| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Immunology, University of Konstanz, Konstanz, Germany; and
Institute of Biochemistry, Christian-Albrechts University of Kiel, Kiel, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-tocopherol. We have shown that MMP-19 associates with the
cell surface of myeloid cells. This cell surface association was not
affected by phospholipase C. However, acidic treatment of the
THP-1-derived cell membranes abolished the immunoprecipitation of
MMP-19 thereof. Moreover, a high salt treatment of THP-1 cells
diminished the MMP-19 detection on the cell surface. This implicates a
noncovalent attachment of MMP-19 to the cell surface. Because a
truncated form of the MMP-19, in which the hemopexin-like domain was
deleted (
hpMMP-19), does not associate with the surface,
the hemopexin-like domain appears to be critical for the cell surface
attachment of human MMP-19. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MMPs are regulated via mechanisms that involve transcription, posttranscriptional processing, and inhibition by MMP-specific and unspecific inhibitors. Most of the MMPs are secreted in inactive forms (zymogens) that are activated by an extracellular mechanism called the cysteine switch. The activity of MMPs can be regulated by specific inhibitors called tissue inhibitors of metalloproteinases. In general, MMPs function in a highly regulated fashion as a part of the physiological turnover of the extracellular matrix. Although the precise control of MMP expression ensures their normal physiological role, their dysregulation, due to intrinsic and extrinsic factors, may have pathological consequences. An imbalance in the regulation of the expression of MMPs and tissue inhibitors of metalloproteinases has been implicated in cancer, metastasis, inflammation, and angiogenesis. (1, 2, 3).
The cDNA of human MMP (hMMP)-19 has been isolated from a synovium of a patient suffering from rheumatoid arthritis as an autoantigen (4). The genomic structure of the human and murine orthologs, consisting of nine exons, is unusual (5, 6). Investigation of MMP-19 expression has revealed a unique expression pattern in human tissues, although its mRNA levels are relatively high in many organs (7, 8). Under normal physiological conditions, MMP-19 is strongly expressed in the tunica media by vascular smooth muscle cells but not by endothelial cells (9). Although MMP-19 is not present in the quiescent endothelium of arteries and veins, we found its expression in certain capillaries of an acutely inflamed synovium (10). Compared with other MMPs, MMP-19 seems to undergo specific regulation during progression of rheumatoid arthritis and, in addition, in the ovary during gonadotropin-induced ovulation (11, 12). The regulation of MMP-19 expression in mammary glands as well as benign and malignant tumors thereof is also unique. For example, the expression of MMP-19 is down-regulated in malignant carcinomas (13).
The capability of MMP-19 to associate with the cell surface is
unexpected, because it has no obvious transmembrane domain such as
membrane-type MMP. A similar phenomenon has been previously described
for gelatinase A (MMP-2) and its hemopexin fragment, PEX. For example,
both the MMP-2 and the PEX fragment bind to the
v
3 integrin
(14, 15).
Recently, we have reported that activated PBMCs express hMMP-19 on their cell surface. Because MMP-19 can, for instance, cleave collagen type IV, tenascin C, fibrin, and fibrinogen, its activity may be important for the physiology of PBMC, especially for their transmigration through blood vessel walls (16). In the work reported in this study, we have focused on the analysis of the MMP-19 expression in PBMC, especially in myeloid cells. MMP-19 is expressed by myeloid cells in an adhesion-dependent manner and associates with the cell surface by an interaction with the hemopexin-like domain.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The isolation and analysis of mRNA coding for MMP-19 was performed using an RT-PCR assay based on the primer dropping method (17). This semiquantitative PCR was performed in 100-µl (100 ng RNA) reaction volumes on a PREM thermal cycler (LEP Scientific, Serva Feinbiochemica, Heidelberg, Germany). All preparations were free of genomic DNA. Each reaction contained 3 µl of cDNA from the previously described reverse transcription reaction (corresponding to cDNA synthesized from 100 ng of total RNA) or 3 µl of negative control, 10 mmol/L Tris-HCl (pH 9), 50 mmol/L KCl, 1.5 mmol/L MgCl2, 0.1% Triton X-100, 0.2 mg/ml BSA, 0.2 mmol/l of all four dNTPs, 2.5 U of Taq-Polymerase (Appligene Oncor, Pleasanton, CA) and 0.5 µmol/L MMP-19-specific primers. Primer and template cDNA were separated from the other reaction components by a layer of paraffin wax until the first denaturation step. Each PCR cycle consisted of a denaturation step at 94°C for 45 s, an annealing step at 62°C for 45 s, and an elongation step at 72°C for 45 s. A total of 5 µl of the GAPDH-specific primer pair (0.5 µmol/L each) were added after 15 cycles and PCR was continued for another 15 cycles. The number of cycles for the MMP-19 and the GAPDH-specific primer pairs were chosen to be in the exponential range of the amplification reaction to obtain semiquantitative results (data not shown). All RT-PCR experiments were repeated at least once. Primers were as follows: MMP-19, 5'-TGCCCACAGAACCCAGTCC-3' and 5'-GGTATTCCCACCTGATGGGGTAG-3' (product length, 626 bp); GAPDH, 5'-ACCATCTTCCAGGAGCGAG-3' and 5'-ACGTTGGCAGTGGGGACAC-3' (product length, 498 bp).
