Bronchial Tissue Kallikrein Activity Is Regulated by
Hyaluronic Acid Binding
Rosanna Forteza, Isabel Lauredo, William M. Abraham, and Gregory E. Conner
Division of Pulmonary Disease and Critical Care Medicine, University of Miami at Mt. Sinai Medical Center, Miami Beach;
and Division of Pulmonary and Critical Care Medicine, University of Miami School of Medicine, Miami, Florida
Tissue kallikrein (TK) is secreted by serous cells of tracheobronchial submucosal glands and plays a
role in allergic airway responses. To better understand the regulation of TK, we used primary cultures of
submucosal gland cells that release TK upon stimulation. Media from cultures stimulated with chymase
(1027 M) showed increased TK activity (0.50 6 0.22 mU/ml mean 6 standard error) in comparison with
the control group (0.08 6 0.02 mU/ml). The increased TK activity was significantly correlated with increases in the levels of the serous cell marker, secretory leukoprotease inhibitor. Anion exchange chromatography of the conditioned culture media showed that TK activity eluted as a broad peak between 1.6 and
1.8 M NaCl, unlike the reported elution (0.3 to 0.6 M NaCl) of kallikreins from other tissues, suggesting
that secreted bronchial TK was bound to a negatively charged molecule. Hyaluronidase digestion increased TK activity in both pre– and post–chymase-stimulated culture media, whereas no such change was
seen after samples were digested with heparinase or chondroitinase ABC. Further, after hyaluronidase digestion of media, TK eluted from an anion exchange column between 0.3 and 0.6 M NaCl. Enzymatic detection of TK after nondenaturing gel electrophoresis showed that hyaluronidase digestion also reduced the
electrophoretic heterogeneity of TK to a single band, whereas adding back hyaluronic acid (HA) to hyaluronidase-digested samples restored the original heterogeneity. Finally, TK activity bound to HA-Sepharose and could be eluted with HA. These studies show that primary cultures of ovine submucosal gland
cells secrete TK in a regulated fashion, and that secreted TK binds to HA. This binding reduces TK enzymatic activity; therefore, factors that affect HA turnover could modify the TK activity in the airway lumen.
These events could be important in the regulation of kinin-mediated airway inflammation. Forteza, R., I.
Lauredo, W. M. Abraham, and G. E. Conner. 1999. Bronchial tissue kallikrein activity is regulated
by hyaluronic acid binding. Am. J. Respir. Cell Mol. Biol. 21:666–674.
Tissue kallikreins (TKs) (EC 3.4.21.35) are a family of
serine proteinases that are involved in the post-transcriptional processing of polypeptide precursors to their biologically active form (1, 2). TKs are secreted in a variety of
glandular tissue, including salivary glands (3–5), colon (4),
stomach (6), uterus (7), pituitary gland (8), and pancreas
(9). It is also synthesized by neutrophils (10), kidney (11),
and endothelial cells (12). TK function and regulation var(Received in original form February 22, 1998 and in revised form June 3,
1999)
Address correspondence to: Rosanna M. Forteza, M.D., Department of
Research, Mount Sinai Medical Center, 4300 Alton Road, Miami Beach,
FL 33140. E-mail: [email protected]
Abbreviations: Alcian blue–periodic acid-Schiff, AB-PAS; bronchoalveolar
lavage fluid, BALF; biotinylated HA binding protein, bHABP; Dulbecco’s
modified Eagle’s medium, DMEM; glycosaminoglycan, GAG; hyaluronic
acid, HA; human urinary kallikrein, HUK; immunoglobulin, Ig; phosphate-buffered saline, PBS; room temperature, RT; soybean trypsin inhibitor, SBTI; secretory leukoprotease inhibitor, SLPI; tissue kallikrein, TK.
Am. J. Respir. Cell Mol. Biol. Vol. 21, pp. 666–674, 1999
Internet address: www.atsjournals.org
ies with tissue localization (2). For example, in the kidney,
TK regulates local blood flow through kinin generation,
and its expression is affected by NaCl intake (13). In contrast, in pituitary glands TK expression is induced by estrogen (8) and its main function seems to be prolactin processing (14). Thus, the variability in localization, regulation,
and substrate specificity of TK (2) suggest that its characterization in the specific tissue of interest is necessary to
fully understand its biologic functions.
TK has been identified as the major kininogenase in the
airways (3). TK proteolyzes both high and low molecularweight kininogen in a highly selective and limited fashion
to yield lysyl–bradykinin (kallidin), a potent vasoactive
peptide that influences a number of biologic processes including vasodilation, vascular permeability, and bronchoconstriction—all of which contribute to asthma pathophysiology. Immunohistochemical studies localize bronchial TK
in the serous cells of submucosal glands of human and guinea
pig trachea (15, 16), and TK-like activity is increased in
human nasal and bronchoalveolar lavage (BAL) after allergen challenge (17–19), but its biochemical characteris-
Forteza, Lauredo, Abraham, et al.: Regulation of Bronchial Kallikrein by Hyaluronic Acid
tics, as well as control mechanisms of synthesis and secretion, are poorly understood.
Studies from our laboratory using allergic sheep indicate that airway challenge with a wide range of inflammatory stimuli that cause bronchoconstriction and/or airway
hyperresponsiveness (AHR), such as allergen, metabisulfite,
ozone, and bacterial supernatants, are associated with increased concentrations of immunoreactive kinins and increased TK-like activity in BAL fluid (BALF) (15, 20–22).
That kinins contribute to these events is supported by the
finding that bradykinin B2 receptor antagonists can block
the bronchoconstriction and airway hyperresponsiveness
caused by these agents (23–27).
Collectively, these observations indicate that in the lung,
increased TK activity contributes to airway inflammation,
which leads to bronchoconstriction and AHR. Therefore,
understanding the mechanisms leading to the increased TK
activity is important. Primary cultures from ovine tracheal
gland cells were used in the present study to explore the
regulation of bronchial TK activity and secretion.
Materials and Methods
All materials were purchased from Sigma Chemical Co.
(St. Louis, MO) unless otherwise specified.
