also referred to as IID7, against the uPAR D2 domain) (20), both purified IgG1 from American Diagnostica, except for immunohistochemistry (see below), which used MAb IID7 as the hybridoma culture medium (a kind gift from Dr. Viktor Magdolen, Technical University of Munich, Munich, Germany); mouse MAbs R2 and R4 (a kind gift of Dr Gunilla Høyer-Hansen, The Finsen Laboratory, Copenhagen, Denmark), both IgG1 recognizing different epitopes in the D3 domain (12, 18); a mouse MAb directed to HAT as previously characterized (37); a nonspecific MAb of the IgG1 subclass from Ancell (Bayport, MN); His-probe (H-15), a rabbit polyclonal antibody against the [His]6 tag, from Santa Cruz Biotechnologies (Santa Cruz, CA); FITC-conjugated F(ab)2 goat antibody against mouse IgG from Dako (Glostrup, Denmark); and horseradish peroxidase (HRP)-conjugated antibodies against mouse or rabbit IgG (ImmunoPure) from Pierce (Rockford, IL).

Cell culture reagents were from GIBCO BRL (Paisley, Scotland), except for heat-inactivated FCS from Hyclone (Logan, UT). Chemicals for cell solubilization, SDS-PAGE, and protein transfer were essentially from Sigma-Aldrich or from Bio-Rad (Hercules, CA).

Cells. 16HBE14o– cells, hereafter designated as 16HBE cells, are human SV40-transfected bronchial epithelial cells (9) and were a gift from Dr. C. Gruenert (University of Vermont, Colchester, VT). Cells were cultured to confluence in 24-well plates; when needed, cells were surface biotinylated exactly as previously described (3). The U937 human promonocytic cell line (34) (CRL-1593.2; American Type Culture Collection, Manassas, VA) was differentiated into monocytes and macrophages by exposure to 15 nM PMA (Sigma-Aldrich) for 48 h at a density of 0.5 x 106 cells/ml. Differentiated adherent U937 cells were recovered as described before, as was biotinylation of membrane proteins in some experiments (2). The human embryonic kidney epithelial cell line 293T (HEK cells) was purchased from GenHunter (Nashville, TN). HEK cells were maintained in culture in DMEM containing 10% (vol/vol) FCS. It must be noted that tests to rule out the presence of mycoplasmal contamination in cultured cell lines were not performed.

Monocytes (purity >85%) were isolated from peripheral blood obtained from human volunteers with their informed consent (Etablissement Français du Sang, Paris, France), as previously described (2).

Exposure of recombinant uPAR or uPAR-expressing cells to soluble HAT. rhuPAR[His]6 adjusted to 1 µg/ml (~20 nM) in HBSS either was mixed with varying concentrations of soluble HAT, for enzyme-to-substrate molar ratios (E/S) in the range from 0.5/1 to 25/1, or was left untreated as control. After incubation for 5–60 min at 37°C, the enzyme was neutralized by addition of 5 mM benzamidine, a Ser proteinase inhibitor that readily blocks HAT activity (40). The samples were then processed for SDS-PAGE as previously described (2).

16HBE cell monolayers covered with 200 µl of HBSS containing CaCl2 and MgCl2, 1 mM each (HBSS-CaMg), or monocytic cells suspensions adjusted to 1 x 106 cells/ml in HBSS received 10–500 nM of soluble HAT, and incubation continued at 37°C for 5–60 min, before enzymatic activity was blocked with benzamidine as above. Alternatively, cells were incubated with 1 U/ml PI-PLC for 30 min under similar conditions. Extracellular fluids were collected and centrifuged at 18,000 g for 30 min at 4°C to eliminate cell debris. In some experiments, the extracellular fluids obtained from 16HBE cell monolayers exposed to PI-PLC were subsequently exposed to 50–500 nM HAT for 30 min at 37°C before HAT was blocked by benzamidine. When needed, biotinylated 16HBE cell monolayers or monocytic cell pellets were solubilized at 4°C for 30 min in a radioimmunoprecipitation assay medium [150 mM NaCl, 25 mM HEPES, 5 mM EDTA, 0.1% (wt/vol) SDS, 1% (vol/vol) Nonidet P-40, 0.5% (wt/vol) deoxycholate, pH 7.5] containing a cocktail of protease inhibitors (1 mM PMSF, 10 µM aprotinin, 5 mM benzamidine, 100 µM leupeptin, 1.5 µM soybean trypsin inhibitor, and 5 mM N-ethylmaleimide), before removal of insoluble material by centrifugation at 18,000 g for 30 min at 4°C. Protein concentration in the cell lysates was measured by the bicinchoninic acid protein assay (Pierce). Nonbiotinylated monocytic cells were resuspended at 1 x 106/ml in HBSS containing 0.5% (wt/vol) BSA (HBSS-BSA) and kept at 4°C until fluorescence-activated cell sorting (FACS) analysis (see below).

Coexpression of membrane HAT and uPAR in HEK cells. The HEK cell line was used for transient coexpression of membrane HAT and uPAR. HAT and h-uPAR cDNAs were cloned into the pEF4/Myc-His A expression vector (Invitrogen, Carlsbad, CA) using BglII and NotI and EcoRI and NotI sites, respectively. For this, PCR primer sequences were as follows: for h-uPAR (accession no. NM_002659), h-uPAR-F0228(EcoRI) GGTT GAATTC GAC ATG GGT CAC CCG CCG CT and h-uPAR-R1126(NotI) GGTT GCGGCCGC GGG ATT TCA GGT TTA GGT CCA GAG G; for HAT (accession no. AB002134), HAT-F0059(BglII) GGTT AGATCT AAA ATG TAT AGG CCA GCA CGT GT and HAT-R1298(NotI) GGTT GCGGCCGC CTA GAT CCC AGT TTG TTG CCT. The PCR fragment of h-uPAR cDNA was amplified from human lung cDNA (BD Biosciences Clontech, Palo Alto, CA), and the PCR fragment of HAT cDNA was amplified from human trachea cDNA synthesized with human poly(A+) RNA (BD Bioscience Clontech). The enzymatically inactive variant of transmembrane HAT cDNA was constructed by using PCR-based single base pair substitution at position 368, resulting in alanine for serine within the catalytic triad, and is designated hereafter as HAT-S368A. For this, PCR primer sequences were as follows: for HAT-S368A-F; AGG GTG ACG CTG GTG GCC CAC TAG TAC AAG; for HAT-S368A-R, GCC ACC AGC GTC ACC CTG ACA TGC GTC CAC.

Transmembrane HAT and GPI-anchored uPAR were coexpressed by transfection of the expression plasmids described above in the HEK cells. For this, cells were plated on two 12-well poly-D-lysine-coated plates (Becton Dickinson Labware, Bedford, UK) and grown to 90–95% confluence. Cells were then transfected using Lipofectamine 2000 (Invitrogen) with 2 µg of plasmids (1 µg of uPAR together with 1 µg of either active HAT or inactive HAT-S368A or with 1 µg of an empty plasmid as control), according to standard procedures. After 4-h incubation with the transfection mix at 37°C in a CO2 incubator, the medium was changed to 1 ml of DMEM containing 10% FCS, and cells were further incubated for 24 h, the period of time assigned to the experimental coexpression of uPAR with either active HAT or inactive HAT-S368A. Extracellular fluids were then collected and centrifuged at 18,000 g for 30 min at 4°C to remove nonsoluble material and kept at –80°C until either immunoblotting analysis or measurement of HAT activity (see Immunoblot analysis of recombinant, cellular, or suPAR). After removal of the culture fluids, cell monolayers were either surface biotinylated and solubilized in view of immunoblotting analysis of uPAR exactly as described above for the 16HBE cell line or washed with 1 ml/well PBS and then lysed by using the M-PER reagent (Pierce) (0.5 ml/well, 60 min at 4°C), before centrifugation at 18,000 g for 30 min at 4°C and storage of the cleared lysates at –80°C until measurement of the cell-associated HAT activity.