Primary cells and cell lines
Blood-derived cells were fractioned by centrifugation with an LSM medium (ICN Pharmaceuticals, Costa Mesa, CA) according to the manufacturer’s recommendations. The isolated cells, consisting of monocytes and macrophages as well as T and B lymphocytes were resuspended in RPMI 1640 medium supplemented with 0.5–50% autologous serum. For the adhesion study, freshly isolated PBMC were cultured in ultra-low attachment plates (Sigma, Deisenhofen, Germany) in presence of autologous serum (20%).
The CD14+ cell fraction of PBMC was isolated using MACS columns according to the manufacturer’s recommendations (Miltenyi Biotec, Bergisch Gladbach, Germany).
Human peritoneal macrophages, obtained from a patient with chronic renal failure undergoing peritoneal dialysis, were washed with PBS and directly subjected to flow cytometry analysis.
The myeloid cell lines (THP-1 and HL-60) and the lung fibroblasts (SV80), as well as the other cells (Hela, Hep G2, Jurkat, and ECV304), were cultured in RPMI 1640 medium supplemented with 5% FCS, 25 mM HEPES, and 2 mM glutamine. For the induction of gene expression, 20 ng/ml PMA was used.
Anti-MMP-19 Abs
Two specific anti-MMP-19 Abs were used. First, the immunoaffinity-purified, polyclonal, rabbit anti-MMP-19 Abs designated anti-Pep21 that recognize a unique 21-aa-long peptide derived from the MMP-19 hinge region. These Abs were prepared by immunization with the keyhole limpet hemocyanin-coupled peptide and by immunoaffinity purification using immobilized peptide Pep21 (4). Second, the mAb CK8/4 was prepared by immunization of a mouse with the full-length hMMP-19. The selected Ab CK8/4 recognizes an amino acid sequence, NH2-LQKPLEGSNNFKPDITEALRAFQEASEL, as it was determined by digestion of captured MMP-19 on the CK8/4 affinity column followed by elution and matrix-assisted laser desorption ionization-time of flight analysis (K. Kolb, M. Kohlmann, C. Colb, M. Przybylski, and R. Sedlacek, manuscript in preparation). This amino acid stretch is localized in the propeptide domain of hMMP-19, i.e., unprocessed MMP-19. Further analysis with synthetic peptides has shown that the CK8/4 does not cross-react with known MMPs. The Ab was purified using protein A-Sepharose (Pharmacia, Freiburg, Germany).
Labeling of the cell surface and immunoprecipitation
A modification of the cell surface biotinylation method according to the Pierce’s protocol was used. Briefly, a solution of 1 mg/ml EZ-Link Sulfo-NHS-LC-Biotin (Pierce, Sankt Augustin, Germany) was used to label 4 x 108 cells. Labeling was performed in the dark at 4°C for 45 min. After labeling, cells were washed four times with PBS/0.15% glycine at 4°C and then used for immunoprecipitation.