Primary Cultures of Tracheal Submucosal Gland Cells
Primary cultures of tracheal submucosal gland cells were
prepared according to the method of Tournier and colleagues (28) with some modifications. Briefly, sheep trachea was cut in pieces and bathed in Dulbecco’s modified
Eagle’s medium with 20 mM N-2-hydroxyethylpiperazineN9-ethane sulfonic acid (DMEM) containing penicillin
(100 U/ml), streptomycin (100 mg/ml), and amphotericin
(0.25 mg/ml). The tracheal segments were cut longitudinally along the posterior membrane. The luminal surface
was gently scraped with gauze and rinsed three times with
DMEM to remove the surface epithelium. The tissue
above the cartilage was dislodged, cut in pieces, rinsed
twice with DMEM, and incubated with 0.1% protease
(dispase from Bacillus polymyxa) at 48C for 3 d. The tissue
fragments were removed from the protease solution,
scraped with a disposable cell scraper and vigorously
shaken in DMEM containing 5% heat-deactivated fetal
calf serum (FCS). The cell suspension was centrifuged at
150 3 g and resuspended in 50% DMEM and 50% Ham’s
Nutrient F-12 with 5% FCS. The cell suspension was
plated on a Costar 24-well plate previously coated with
collagen (type VI, 500 ml/well, 0.05% in 0.2% acetic acid)
and incubated for 24 h at room temperature (RT). The
plates were then transferred into a 5% CO2 humidified incubator at 378C for 24 h. The media were replaced with serum-free media made with 50% DMEM and 50% Ham’s
Nutrient F-12 containing insulin (10 mg/ml; Collaborative
Biomedical Products, Bedford, MA), transferrin (5 mg/ml),
endothelial cell growth supplement (7.5 mg/ml; Collaborative Biomedical Products), triiodothyronine (20 ng/ml),
penicillin (100 U/ml), streptomycin (100 mg/ml), and amphotericin (0.25 mg/ml) (28). Plates were then incubated at
328C to promote the preferential growth of gland cells
(29), and the media changed daily. For the experiments,
667
we used confluent monolayers that contained approximately 5 3 105 cells per well (2 cm2).
Stimulation of the Cultures with Chymase
For stimulation experiments, primary cultures of submucosal gland cells obtained from five sheep were used. To
establish baseline secretion rates (28, 30), confluent cultures were washed three times with phosphate-buffered saline (PBS) and twice with DMEM and incubated in RPMI
1640 without phenol red at 378C in a CO2 humidified incubator for 24 h.
The media were removed, and 400 ml of RPMI were
added to each well, and cultures were incubated under the
same conditions for 90 min. This procedure was then repeated at the same intervals (28). After collection of the media, the cultures were stimulated by adding 400 ml of 1027 M
chymase (a gift from Dr. Richard Tanaka, Arris Pharmaceutical, San Francisco, CA) in RPMI. RPMI only (400 ml) was
added to control wells. Supernatants were collected 1.5 and
24 h after stimulation. All samples were centrifuged (250 3 g,
10 min at 4°C) and frozen at 2208C for further analysis.
TK Enzyme Activity
Activity was measured with a spectrophotometric assay
using a synthetic peptide substrate that, in the presence of
soybean trypsin inhibitor (SBTI), is highly specific for TK
(31) with the exception of the mast-cell product tryptase.
Mast cells are not present in our cultures, but mast-cell
tryptase could be a contaminant in the chymase preparations. For that reason, we added the tryptase inhibitor
APC 366 to SBTI. Briefly, in an ultralow-binding 96-well
plate, samples (100 ml) were incubated with 25 ml of trypsin
(5 mg/ml) for 15 min at 378C to activate prokallikrein (1).
After adding 25 ml of SBTI (2 mg/ml) and 50 ml of APC
366 (1 mg/ml; a gift from Dr. Richard Tanaka, Arris Pharmaceutical), 100 ml of DL-Val-Leu-Arg-pNA (2 mM; ICN,
Irvine, CA) in 200 mM Tris (pH 8.2) was added, and absorbance was monitored at 405 nm using a Dynatech MRX
microplate reader. One enzyme unit was defined as one
mole of substrate degraded per minute at pH 8.2 and 228C.
Chymase showed no kallikrein-like activity in our assay.
Immunohistochemistry
Samples of tissue from trachea to bronchi of the fifth generation were fixed in 4% paraformaldehyde in PBS and
embedded in paraffin. After rehydration using xylene, followed by decreasing concentrations of ethanol, slides were
incubated in 3% hydrogen peroxide for 15 min to quench
endogenous peroxidase activity, known to exist in goblet
cells and submucosal glands (32, 33). Slides were washed
in PBS (5 min), blocked with 20% goat serum in PBS for 1 h
and then incubated with an antibody to human urinary
kallikrein (HUK) (1/1,000 in 20% goat serum, Calbiochem, La Jolla, CA) at 4 8C overnight. After washing in
PBS (three times for 5 min each), slides were incubated
with peroxidase-labeled goat antirabbit immunoglobulin
(Ig)G (ABC kit; Vector Laboratories, Burlingame, CA) at
RT for 30 min and washed with PBS (five times for 5 min
each). Visualization was achieved by incubating with diaminobenzamidine (1 mg/ml) in PBS 1 H2O2 0.01% at RT
for 8 min. Slides were washed, counterstained with Harris
668
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 21 1999
hematoxylin, dehydrated, and mounted. Controls omitting
the first antibody were prepared identically.
Immunofluorescence
Primary cultures grown on glass coverslips were fixed in
4% paraformaldehyde in PBS (1 h), permeabilized with
cold methanol (2208C, 10 min), washed in PBS (three times
for 5 min each), and blocked with 20% goat serum at RT
for 1 h. Blocking solution was removed and rabbit antiHUK (1/500 in 20% goat serum) was added and incubated
overnight at 48C.
After washing with PBS (three times for 10 min each),
fluorescein-labeled antirabbit IgG was applied to the coverslips and incubated for 30 min at RT. Slides were washed
with PBS, mounted in aqueous mounting medium (Slow
Fade kit; Molecular Probes, Eugene, OR), and examined
using a Nikon Labophot microscope.