Measurement of HAT enzymatic activity. The HAT hydrolytic activity was measured essentially as previously described (40). Briefly, cell lysates and extracellular culture fluids were diluted 20- and 10-fold, respectively, into a 50 mM Tris·HCl (pH 8) buffer containing 0.01% (wt/vol) BSA, whereas standard HAT samples were prepared by dilution of a stock solution of recombinant soluble HAT, for final concentrations in the range from 0.08 to 4 nM. Test samples or standard samples (40 µl) placed into wells of a 96-wells plate were added with 20 µl of a 0.3 mM solution of the fluorogenic substrate Boc-Phe-Ser-Arg-methylcoumarinamide (Peptide Institute, Osaka, Japan). Assay was continued for 5 min, and the fluorescence emitted by free aminomethylcoumarin was monitored at an excitation wavelength of 360 nm and an emission wavelength of 460 nm. HAT activity was expressed as the increase of fluorescence intensity per minute, and HAT molar concentration was then extrapolated from the curve derived from the standard HAT solutions.

Determination of cleavage sites for HAT within uPAR. RhuPAR[His]6 (0.1 mg/ml, ~2 µM) in HBSS was incubated overnight at 37°C with 10 nM of HAT (E/S = 1/200) or left untreated as control. After protein separation by SDS-PAGE under reducing conditions, followed by transfer onto polyvinylidene fluoride membranes (Immobilon-PSQ; Millipore, Bedford, MA), proteins were visualized by amido black staining, and those of interest were excised and subjected to amino-terminal microsequencing on an Applied Biosystem ABI 494 protein sequencer.

Immunoblot analysis of recombinant, cellular, or suPAR. Biotinylated cell membrane proteins were separated from nonbiotinylated proteins on NeutrAvidin-agarose beads (Pierce) as previously reported (2). rhuPAR[His]6 (15 ng/lane), biotinylated cell membrane proteins (extracted from 5 or 10 µg of total cell proteins/lane for the HEK and U937 cells or for the 16HBE cells, respectively), or proteins present in the extracellular fluids (30 or 40 µl/lane, corresponding to the equivalent of 1, 2.5, or 20 µg of total cell proteins, for the HEK, the U937, or the 16HBE cells, respectively), including the PI-PLC-derived suPAR further exposed to HAT, were separated by SDS-PAGE under disulfide reducing conditions as previously described (2). After SDS-PAGE, proteins were transferred to Immobilon-P polyvinylidene fluoride membranes, which were then blocked with dried skimmed milk and probed with a primary anti-uPAR MAb at the concentrations indicated. Bound antibodies were detected by an HRP-coupled second antibody with the HRP activity revealed by chemiluminescence using the enhanced chemiluminescence reagent kit (Amersham Biosciences, Little Chalfont, UK). Calibration for relative molecular mass determination was performed with the Kaleidoscope prestained or Precision Plus All blue standards (Bio-Rad).

Detection of suPAR by quantitative ELISA. The concentration of suPAR in cell-free 16HBE or U937 extracellular fluids was determined by a specific quantitative ELISA kit (Quantikine ELISA for h-uPAR, R&D Systems), according to the manufacturer's instructions. The lower detection limit of the assay is 62.5 pg/ml suPAR.

FACS analysis for binding of antibodies or uPA to monocytic cells. After exposure to HAT, U937 cells or blood monocytes were resuspended in fresh HBSS, distributed at 105 cells/well in conical-bottomed, 96-wells microplates (Nunc, Roskilde, Denmark), sedimented, and then immunolabeled for 1 h at 4°C in 0.1 ml of HBSS-BSA containing a primary anti-uPAR MAb at 5–10 µg/ml. After samples were washed, immunofluorescence staining and FACS analysis then proceeded at 4°C exactly as described before (2), using a FACScan cytometer (Becton Dickinson, Franklin Lakes, NJ). For FITC-uPA binding assays, cells were resuspended in 0.1 ml of HBSS-BSA containing 50 nM FITC-uPA and then incubated for 1 h at 4°C before extensive washings, as previously described (2). Quantitative binding of specific IgG or of FITC-uPA to the surface of the selected viable cells (defined as those remaining negative for labeling with propidium iodide) was expressed as the geometric mean of fluorescence intensity histograms, after background binding provided by nonspecific IgG isotypes or FITC-F(ab)2 had been subtracted. It has to be noted that cells remained unfixed through the whole procedure, to avoid undesired modifications of the MAb epitope expression due to fixatives.

Immunohistochemical staining for uPAR and HAT in human lung tissue. Normal lung tissue and pulmonary parenchyma with chronic nonspecific inflammatory changes were obtained from patients undergoing pulmonary resection for malignant or inflammatory disease. Specimens were formalin fixed, paraffin embedded, and sectioned into 2-µm-thick slices.

Immunohistochemical staining for uPAR and HAT were performed with the EnVision method and kit (Dako). Briefly, lung sections were dewaxed in xylene and rehydrated in descending graded ethanol. For uPAR staining, sections were further subjected to heat-induced epitope retrieval by pressure cooking for 4 min in a 10 mM citrate buffer, pH 6.0. Endogenous peroxidase activity was inhibited by a 20-min incubation of the slides with endogenous enzyme block, followed by an overnight incubation at 4°C with anti-HAT (10 µg/ml) or anti-uPAR (IID7, 1/10 dilution of the hybridoma conditioned medium) MAbs, both diluted in antibody diluent. As negative control, the primary antibody was omitted. Slides were then incubated for 30 min at room temperature with a labeled polymer-HRP-Ab, followed by detection of immunocomplexes for 10 min at room temperature using the 3,3'-diaminobenzidine chromogenic substrate. Hematoxylin was used for nuclear counterstaining.

Vitronectin-uPAR interaction assay. The capacity of HAT-treated rhuPAR[His]6 to bind vitronectin was assayed essentially according to a previously reported procedure (30). Briefly, Maxisorb 96-well plates (Nunc) were coated with 5 µg/ml vitronectin in PBS (0.1 ml/well) overnight at 4°C and then saturated with 1% BSA in PBS (PBS-BSA, 0.2 ml/well) for 1 h at room temperature. Wells then received 5 nM uPA together with 10 nM rhuPAR[His]6, which has been previously exposed (concentration of 20 nM) to 10–500 nM HAT (E/S of 0.5/1 to 25/1) for 30 min at 37°C, before HAT was blocked with benzamidine or left unexposed to the proteinase. All reactants were made in PBS-BSA (0.1 ml/well), and incubation continued for 1 h at 4°C. After washings in PBS containing 0.1% (vol/vol) Tween 20, wells were probed for bound rhuPAR[His]6 using either the His-Probe antibody (against the [His]6 tag) or the R2 or the R4 MAbs (both against D3) at 2 µg/ml, followed by a HRP-coupled anti-IgG antibody at a 1/1,000 dilution. The immune complexes were then quantified through a colorimetric assay for HRP activity with the use of the chromogenic substrate 3,3',5,5'-tetramethylbenzidine (KPL, Gaithersburg, MD). After acidification of the medium, optical density in wells was measured at 405 nm in an ELISA plate reader. Vitronectin-bound uPAR was calculated by substracting the background optical density measured in wells that did not receive rhuPAR[His]6 and expressed as the percentage of the binding measured with the untreated rhuPAR[His]6.