For immunoprecipitation, the purified mAb CK8/4 (2 µg) was coupled to Dynabeads M-450 rat anti-mouse IgG1 (Deutsche Dynal, Hamburg, Germany) according to the manufacturer’s recommendations. To precipitate MMP-19, 5 x 106 THP-1 cells were washed twice with PBS followed by lysis with RIPA buffer (1% Nonidet P-40, 0.1% SDS, 1% Na-deoxycholate, 10 mM Tris-HCl, pH 8, 140 mM NaCl, and Complete (Boehringer Mannheim, Mannheim, Germany) protease inhibitor without EDTA) for 45 min on ice. The cell remnants were removed by centrifugation. A 100-µl aliquot of Dynabead Ab solution was added to the cell lysates and this suspension was gently mixed at 4°C for 2 h. After polyacrylamide electrophoresis, the precipitated proteins were identified with both specific Abs (CK8/4 and anti-Pep21) and by using streptavidin-peroxidase.
To test the type of cell surface association of MMP-19, THP-1 cells were pretreated with a high-salt buffer (500 mM NaCl, 10 mM Na2HPO4, pH 11) and then lysed with TSA buffer (10 mM Tris-HCl, pH 8, 140 mM NaCl, Complete inhibitor without EDTA) containing 1% Nonidet P-40. Cell membranes were then isolated by ultracentrifugation using a two-step sucrose gradient as described (18).
Construction and expression of deletion mutants of MMP-19
In addition to the full-length MMP-19, two truncated forms of
MMP-19 were constructed in which either the hemopexin-like domain
(hpMMP-19) or the catalytic
domain (catMMP-19) was deleted.
All three constructs were tagged with an HA tag (YPYDVPDYA), a
nonapeptide derived from human influenza hemagglutinin protein. The HA
tag was inserted into the hinge region. The constructs were produced by
PCR using TurboPfu (Stratagene, La Jolla, CA) and following
primer pairs: 1) for catMMP-19 (HA tag between the
catalytic domain and the hinge region), proHA-r
(5'-AGGGGCGTAGTCGGGCACGTCGTAGGGGTATTTAAGGGTCTTCTGGTTGAAG-3')
and hemo-f (5'-GACCCTTGCAGTAGTGAACTG-3'), 2) for
hpMMP-19 (C-terminal HA tag in
the hinge region), rasih-HA-r
(5'-TTCTCGAGTCAGGCGTAGTCGGGCACGTCGTAGGGGAAGGGTCTGGCATGGGACTGGGTTC-3')
plus r-orf-5-Eco
(5'-AAGAATTCCACCARGAACTGCCAGCAGCTGTGG-3'), and 3) for the
full-length hMMP-19-HA (the HA tag between the catalytic domain and the
hinge region), hingeHA-r
(5'-AGGGGCGTAGTCGGGCACGTCGTAGGGGTATGGCATGGGACTGGGTTCTGTG-3') plus
hemo-f. pcDNA3 with subcloned MMP-19 was used as a PCR
template. To construct the catMMP-19-HA and the whole MMP-19-HA a
circular PCR amplification of the complete vector was performed.
Circularization of the PCR products was performed as described
elsewhere (18). Plasmids were then transfected into THP-1
cells using electroporation. Stabile-transfected clones were selected
in RPMI 1640 medium with 10% FCS and 0.5 mg/ml G418.