Anion Exchange Chromatography
Conditioned cell culture medium (200 ml) was applied to
an anion exchange column (1 ml UnoQ; Bio-Rad) previously
equilibrated in 50 mM Tris (pH 8.0), 0.5% 3-(3-cholamidopropyl) dimethyl-ammonio-1 propanesulfonate (CHAPS),
in 0.15 M NaCl at a 250- ml/min flow rate. Washing was
performed with equilibration buffer until absorbance at
280 nm (A280) returned to baseline values. Elution was
achieved with a gradient of 0.1 to 2.0 M NaCl in 50 mM
Tris (pH 8.0, 0.5% CHAPS). A column was also run using
conditioned media (150 ml) that had been digested with
hyaluronidase (0.5 U/ml at 378C overnight). Chromatography was then performed as previously described, except
that elution gradient was between 0.1 and 1.0 M NaCl.
Hyaluronic Acid Affinity Chromatography
Hyaluronic acid (HA) was cross-linked to activated Sepharose according to the method of Tengblad (20) with some
modifications (34). Briefly, 5 ml epoxy-activated hydrophilic (EAH)-Sepharose (Pharmacia, Uppsala, Sweden), 10
mg HA (Worthington, Freehold, NJ), and 0.5 g N-ethylN-(3-dimethylaminopropyl)-carbodiimide (Pierce, Rockford,
IL) were mixed with 20 ml of distilled water and the pH
was adjusted to 4.5. The mixture was rotated end-over-end
at RT overnight. Beads were washed with 0.1 M Na acetate in 0.1 M NaCl (pH 4.0) and 0.1 M Tris in 0.5 M NaCl
five times alternately and then equilibrated in PBS. The
efficiency of the resulting HA-Sepharose was tested using
a biotinylated HA-binding protein (bHABP), a gift from
Dr. V. Lockeshwar, University of Miami. HA-Sepharose
(100 ml) was incubated with buffer and with 1, 10, 100, and
1,000 mg of bHABP, and rotated end-over-end for 1 h at RT.
Samples were spun gently (30 3 g) and supernatants
were tested by dot blot using avidin–horseradish peroxidase (HRP). The resulting beads showed a binding capacity
of at least 10 mg bHABP/ml. Newly prepared HASepharose was packed in a 5-ml column. The culture supernatants were pooled, and an aliquot (85 ml) was applied to the column at a flow rate of 200 ml/min. After
washing with 0.5 M NaCl in PBS, the column was eluted
with HA (1 mg/ml, 0.5 M NaCl) in PBS. Another aliquot
(20 ml) was run through a 1 ml heparin-Sepharose column
(Sigma) as a negative control.
Detection of TK Activity in Acrylamide Gels
Using an overlay membrane method described by Smith
(35), samples were electrophoresed in 12% polyacrylamide gels (precast minigels; Bio-Rad) under nondenaturing conditions, followed by the application, on top of the
gel, of a moist cellulose acetate membrane preimpregnated with DH-Val-Leu-Arg amidofluorocoumarin (AFC)
(2 mM; Novabiochem, CA), in 4 mg/ml SBTI and 10 mM
Tris (pH 8.2). After incubation for 30 min at 37 8C, the
membrane was peeled off, examined under a long-wave
ultraviolet lamp (365 nm), and photographed.
Glycosidase Digestion
Aliquots from pooled supernatants, obtained from unstimulated as well as chymase-stimulated cultures (n 5 3) (500
ml), were digested by incubation at 37 8C overnight with
heparinase in PBS, ABC (ICN) in 100 mM Tris-Cl (pH
6.5), and hyaluronidase (Worthington) in 100 mM K-Acetate (pH 5.6), each at a concentration of 0.5 U/ml. All enzymatic digestions were done in the presence of a cocktail
of protease inhibitors, consisting of leupeptin (10 27 M),
pepstatin (0.1 mg/ml), and SBTI (1 mg/ml), as well as ethylenediaminetetraacetic acid (EDTA) (1 mM). Controls
were done using each buffer without glycosidase. None of
the three glycosidases showed TK-like activity in the presence of the inhibitory cocktail.
Reversibility of Hyaluronidase Digestion
HA (10 mg/ml; Fluka, Milwaukee, WI) was added to 100 ml
of a previously hyaluronidase-digested sample and incubated for 1 h at 378C. The resulting sample was used for
TK activity measurement as well as visualization, using the
overlay membrane method (35).
Secretory Leukoprotease Inhibitor Levels
Secretory leukoprotease inhibitor (SLPI) was measured
using an ELISA kit (R&D, Minneapolis, MN) according
to the manufacturer’s instructions.
Statistical Analysis
Data in the text and figures are expressed as means 6
standard error, or means 6 95% confidence interval. The
statistical significance of differences between means was
determined by analysis of variance (ANOVA). P , 0.05
was considered significant. Linear regression was fit using
SigmaPlot (Jandel Scientific, San Francisco, CA) and the
correlation coefficient calculated using the method of least
squares.
Results
Primary Cultures of Ovine Gland Cells
Contain TK-Secreting Cells
We previously demonstrated increased TK activity in
sheep BALF (23–27) after provocation. To confirm that
the enzyme is synthesized and secreted by submucosal
glands in the sheep respiratory tract, we localized TK by
immunohistochemistry using commercially available antibodies to HUK that have been successfully used previously
in sheep tissue for the labeling of salivary TK (36). Airway
tissues from trachea to fifth-generation bronchi showed
Forteza, Lauredo, Abraham, et al.: Regulation of Bronchial Kallikrein by Hyaluronic Acid
positive labeling in the cells of the submucosal glands, but
not in any surface epithelial cells (Figure 1). Morphology
and staining with Alcian blue–periodic acid-Shiff (ABPAS) suggested that its localization was limited to serous
cells, as reported by Basbaum and associates (32). These
data are consistent with the reported localization of TK in
human and guinea-pig airways (15, 16), and suggest that
cell cultures from submucosal glands could be used as an in
vitro system to explore the regulation of TK activity.