Expression of data. Results are expressed as means ± SE for the indicated number of experiments or measures performed independently. Statistical significance between the different values was analyzed with the Student's t-test with a threshold of P ≤ 0.05.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Soluble HAT cleaves h-uPAR after arginine residues in the D1-D2 linker sequence. To assess the direct proteolytic activity of HAT on uPAR, the recombinant form of the human receptor rhuPAR[His]6 was exposed to increasing concentrations of the recombinant, soluble form of the proteinase, and molecular species thus generated were analyzed by immunoblotting after electrophoretic separation was performed under disulfide bond reducing conditions, using a panel of domain-specific anti-uPAR antibodies (Fig. 1). This approach indicated that the full-length, three-domain rhuPAR[His]6 (molecular mass ~58 kDa), recognized both by the anti-D1 (3931) and the anti-D2 (3932) MAbs, progressively disappeared with increasing concentrations of HAT applied for 30 min, whereas two major molecular species were produced. On the basis of their mass and antigenic reactivity (27), these fragments corresponded to the truncated D2D3 species (~43.5 kDa, recognized by the anti-D2 MAb only; Fig. 1A) and to the free D1 domain (~19 kDa, recognized by the anti-D1 MAb only; Fig. 1B), respectively, and reflected the cleavage of rhuPAR[His]6 between the D1 and D2 domains, which could be already observed for HAT concentrations as low as 10 nM (E/S of 0.5/1). For HAT concentrations higher than 100 nM, a further degradation of the free D1 domain was noted (Fig. 1B), whereas the D2D3 species resisted further endoproteolysis up to 500 nM HAT (Fig. 1A). Cleavage of rhuPAR[His]6 into the D1 and D2D3 species could be seen as early as 5 min after exposure to 75 nM HAT (E/S of 4/1) and was similarly observed when proteins were analyzed with intact intradomain disulfide bonds (data not illustrated).


Figure 1
View larger version (73K):
[in this window]
[in a new window]

  		Fig. 1. Soluble human airway trypsin-like protease (HAT) cleaves soluble recombinant human urokinase-type plasminogen activator receptor (uPAR) within the D1-D2 linker sequence. rhuPAR[His]6 was exposed at 37°C to 10–500 nM HAT or was incubated without proteinase [untreated (NT)] for 30 min. Proteins (15 ng uPAR/well) were separated by SDS-PAGE under reducing conditions and immunoblotted with the anti-uPAR(D2) MAb 3932 (0.05 µg/ml; A), the anti-uPAR(D1) MAb 3931 (1 µg/ml; B), or the anti-[His]6 tag His-Probe antibody (0.1 µg/ml; C), which labels the rhuPAR[His]6 carboxy terminus. Portions of films corresponding to the location of the relevant antigens and representative blots of 6 experiments are depicted, with positions and relative molecular mass (Mr) values of the calibration standard proteins indicated on right. D1D2D3, full-length, 3-domain uPAR; D2D3, cleaved, truncated uPAR; D1, released free D1 domain.

 
Some proteinases known to cleave the D1-D2 linker sequence have been shown to also target the carboxy-terminal region of uPAR, as judged by the removal of the polyhistidine epitope and abrogation of the immunoblot labeling of the intact uPAR and D2D3 species with the anti-[His]6 antibody (2, 3). We thus examined whether HAT was similarly active on the carboxy-terminal region of uPAR. Despite the efficient conversion of the full-length molecule into the truncated D2D3, HAT-treated rhuPAR[His]6 showed a rather similar reactivity whether probed with the anti-D2 or the anti-[His]6 antibodies (compare Fig. 1A and Fig. 1C), suggesting that, in contrast to its marked proteolytic efficacy on the D1-D2 linker region, HAT up to 500 nM has a poor capacity to target the carboxy-terminal region of domain D3 in suPAR.

Determination of the peptide bonds targeted by HAT within the uPAR D1-D2 linker sequence was approached through amino-terminal sequencing of the truncated D2D3 species generated through exposure of rhuPAR[His]6 to HAT (E/S of 1/200), revealing two amino-terminal sequences, AVTYSR and SRYLE, corresponding to cleavage of rhuPAR[His]6 by HAT after arginine (Arg)83 and Arg89, respectively (27).

Soluble HAT cleaves domain D1 from membrane uPAR on human bronchial epithelial and monocytic cells. Because the airway epithelium is one site of expression of uPAR in the human respiratory tract (Ref. 8, and see below), we next evaluated the capacity of HAT to interact with epithelial membrane uPAR using confluent monolayers of the 16HBE bronchial epithelial cell line (9), a nontumor cell line that constitutively expresses the receptor (3). It must be noted that, although bronchial or small airway human epithelial cells in primary cultures express HAT mRNA (22), as expected from the in situ detection of the protein in the human airway epithelial lining (Ref. 37, and see below), established bronchial human cell lines maintained in culture do not (22). We observed that this also applied to the 16HBE cell line (data not shown). Because the 27-kDa HAT species has been identified as a major soluble trypsin-like proteinase present and active in the airway extracellular fluids of patients with chronic inflammatory diseases (21, 40, 41), experiments on uPAR-expressing 16HBE cells were thus conducted with the recombinant soluble HAT. Exposure of surface biotin-labeled 16HBE monolayers to exogenous HAT provided evidence for the occurrence of a specific, limited cleavage of membrane uPAR in the D1-D2 linker sequence (Fig. 2). With anti-D2 3932 MAb for immunoblotting of the isolated biotinylated membrane proteins, HAT was indeed shown to induce a proteolytic transition from the full-length membrane D1D2D3 species (~63 kDa) to the truncated D2D3 species (~49.5 kDa), which was found to accumulate in the cell plasma membrane up to the highest HAT concentration tested (Fig. 2A). Immunoblot analysis coupled to quantitative ELISA applied to the extracellular fluids recovered from cell monolayers exposed to the proteinase showed no evidence of release of suPAR, unless cells were exposed to HAT concentrations ≥100 nM, for which only traces of soluble D2D3 species could be detected (Fig. 2B), amounting to 0.75 ± 0.1 ng/ml antigenic suPAR after exposure to 500 nM HAT for 30 min, compared with 0.35 ± 0.01 ng/ml suPAR measured in the extracellular fluids of control nontreated monolayers (Fig. 2C). This clearly contrasted with the effects of PI-PLC, which was found to release larger amounts of antigenic suPAR, i.e., 3.5 ± 0.7 ng/ml (Fig. 2C). It must be noted that PI-PLC, which hydrolyzes GPI structures between the inositol phosphate ring and the diacylglycerol moiety (27), released all the uPAR species initially present at the surface of the 16HBE monolayers, i.e., the major three-domain uPAR and traces of truncated D2D3, without evidence for increased D1 shedding, as opposed to HAT. An important feature in our data is that none of the anti-D1 MAbs available for the present study could detect soluble D1 in the extracellular fluids, likely due to a further degradation of this domain after its shedding from the cell membrane (data not shown). Truncation of uPAR with removal of D1 could be already observed at 15 min of exposure of the cell monolayers to 150 nM HAT, whereas it was suppressed when the proteinase was added together with the Ser proteinase inhibitor benzamidine (data not shown).


Figure 2
View larger version (52K):
[in this window]
[in a new window]

  		Fig. 2. Soluble HAT cleaves membrane uPAR on human bronchial epithelial cells. A and B: epithelial 16HBE14o– cells (16HBE cells) as confluent monolayers were surface biotinylated and then left untreated (NT) or incubated with 10–500 nM HAT or with 1 U/ml phosphatidylinositol-specific phospholipase C (PI-PLC) for 30 min at 37°C, before removal of the extracellular fluids and separate solubilization of both fluids and cells. Biotinylated membrane proteins (A) or extracellular fluid proteins (B) were separated by SDS-PAGE under reducing conditions and immunoblotted with the anti-D2 3932 MAb (0.25 µg/ml). mD1D2D3 and mD2D3 stand for the membrane intact or truncated forms of uPAR, respectively; sD1D2D3 and sD2D3 stand for their soluble counterparts. C: concentrations of soluble uPAR (suPAR) in the extracellular fluids were assayed by quantitative ELISA. Histograms represent means + SE of 6 independent experiments. A statistically significant increase in suPAR was noted for cells exposed to 500 nM HAT compared with control cells. D: extracellular fluids (30 µl) obtained from exposure of 16HBE monolayers to PI-PLC were then further exposed to 50–500 nM HAT for 30 min at 37°C before protein separation by SDS-PAGE and immunoblotting as indicated for B. In A, B, and D, portions of films representative of 3–6 experiments are depicted.

 
Interestingly, it appeared that the soluble recombinant form of uPAR already showed an extensive cleavage of domain D1 for HAT concentrations (i.e., 10–50 nM) lower than those required to initiate removal of D1 from membrane uPAR (compare Fig. 1A to Fig. 2A). To validate this assumption on a soluble form of uPAR actually shed from a cell surface, the three-domain suPAR released from 16HBE cell monolayers by PI-PLC (see Fig. 2B) was further exposed to 50–500 nM HAT. As illustrated in Fig. 2D, exposure of this suPAR to 50 nM HAT already resulted in an extended removal of D1, which was nearly complete with 100 nM of the proteinase, thus confirming that the soluble receptor exhibits a higher susceptibility for cleavage of the D1-D2 linker sequence than membrane uPAR.

During lung inflammation, monocytes and macrophages recruited in the airways represent yet another major uPAR-expressing cell type (32), which can equally be exposed to soluble HAT. Flow immunocytometry analysis allowed us to evaluate the structural status of membrane uPAR on exposure of macrophage-like, differentiated monocytic U937 cells to the proteinase. Thus 10–500 nM HAT decreased the binding of the anti-D1 MAb 3931, resulting in a maximal reduction of 70 ± 10% compared with the binding on cells unexposed to the proteinase, whereas the binding of the anti-D2 MAb 3932 remained unchanged or even slightly increased (Fig. 3A, left), as was the binding of a MAb directed at the CD11b/{alpha}M integrin subunit taken as a further control for the specificity of the effect of HAT on D1 expression (data not shown). A similar specific decrease in expression of the D1 domain was confirmed on freshly isolated peripheral blood human monocytes exposed to soluble HAT (Fig. 3A, right). As previously noted (2), exposure of differentiated U937 cells to PI-PLC resulted in an ~50% reduction in binding of all anti-uPAR MAbs. Similar to what was observed with epithelial cells, immunoblotting analysis of biotinylated membrane proteins extracted from U937 cells exposed to HAT indicated a decreased expression of the intact membrane D1D2D3 form of the receptor (~61 kDa), whereas a truncated D2D3 species (~43 kDa) accumulated at the cell surface (Fig. 3B, top left). In parallel, immunoblotting analysis and quantitative ELISA performed on the extracellular fluids showed a very limited shedding of the membrane D2D3 species formed on cell exposure to 500 nM HAT (Fig. 3B, bottom left), which released 6.7 ± 1.7 ng/ml antigenic suPAR to be compared with 1.3 ± 0.2 ng/ml measured in the extracellular fluids of control monolayers, whereas much larger amounts of suPAR (i.e., 92 ± 2.4 ng/ml) were detectable after exposure of cells to PI-PLC (Fig. 3B, right).


Figure 3
View larger version (43K):
[in this window]
[in a new window]

  		Fig. 3. Soluble HAT cleaves membrane uPAR at the surface of monocytic cells. Differentiated U937 cells or isolated blood monocytes were exposed to 10–500 nM HAT or incubated as control in the absence of proteinase (NT) for 30 min at 37°C. A: membrane uPAR expression was monitored by fluorescence-activated cell sorting analysis using the anti-uPAR MAbs 3931 (anti-D1, black bars) or 3932 (anti-D2, gray bars), as described in MATERIALS AND METHODS. Results are expressed as a percentage of the mean fluorescence intensity (MFI) value obtained for cells incubated in the absence of proteinase and are illustrated as means + SE of 3 (monocytes; right) or 4 (U937 cells; left) independent experiments. *Statistically significant (P ≤ 0.05) difference between HAT-treated and nontreated cells. B: surface-biotinylated U937 cells and extracellular fluids were separately solubilized after exposure of cells to 500 nM HAT, to 1 U/ml PI-PLC, or to medium alone (NT) for 30 min at 37°C, and biotinylated membrane proteins (top left) and extracellular proteins (bottom left) were separated by SDS-PAGE under reducing conditions and immunoblotted with the anti-D2 3932 MAb (0.05 µg/ml). Portions of films representative of 3 experiments are depicted. Concentrations of suPAR in the extracellular fluids were assayed by quantitative ELISA (right); histograms represent means + SE of 3 independent experiments. A statistically significant increase in suPAR was noted for cells exposed to 500 nM HAT compared with control cells.