Flow cytometry and analysis of MMP-19 cell surface expression
The flow-cytometry analysis was performed by using a FACScan device (BD Immunocytometry Systems, San Jose, CA) using Lysis II systems. Briefly, suspension of 1 x 106 cells were stained for 1 h in 50 µl of PBS with 2% newborn calf serum, 0,1% NaN3, and immunoaffinity-purified, polyclonal, rabbit anti-MMP-19 Abs. The incubation with primary Abs was followed by incubation with Cy3-conjugated goat anti-rabbit IgG (Dianova, Hamburg, Germany). To test the type of the cell surface association of MMP-19, THP-1 cells were also pretreated with an acidic buffer (50 mM Na-acetate, 85 mM NaCl, 5 mM KCl, 1% FCS) to release the molecules that were noncovalently attached to the cell surface as previously described (19). Phospholipase C (PLC) treatment of THP-1 cells was done using PLC from Bacillus cereus (Roche, Basel, Switzerland) in Earle’s medium according to the manufacturer’s recommendations. Briefly, 1.5 x 106 cells were washed twice and then resuspended in 100 µl of Earle’s medium and 0.5 U of PLC was added. After incubation at 37°C for 1 h, cells were washed twice and analyzed by flow cytometry as described above.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To investigate the regulation of MMP-19 gene expression
and to establish a model for further studies of MMP-19 function, we
first analyzed the expression of MMP-19 mRNA in various cell
types using a semiquantitative RT-PCR assay (17). Among
the cell lines studied, the lung fibroblast cell line SV80 as well as
two myeloid cell lines, THP-1 and HL-60, expressed MMP-19 at
the mRNA level. However, the unstimulated promyelocytic cell line HL-60
exhibits a weaker signal than monocytic THP-1 cells (Fig. 1
).
|
). Expression of MMP-19 in PBMCs
The above analysis of MMP-19 expression in various cell
lines suggests that myeloid cells are the major producers of MMP-19 in
blood-derived cells. In contrast, based on Northern analysis (7, 8) as well as flow cytometry analysis (4), it has
been reported that unstimulated PBMC do not express MMP-19.
To address whether this observation points to a specific endogenous
stimulus driving the expression of MMP-19, we isolated PBMC using a
Ficoll (Sigma) gradient and cultured them in a medium
supplemented with autologous serum. Freshly isolated PBMC appear to be
negative for the MMP-19 mRNA while the cultured PBMC show
significant induction of MMP-19 mRNA (Fig. 2A). This up-regulation can be
seen after 16 h and correlates positively with the concentration
of the autologous serum present in medium such that the higher the
content of serum used the stronger the MMP-19 expression
observed.
|
Based on the results described above, we focused further on
studying the MMP-19 regulation in myeloid cells. First, we
compared the expression of MMP-19 mRNA among unsorted PBMC,
and PMBC which were separated into CD14+
(predominantly monocytes and macrophages) and the
CD14- (lymphocytes) populations. The analysis
was performed immediately after the sorting of freshly isolated PBMC
(Fig. 2B). In all three populations, no mRNA message of
MMP-19 could be identified under these conditions. In the
second set of experiments, the above cells were sorted into
CD14+ and CD14-
populations, and then cultured for 2 days in RPMI 1640 medium
supplemented with an autologous serum (Fig. 2C). Under these
conditions it was evident that MMP-19 mRNA expression was induced in
the CD14+ fraction without additional
stimulation. The weak band in the CD14- fraction
probably originates from cross-contamination with myeloid cells because
the magnetic sorting achieved only 95% purity of the
CD14- population according to a simultaneous
flow cytometry analysis (data not shown). Third, similar results were
obtainedwhen PBMC were cultured for 2 days under identical
conditions and then sorted into the CD14+ and
CD14- populations. As shown in Fig. 2D, the entire MMP-19 mRNA message in PBMC
originated from CD14+ cells.
The expression of MMP-19 is up-regulated during differentiation of myeloid cells
As can be seen in Fig. 1, THP-1 and HL-60 cells differ in the
level of MMP-19 expression. This difference could mirror
their distinct differentiation potential. While the THP-1 cells develop
into macrophages, HL-60 can differentiate into several cell types. Upon
stimulation with PMA, HL-60 cells differentiate into monocytes, which
proceed further into macrophages (20, 21). To gain insight
into how MMP-19 is regulated during differentiation of
myeloid cell types both myeloid cell lines were stimulated with PMA.
This treatment induces differentiation with concurrent inhibition of
proliferation in HL-60 cells (20). Expression of
MMP-19 between these myeloid cell lines was then compared
using a semiquantitative RT-PCR (Fig. 2E). Although the
expression of MMP-19 is barely visible in the unstimulated
HL-60 cells, both cell lines show strong and comparable expression
after the PMA stimulation, indicating that macrophages are major
producers of MMP-19.