Primary cultures of sheep tracheal submucosal tissue
were prepared as described in MATERIALS AND METHODS
and stained with AB-PAS to determine the percentage of
gland cells. After 10 d growth, 60 6 4% of cells stained
positively with AB-PAS (100 cells were counted in each of
five slides). To quantitate the fraction of cells expressing
TK and to better assess the intracellular localization of
TK, antibodies to HUK were used for immunofluorescence microscopy. The cultures showed that 41 6 23%
(100 cells were counted in each of six slides) contained immunoreactive TK. Fluorescent staining was clearly observed in secretory granules of cells (Figure 2), confirming
that submucosal gland cell cultures can be used to study
TK regulation. The data also indicate that TK is packed in
secretory granules, thus suggesting that TK could be released in response to secretagogue stimulation.
669
Chymase Stimulates Bronchial TK Secretion
To determine whether bronchial TK is regulated in a fashion similar to other secreted proteins of airway submucosal glands (e.g., lactoferrin, lysozyme, etc.), we treated
cultures with chymase, a known potent secretagogue of
respiratory tract glands (37). As illustrated in Figure 3,
chymase (1027 M) induced a significant increase (P , 0.01,
n 5 5) in the TK activity measured in the stimulated culture media (0.50 6 0.22 mU/ml), in comparison with the
control group (0.08 6 0.02 mU/ml). Chymase stimulation
also increased SLPI levels (a specific serous-cell product)
(38), and there was a significant positive correlation ( r 5
0.93) between TK activity and SLPI levels (Figure 4). The
significant relationship between SLPI and TK secretion
provided further evidence that TK was synthesized and secreted by serous cells and that its secretion was regulated
in a fashion similar to other serous-cell products (32, 37).
Treatment of media obtained from unstimulated cell cultures with chymase (10 24 M) did not alter TK activity,
showing that activation of prokallikrein by chymase was
not responsible for the increases in TK activity observed
after secretagogue stimulation.
Figure 2. TK is expressed in
cultured sheep tracheal submucosal gland cells and is localized in secretory granules:
(a) Phase-contrast micrograph
of primary cultures of ovine
submucosal gland cells. (b) Immunofluorescence of the same
field, using rabbit anti-HUK
antibodies and fluoresceinlabeled goat antirabbit IgG. (c)
Control cells incubated without anti-HUK antibodies. Bar 5
10 mm.
Figure 1. TK is localized in bronchial submucosal glands, but not
surface epithelial cells. Immunohistochemistry of sheep bronchi
(fifth generation) paraffin sections labeled with rabbit antibodies
to HUK and visualized with HRP-labeled goat antirabbit IgG
(a). Control sections were incubated without the first antibody
(b). Bar 5 20 mm.
670
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 21 1999
Figure 3. Secretion of TK is stimulated by chymase treatment of
submucosal gland cell cultures. Baseline values (BSL) of TK activity were determined in media from primary cultures of ovine
submucosal gland cells after 1.5 h incubation. TK activity was
then measured 1.5 and 24 h after stimulation with chymase (1027 M).
TK activity was measured on at least four replicates in five individual experiments. Open bars are controls without chymase;
filled bars are media from stimulated cells.
Bronchial TK Is Associated with Hyaluronic Acid
Anion exchange chromatography. To characterize the TK
activity secreted by the cultures, the enzyme was concentrated and separated from media using anion exchange
chromatography. TK activity eluted between 1.6 and 1.8 M
NaCl with a minor peak observed at 0.3 M NaCl (Figure
5). This finding differed from reports of pancreatic, urinary, and salivary kallikreins that eluted between 0.3 and
0.6 M NaCl under similar conditions (13, 39, 40). Similar
tight binding was observed using another anion exchange
column (Hitrap Q Pharmacia; data not shown). One possible explanation for bronchial TK chromatographic behav-
Figure 4. TK activity in secretions of cultured submucosal gland
cells correlated with levels of secreted SLPI. TK activity was
measured spectrophotometrically, and SLPI levels were measured by enzyme-linked immunosorbent assay (r 5 0.93). Levels
were determined in four replicated wells. Values shown are an
average of two experiments.
Figure 5. Bronchial TK behaved as a highly negatively charged
molecule in anion exchange chromatography. TK in culture media was bound tightly to a UnoQ column and eluted at 1.6 to 1.8 M
NaCl. A minor peak is observed at 0.6 M NaCl. (Filled triangles)
TK activity, ( filled circles) absorbance at 280 nm, and (filled
squares) NaCl concentration.
ior is that it was bound to a negatively charged molecule in
this culture supernatant.
Glycosidase digestion. Submucosal glands produce a variety of negatively charged glycosaminoglycans (GAGs),
such as HA and chondroitin sulfate (32, 41, 42). To test
whether such molecules were bound to TK in the culture
media, we digested media with either hyaluronidase, chondroitinase ABC, or heparinase. Comparison of TK enzyme
activity in the hyaluronidase-digested and untreated samples demonstrated that digestion dramatically increased
bronchial TK activity from 0.23 6 0.02 to 19.81 6 3.65
mU/ml. Only hyaluronidase treatment caused a change in
enzyme activity, while heparinase and chondroitinase treatment did not (Figure 6). Hyaluronidase digestion of samples increased the TK activity in both unstimulated and
chymase-stimulated samples (n 5 2, unstimulated from
0.04 6 0.01 to 0.26 6 0.03 mU/ml and postchymase from
0.32 6 0.08 to 0.52 6 0.02 mU/ml). Hyaluronidase digestion did not, however, change the magnitude of the increase in activity induced by chymase.
Figure 6. Hyaluronidase digestion of culture media remarkably
increased TK activity. Pooled media from submucosal gland cultures (n 5 3) were digested with either (a) control, (b) hyaluronidase, (c) chondroitinase ABC, or (d) heparinase. Bars are
means 6 95% confidence interval.
Forteza, Lauredo, Abraham, et al.: Regulation of Bronchial Kallikrein by Hyaluronic Acid
Figure 7. Hyaluronidase digestion changed the elution profile of
bronchial TK in anion exchange chromatography. TK activity in
culture media predigested with hyaluronidase bound to a UnoQ
column and eluted at 0.35 NaCl concentration. (Filled triangles)
TK activity, ( filled circles) absorbance at 280 nm, and (filled
squares) NaCl concentration.
