|
|
||||||||
Beatrice and Samuel A. Seaver Laboratory, Department of Medicine, Cornell University Medical College, New York, NY 10021
| Abstract |
|---|
|
|
|---|
12,000 transcripts evaluated, 4 were differentially expressed. One
transcript represented secretory leukocyte protease inhibitor. In this
study, we report another differentially expressed transcript, mouse
matrix metalloprotease-9 (MMP-9). Like secretory leukocyte protease
inhibitor, MMP-9 was expressed constitutively in the
Lpsd macrophage cell line and not in the
Lpsn cell line. Similarly, two additional
macrophage cell lines that respond readily to LPS by producing nitric
oxide and TNF expressed no MMP-9 under LPS-free conditions. However, in
all four cell lines, LPS induced MMP-9 or augmented its expression. In
primary macrophages, concentrations of LPS in the ng/ml range augmented
the expression of MMP-9 mRNA. Paradoxically, macrophages from
Lpsd mice expressed more MMP-9 transcripts than
macrophages from Lpsn mice. In contrast, the
induction of TNF in response to LPS was much more pronounced in
Lpsn macrophages. The present findings with
MMP-9 suggest that homozygosity at Lpsd does not
so much prevent a response to LPS as dysregulate it, resulting in the
suppression of some LPS signaling pathways and the preservation of
others. | Introduction |
|---|
|
|
|---|
One cDNA identified by that analysis proved to be an LPS-inducible LPS
response inhibitor, novel characteristics of a previously identified
protein, secretory leukocyte protease inhibitor
(SLPI)4, 13 . Only 3 other
differentially expressed transcripts were detected among the
12,000
compared. One differentially expressed transcript encoded matrix
metalloproteinase-9 (MMP-9) (gelatinase), whereas the remaining two
corresponded to previously unidentified genes 14 . Characterization of
MMP-9 expression is reported in this study. LPS hyporesponsive C3H/HeJ
(Lpsd) primary macrophages responded to LPS by
increasing MMP-9 expression better than C3H/HeN
(Lpsn) primary macrophages. This paradoxical
LPS-normoresponsive phenotype in otherwise LPS-hyporesponsive cells
suggests that LPS responses must proceed down distinct pathways that
are regulated in different fashions by the Lps gene product.
| Materials and Methods |
|---|
|
|
|---|
Charles River Breeding Laboratories (Wilmington, MA) and the Jackson Laboratories (Bar Harbor, ME) supplied C3H/HeN (Lpsn) and C3H/HeJ (Lpsd) female mice, respectively.
Cells
ANA-1 cells were kindly provided by Dr. L. Varesio (National Cancer Institute, Frederick, MD), and HeNC2 (from C3H/HeN (Lpsn) mice) and GG2EE cells (from C3H/HeJ (Lpsd) mice) were provided by Dr. D. Radzioch (McGill University, Montreal, Canada). All are bone marrow-derived macrophage lines 15, 16 . The RAW 264.7 mouse macrophage cell line was from American Type Culture Collection (Manassas, VA). Primary macrophages were collected from the peritoneal cavity 4 days after i.p. injection of 2 ml of 4% Brewers thioglycollate broth (Difco, Detroit, MI). Cells were maintained in RPMI 1640 supplemented with 10% heat-inactivated FBS (HyClone, Logan, UT), 2 mM L-glutamine, 200 U/ml penicillin, and 200 µg/ml streptomycin at 37°C in 5% CO2/95% air. Complete culture medium was routinely monitored for LPS contamination by the chromogenic Limulus amebocyte lysate assay (BioWhittaker, Walkersville, MD) and found to contain <25 pg LPS/ml.
Differential display of mRNA
DNA-free total RNA from HeNC2 and GG2EE cells was used for differential display with a random 10-mer and T12M(C, G, T, A), in which M signifies a mixture of A, C, and G, as detailed elsewhere 13 . Differentially expressed products in the sequencing gel were extracted with H2O and reamplified twice by PCR using the same set of primers under the same conditions. The reamplified PCR product was gel purified using a Gel Extraction kit (Qiagen, Chatsworth, CA), subcloned into the PCR TA cloning vector (Invitrogen, San Diego, CA) and sequenced.