Regulation of MMP-19 expression through adhesion
Induction of the MMP-19 expression due to addition of serum cannot be explained by a simple stimulatory effect of the serum because an autologous serum was used. Various factors that are present in a serum such as cytokines, growth factors and adhesion molecules could be possible stimuli leading to up-regulation of MMP-19 in the culture. Although adhesion mechanisms (e.g., between blood-derived cells and endothelium) are very complex in vivo, a simplified interaction among serum adhesion factors, cells, and the surface of the cell culture vessel can also induce gene expression. If the adhesion molecules are involved in regulation of the MMP-19 expression, a weakening of the adhesion to the vessel surface should inhibit the up-regulation of MMP-19.
Although PBMC cultured in the medium supplemented with the autologous
serum do up-regulate MMP-19 mRNA, the expression level was
minimal when the ultra-low attachment cell culture vessels were used
(Fig. 3A). Moreover, we
treated PBMC with -tocopherol (Sigma), which is known to inhibit
adhesion in cells such as monocytes and macrophages. Blocking of cell
adhesion by -tocopherol involves a protein kinase C pathway and is
independent of its action as an antioxidant, because -tocopherol as
well as other chemically similar substances do not exhibit such
antiadhesion effect (22, 23, 24).
|
-tocopherol while full inhibition of the MMP-19 induction
is reached at 50 µM -tocopherol. -tocopherol did not influence
the expression of MMP-19. The same inhibition could be
observed when cultured PBMC were stimulated with 20 ng/ml PMA (Fig. 3B). HL-60 cells, stimulated with 20 ng/ml PMA, also showed
no MMP-19 mRNA induction when they were cultured and treated
in the same way as PBMC (Fig. 3C).
Both reduction of the adhesiveness by the low-attachment surface and
the blockade of adhesion by -tocopherol indicate that the MMP-19
expression in myeloid cells is regulated by cell adhesion.
MMP-19 associates with cell surface of myeloid cells
Previously, we have shown that hMMP-19 associates with the cell surface of activated PBMC and Chinese hamster ovary cells transfected with MMP-19 (4). Based on these data and the expression analysis above, we have further analyzed the cell surface association of MMP-19 in blood-derived cells.
PBMC were isolated from human fresh blood using a Ficoll gradient and
then cultured for 2 days in the presence of 20% autologous serum. The
association of MMP-19 with the cell surface was studied using flow
cytometry (FACScan) and by immunoprecipitation of biotin-labeled cell
surface molecules. In accordance with our previous studies, MMP-19
could not be found in freshly isolated PBMC, whereas a clear cell
surface expression was seen in cultured cells (Fig. 4, A and B).
|
C). However, it was present on the cell surface of
CD14+ cells, i.e., monocytes and macrophages
(Fig. 4D). Peritoneal macrophages also exhibit an apparent
cell surface expression of MMP-19 (Fig. 4E). Similar results
were obtained upon analysis of the myeloid cell lines. Whereas MMP-19
can be detected on the cell surface of both THP-1 and HL-60 cells (Fig. 4, F and G), it is not expressed at the surface
of Jurkat cells, even after stimulation with PMA (Fig. 4, H
and I).
To confirm the results obtained with FACScan analysis we used a new
high-affinity mAb CK8/4 recognizing the propeptide domain of MMP-19.
The Ab was used to precipitate MMP-19 from a whole cell lysate and also
from the cell surface labeled with biotin. The precipitated proteins
were analyzed by immunoblotting analysis. A small portion of the
expressed MMP-19 could be detected using immunoblotting after 10 times
concentration of the THP-1 cell culture medium obtained after a 3-day
culture. MMP-19 could not be detected in the supernatant without
concentration using acetone precipitation (data not shown). A
specific MMP-19 band was detected with both the CK8/4 mAb and the
polyclonal monospecific anti-MMP-19 (anti-Pep21) Abs (Fig. 5A). The anti-Pep21 Abs,
recognizing the hinge region of MMP-19, verified the specificity of the
precipitation by CK8/4. Moreover, the cell surface THP-1 cells were
biotinylated and then lysed with RIPA buffer. The mAb CK8/4 was
used to precipitate MMP-19, which in the case of its cell surface
localization has to be biotinylated. Indeed, the biotinylated
MMP-19 was clearly detected after incubation of the Western blot with
streptavidin-coupled peroxidase (Fig. 5B).
|
Based on both the flow cytometry analysis and the precipitation of biotin-labeled cell surface molecules, we conclude that MMP-19 associates with the surface of human monocytes and macrophages as well as THP-1 and HL-60 cell lines, i.e., human cells of myeloid origin.
