* Human Health Research Center, INRS-Institut Armand-Frappier, University of Quebec, Laval, Québec, Canada; and
Department of Pathology, University of Calgary and Calgary Laboratory Services, Calgary, Alberta, Canada
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: ICAM-1 • MMP-9 • cytokines • B cells • IL-6 • IL-13
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The expression of MMP by tumor cells at late stages of metastases is thought to occur during contact with the vascular endothelium via cell-adhesion molecules [4 5 6 ]. Our recent finding that the vascular endothelial cell-adhesion molecule intercellular adhesion molecule-1 (ICAM-1) controls the production of MMP-9 in lymphoma cells [4 ] and that ICAM-1-deficient mice are resistant to the spread of lymphoma [7 ] indicates a strong relationship between adhesion molecules and extracellular proteases. However, the MMPs are not the only extracellular proteases that could be involved in lymphomas. Cathepsins and leukocyte elastase (LE), two extracellular proteases expressed in normal leukocytes, are among the key candidates that could modulate the spread of lymphoma cells. Although the pathophysiological role of these enzymes, most notably LE, has been well documented in pulmonary disorders such as emphysema and cystic fibrosis [8 ], any participation of LE in cancer has never been established.
Several studies support the hypothesis that LE could be involved in metastases, most notably during the spread of lymphoma cells to peripheral organs. Not only can LE cleave cell-surface ICAM-1 [9 ], it can also directly influence the architecture of target organs by degrading components of the extracellular matrix. Kossakowska et al. [10 ] have demonstrated that cells isolated from aggressive NHL, but not those isolated from hyperplastic tonsils, had an increased ability to penetrate a physical barrier constituted of elastin. LE could also indirectly affect remodeling of the extracellular matrix at late stages of metastasis by activating MMP-9 [11 ] or by degrading TIMP-1 complexed with active MMP-9 [12 ]. LE can also degrade interleukin (IL)-2 to yield IL-2 polypeptides that inhibit IL-2-induced release of lymphoid cells from laminin, collagen IV, and fibronectin [13 ]. Moreover, the impact of LE in lymphomas could go well beyond its ability to enhance the migration of tumor cells through the extracellular matrix; apart from cleaving ICAM-1, LE has also been shown to cleave many regulators of the anti-tumoral response, such as immunoglobulin G (IgG), CD4, and CD8 [14 ]. However, the fact that aggressive lymphomas express high levels of MMP-9, an enzyme with an intrinsic elastolytic activity [15 ], implies that the ability of lymphoma cells to penetrate through the elastin barrier may not necessarily be mediated through the secretion of LE. Thus, in the present work, we sought to determine if LE is expressed in lymphoma cells and tissues and how this expression is regulated.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reagents and antibodies (Abs)
Purified, human LE (HLE) was obtained from Calbiochem (La Jolla,
CA). Sputum samples of patients with cystic fibrosis containing high
concentrations of HLE were also used as a source of HLE for some
experiments. The samples were prepared as described previously
[9
]. The rat anti-mouse ICAM-1 monoclonal antibody (mAb)
YN-1 was purified from hybridoma culture supernatants by chromatography
on protein G-Sepharose (Pharmacia Fine Chemicals, Piscataway, NJ) using
standard protocols. Although the epitope recognized by this antibody is
yet unknown, it is likely located in or nearby the first domain of
ICAM-1, because this antibody can inhibit lymphocyte
function-associated antigen-1 (LFA-1)-mediated intercellular adhesion
[18
]. Detection of YN-1 binding was carried out
using a phycoerythrin (PE)-conjugated, donkey anti-rat antibody
(Jackson ImmunoResearch Laboratory, Mississsauga, Ontario, Canada). The
biotinylated 3E2 anti-mouse ICAM-1 mAb was obtained from Pharmingen
(San Diego, CA). Mouse monoclonal IgG1 specific for HLE was obtained
from Calbiochem. Biotinylated anti-mouse IgG (Fc-specific) was obtained
from Sigma Chemical Co. (St. Louis, MO). Streptavidin (SA)-PE
was obtained from Beckman-Coulter (Ville St-Laurent, Quebec, Canada).
Recombinant cytokines IL-3, IL-5, IL-6, IL-11, and IL-13 were obtained
from R&D Systems (Minneapolis, MN). pBK-RSV vector was obtained from
Stratagene (La Jolla, CA).
Culture of cell lines
The mouse T lymphoma cell lines 267 and 718 were established in
our laboratory from radiation-induced T-cell lymphomas in C57BL/Ka- and
ICAM-1-deficient mice, respectively, as described previously
[3
]. Consequently, the 267 cells expressed the common
form of ICAM-1 on its surface, and the 718 cells only expressed a low
but detectable level of ICAM-1 isoforms ([3
] and
unpublished results). Both cell lines expressed surface T-cell
receptors, CD90, CD102 (ICAM-2), and CD31. The human Burkitt’s
lymphoma Raji cell line was obtained from the American Type Culture
Collection (Manassas, VA). All cells were maintained in culture using
RPMI 1640 medium supplemented with 10% fetal calf serum (FCS), 2 mM
glutamine, 10 mM HEPES, and antibiotics (complete medium).
Transfection of ICAM-1 in 718 lymphoma cells
To restore the expression of ICAM-1 in 718 cells, the cDNA
encoding the common form of murine ICAM-1 was subcloned into the SR
eukaryotic expression vector (kindly provided by Dr. François
Denis, INRS-Institut Armand-Frappier, Quebec, Canada). The 718 lymphoma
cells were transfected using Superfect (Stratagene) according to the
manufacturer’s instructions. Stable cell lines expressing distinct
ICAM-1 levels were established using puromycin selection. The ICAM-1
expression was assessed by flow cytometry using the biotinylated 3E2
anti-ICAM-1 mAb and SA-PE.
Binding of LE to the surface of lymphoma cells
Aliquots of 3–4 x 105 lymphoma cells were
transferred to 96-well plates and washed twice with serum-free RPMI
medium. The cells were then incubated at 37°C with the indicated
concentrations of purified recombinant HLE or dilutions of
LE-containing human sputum. After washing with phosphate-buffered
saline (PBS), LE binding was measured by flow cytometry using
LE-specific antibodies, biotinylated anti-mouse IgG, and SA-PE.
Controls included cells incubated sequentially with LE, with the
anti-mouse IgG, and with SA-PE.
Flow cytometry
Cells were stained at 4°C and washed with PBS containing 0.5%
bovine serum albumin (BSA) and 0.2% sodium azide (PBA). Prior to
staining, cells were incubated with 10 µg/ml human IgG (Sigma) for 20
min at 4°C to block nonspecific binding. Fluorochrome- or
biotin-labeled mAbs were then added at appropriate concentrations and
incubated for another 20 min. Cells were then washed four times with
PBA. For indirect staining with SA-Red 670 (Gibco-BRL, Grand Island,
NY), cells were washed three times following the reaction with the
first mAb and then incubated 20 min on ice with the fluorescent
conjugate. Results shown are representative of at least three
independent experiments. Flow cytometric analyses were performed on
1–5 x 105 events using a Coulter XL-MCLTM flow
cytometer (Coulter Electronics, Hialeah, FL) and recorded on a
logarithmic scale.
Stimulation of Raji cells
Raji cells were cultured at 107/ml in serum-free
RPMI medium for 12 h in the presence or absence of phorbol
12-myristate 13-acetate (PMA; Sigma), 50 nM; db-cAMP (Sigma), 1 mM; or
with recombinant cytokines, i.e., IL-3, 10 ng/ml; IL-5, 10 ng/ml; IL-6,
100 ng/ml; IL-11, 10 ng/ml; and IL-13, 100 ng/ml. Cells were then
centrifuged, and the pellets were frozen for RNA extraction.
Lymphoma specimen collection
NHL tissue samples in excess of that needed for diagnostic
purposes were received in the Department of Pathology at Foothills
Hospital (Calgary, Alberta, Canada), snap-frozen in liquid nitrogen,
and kept at -70°C. The cases were classified according to the REAL
Working Formulation [19
]. All except Case 11 were of
B-cell origin. Case 11 represents a large T-cell lymphoma. Cases 6, 8,
13, 14, 16, and 18–32 were large B-cell, aggressive lymphomas. Cases 1
and 10 were small lymphocytic, indolent lymphomas. Cases 2 and 3 were
low-grade, follicular lymphomas. Case 4 was a mantle cell lymphoma.
Cases 12 and 15 represented angioimmunoblastic lymphadenopathy, which
are not lymphomas. RNA extraction and reverse transcription (RT) were
carried out as described previously using the acid guanidium
thiocyanate-phenol-chloroform-extraction method and the Superscript RT
RNAse H-RT (Gibco-BRL), as described previously [10
].
RNA isolation and analysis
Unless otherwise indicated, total cellular RNA was isolated
using the Trizol reagent (Gibco-BRL) according to the manufacturer’s
instructions. First-strand cDNA was prepared from total cellular RNA
using the Superscript RT RNAse H-RT. After RT, polymerase chain
reaction (PCR) amplification was carried out using the PCR core kit
(Boehringer Mannheim, Laval, Quebec, Canada) with specific primers for
LE, human MMP-9 [10
], IL-13 [20
],
ß-actin (Stratagene), and glyceraldehyde 3-phosphate dehydrogenase
(GAPDH; kindly provided by Dr. Daniel Oth, INRS-Institut
Armand-Frappier; see Table 1
). The design of human, LE-specific primer pairs was carried out
using DNA sequences obtained from Genbank (National Center for
Biotechnology Information, Bethesda, MD) and was chosen to have
approximately 50% GC content. The sense and anti-sense primers
were located in exons III and V, respectively. Thirty cycles of
amplification were performed in an MJ Research Thermal Cycler (model
PTC-100TM). Each cycle of amplification consisted of a denaturating
step at 94°C for 1 min, a primer annealing step at 60°C for 2 min,
and a strand-elongation step at 72°C for 1 min. The amplification for
each gene was in the linear curve. The reaction mixture (5–15 µl)
was size-separated on a 1.5% agarose gel, and specifically amplified
products were detected by ethidium-bromide staining and ultraviolet
transillumination. Loading was equalized to the internal control mRNA
(GAPDH or ß-actin). Semi-quantitative analysis was carried out using
a computerized densitometric imager (model GS-670, BioRad, Mississauga,
Ontario, Canada).
|
View this table: [in a new window] |
Table 1. Sequence of the Oligoprimers Used for the Detection of MMP, LE, and
IL-13 in NHL
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (22K): [in a new window] |
Figure 1. Binding of LE to lymphoma cells. (A) Dose-dependent binding of
recombinant HLE to lymphoma cells. Lymphoma cells (267) were incubated
with the indicated concentrations of purified LE at 37°C for 30 min,
and binding of LE was detected by flow cytometry using antibodies to
LE. (B) Time-dependent binding of LE to 267 lymphoma cells. Lymphoma
cells were incubated with 5 µg recombinant LE for the indicated
periods of time, and binding of LE was detected by flow cytometry using
antibodies to LE. (C) Release of active recombinant LE in the
supernatant after binding to lymphoma cells. Lymphoma cells were
incubated with 5 µg recombinant LE for the indicated periods
of time. LE activity ( ) in the supernatants was measured by
fluorescent-activated substrate conversion. Binding of LE ( ) was
measured as described. Data are representative of at least three
separate experiments.
|
 View larger version (23K): [in a new window] |
Figure 2. Correlation between LE binding and ICAM-1 expression in lymphoma cells.
ICAM-1-deficient 718 cells were transfected with cDNA encoding murine
ICAM-1. Stable clones expressing different levels of ICAM-1 (lower
panel) were then characterized for their ability to bind LE (upper
panel). Binding was carried out at 37°C for 30 min using recombinant
LE and determined by flow cytometry using anti-LE antibodies. Results
represent the mean fluorescent intensity (MFI) measured on 5000 cells.
Results are representative of at least two independent experiments.
|
 View larger version (48K): [in a new window] |
Figure 3. Up-regulation of LE in Raji cells upon exposure to IL-6 and IL-13. Raji
cells were stimulated with various cytokines, PMA (50 nm), or db-cAMP
(1 mM) for 8 h. LE and MMP-9 mRNA expression was then measured by
RT-PCR. Results are representative of at least three independent
experiments.
|
 View larger version (26K): [in a new window] |
Figure 4. Expression of IL-13 and LE in NHL lymphoma tissues. Semi-quantitative
analysis of the expression of human IL-13 and LE was measured by RT-PCR
of RNA extracted from lymphoma tissues collected from 20 different
cases of NHL. Results are representative of three independent
experiments.
|
We next investigated the expression of LE at the protein level in lymphoma tissue. Figure 5A shows a low power view of a large cell, peripheral T-cell NHL, where LE expression was found to be present within the tumor. The blue area represents a residual, normal lymphoid follicle where no staining of LE could be detected. Figure 5B shows high magnification of the cells expressing HLE. Small (reactive) lymphocytes do not appear to express HLE.
 View larger version (73K): [in a new window] |
Figure 5. HLE expression in peripheral, large T-cell NHL. (A) Section of a lymph
node shows partial replacement by peripheral, large T-cell NHL.