 
Coexpression of membrane HAT and uPAR in transfected human epithelial cells results in shedding of the D1 domain. The epithelial HEK cell line, which constitutively expresses neither HAT nor uPAR at the mRNA and at the protein level (Ref. 38, and data not shown), was used to establish a cellular model of transient cotransfection and coexpression of both transmembrane HAT and GPI-anchored human membrane uPAR, to evaluate the capacity of HAT to alter the structure of the receptor when both are present at the surface of the same cell population. Immunoblotting analysis of HAT-transfected HEK cell lysates with a polyclonal anti-HAT antibody revealed the presence of HAT with a mass of ~46 kDa (data not shown), corresponding to that expected for the full-length transmembrane form of HAT (39). When cells were transfected for expression of uPAR alone and were surface biotinylated at the end of the 24-h culture period, immunoblotting analysis after separation of biotinylated membrane proteins revealed a pattern of membrane uPAR expression similar to that observed in cells constitutively expressing this receptor, i.e., a predominant full-length, three-domain uPAR (~68 kDa), reacting with both the anti-D2 MAb 3932 and the anti-D1 MAb 3931 (these latter data not shown), along with a minor truncated D2D3 species (~52 kDa) only reacting with the anti-D2 MAb (Fig. 4A, lane uPAR/EF4). It has to be noted that traces of membrane D2D3, as well as some low levels of uPAR shedding, are a common feature of many cell types maintained in culture (25) (see lanes and histograms labeled NT in Fig. 2, A and C, and in Fig. 3B for 16HBE and U937 cells, respectively). These events may arise through the activity of minute amounts of uPA, plasmin, and/or MMP endogenously produced during cell culture (1, 3, 12). Most importantly, coexpression of HAT and uPAR in cultured HEK cells resulted in an extensive shedding of domain D1 from the cell membrane, as shown by a shift from full-length membrane uPAR to a now major truncated membrane D2D3 species (Fig. 4A, lane uPAR/HAT), whereas immunoblotting analysis of the extracellular fluids using the anti-D1 MAb revealed the presence of some shed D1 domain (~18 kDa), which was undetectable in the fluids of HEK monolayers transfected with uPAR alone (Fig. 4B, compare lanes uPAR/HAT and uPAR/EF4). Finally, coexpression of uPAR with an enzymatically inactive variant of transmembrane HAT, bearing a Ser368-to-Ala mutation at the catalytic site, resulted in an essentially uncleaved pattern for membrane uPAR similar to that observed in cells not transfected with HAT, with no detectable shedding of D1 (Fig. 4, A and B, lanes uPAR/HAT-SA). Measurement of HAT enzymatic activity in the various fractions indicated that it was below the detection threshold (<0.1 nM) in all of the extracellular culture fluids and in the lysates of HEK cells transfected with inactive HAT and/or with uPAR, whereas it amounted to ~1 nM (mean of 2 determinations) in the lysates of HEK cells transfected for active HAT expression, with or without uPAR.


Figure 4
View larger version (27K):
[in this window]
[in a new window]

  		Fig. 4. Membrane coexpression of HAT and uPAR in the human embryonic kidney epithelial cell line 293T (HEK cells) results in shedding of the D1 domain. Epithelial HEK cells were cotransfected with both a plasmid vector encoding for human uPAR and either an empty plasmid vector (uPAR/EF4), a plasmid vector encoding for an enzymatically active HAT (uPAR/HAT), or an inactive HAT bearing a point Ser368 to Ala mutation in its catalytic site (uPAR/HAT-SA). Cells were grown in culture for 24 h in fresh medium in the presence of 10% FCS, before they were separated from the extracellular culture fluids, surface biotinylated, and solubilized. Biotinylated membrane proteins (A) or extracellular culture fluid proteins (B) were separated by SDS-PAGE under reducing conditions and immunoblotted with the anti-D2 3932 MAb (0.5 µg/ml; A) or the anti-D1 3931 MAb (1 µg/ml; B). Note that, in B, some bovine serum proteins, mostly albumin, were nonspecifically labeled by the secondary, horseradish peroxidase (HRP)-coupled antibody.

 
HAT and uPAR are both expressed in the human airway epithelium. To ascertain the (patho)physiological relevance of the epithelial HAT/uPAR coexpression model described above, we examined normal and inflamed lung tissue sections for HAT and uPAR expression by immunohistochemistry (Fig. 5). When normal airway sections were probed with a specific MAb raised against uPAR, the bronchial epithelium showed very faint staining, whereas uPAR was significantly expressed in the inflamed bronchial epithelium (compare Fig. 5, A and D). Note that, in both patho(physio)logical situations, alveolar macrophages and/or inflammatory cells showed a prominent staining. Regarding HAT-specific staining, we observed a diffuse, apically enhanced staining of the bronchial epithelial cells in both normal and inflamed lung tissues (Fig. 5, B and E). These data thus confirmed that uPAR and HAT are both expressed in human bronchial epithelium in vivo.


Figure 5
View larger version (76K):
[in this window]
[in a new window]

  		Fig. 5. Immunohistochemical staining of normal and inflamed lung tissues reveals epithelial airway expression of HAT and uPAR. All panels show immunoperoxidase staining and depict medium power fields (x200). A–C: serial sections of a normal lung showing a medium-sized bronchiole stained for uPAR (A) or HAT (B) expression and negative control stained with omission of the primary antibody (C). Note diffuse bronchial epithelial staining for HAT and strong uPAR staining of macrophages in adjacent alveoli. Representative images are from 1 lung specimen. D–F: serial sections on a larger bronchus with chronic inflammation stained for uPAR (D) or HAT (E) expression, with negative control (F). Representative images are from 1 of 5 lung specimens. Note significant diffuse bronchial epithelial staining of both HAT and uPAR and strong uPAR staining in inflammatory cells.

 
HAT disrupts uPAR interaction with its major ligands urokinase and vitronectin. Because domain D1 in uPAR has been reported to be mandatory for the high-affinity interactions of the receptor with its ligands uPA and vitronectin (4, 27, 28), we next investigated the capacity of the cellular uPAR or suPAR to interact with the ligands after its exposure to HAT.

For this purpose, we first monitored by flow cytometry analysis the impact of HAT on the binding of FITC-uPA to differentiated U937 cells (Fig. 6A). Exposure of cells to the proteinase resulted in a HAT concentration-dependent reduction in the capacity to bind uPA, which could be significantly observed with HAT as low as 50 nM, reached a maximal inhibition of 85 ± 3% (n = 4), and appeared to be closely associated with a decreased membrane expression of the D1 domain, as judged through the reduced binding of the anti-D1 MAb 3931.


Figure 6
View larger version (24K):
[in this window]
[in a new window]

  		Fig. 6. Cleavage by HAT impairs uPAR interactions with urokinase and vitronectin (Vn). A: differentiated U937 cells were exposed to 10–500 nM HAT, to 1 U/ml PI-PLC, or left untreated as control for 30 min at 37°C. Cells were then probed with the anti-D1 3931 MAb followed by a FITC-conjugated anti-mouse IgG antibody (black bars) or in the presence of 50 nM FITC-uPA (open bars), and binding was assessed by flow cytometry. Specific MFI was expressed as the percentage of values obtained for cells incubated in the absence of proteinase, and results are expressed as means + SE of 4 independent experiments. B: rhuPAR[His]6 was either left untreated or exposed to 10–500 nM HAT for 30 min at 37°C before blocking the enzymatic activity, as described in MATERIALS AND METHODS. Vitronectin-coated wells (5 µg/ml) were incubated for 1 h at 4°C with the diluted reaction mixtures containing 10 nM rhuPAR[His]6, in the presence of 5 nM uPA. Bound uPAR was quantified by sequential incubations with the anti-[His]6 tag His-Probe antibody (2 µg/ml), a HRP-conjugated donkey anti-rabbit antibody (1/1,000 dilution), and finally the chromogenic substrate 3,3',5,5'-tetramethylbenzidine before measuring optical density at 405 nm. Vitronectin-bound rhuPAR[His]6 was calculated by subtracting the background optical density measured in rhuPAR[His]6-free wells and is expressed as the percentage of the binding measured with intact, nontreated rhuPAR[His]6. Histograms represent means + SE of 5 independent experiments (top). In parallel, cleavage of rhuPAR[His]6 by HAT was monitored by immunoblotting using the anti-D2 3932 MAb as described in the legend to Fig. 1 (bottom). *Statistically significant (P ≤ 0.05) difference between HAT-treated and nontreated cells.