Because MMP-19 belongs to the "classical" MMPs, which do not have a
transmembrane region, the nature of its cell surface association is
unclear. Identification of the mode of association with the cell
surface could help to delineate the role of MMP-19 in myeloid cells.
There are several theoretical possibilities of how MMP-19 could
associate to the cell surface. First, MMP-19 could be anchored to the
cell membrane with its signal sequence as a type II transmembrane
protein. This seems to be unlikely because MMPs are usually secreted
and the signal sequences are cleaved off during the transport through
the endoplasmatic reticulum. In addition, MMP-19 has a predicted
cleavage site of its signal sequence between the amino acid residues 18
and 19. Second, MMP-19 has an unusual C-terminal tail of 37 amino acids
that is rich in threonine residues and has a partially hydrophobic
character (4, 7, 8). Such a sequence could represent a
hypothetical signal for the addition of a GPI anchor. Third, MMP-19
could be bound to the cell surface via interaction with a cell surface
receptor, for instance, as in the case of MMP-2, through interaction
with the integrin v3
(14).
To examine the possible cell surface association of MMP-19 via a GPI
anchor, THP-1 cells were treated with PLC. This enzyme exhibits
specificity for phosphatidylinositol and lysophosphatidylinositol and
should thus liberate most of the GPI-anchored molecules from the cell
surface. As can be seen in the Fig. 6, A–D, addition of PLC to THP-1 cells did not
affect the MMP-19 cell surface association, although CD14, a
GPI-anchored molecule that is specifically expressed on monocytes and
macrophages, was released. In addition to the PLC treatment we isolated
cell membranes of THP-1 cell using two-step centrifugation over sucrose
gradient. The membranes were then treated with an alkaline, high-salt
solution to dissociate cell surface molecules that are not covalently
bound. It can be seen in Fig. 6E that such a treatment
prevented precipitation of the membrane-bound MMP-19 with the mAb
CK8/4. A similar result was achieved by washing of THP-1 cells with an
acetate buffer which also diminished the cell surface association of
MMP-19 (Fig. 6, F–G). The above experiments
indicate that MMP-19 is not covalently bound to the cell surface.
|
v3 via its
hemopexin-like domain (14, 15).
To test this possibility, we prepared MMP-19 constructs lacking either
the catalytic (catMMP-19) or hemopexin-like
(hpMMP-19) domains. To distinguish the
recombinant MMP-19 from an endogenous one we placed a hemagglutinin
protein tag (HA tag) into the hinge region of MMP-19. THP-1 cells that
naturally express MMP-19 at the cell surface were then transfected with
recombinant MMP-19 cDNA. The transcription of the constructs was
monitored by RT-PCR (data not shown). A FACScan analysis was used to
analyze the cell surface association (Fig. 7, A–D). We used a
full-length MMP-19 that associates with the cell surface as a
control (Fig. 7B). The deletion of the hemopexin-like domain
had a profound effect on the MMP-19 localization. As can be seen in the
Fig. 7C, the hpMMP-19
is not able to associate with the surface. The second construct, the
catMMP-19, still associates with
the surface although with a slightly diminished abundance (Fig. 7D).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
First, the expression of hMMP-19 at the mRNA level was studied. Among the analyzed cell lines, the hepatocytic cell line HepG2, the cervix carcinoma line HeLa, the epithelial cell line ECV304 (previously reported as an endothelial cell line), and T cell lines Molt 4, and Jurkat (data not shown) did not produce a detectable amount of MMP-19 mRNA. The absence of MMP-19 mRNA in the HepG2 cells corresponds with the lack of mRNA of MMP-19 in liver (7, 8). The expression of MMP-19 in the lung fibroblastic cell line SV80 correlates with its expression in lung (7, 8).