To assess the effects of the digestion, samples of culture
media were run on nondenaturing gels, and TK enzyme
activity was visualized in the gel by a membrane-overlay
technique. While recombinant HUK demonstrated a single sharp band of activity, undigested samples obtained
from cultures showed a smear of activity consistent with
numerous charged molecules having TK activity (this observation was also made in sheep unconcentrated tracheal
lavage; data not shown). Only hyaluronidase digestion altered this pattern of TK activity in the gels, reducing the
heterogeneity and average mobility of the TK-treated samples, whereas chondroitinase and heparinase digestion had
no effect (Figure 7). Hyaluronidase digestion (3 h) increased
TK activity, whereas longer digestion (18 h) completely re-
Figure 8. Hyaluronidase treatment altered the electrophoretic
mobility of bronchial TK in nondenaturing conditions to that of
recombinant TK. Addition of HA restored the heterogeneity observed in undigested samples. Enzyme activity was visualized after electrophoresis by overlay with a membrane containing fluorescent substrate. (a) Lane 1: untreated sample, lane 2: after 3 h,
and lane 3: after 18 h of hyaluronidase digestion. Lane 4: recombinant HUK. (b) Lane 1: hyaluronidase-digested sample, lane 2:
sample before digestion, and lane 3: addition of HA to the digested sample.
671
Figure 9. Bronchial TK activity specifically bound immobilized
HA. Culture supernatant was applied to a HA-Sepharose column, washed with 0.5 M NaCl, and eluted with HA in 0.5 M
NaCl. (Filled triangles) TK activity, and (filled circles) absorbance at 280 nm.
duced the heterogeneity and increased the electrophoretic
mobility of bronchial TK to resemble that of recombinant
HUK. This result could be reversed by adding HA to a hyaluronidase-digested sample, restoring the heterogeneity
and reducing the TK activity (Figure 8b).
These results were consistent with HA being the molecule that conferred the high negative charge to the TK
eluted in the anion exchange chromatography, and suggest
that HA binding altered the expression of TK catalytic activity.
Anion exchange chromatography of hyaluronidasedigested samples. The finding that hyaluronidase treatment resulted in a single band of enzyme activity suggests
that anion exchange chromatography of hyaluronidasetreated samples would demonstrate a change in the elution position of TK. Hyaluronidase-digested conditioned
media was applied to a UnoQ column, and chromatography was done as before. TK activity eluted as a single peak
at about 0.35 M NaCl (Figure 7). This was consistent with
the elution profile of other tissue kallikreins (40), and confirmed the observation that hyaluronidase treatment alters
the molecular nature of bronchial TK secreted by serousgland cells.
HA affinity chromatography. To confirm that bronchial
TK specifically binds HA, undigested conditioned culture
media were applied to a HA-Sepharose column. The majority of the enzyme activity in the sample (70%) bound to
the column in 0.15 M NaCl, remained during 0.5 M NaCl
washes, and eluted with 1 mg/ml HA in 0.5 M NaCl (Figure 9). Equivalent samples, run on a heparin-Sepharose
column, showed no binding of TK activity. The ability of
TK to bind immobilized HA, and the fact that addition of
HA to the elution buffer allowed TK recovery by competition, strongly suggest that TK is a HA binding protein.
Discussion
The results of this study show that bronchial TK binds to
HA and that the association of TK with HA reduces TK
activity. The response of submucosal gland cells to chy-
672
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 21 1999
mase stimulation indicates that bronchial TK is a regulated secretory product.
The immunolocalization of TK in submucosal glands of
sheep trachea and bronchi is consistent with findings in
human and guinea pig tissue (15, 16) and encouraged us to
test cultures of these cells as an in vitro model to explore
TK regulation. Submucosal gland cultures offered several
advantages to study the behavior of TK. The culture media contained fewer proteins than did BALF, and many
potential interactions of TK with serum-derived proteinase inhibitors present in BALF (e.g., protein C inhibitor,
a2 macroglobulin, and a1 protease inhibitor) were avoided.
In addition, the absence of inflammatory cells ruled out
any contribution of neutrophil kallikrein to our measurement of TK activity.
The primary cultures used in this study were grown in
the absence of retinoic acid to favor the serous secretory
phenotype expression (43) and at confluence, 60% of the
cells were secretory, as estimated by AB-PAS staining. Although our cultures had a mixed population of epithelial
cells, TK immunolocalization in sheep airways is consistent with that reported for human and guinea pig trachea,
which indicated that TK was localized in gland cells and
that these cells were the source of the secreted enzyme in
our primary cultures. Immunofluorescence of cultures
showed that TK was packed in secretory granules, as reported for salivary and submaxillary glands (4, 5, 31, 44–
48). The positive correlation of released SLPI (a specific
serous-cell marker) and TK provided further evidence
that TK was synthesized by serous cells. When exposed to
chymase for 1.5 h, our cultures showed an increase of TK
activity. Media showed low levels of activity 24 h later,
similar to the levels found before stimulation, probably reflecting constitutive release. Salivary kallikreins are reported to be secreted in both constitutive and regulated
pathways (5, 46, 47, 49); our results suggest that this is also
true for bronchial TK.
To characterize the TK activity, we used anion exchange chromatography as a first-step purification. Kallikreins have been purified from different exocrine tissues
in various species using this approach, and typically showed
an elution profile of simple or double peaks, between 0.3
and 0.6 M NaCl (13, 39, 40). Our initial experiments
showed a different profile, in which the main peak of TK
activity eluted at high salt concentrations, suggesting that
the enzyme had a much tighter binding with the anion exchange beads under these conditions. Possible explanations for this phenomena are: (1) the isoelectric point of
this submucosal gland kallikrein was lower than previously
reported (between 4 and 5); (2) we had identified a different enzyme expressing TK-like activity; or (3) the secreted
bronchial TK was associated with a negatively charged
molecule. The biochemical characteristics of the enzyme
(Km and sensitivity to inhibitors, data not shown), the tissue localization, and its recognition by antibodies to HUK
led us to test the third hypothesis. From the molecules
secreted by submucosal gland cells, GAGs, such as heparin, chondroitin sulfate and HA, best suit these characteristics.