 
We then assayed the capacity of rhuPAR[His]6 to bind to vitronectin-coated microplates through an ELISA-type procedure and found, as previously reported (30), that the interaction of suPAR with immobilized vitronectin was strictly dependent on the presence of uPA (data not shown). Exposure of rhuPAR[His]6 to increasing concentrations of HAT before the assay induced a drastic decrease, up to ~95%, in the binding of suPAR to vitronectin, as judged through the immunodetection of bound uPAR using an antibody directed at the carboxy-terminal [His]6 tag (Fig. 6B, top). Abrogation of binding of rhuPAR[His]6 to vitronectin was concomitant with its cleavage in the D1-D2 linker sequence and the appearance of the D2D3 species, as evaluated by immunoblotting with an anti-D2 MAb (Fig. 6B, bottom). It must be noted that, when probed with either of two MAbs, R2 and R4, reacting with different epitopes internal to the D3 domain, binding of rhuPAR[His]6 to vitronectin was similarly decreased up to 98% ± 1% (n = 4) after exposure to 500 nM HAT.


   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
Besides their implications in beneficial immune defense and repair processes, extracellular proteinases can also be highly deleterious to tissues and organs through uncontrolled production and/or activity and can lead or participate to pathogenic inflammatory processes, particularly within the lungs (6, 13, 19, 23, 26, 36). Recently, the respiratory epithelium has been identified as a major source of proteinases (26, 36), including a trypsin-like Ser proteinase such as HAT (40). Although HAT is a typical member of the membrane-associated type II transmembrane Ser proteinase family with a mass of ~47 kDa (35, 39), a significant HAT-related activity can be detected in the extracellular airway fluids of patients with chronic airway inflammatory diseases, particularly in patients with asthma (21, 40, 41). In these fluids, HAT actually exist as a 27-kDa soluble enzyme, which accounts for most, if not all, of the measurable trypsin-like enzymatic activity (21, 40, 41). To date, data are scarce regarding the molecular mechanism that allows the proteolytic shedding of the 27-kDa HAT catalytic domain from the membrane-tethering domain (39). Similarly, there is little information about the regulation of expression and activity of membrane-bound HAT, except that an inflammatory environment can upregulate HAT epithelial expression (14). However, recent evidence supports the assumption that HAT might play an important role in respiratory homeostasis during the inflammatory response, since its capacity to degrade fibrinogen, to activate pro-uPA, and to activate membrane receptors such as PAR-2 (22, 41) suggests that HAT may both participate in the control of fibrin deposition and modulate the behavior of inflammatory cells in the airways (7, 21, 22). Identification of proteins, particularly membrane receptors, which are targets for this new Ser proteinase, should thus help to delineate the role of HAT in (patho)physiological processes in the respiratory tract.

In this regard, we have now demonstrated that HAT interacts with both soluble and membrane forms of uPAR and produces limited endoproteolysis of this receptor, as expressed by human bronchial epithelial cells and monocytes and macrophages. uPAR is now recognized as a key element in (patho)physiological processes involving cell migration and tissue remodeling, particularly during the host inflammatory response to infections (4, 24). Because this protein 1) is expressed in all cell types involved in lung inflammation, 2) is a multiligand binding protein interacting with partner molecules such as uPA, integrins, and vitronectin, and 3) is a proteolytically regulated receptor (4, 13, 25), the activity of HAT on this ubiquitous receptor could be of major biological relevance, through many impacts on uPAR pleiotropic functions.

Regarding the interaction of HAT with uPAR, a number of major features emerge from our studies. First, soluble HAT efficiently cleaves the linker sequence separating domains D1 and D2 in uPAR. As for uPA and plasmin (3, 12, 27), and in agreement with its known trypsin-like activity (40), HAT targets two peptide bonds in D1-D2 with an Arg residue at the P1 position, i.e., Arg83-Ala84 and Arg89-Ser90. This cleavage appears to occur even more readily (i.e., at lower HAT concentrations) on suPAR than on the GPI-anchored membrane uPAR. It is well established that the GPI moiety per se, and/or its membrane microenvironment, influences the overall conformation of GPI-linked membrane proteins and presentation of antigenic epitopes or of cleavage sites (5) and uPAR and its D1-D2 linker sequence are exemplary in this regard (1, 12). Indeed, it has been observed that uPA is more efficient in the cleavage of the D1-D2 sequence when it is presented in the context of the GPI-linked uPAR as opposed to the suPAR devoid of the GPI moiety (12), whereas HAT appears to have the opposite property (this study); plasmin shows an equal high efficiency for cleavage of D1-D2 in soluble or membrane uPAR (3, 12). Similarly to uPA (3), HAT also appears to be poorly active on the carboxy-terminal domain of uPAR and thus has very little capacity to release suPAR, as opposed to plasmin, which is highly efficient in releasing D2D3 species through cleavage of Arg281-Ser282 at the carboxy terminus of D3 (3). Altogether, this indicates that, despite their common primary trypsin-like proteinase specificity with preferential cleavage following Arg residues, more complex molecular mechanisms, yet to be identified, determine the interactions of these proteinases with uPAR. Finally, we established that, although soluble HAT may have the capacity to transregulate the structure of uPAR in juxtacrine and/or paracrine manners on nearby epithelial or leukocytic cells, membrane-associated HAT appears to be endowed with the same capacity toward uPAR coexpressed in epithelial cells. This autocrine activity appears to be highly efficient in shedding domain D1 from the surface of the cells, and these data represent so far the first demonstration that membrane-associated HAT can exert its proteinase activity on proximal membrane substrates.

Cleavage of uPAR by soluble HAT occurs in vitro for enzyme concentrations compatible with those actually detected in human pathological airway extracellular fluids (e.g., sputa), which are in the range of 5–100 nM (21, 41). These concentrations are clearly sufficient to produce the cleavage of the D1 domain from suPAR, whereas they are at the lower range for shedding of D1 from membrane uPAR. However, sputum is likely to be a somehow diluted milieu in the airways, and thus one could expect that the local concentrations of HAT, in the immediate vicinity of epithelial cells producing the proteinase, are higher than the concentrations in sputa. Obviously, HAT is one of many proteinases present in pathological airways (19, 23, 26, 36, 41) and which can be active on uPAR. Indeed, among those proteolytic systems, we and others previously showed that the uPA/plasmin/MMP cascade on one hand and the neutrophil elastase/cathepsin G combination on the other hand can alter uPAR by cleavage in the D1-D2 linker sequence and/or at the D3 carboxy terminus (1–3, 12), when proteinases are present in a range of concentrations similar to that observed for HAT, i.e., ~10–500 nM (2, 3, 12). However, some features plead for a peculiar role for HAT: 1) it is poorly inhibitable by either {alpha}1-proteinase inhibitor or secretory leukocyte proteinase inhibitor at the concentrations that these natural anti-proteinases can reach in the airways (40); 2) in some pathological airway extracellular fluids, this proteinase makes the only detectable trypsin-like activity (21, 40), which can furthermore represent a higher proteolytic burden compared with the leukocyte elastase activity (41), and 3) coexpression of membrane-associated HAT and uPAR appears to lower to the nanomolar range the HAT concentrations active on uPAR. Furthermore, we have now demonstrated the coexpression of uPAR and HAT in the human bronchial epithelium, particularly in inflamed tissues. Indeed, although uPAR was reported to be absent from the normal bronchial epithelium in mice (33), it is expressed (albeit at low levels) in the human normal bronchial epithelium (Ref. 8, and this study). Interestingly, uPAR expression is significantly increased in inflamed bronchial epithelium (Ref. 8, and this study), including that of asthmatic patients (8), whose airway secretions are also particularly enriched for soluble HAT (41). Altogether, these observations thus support the potential for autocrine-juxtacrine interactions between uPAR and HAT in the airways and strengthen the assumption that HAT may be an effector on uPAR structure and functions in in vivo settings, albeit the (patho)physiological role(s) of uPAR cleavage in the lung remain yet to be characterized.

Interplay between HAT and the fibrinolytic system has already been established, since HAT can activate pro-uPA (22) whereas its fibrinogenolytic activity impairs the formation of fibrin clots (41). From the results reported in the present study, i.e., that cleavage of membrane uPAR by HAT drastically reduces the capacity of cells to bind uPA, we now propose that a persistent exposure of cells to concentrations of HAT similar to those expected in an inflammatory context (21, 41) may result in a depressed pericellular plasminogen/MMP activation cascade. Furthermore, and although intact suPAR has been shown to retain the ability to bind uPA and to initiate plasminogen activation on cells or ECM through binding of the uPAR/uPA complexes to vitronectin (28), our observation that HAT is particularly active on suPAR for removal of D1 and for abrogation of the suPAR/uPA/vitronectin trimolecular interactions thus strengthens the proposal for HAT as a regulator of the plasminogen/MMP activation.

The observation that cleavage of uPAR by HAT prevents interactions of the receptor with vitronectin may then have other cellular implications. Vitronectin is indeed a major component of the fibrin-rich provisional ECMs that form during the early stages of inflammation in many organs, including the lungs, and is suggested to promote adherence and spreading both of fibrinolytic leukocytes in the areas where fibrin deposits accumulate and must be tightly controlled (32) and of epithelial cells during repair and healing (16). Because uPAR is, together with some integrin-type adherence receptors, a major cell binding site for vitronectin (28), the disruption of uPAR-uPA-vitronectin interactions as a result of HAT activity could thus participate in the processes, whether normal or pathological, of ECM remodeling and epithelium regeneration. In parallel, other routes for HAT to modulate the inflammatory response in the airways via uPAR cleavage and D1 shedding may include 1) the regulation of the physical and functional association of uPAR with integrins during cell adherence and migration (15, 25), 2) the exposure or disruption of the chemotactic epitope located over the sequence Ser88-Tyr92 within the D1-D2 linker sequence of uPAR, and which is active on various inflammatory cells (4), and 3) the control of airway epithelial cell proliferation triggered by binding of uPA to uPAR (29). Altogether, the regulation of uPAR by HAT during inflammatory airway disorders could certainly have a strong impact on leukocyte and epithelial cell recruitment and activation and thus on repair and healing after tissue injury (4, 6, 13, 24). The future availability of animal models of HAT deficiency or overexpression, as well as a better knowledge of the biology of this new epithelial proteinase, including identification of specific endogenous inhibitors, should further help clarifying its role(s) in respiratory physiology and pathology.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was funded by the nonprofit association Vaincre la Mucoviscidose, by the Centre National de la Recherche Scientifique (D. Pidard), and by a fellowship from the Ministère de la Recherche (France) and from the Alexander von Humboldt Foundation (Germany) (N. Beaufort).


   ACKNOWLEDGMENTS
 
The authors express gratitude to Dr. Jacques D'Alayer (Plate-Forme d'Analyse et de Microséquençage des Protéines, Institut Pasteur, Paris) for performing protein amino-terminal sequencing and to Dr. Anne-France Petit-Bertron (UP Cytokines and Inflammation, Institut Pasteur, Paris, France) for the preparation of purified human monocytes. The authors are grateful to Dr. Gunilla Høyer-Hansen (The Finsen Laboratory, Copenhaguen, Denmark) for providing R2 and R4 anti-uPAR antibodies used in this study, as well as to Dr. Viktor Magdolen and Pr. Manfred Schmitt (Department of Obstetrics and Gynaecology, Technical University of Munich, Munich, Germany) for providing reagents and valuable advice regarding the immunohistochemistry assay.

Present addresses: N. Beaufort, Klinische Forschergruppe der Frauenklinik der Technische Universität München, 81675 München, Germany; D. Pidard, Inserm U698, Hôpital Bichat-Claude Bernard, 75018 Paris, France.

The data have been presented in part at the 28th European Cystic Fibrosis Conference, Hersonissos, Crete, Greece, June 22–25, 2005.


   FOOTNOTES
 

Address for reprint requests and other correspondence: D. Pidard, Inserm U.698, CHU Xavier-Bichat, 46, rue Henri-Huchard, 75877 Paris Cedex 18, France (e-mail: pidard{at}bichat.inserm.fr)