The two cell lines of myeloid origin, THP-1 and HL-60, produce MMP-19, although the expression in HL-60 cells is only weak under normal conditions. These myeloid lines vary in their differentiation capacity. However, they can be stimulated to differentiate along the monocytes/macrophages cell lineage after their stimulation with PMA. While THP-1 cells differentiate into macrophages, HL-60 cells stimulated with PMA possess the capacity to differentiate into monocytes and then macrophages. Moreover, a treatment of HL-60 cell with PMA induces their differentiation into monocytes and macrophages (not to neutrophils and granulocytes) and inhibits their proliferation at the same time (20, 21).
After stimulation with PMA, both cell lines up-regulate mRNA for MMP-19. Because the up-regulation of MMP-19 is first observable after a 6-h delay and is stable for at least 2 days, this is most likely no direct effect of phorbol ester on the protein kinase C activation pathway, and the up-regulation appears to be in accord with the previously published fact that PMA blocks the proliferation of HL-60 cells. Thus the induction of MMP-19 expression in the HL-60 cells can be attributed, at least partially, to their differentiation into monocytes and macrophages. A clear expression of MMP-19 can be also seen in peritoneal macrophages isolated from a patient with chronic renal failure. A similar phenomenon was also observed for macrophage elastase (MMP-12), which is expressed almost exclusively in differentiated macrophages (25, 26, 27). Moreover, it has been shown that the proteinase repertoire changes during monocytes differentiation. While monocytes express more serine proteases and less MMP, maturation of macrophages is accompanied by a decrease of serine protease production and up-regulation of MMP (28, 29, 30).
The delay in the up-regulation of MMP-19 mRNA in the myeloid
cell lines led us to assume that an indirect mechanism could be
involved. Therefore, we focused on an adhesion-dependent regulation of
the MMP-19 expression. The first evidence for this
hypothesis came from the dependence of MMP-19 expression on
the autologous serum concentration in cell culture of PBMC. Using the
autologous serum we eliminated simple stimulatory effects of allogeneic
or xenogeneic sera. In accord with this observation, PBMC and HL-60
cells cultured on ultra-low attachment dishes exhibited a marked
decreased in MMP-19 expression. In addition, -tocopherol,
which inhibits cellular adhesion in certain cell types, among them
macrophages, also decreased the expression of MMP-19 in a
concentration-dependent manner. This compound is known to suppress
expression of several adhesion molecules, including ICAM-1, VCAM-1,
endothelial leukocyte adhesion molecule-1 (E-selectin), and CD11b/CD18
in several cell types. Indeed, it has been reported that -tocopherol
inhibits several processes, including cell proliferation, platelet
aggregation, and monocyte adhesion (22, 23, 24, 31).
The effect of -tocopherol on the MMP-19 expression is not
due to its antioxidative properties because -tocopherol, another
antioxidant compound, does not affect the regulation of
MMP-19. -tocopherol has identical antioxidative
properties to -tocopherol and has the same biological availability
(22, 23, 24).
The phenomenon of adhesion-coupled induction of expression is also
known for other MMPs. For instance, blocking of the interaction between
51 integrin and
fibronectin inhibits the expression of gelatinase B and, moreover, the
ligation of the fibronectin receptor Mac-1
(m2) increases the
production of this MMP (32, 33).
Regarding the MMP-19 expression in PBMC, we have analyzed which cell types exhibit a surface expression of this enzyme. Freshly isolated, nonactivated PBMC do not have cell surface MMP-19. In contrast, MMP-19 is also found in freshly isolated peritoneal macrophages and in myeloid cell lines HL-60 and THP-1.
Because MMP-19 has no obvious transmembrane domain, an alternative mechanism for its cell surface association must exist. The 37-aa-long C-terminal domain of MMP-19 exhibits hydrophobic characteristics and could, at least theoretically, contain a signal for the addition of a GPI anchor. However, PLC treatment did not affect its surface association despite the fact that CD14 was released from THP-1 cells.
Moreover, any covalent binding of MMP-19 to the cell membrane appears impossible because treatment of the cell membranes with an alkaline high-salt buffer abolished the membrane association of MMP-19. The cell surface association of MMP-19 was also reduced after washing with acidic acetate buffer.
hMMP-19 appears to be bound via association with a cell surface
receptor-like molecule. Such an association has already been described
for several other MMP. For instance, gelatinase A (MMP-2) binds to the
vitronectin receptor integrin
v3 via its
hemopexin-like domain (14).
To determine which structural domain of MMP-19 is involved in the cell surface association, we prepared two mutant constructs of MMP-19. The construct without the catalytic domain was still detectable on the cell surface. There was only a slight decrease the MMP-19 abundance on the surface. The second construct was made by deletion of the whole hemopexin-like domain. This deletion had a dramatic effect on binding to the cell surface. hMMP-19 lacking the hemopexin-like domain could not be detected on the cell surface. This kind of association is reminiscent of that of MMP-2. Nevertheless, the receptor or the cell surface association partner remains to be identified. Its identification could help to understand the distinct migration capabilities of blood-derived cells.
Taken together the results presented in this study show that MMP-19 is expressed in myeloid cells and that its up-regulation occurs in an adhesion-dependent manner. The MMP-19 expressed by myeloid cells associates with their cell surface via the hemopexin-like domain. Nevertheless, the association of MMP-19 with the cell surface is not a general phenomenon, because fibroblasts and endothelial cells do not possess surface MMP-19 (our unpublished observation). Endothelial cells isolated from umbilical arteries accumulate MMP-19 in cellular protrusions (10).
However, the association of MMP-19 with cell surface may be critical for the spatial and temporal control of the activity of MMP-19, which in turn could affect the migratory behavior of myeloid cells, especially monocytes and macrophages.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Radislav Sedlacek, Institute of Biochemistry, University of Kiel, Olshausenstrasse 40, 24098 Kiel, Germany. E-mail address: Radek.Sedlacek{at}biochem.uni-kiel.de
3 Abbreviations used in this paper: MMP, matrix metalloproteinase; hMMP, human MMP; PLC, phospholipase C.
Received for publication July 30, 2001. Accepted for publication November 28, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
v3. Cell 85:683.[Medline]
-tocopherol. J. Nutr. 130:1649.-tocopherol inhibition of vascular smooth muscle cell proliferation occurs at physiological concentrations, correlates with protein kinase C inhibition, and is independent of its antioxidant properties. Proc. Natl. Acad. Sci. USA 92:12190. and IL-1 selectively induce expression of 92-kDa gelatinase by human macrophages. J. Immunol. 157:4159.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  H. Piccard, P. E. Van den Steen, and G. Opdenakker Hemopexin domains as multifunctional liganding modules in matrix metalloproteinases and other proteins J. Leukoc. Biol., April 1, 2007; 81(4): 870 - 892. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sadowski, S. Dietrich, F. Koschinsky, and R. Sedlacek Matrix Metalloproteinase 19 Regulates Insulin-like Growth Factor-mediated Proliferation, Migration, and Adhesion in Human Keratinocytes through Proteolysis of Insulin-like Growth Factor Binding Protein-3 Mol. Biol. Cell, November 1, 2003; 14(11): 4569 - 4580. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Mitola, M. Strasly, M. Prato, P. Ghia, and F. Bussolino IL-12 Regulates an Endothelial Cell-Lymphocyte Network: Effect on Metalloproteinase-9 Production J. Immunol., October 1, 2003; 171(7): 3725 - 3733. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. S. Laursen, M. T. Overgaard, K. Weyer, H. B. Boldt, P. Ebbesen, M. Christiansen, L. Sottrup-Jensen, L. C. Giudice, and C. Oxvig Cell Surface Targeting of Pregnancy-associated Plasma Protein A Proteolytic Activity. REVERSIBLE ADHESION IS MEDIATED BY TWO NEIGHBORING SHORT CONSENSUS REPEATS J. Biol. Chem., November 27, 2002; 277(49): 47225 - 47234. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Abs were used as
controls. B, The cell surface of THP-1 cells was
biotinylated and then lysed with RIPA buffer. MMP-19 labeled with
biotin was precipitated with the mAb CK8/4 and identified using
streptavidin-coupled peroxidase.