Digestion with hyaluronidase suggested that HA was
responsible for bronchial TK behavior in anion exchange
chromatography. Although digestion of HA could increase
accessibility of the HA-derived anions to the support, the
net number of HA charges present in the sample does not
change after digestion and would not be expected to alter
the elution of the enzyme on the basis of the capacity of
the column. In addition, hyaluronidase-dependent alteration of TK activity and electrophoretic mobility was consistent with this concept. Addition of HA to digested samples restored the original electrophoretic heterogeneity,
confirming the specificity of the hyaluronidase digestion
and indicating that the pattern of TK electrophoretic mobility, observed in untreated samples, is due to HA association. Together, these data suggest that bronchial TK is a
newly recognized HA binding protein.
In light of these results, we repeated secretagogue stimulation with chymase in a new batch of cells and compared
the activities after hyaluronidase digestion. Hyaluronidase
digestion increased the amount of TK activity almost identically in all samples (6-fold), whereas the differences before and after chymase remained constant (2-fold). This
suggests that (1) HA content in the supernatant remained
constant after chymase stimulation; (2) most of the TK
present in the supernatant was bound to a large reservoir
of HA; and (3) TK was a regulated product of these cultures. In these experiments, activation achieved by hyaluronidase was about 6-fold in contrast with the 19-fold obtained in the initial glycosidase digestion experiments (see
RESULTS). It is not clear why this difference occurred, and
we can only offer the following potential explanations to
these differences: The changes in TK activity after hyaluronidase treatment is determined by both the total
amount of TK and HA and the HA/TK ratio present in
the supernatant. Little is known about the factors controlling the rate of HA degradation in the airways, but it is
likely that factors degrading HA (including hyaluronidase) are produced by different epithelial cells. Therefore,
the total amount of HA and TK, as well as their binding
ratio, will depend on the percentage of different epithelial
cells in culture. Since these are primary cell cultures, we
have no control over this distribution, and these differences of the percentages of cells could easily account for
the observed discrepancies. Obviously, variation in additional HA binding proteins in the culture (i.e, cell surface
receptors such as CD44 or soluble binding proteins such as
lactoferrin) can alter the binding ratio between TK and
HA in an unpredictable manner.
Bronchial TK is secreted as a proenzyme, and a fraction
of TK is present in BALF as prokallikrein. Whether HA is
bound to kallikrein or prokallikrein could not be addressed in the present study because hyaluronidase digestion must be performed in the presence of protease inhibitors including SBTI, prohibiting proteolytic activation of
prokallikrein. For this reason, the data refer only to the association of active TK with HA.
The biologic reason for bronchial TK to be bound to
HA is not clear, but association of GAGs with proteases
and protease inhibitors can regulate their functions by (1)
immobilizing the enzyme, thereby restricting its range of
action; (2) stearically blocking its activity; (3) providing a
reservoir for delayed release; or (4) protecting it from proteolytic degradation (22, 50–55). These previously de-
Forteza, Lauredo, Abraham, et al.: Regulation of Bronchial Kallikrein by Hyaluronic Acid
scribed functions of GAG–proteinase association could
apply to bronchial TK–HA interactions. Here, we have
shown that HA binding blocks TK enzyme activity. Many
trypsin-like proteases, such as Cathepsin G, elastase, and
tryptase, potentially can cleave TK. It is therefore also
possible that HA binding could protect and increase the
half-life of TK in the airways, thus serving as a reservoir
for delayed release. HA has been reported to be elevated
in BALF of patients with persistent asthma in comparison
with normal or intermittent asthmatic subjects (56), indicating that HA turnover is, in fact, altered in these patients. No other kallikreins to our knowledge have been
reported to bind GAGs, but it is possible that in secretions
rich in HA (such as saliva or seminal fluid) TK, it is also
associated with this GAG.
Hypersecretion is a common feature in airway diseases
such as asthma, chronic bronchitis, bronchiectasis, and cystic fibrosis, and several products have the ability to induce
submucosal gland secretion. Cell products (e.g., elastase
from polymorphonuclear leukocytes and chymase from
mast cells) (36, 42), sensory nerve stimulation (through the
release of neuropeptides such as substance P) (57), and autonomic stimulation (mediated by muscarinic M1 and M3
receptors) (58), are well characterized secretagogues. In
light of our findings, all these mechanisms could be expected to increase TK release and, therefore, kinin generation when substrate is available, thus providing a positive
stimulus for kinin-induced airway inflammation.
Studies by Christiansen and coworkers (17, 18) showed
that TK is an important mediator in rhinitis and asthma.
They also noted that bronchial TK was resistant in vitro to
inhibition by most of the serine protease inhibitors present
in BALF, suggesting that there was no effective inhibition
for TK in the airways (59). Although our own in vivo studies have identified potential inhibitors of TK activity, the
studies described here clearly show that HA is an important and determining factor in the regulation of TK catalytic activity in the bronchial lumen. The mechanisms leading to TK upregulation during airway inflammation are
unclear, and strategies to decrease its activity are important in defining its exact role in airway pathophysiology.
Acknowledgments: The authors thank Dr. Matthias Salathe for his help and
critical comments on the manuscript, and Dr. Adam Wanner for his suggestions
and support. This work was supported by NIH-NHLBI K01-03534.
References
1. Geiger, R., and W. Miska. 1988. Human tissue kallikrein. Methods Enzymol. 163:102–115.
2. Mason, A. J., B. E. Evans, D. R. Cox, and R. I. Richards. 1983. Structure of
mouse kallikrein gene family suggests a role in specific processing of biologically active peptides. Nature 303:300–307.
3. Schenkels, L. C., E. C. Veerman, and A. Nievw Amerogen. 1995. Biochemical composition of human saliva in relation to other mucosal fluids. Crit.
Rev. Oral Biol. Med. 6:161–175.
4. Berg, T., L. Johansen, and K. Poulsen. 1990. Exocrine and endocrine release of kallikrein after reflex-induced salivary secretion. Acta Physiol.
Scand. 139:29–37.
5. Anderson, L. C., J. R. Garret, X. S. Zhang, G. B. Proctor, and D. K. Shori.
1995. Differential secretion of proteins by rat submandibular acini and
glandular ducts on graded autonomic nerve stimulation. J. Physiol.
(Lond.) 485:503–511.
6. Naidoo, S., R. Ramsaroop, R. Bhoola, and K. D. Bhoola. 1997. The evaluation of tissue kallikrein in Helicobacter pylori-associated gastric ulcer disease. Immunopharmacology 36:263–269.