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.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Andolfo A, English WR, Resnati M, Murphy G, Blasi F, Sidenius N. Metalloproteases cleave the urokinase-type plasminogen activator receptor in the D1-D2 linker region and expose epitopes not present in the intact soluble receptor. Thromb Haemost 88: 298–306, 2002.[ISI][Medline]
  2. Beaufort N, Leduc D, Rousselle JC, Magdolen V, Luther T, Namane A, Chignard M, Pidard D. Proteolytic regulation of the urokinase receptor/CD87 on monocytic cells by neutrophil elastase and cathepsin G. J Immunol 172: 540–549, 2004.[Abstract/Free Full Text]
  3. Beaufort N, Leduc D, Rousselle JC, Namane A, Chignard M, Pidard D. Plasmin cleaves the juxtamembrane domain and releases truncated species of the urokinase receptor (CD87) from human bronchial epithelial cells. FEBS Lett 574: 89–94, 2004.[CrossRef][ISI][Medline]
  4. Blasi F, Carmeliet P. uPAR: a versatile signalling orchestrator. Nat Rev Mol Cell Biol 3: 932–943, 2002.[CrossRef][ISI][Medline]
  5. Bütikofer P, Malherbe T, Boschung M, Roditi I. GPI-anchored proteins: now you see 'em, now you don't. FASEB J 15: 545–548, 2001.[Abstract/Free Full Text]
  6. Chapman HA. Disorders of lung matrix remodeling. J Clin Invest 113: 148–157, 2004.[CrossRef][ISI][Medline]
  7. Chokki M, Eguchi H, Hamamura I, Mitsuhashi H, Kamimura T. Human airway trypsin-like protease induces amphiregulin release through a mechanism involving protease-activated receptor-2-mediated ERK activation and TNF-{alpha}-converting enzyme activity in airway epithelial cells. FEBS J 272: 6387–6399, 2005.[CrossRef][Medline]
  8. Chu EK, Cheng J, Foley JS, Mecham BH, Owen CA, Haley KJ, Mariani TJ, Kohane IS, Tschumperlin DJ, Drazen JM. Induction of the plasminogen activator system by mechanical stimulation of human bronchial epithelial cells. Am J Respir Cell Mol Biol 35: 628–638, 2006.[Abstract/Free Full Text]
  9. Cozens AL, Yezzi MJ, Kunzelmann K, Ohrui T, Chin L, Eng K, Finkbeiner WE, Widdicombe JH, Gruenert DC. CFTR expression and chloride secretion in polarized immortal human bronchial epithelial cells. Am J Respir Cell Mol Biol 10: 38–47, 1994.[Abstract]
  10. Dello Sbarba P, Rovida E. Transmodulation of cell surface regulatory molecules via ectodomain shedding. Biol Chem 383: 69–83, 2002.[CrossRef][ISI][Medline]
  11. Gyetko MR, Sud S, Kendall T, Fuller JA, Newstead MW, Standiford TJ. Urokinase receptor-deficient mice have impaired neutrophil recruitment in response to pulmonary Pseudomonas aeruginosa infection. J Immunol 165: 1513–1519, 2000.[Abstract/Free Full Text]
  12. Høyer-Hansen G, Pessara U, Holm A, Pass J, Weidle U, Danø K, Behrendt N. Urokinase-catalysed cleavage of the urokinase receptor requires an intact glycolipid anchor. Biochem J 358: 673–679, 2001.[CrossRef][ISI][Medline]
  13. Idell S. Coagulation, fibrinolysis, and fibrin deposition in acute lung injury. Crit Care Med 31: S213–220, 2003.[CrossRef][ISI][Medline]
  14. Iwakiri K, Ghazizadeh M, Jin E, Fujiwara M, Takemura T, Takezaki S, Kawana S, Yasuoka S, Kawanami O. Human airway trypsin-like protease induces PAR-2-mediated IL-8 release in psoriasis vulgaris. J Invest Dermatol 122: 937–944, 2004.[CrossRef][ISI][Medline]
  15. Kugler MC, Wei Y, Chapman HA. Urokinase receptor and integrin interactions. Curr Pharm Des 9: 1565–1574, 2003.[CrossRef][ISI][Medline]
  16. Lazar MH, Christensen PJ, Du M, Yu B, Subbotina NM, Hanson KE, Hansen JM, White ES, Simon RH, Sisson TH. Plasminogen activator inhibitor-1 impairs alveolar epithelial repair by binding to vitronectin. Am J Respir Cell Mol Biol 31: 672–678, 2004.[Abstract/Free Full Text]
  17. Legrand C, Polette M, Tournier JM, de Bentzmann S, Huet E, Monteau M, Birembaut P. uPA/plasmin system-mediated MMP-9 activation is implicated in bronchial epithelial cell migration. Exp Cell Res 264: 326–336, 2001.[CrossRef][ISI][Medline]
  18. List K, Høyer-Hansen G, Rønne E, Danø K, Behrendt N. Different mechanisms are involved in the antibody mediated inhibition of ligand binding to the urokinase receptor: a study based on biosensor technology. J Immunol Methods 222: 125–133, 1999.[CrossRef][ISI][Medline]
  19. Lucey EC, Stone PJ, Snider GL. Consequences of proteolytic injury. In: The Lung: Scientific Foundations (2nd ed.), edited by Crystal RG, West JB, Barnes PF, Weibel E. Baltimore, MD: Lippincott-Raven, 1997.
  20. Luther T, Magdolen V, Albrecht S, Kasper M, Riemer C, Kessler H, Graeff H, Müller M, Schmitt M. Epitope-mapped monoclonal antibodies as tools for functional and morphological analyses of the human urokinase receptor in tumor tissue. Am J Pathol 150: 1231–1244, 1997.[Abstract]
  21. Matsushima R, Takahashi A, Nakaya Y, Maezawa H, Miki M, Nakamura Y, Ohgushi F, Yasuoka S. Human airway trypsin-like protease stimulates human bronchial fibroblast proliferation in a protease-activated receptor-2-dependent pathway. Am J Physiol Lung Cell Mol Physiol 290: L385–L395, 2006.[Abstract/Free Full Text]
  22. Miki M, Nakamura Y, Takahashi A, Nakaya Y, Eguchi H, Masegi T, Yoneda K, Yasuoka S, Sone S. Effect of human airway trypsin-like protease on intracellular free Ca2+ concentration in human bronchial epithelial cells. J Med Invest 50: 95–107, 2003.[Medline]
  23. Miller HR, Pemberton AD. Tissue-specific expression of mast cell granule serine proteinases and their role in inflammation in the lung and gut. Immunology 105: 375–390, 2002.[CrossRef][ISI][Medline]
  24. Mondino A, Blasi F. uPA and uPAR in fibrinolysis, immunity and pathology. Trends Immunol 25: 450–455, 2004.[CrossRef][ISI][Medline]
  25. Montuori N, Visconte V, Rossi G, Ragno P. Soluble and cleaved forms of the urokinase-receptor: degradation products or active molecules? Thromb Haemost 93: 192–198, 2005.[ISI][Medline]
  26. Parks WC, Shapiro SD. Matrix metalloproteinases in lung biology. Respir Res 2: 10–19, 2001.[CrossRef][ISI][Medline]
  27. Ploug M. Structure-function relationships in the interaction between the urokinase-type plasminogen activator and its receptor. Curr Pharm Des 9: 1499–1528, 2003.[CrossRef][ISI][Medline]
  28. Preissner KT, Kanse SM, May AE. Urokinase receptor: a molecular organizer in cellular communication. Curr Opin Cell Biol 12: 621–628, 2000.[CrossRef][ISI][Medline]
  29. Shetty S, Rao GN, Cines DB, Bdeir K. Urokinase induces activation of STAT3 in lung epithelial cells. Am J Physiol Lung Cell Mol Physiol 291: L772–L780, 2006.[Abstract/Free Full Text]
  30. Sidenius N, Andolfo A, Fesce R, Blasi F. Urokinase regulates vitronectin binding by controlling urokinase receptor oligomerization. J Biol Chem 277: 27982–27990, 2002.[Abstract/Free Full Text]
  31. Sidenius N, Sier CF, Blasi F. Shedding and cleavage of the urokinase receptor (uPAR): identification and characterisation of uPAR fragments in vitro and in vivo. FEBS Lett 475: 52–56, 2000.[CrossRef][ISI][Medline]
  32. Simon DI, Rao NK, Xu H, Wei Y, Majdic O, Rønne E, Kobzik L, Chapman HA. Mac-1 (CD11b/CD18) and the urokinase receptor (CD87) form a functional unit on monocytic cells. Blood 88: 3185–3194, 1996.[Abstract/Free Full Text]
  33. Solberg H, Ploug M, Høyer-Hansen G, Nielsen BS, Lund LR. The murine receptor for urokinase-type plasminogen activator is primarily expressed in tissues actively undergoing remodeling. J Histochem Cytochem 49: 237–246, 2001.[Abstract/Free Full Text]
  34. Sundstrom C, Nilsson K. Establishment and characterization of a human histiocytic lymphoma cell line (U-937). Int J Cancer 17: 565–577, 1976.[ISI][Medline]
  35. Szabo R, Wu Q, Dickson RB, Netzel-Arnett S, Antalis TM, Bugge TH. Type II transmembrane serine proteases. Thromb Haemost 90: 185–193, 2003.[ISI][Medline]
  36. Taggart CC, Greene CM, Carroll TP, O'Neill SJ, McElvaney NG. Elastolytic proteases. Inflammation resolution and dysregulation in chronic infective lung disease. Am J Respir Crit Care Med 171: 1070–1076, 2005.[Free Full Text]
  37. Takahashi M, Sano T, Yamaoka K, Kamimura T, Umemoto N, Nishitani H,