Northern blot analysis
Total RNA (20 or 25 µg/lane) from primary macrophages or cell lines was electrophoresed on a 1% agarose gel with 0.2 M 3[N-morpholino] propanesulfonic acid (pH 7.0), 0.5 M sodium acetate, 10 mM EDTA (1x MOPS), and 2% formaldehyde, confirming equal loading by means of staining with 2.5 µg/ml ethidium bromide. RNA was transferred in 20x SSC onto nylon membranes (NEN Research Products, Boston, MA). Membranes were hybridized for 18 h at 42°C with probe labeled (106 cpm/ml) with Priming-A-Gene kit from Promega (Madison, WI) in 5x SSC, 5x Denhart buffer, 50% formamide, plus 1% SDS and washed extensively before autoradiography. Control probes consisting of ß-actin or glyceraldehyde-3-phosphate dehydrogenase (G3PDH) cDNAs were amplified with ß-actin or G3PDH amplimers (Clontech Laboratories, Palo Alto, CA) from templates provided with the amplimers.
dsDNA sequencing
Subcloned DNA was sequenced by the dideoxynucleotide chain termination method using Sequence Version 2.0 DNA sequencing kit (United States Biochemical, Cleveland, OH) according to the manufacturers instructions.
Additional materials
LPS prepared from Escherichia coli 0111:B4 by phenol
extraction was from List Biological Laboratories (Campbell, CA) and
contained <0.8% protein; this LPS was used unless indicated
otherwise. LPS prepared from E. coli K235 by phenol-water
extraction was kindly provided by Dr. Stefanie Vogel (Uniformed
University, Bethesda, MD) with <0.008% protein. Other reagents were
mouse rIFN-
(1.1 mg protein/ml; sp. act. 5.2 x 106
U/mg; LPS content <10 pg/ml) from Genentech (South San Francisco, CA);
oligonucleotide primers from Oligos Etc. (Guilford, CT); reverse
transcription buffer and Moloney murine leukemia virus reverse
transcriptase from Life Technologies (Grand Island, NY); restriction
enzymes from New England Biolabs (Beverly, MA); AmpliTaq DNA
polymerase, dNTPs, and PCR buffer solutions from Perkin-Elmer Cetus
(Foster City, CA); guanidinium isothiocyanate, formaldehyde, and
formamide from Fluka Chemica-Biochemica (Ronkonkoma, NY); and all other
reagents from Sigma (St. Louis, MO). Plasmid DNA preparation columns
were from Qiagen, and tissue culture dishes were from Corning Glass
Works (Corning, NY).
| Results |
|---|
|
|
|---|
The identically derived HeNC2 (Lpsn) and GG2EE
(Lpsd) macrophage cell lines maintain the
LPS-responsive (HeNC2) or LPS-hyporesponsive (GG2EE) phenotypes of
their strains of origin as judged by LPS-induced nitric oxide and
TNF-
production 13 as well as the resultant cytotoxicity toward
tumor cells 16 . Differential display analysis of these two cell lines
began with reverse transcription of their mRNA with
oligodeoxythymidylate primers T12M(G, C, T, or A), in which
M connotes a mixture of A, C, and G. Specific cDNA fragments were
amplified by PCR with the same poly(dT) primer and 1 of 20 arbitrary
10-mers, whose binding positions in different transcripts were expected
to vary in distance from the poly(A) tail, producing a ladder when
amplified DNA was resolved on a polyacrylamide gel. From >12,000
resolved amplificands, 4 met the following criteria: in replicate
experiments, they were consistently detectable from only one cell line,
and they proved on subsequent analysis to represent distinct genes. One
of these, SLPI, was the subject of a previous report 13 .
The cDNA fragments designated clones 3 and 23 were amplified by
T12MA and GTTGGACCTA only from GG2EE
(Lpsd) and not from HeNC2 (Lpsn)
cells (Fig. 1
A). These two
fragments were excised from the polyacrylamide gel, reamplified,
purified from agarose gels, and used as probes in Northern blot
analyses. Clone 3 hybridized to mRNA from GG2EE cells
(Lpsd), but not from HeNC2 (Lpsn)
cells, whereas clone 23 identified two bands, one equally, and the
other differentially expressed (Fig. 1
B). This finding
suggested that at the position in which fragment 23 had been excised,
two different cDNA species had contributed amplificands of
indistinguishable size. Accordingly, cDNA fragments 3 and 23 were both
subcloned. Individual colonies of each were selected, whose DNA
hybridized selectively and at the same Mr with
mRNA from GG2EE cells (Lpsd) but not HeNC2 cells
(Lpsn). These subclones were sequenced and found to
be flanked by primer binding sites corresponding to those used in the
original RT-PCR analysis and to be identical with each other. The
subclones of clones 3 and 23 were completely concordant in sequence
with mouse MMP-9 (gelatinase B; type IV basement membrane collagenase)
(GenBank accession no. Z27231).
|
We compared the basal levels of MMP-9 transcript in HeNC2, GG2EE,
and two additional macrophage cell lines derived in a different manner
from non-C3H backgrounds, RAW 264.7 and ANA-1. Only the GG2EE cells
(Lpsd) expressed readily detectable MMP-9 mRNA under
LPS-free conditions (Fig. 2
). The three
Lpsn lines were all negative for basal MMP-9
expression.