Anti-HLE antibodies bind to macrophages and polymorphonuclear (pmn)
leukocytes within the tumor. Central area represents residual lymphoid
follicle and shows no evidence of HLE expression. Original
magnification x40. (B) Section from the tissue adjacent to the NHL
described in A, stained with anti-HLE antibodies. Macrophages and pmn
outside the tumor express HLE. Original magnification
x100.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To our knowledge, the present study is the first to directly address the expression of LE in tumors. The importance of extracellular proteases, such MMPs, has been well-recognized at different stages of tumor development or during the spread of tumor cells to the periphery. A first indication that elastolytic activity plays a role in lymphoma was demonstrated in a previous study showing that lymphoma cells had the ability to penetrate through an elastin-rich matrix in vitro, suggesting that lymphoma cells possess an intrinsic elastolytic activity [10 ]. We propose that this elastolytic activity results from the up-regulation of LE expression by the lymphoma cells. Our in vitro data clearly established that lymphoma cells have the ability to express LE upon exposure to specific cytokines, such as IL-6 and IL-13, which we found to be expressed in NHL. Alternatively, lymphoma cells may use their ability to recruit, via ICAM-1, soluble forms of LE released in the extracellular space by peritumoral cells, most notably, those of the monocytic lineage, such as macrophages or lymphocytes. Indeed, there is increasing evidence that peritumoral cells contribute significantly to the local production of extracellular proteases, thereby favoring tumor growth [25 26 27 ]. Our in vivo data showing that LE was expressed by infiltrating leukocytic cells surrounding the lymphoid tumor are consistent with this possibility. Notwithstanding these observations, we believe that the biological impact of the presence of LE within the tumor is independent of its cellular origin. The concept that extracellular proteases play an important part in the remodeling of the tissue architecture, regardless of their cellular origin, has been well-demonstrated by the studies of the late Paul Basset [25 ].
We found that IL-13 increased the mRNA levels of LE and MMP-9 genes in Raji cells. Such a concomitant expression of both of these proteases was also observed upon exposure to IL-6, suggesting that LE and MMP-9 genes could share common, transcriptional, and regulatory elements in their promoters. One candidate is PU.1, a transcription factor of the ets family present in B cells and macrophages that have been shown to regulate the expression of LE and MMP-9 in monocytes and lymphocytes [28 , 29 ]. PU.1 involvement is supported by the fact that IL-13-mediated signals in B lymphocytes, such as those leading to the Ig heavy-chain class switching to IgE, involve PU.1 in cooperation with STAT6 [30 ]. However, there is also an argument against this possibility, namely the recent observation that PU.1 is part of a silencing sequence located in the promoter of gelatinase A [31 ]. This enzyme, a member of the gelatinase subfamily of MMPs with structural and functional homology with MMP-9, has also been implicated indirectly in the invasive behavior of lymphomas [10 ]. Furthermore, the fact that intracellular signals mediated through the IL-6 receptor seem to be intact in PU.1-deficient mice also suggests that other pathways may be involved in the activation of LE by these cytokines. A second candidate is the GABP, a heteromeric, transcription-factor complex that consists of GABP alpha, an ets factor, and GABP beta, a Notch-related protein. The binding of the GABP protein to the LE promoter ets site strongly activates its expression alone and in cooperation with the transcription factors c-Myb and C/EBP [32 ]. Future experiments using reporter constructs containing promoter elements of MMP-9 and LE genes will be needed to answer this question.
We found no direct correlation between the expression of IL-13 and that of LE. This was expected in cases where LE expression was not accompanied by IL-13 expression, because LE could also be up-regulated upon stimulation by other cytokines often expressed in malignant lymphomas, such as IL-6 [10 ]. In cases where we found no detectable LE expression in the presence of IL-13, we could envisage that up-regulation of LE by IL-13 was inefficient because of inappropriate numbers of functional, cell-surface receptors at that stage of the tumor development, as discussed above. However, IL-13 could induce the secretion of LE by peritumoral cells, which can be recruited at the surface of lymphoma cells via their receptors, most notably ICAM-1. If so, it is likely that the expression of elastolytic activity would not be restricted to lymphomas but could also be implicated in other types of cancer associated with a density of ICAM-1 adhesion molecules at their surface, such as melanoma [33 ]. Although ICAM-1 has been shown to act as an accessory molecule in the killing of tumor cells by cytolytic T lymphocytes (CTL) or natural killing (NK) [34 35 36 ], our data could provide an alternative explanation for the association of ICAM-1 with poor prognosis in stage I tumors and for inhibition of experimental metastasis by melanomas by interfering with its expression [34 ].
Our study is the first to show the up-regulation of LE expression upon exposure to IL-13 or IL-6. It is interesting that there has been an increasing number of studies pointing out a possible role of IL-13 in tumor development, most notably lymphomas. Using cDNA microarrays, Kapp et al. [24 ] found that IL-13 was highly expressed in Hodgkin’s disease (HD)-derived cell lines. Although the authors concluded that the expression of IL-13 seemed to be restricted to HD lymphomas, our results clearly show that IL-13 can also be expressed in NHL cases, including T-cell lymphomas. The same study by Kapp et al. showed that in situ hybridization of lymph-node tissue from HD patients demonstrated that elevated levels of IL-13 were expressed specifically by Hodgkin’s/Reed-Sternberg (H/RS) tumor cells and that neutralizing antibodies to IL-13 inhibited the proliferation of these cells, indicating that IL-13 could have multiple role in lymphomas. Moreover, Riemann et al. [37 ] had shown previously that IL-13 up-regulated protein expression as well as enzymatic activity of aminopeptidase N and dipeptidylpeptidase IV, two transmembrane type II molecules thought to be involved in metastasis. Experimental studies using gene-transfer strategies or genetically engineered mouse models should help to determine the molecular mechanisms controlled via IL-13 that contribute to the aggressiveness of lymphoma cells.
We have clearly established that LE can be expressed by lymphoma cells or can be recruited to its surface via ICAM-1 when found in its soluble form. The exact binding site(s) on ICAM-1 is unknown. The common form of ICAM-1 is a large molecule containing five Ig-like domains. That YN-1 antibody did not interfere with binding of LE on ICAM-1 is not surprising; in fact, it is likely that LE binds to at least two distinct binding sites, because cleavage of purified ICAM-1 generates several fragments [9 ]. The molecular mechanism by which LE could modulate the development of lymphoma is as yet unknown but is likely to involve local extracellular matrix remodeling and/or protection from the anti-tumoral response. Although constitutive expression of high levels of LE in murine lymphoma cells is difficult to obtain, experiments using gene transfection could help to resolve that question (P. De Noncourt, Y. St-Pierre, unpublished observations). However, so far our experiments using lymphomas transfected with a cDNA encoding LE have shown that transfection of LE in lymphoma cells did not have any significant effect on their growth after their intrathymic injection (unpublished results), as we had previously observed in the case of MMP-9 transfection [3 ]. This would suggest that the biological role of LE in lymphomas could be distinct from that of MMP-9 and that the elastolytic activity of MMP-9 is different from that of LE. This would be consistent with our previous observations showing that only LE, not MMP-9, can cleave ICAM-1 even at high concentrations [9 ]. Because LE can bind to ICAM-1 and cleave several key receptors involved in the recognition of tumor cells, it will be important to determine whether increased binding of LE on lymphoma cells can alter their recognition by anti-tumoral cells. The use of genetically engineered mouse models with altered LE gene expression will help to better understand its role in lymphomagenesis.
The localization of LE in lymphoma cells stimulated with IL-6 or IL-13 is yet unknown. We hypothesize that LE can be stored in an endosomal compartment, associated with the plasma membrane, or secreted. Whether this localization depends on the stimulus will require an extensive investigation using confocal microscopy. Moreover, phylogenetic studies have shown that LE is a member of a separate branch (Class 6) of serine proteases that is expressed by cells of the immune system and that clearly differs from that of well-known digestive enzymes such as trypsin, chymotrypsin, and pancreatic elastase [38 ]. Cytogenetic-positioning studies have shown that Class 6 enzymes are disseminated in three main, multi-gene clusters. As for the LE gene, it is colocalized with azurocidin and proteinase 3 within an approximately 50-kb cluster on human chromosome 19, sharing strong homology in terms of exon-intron organization and transcriptional orientation [39 ]. Because all three genes are often expressed coordinately, it will be interesting to test the role of other members of the Class 6 enzymes in the development of lymphoid and myeloid neoplasia.
Received January 13, 2001; revised April 23, 2001; accepted May 23, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. P. Holt, L. Cheng, and C. Ju Identification and characterization of infiltrating macrophages in acetaminophen-induced liver injury J. Leukoc. Biol., December 1, 2008; 84(6): 1410 - 1421. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. G. Sandler, M. M. Mentink-Kane, A. W. Cheever, and T. A. Wynn Global Gene Expression Profiles During Acute Pathogen-Induced Pulmonary Inflammation Reveal Divergent Roles for Th1 and Th2 Responses in Tissue Repair J. Immunol., October 1, 2003; 171(7): 3655 - 3667. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