673
7. Corthorn, J., C. Figueroa, and G. Valdes. 1997. Estrogen and luminal stimulation of rat uterine kallikrein. Biol. Reprod. 56:1432–1438.
8. Roa, J. P., C. A. Powers, R. Silva, and C. P. Vio. 1993. Cellular mechanisms
of estrogen and dopamine-induced control of glandular kallikrein in the
anterior pituitary gland. Cell Tissue Res. 274:421–427.
9. Bailey, G. S. 1988. Rat pancreas kallikrein. Methods Enzymol. 163:115–128.
10. Wu, H. F., R. D. Venezie, V. W. Cohen, J. W. Jenzano, G. L. Featherstone,
and R. L. Lundblad. 1993. Identification of tissue kallikrein messenger
RNA in human neutrophils. Agents Actions 38:27–31.
11. Geiger, R., and H. Fritz. 1981. Human urinary kallikrein. Methods Enzymol. 80:466–492.
12. Graf, K., M. Grafe, K. W. Auch-Schwel, C. R. Baumgarten, H. Scheffer, A.
Hildebrandt, and E. Fleck. 1994. Tissue kallikrein activity and kinin release
in human endothelial cells. Eur. J. Clin. Chem. Clin. Biochem. 32:495–500.
13. Nolly, H., G. Saed, O. Carretero, G. Scicli, and A. Scicli. 1993. Adrenal kallikrein. Hypertension 21:911–915.
14. Raidoo, D. M., R. Ramsaroop, Naidoo, and K. D. Bhoola. 1996. Regional
distribution of tissue kallikrein in the human brain. Immunopharmacology
32:39–47.
15. Proud, D., and P. Vio. 1994. Localization of immunoreactive tissue kallikrein in human trachea. Am. J. Respir. Cell Mol. Biol. 8:16–19.
16. Poblete, M. T., G. Garces, C. D. Figueroa, and K. D. Bhoola. 1993. Localization of immunoreactive tissue kallikrein in the seromucous glands of
the human and guinea-pig respiratory tree. Histochem. J. 25:839.
17. Christiansen, S. D., D. Proud, R. B. Sarnoff, U. Juergens, C. G. Cochrane,
and B. L. Zuraw. 1992. Elevation of tissue kallikrein and kinin in the airways of asthmatic subjects after endobronchial allergen challenge. Am.
Rev. Respir. Dis. 145:900–905.
18. Christiansen, S. C., D. Proud, and C. G. Cochrane. 1987. Detection of tissue
kallikrein in the bronchoalveolar lavage fluid of asthmatic subjects. J. Clin.
Invest. 79:188–197.
19. Baumgarten, C. R., R. C. Nichols, R. M. Naclerio, and D. Proud. 1986. Concentrations of glandular kallikrein in human nasal secretions increase during experimentally induced allergic rhinitis. J. Immunol. 137:1323–1328.
20. Tengblad, A. 1978. Affinity chromatography of immobilized hyaluronate
and its application to the isolation of hyaluronate binding proteins from
cartilage. Biochem. Biophys. Acta 578:289.
21. Finkbeiner, W., and C. Sommerhoff. 1990. Human tracheobronchial submucosal gland cells in culture. Am. J. Respir. Cell Mol. Biol. 2:41–50.
22. Ying, Q. L., M. Kemme, D. Saunders, and S. R. Simon. 1997. Glycosaminoglycans regulate elastase inhibition by oxidized secretory leukoprotease
inhibitor. Am. J. Physiol. 272:L533–L541.
23. Abraham, W. M., R. Forteza, A. Ahmed, and A. Cortez. 1994. Neural
mechanisms regulate the early and inflammatory cells the late increase in
tissue kallikrein activity after local antigen challenge in allergic sheep. Am.
J. Respir. Crit. Care Med. 149:A533. (Abstr.)
24. Forteza, R., R. M. Burch, and W. M. Abraham. 1994. Increased tissue kallikrein activity, kinins and decreased a1-proteinase inhibitor activity are
linked to ozone induced airway hyperresponsiveness. Am. J. Respir. Crit.
Care Med. 149:A158. (Abstr.)
25. Forteza, R., R. M. Burch, and W. M. Abraham. 1994. Extracellular metabolites of Pseudomonas aeruginosa produce bronchoconstriction by different
mechanisms. Am. J. Respir. Crit. Care Med. 149:687–693.
26. Mansour, E., A. Ahmed, A. Cortes, J. Caplan, R. Burch, and W. M. Abraham. 1992. Mechanism of metabisulfite-induced bronchoconstriction: evidence for bradykinin B2 receptor stimulation. J. Appl. Physiol. 72:1831–1837.
27. Forteza, R., Y. Botvinnicova, A. Ahmed, A. Cortes, R. Gundel, A. Wanner,
and W. M. Abraham. 1996. The interaction of alpha 1 proteinase inhibitor
and tissue kallikrein in controlling allergic ovine airway hyperresponsiveness. Am. J. Respir. Crit. Care Med. 154:36–42.
28. Wu, R., J. Yankaskas, R. Boucher, and M. Knowles. 1985. Growth and differentiation of human nasal epithelial cells in culture. Serum-free, hormone-supplemented medium and proteoglycan synthesis. Am. Rev. Respir.
Dis. 132:311–320.
29. Tournier, J. M., M. Merten, J. Meckler, J. Hinnraeky, and E. Puchelle. 1990.
Culture and characterization of human tracheal gland cells. Am. Rev. Respir.
Dis. 141:1280–1288.
30. Tournier, J. M., G. A. Goldstein, D. E. Hall, C. H. Damsky, and C. B. Basbaum. 1992. Extracellular matrix proteins regulate morphologic and biochemical properties of tracheal gland serous cells through integrins. Am. J.
Respir. Cell Mol. Biol. 6:461–471.
31. Garrett, J. R., Kidd, A. Kyriacou, and R. E. Smith. 1985. Use of different
derivatives of DValLeuArg for studying kallikrein activities in cat submandibular glands and saliva. Histochem. J. 17:805–818.
32. Basbaum, C. B., B. Jany, and W. E. Finkbeiner. 1990. The serous cell. Annu.
Rev. Physiol. 52:97–113.
33. Salathe, M., M. Holderby, R. Forteza, W. M. Abraham, A. Wanner, and
G. E. Conner. 1997. Isolation and characterization of a peroxidase from
the airway. Am. J. Respir. Cell Mol. Biol. 17:95–105.
34. Miura, N. C., T. Tobe, J. Sumiya, Y. Nakan, Y. Sano, T. Mazda, and M. Tomita. 1996. Purification and characterization of a novel hyaluronan-binding protein (HABP) from human plasma: it has three EGF, a Kringle and
a serine protease domain, similar to hepatocyte growth factor activator. J.
Biochem. 119:1157–1165.
674
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 21 1999
35. Smith, R. 1984. Identification of protease isozymes after analytical isoelectrical focusing using fluorogenic substrates impregnated into cellulose membranes. J. Histochem. Cytochem. 32:1265–1274.
36. Trahair, J. F., and G. B. Ryan. 1989. Colocalization of neuron-specific enolase-like and kallikrein-like immunoreactivity in ductal and tubular epithelium of sheep salivary gland and kidney. Histochem. Cytochem. 37:309–314.
37. Sommerhoff, C. P., G. H. Caughey, W. E. Finkbeiner, S. C. Lazarus, C. B.
Basbaum, and J. A. Nadel. 1989. Mast cell chymase: a potent secretagogue
for airway gland serous cells. J. Immunol. 142:2450–2456.
38. Basbaum, C. B. 1986. Regulation of airway secretory cells. Clin. Chest Med.
7:231–237.
39. Johansen, L., H. Bergundhaugen, and T. Berg. 1987. Rapid purification of
tonin, estearase B, antigen gamma and kallikrein from rat submandibular
gland by fast protein liquid chromatography. J. Chromatogr. 387:347–359.
40. Raspi, G. 1996. Kallikrein and kallikrein-like proteinases: purification and
determination by chromatographic and electrophoretic methods. J. Chromatogr. 684:265–287.
41. Baraniuk, J. N., T. Shizari, M. Sabol, and C. B. Underhill. 1996. Hyaluronan
is exocytosed from serous, but not mucous cells, of human nasal and tracheobronchial submucosal glands. J. Investig. Med. 44:47–52.
42. Sommerhoff, S. P., J. A. Nadel, C. B. Basbaum, and G. H. Caughey. 1990.
Neutrophil elastase and cathepsin G stimulate secretion from cultured bovine airway gland serous cells. J. Clin. Invest. 85:682–689.
43. Yoon, J. H., T. Gray, K. Guzman, S. Koo, and P. Nettesheim. 1997. Regulation of the secretory phenotype of human airway epithelium by retinoic
acid, triiodothyronine, and extracellular matrix. Am. J. Respir. Cell Mol.
Biol. 16:724–731.
44. Garret, J. R., G. B. Proctor, X. S. Zhang, D. K. Shori, and B. A. Schulte.
1996. Use of lectin-probes for correlative histochemical and biochemical
assessments of the glycosylation patterns of secretory proteins, including
kallikreins, in salivary gland and saliva. Histol. Histopathol. 11:503–512.
45. Elverdin, J. C., M. O. Kianucki, F. J. Stefano, and C. J. Ferec. 1990. Physiological role of alpha-adrenoreceptor in salivary secretion. Acta Odontol.
Latinoam 5:31–38.
46. Marsuda, Y., Y. Fujimoto, Watanabe, T. Obara, and S. Akihama. 1989. Hu-
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
man salivary kallikrein and submaxillary glands kallikreins. Adv. Exp.
Med. Biol. 247:169–175.
Garret, J. R., X. S. Zhang, G. B. Proctor, and L. C. Anderson. 1996. Sequential secretion of rat submandibular kallikrein and peroxidase during intermittent sympathetic stimulation. J. Auton. Nerv. Syst. 7:26–30.
Jenzano, J. M., H. Su, G. Featherstone, and R. L. Lundblad. 1992. Molecular diversity of tissue kallikrein in human saliva. Agents Actions 38:137–144.
Garrett, J. R., G. B. Proctor, X. S. Zhang, L. C. Anderson, and D. K. Shori.
1998. Constitutive secretion of kallikreins in vivo from rat submandibular
glands. Eur. J. Morphol. 36:86–91.
Hardingham, T. E., and A. J. Fosang. 1992. Proteoglycans: many forms and
many functions. FASEB J. 6:861–870.
Wight, T. N., M. G. Kinsella, and E. E. Qwarnstrom. 1992. The role of proteoglycans in cell adhesion, migration and proliferation. Curr. Opin. Cell
Biol. 4:793–801.
Jackson, R. L., S. J. Bush, and A. D. Cardin. 1991. Glycosaminoglycans: molecular properties, protein interactions, and role in physiological processes.
Physiol. Rev. 71:481–539.
Hay, E. D. 1991. Cell Biology of Extracellular Matrix. Plenum Press, New
York.
Kjellen, L., and U. Lindahl. 1991. Proteoglycans: structures and interactions. Annu. Rev. Biochem. 60:443–475.
Laurent, T. C., U. B. Laurent, and J. R. Fraser. 1996. The structure and
function of hyaluronan: an overview. Immunol. Cell Biol. 74:A1–A7.
Vignola, A. M., P. Chavez, A. M. Campbell, F. Souques, B. Lebel, I.
Enander, and J. Bousquet. 1998. Airway inflammation in mild intermittent
and persistent asthmatics. Am. J. Respir. Crit. Care Med. 157:409–414.
Gashi, A. A., D. B. Borson, W. E. Finkbeiner, J. A. Nadel, and C. B. Basbaum. 1986. Neuropeptides degranulate serous cells of ferret tracheal glands.
Am. J. Physiol. 251:C223–C229.
Culp, D. J., W. Luo, L. A. Richardson, G. E. Watson, and L. R. Latchney.
1996. Both M1 and M3 receptors regulate exocrine secretion by mucous
acini. Am. J. Physiol. C1963–C1972.
Christiansen, S. C., B. L. Zuraw, D. Proud, and C. G. Cochrane. 1989. Inhibition of bronchial kallikrein in asthma. Am. Rev. Respir. Dis. 139:1125–1131.