|
13 , a cytokine whose administration can correct or complement
the LPS response defect in C3H/HeJ mice 17, 18, 19 , we tested the effects
of IFN-
. In LPS-responsive cells (HeNC2, RAW 264.7, and ANA-1),
MMP-9 mRNA was induced within 4 h of exposure to LPS (Fig. 2
(Fig. 2
to block LPS-induced expression of MMP-9
conforms to findings with rat microglia 21 . Superinduction of MMP-9 by LPS in primary macrophages from C3H/HeJ mice
Primary peritoneal macrophages elicited from both C3H/HeN
(Lpsn) and C3H/HeJ (Lpsd) mice by
i.p. injection of thioglycollate broth expressed a low level of MMP-9
(Fig. 3
). The difference in MMP-9
expression between these primary cells and bone marrow-derived cell
lines may reflect differences in their origins, differentiation states,
or past encounters. Expression of MMP-9 was augmented when these cells
were exposed to 100 ng/ml of LPS in vitro, a concentration typically
used in the study of LPS-inducible genes in C3H/HeN macrophages, such
as TNF and inducible nitric oxide synthase 13, 22 . Surprisingly,
up-regulation of MMP-9 was much more pronounced in macrophages from
C3H/HeJ (Lpsd) mice than those from C3H/HeN
(Lpsn) mice at both time points examined, 6 and
18 h (Fig. 3
A). At 6 h, densitometric analysis
indicated that 3.5 ± 0.4-fold (mean ± SEM,
n = 5 experiments) greater levels of MMP-9 mRNA
accumulated in primary macrophages from C3H/HeJ
(Lpsd) mice compared with those from C3H/HeN
(Lpsn) mice.
|
Differential regulation of MMP-9 and TNF in macrophages from C3H/HeN (Lpsn) and C3H/HeJ (Lpsd) mice
To check whether this phenomenon was relatively specific for MMP-9
induction, we next compared MMP-9 and TNF induction in response to LPS.
For this experiment, a different preparation of LPS from
E. coli K235 that contained <0.008% protein was used to
reassure us that what we observed was not due to protein contamination
in the commercial LPS preparation. Thioglycollate broth-elicited
macrophages from C3H/HeJ (Lpsd) and C3H/HeN
(Lpsn) mice were exposed to concentrations of this
LPS ranging from 1 ng/ml to 10 µg/ml for 6 h, at which point the
induction of MMP-9 and TNF transcripts was examined on the same
membrane. As shown in Fig. 4
, macrophages
from C3H/HeJ mice were hyporesponsive to LPS in that they expressed
little TNF mRNA, whereas macrophages from C3H/NeN mice expressed
abundant TNF mRNA in response to LPS. However, induction of MMP-9 by
LPS demonstrated an opposite pattern in these two cells: macrophages
from C3H/NeJ mice expressed MMP-9 mRNA in response to LPS, whereas no
MMP-9 mRNA was detectable in LPS-treated macrophages from C3H/HeN mice
(Lpsn). This result rules out the possibility that
the primary macrophages from C3H/HeJ (Lpsd) mice may
have been activated in vivo 17, 24, 25 and points to differential LPS
responsiveness in C3H/HeJ cells.
|
| Discussion |
|---|
|
|
|---|
, cells
from these mice can respond partially, slowly, or weakly to LPS by
several criteria, including activation of NF-
B 26 and expression
of MnSOD 27 or SLPI 13 . However, the present report may be the
first to identify an LPS response in C3H/HeJ cells that is more robust
than that in C3H/HeN cells under the same conditions. This response is
unlikely to be due to a protein impurity within the LPS preparation
because the induction is polymixin B inhibitable and can be repeated
with a relatively clean LPS preparation (Fig. 4
(Fig. 4There are other metalloproteinases within macrophages that can be regulated by LPS. Duc Dodon and Vogel 28 reported that LPS induced an inhibition in secretory elastase activity from thioglycollate-elicited macrophages. In contrast to the present report, this inhibition was only seen in LPS-responder macrophages and not in macrophages from C3H/HeJ mice.
It is of interest that LPS induces macrophages to release both a
protease, MMP-9, and a protease inhibitor, SLPI. Zhang et al.
29 reported that SLPI interfered with the production of MMP-9
by human monocytes. However, such interference seems unlikely in the
mouse macrophages studied here in which the expression of both SLPI and
MMP-9 rose and fell in parallel, rather than reciprocally, upon
exposure to LPS or IFN-
, respectively. Although SLPI functions as an
LPS-induced LPS-response inhibitor 13 , the role of MMP-9 in the LPS
response is unknown. A proinflammatory contribution is plausible,
considering that MMPs promote cell migration in inflammation 30, 31, 32 ,
tissue remodeling 33 , and wound healing 34 , as well as tumor
invasion 35, 36, 37 , and thus may permit macrophages to migrate toward
sources of LPS and help destroy infected tissue. Alternatively, MMP-9,
like SLPI, may inhibit LPS responses. This speculation is based on the
LPS-induced expression of MMP-9 in C3H/HeJ cells, most of whose LPS
responses are blunted. MMP-9 may degrade extracellular matrix proteins
whose integrin-mediated signals serve as cofactors for macrophage
activation by LPS 38, 39 .
The location of MMP-9 on mouse chromosome 2 (mouse genome, The Jackson Laboratory; http://mgd.hgmp.mrc.ac.uk) excludes its candidacy as the Lps gene on chromosome 4. Whether differential display analysis could yield an immediate candidate for the Lps gene remains unanswered. The inherent shortcomings of the method include that mutations in expressed genes need not lead to striking changes in the levels of their transcripts. Advantages of differential display include its sensitivity and its ability to detect disparities among any number of reference samples, disparities that reflect either silencing or overexpression of a given gene. The even-handedness of the method with respect to the populations under comparison was underscored by our identification of two transcripts (SLPI and MMP-9) that were overexpressed in LPS-hyporesponsive cells and an equal number that were overexpressed in LPS-normoresponsive cells. Microarray assays and/or sequential analysis of gene expression 40 are likely to overshadow differential display because of their greater speed of throughput but may suffer some of the same limitations.
Note added in proof. After this paper was accepted, the lps gene was identified as TLR4 41, 42 . Questions raised here and elsewhere about LPS signaling 43 remain.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Current address: Millennium Pharmaceuticals, 640 Memorial Drive, Cambridge, MA 02139-4815. ![]()
3 Address correspondence and reprint requests to Dr. Aihao Ding, Box 57, Weill Medical College of Cornell University, New York, NY 10021. E-mail address: ![]()
4 Abbreviations used in this paper: SLPI, secretory leukocyte protease inhibitor; MMP-9, matrix metalloproteinase-9; G3PDH, glyceraldehyde-3-phosphate dehydrogenase. ![]()
Received for publication June 12, 1998. Accepted for publication December 7, 1998.
| References |
|---|
|
|
|---|
or a calcium ionophore. Jpn. J. Cancer Res. 78:279.[Medline]
B. J. Leukocyte Biol. 57:174.[Abstract]
This article has been cited by other articles:
![]() |
H. Stapel, S.-C. Kim, S. Osterkamp, P. Knuefermann, A. Hoeft, R. Meyer, C. Grohe, and G. Baumgarten Toll-like receptor 4 modulates myocardial ischaemia-reperfusion injury: Role of matrix metalloproteinases Eur J Heart Fail, November 1, 2006; 8(7): 665 - 672. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Pedron, R. Girard, and R. Chaby TLR4-dependent Lipopolysaccharide-induced Shedding of Tumor Necrosis Factor Receptors in Mouse Bone Marrow Granulocytes J. Biol. Chem., May 30, 2003; 278(23): 20555 - 20564. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Min, A. G. Moore, M. A. Bain, S. N. Breit, and J. G. Lyons Activation of Macrophage Promatrix Metalloproteinase-9 by Lipopolysaccharide-Associated Proteinases J. Immunol., March 1, 2002; 168(5): 2449 - 2455. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. G. Lei and D. C. Morrison Differential Expression of Caveolin-1 in Lipopolysaccharide-Activated Murine Macrophages Infect. Immun., September 1, 2000; 68(9): 5084 - 5089. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Pagenstecher, A. K. Stalder, C. L. Kincaid, B. Volk, and I. L. Campbell Regulation of Matrix Metalloproteinases and Their Inhibitor Genes in Lipopolysaccharide-Induced Endotoxemia in Mice Am. J. Pathol., July 1, 2000; 157(1): 197 - 210. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Ohashi, V. Burkart, S. Flohe, and H. Kolb Cutting Edge: Heat Shock Protein 60 Is a Putative Endogenous Ligand of the Toll-Like Receptor-4 Complex J. Immunol., January 15, 2000; 164(2): 558 - 561. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Ding, N. Thieblemont, J. Zhu, F. Jin, J. Zhang, and S. Wright Secretory Leukocyte Protease Inhibitor Interferes with Uptake of Lipopolysaccharide by Macrophages Infect. Immun., September 1, 1999; 67(9): 4485 - 4489. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. M. C. Wong, Q. Yuan, H. Chen, B. M. Sultzer, and S.-W. Chung A Single Point Mutation at the 3'-Untranslated Region of Ran mRNA Leads to Profound Changes in Lipopolysaccharide Endotoxin-mediated Responses J. Biol. Chem., August 24, 2001; 276(35): 33129 - 33138. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |