|
|
||||||||

*
Japanese Red Cross, Hokkaido Red Cross Blood Center, Yamanote, Nishi-ku, Sapporo, Japan; and
Department of Microbiology, Molecular Genetics and Immunology, University of Kansas Medical Center, Kansas City, KS 66160
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
LPS is a major component of the outer membrane of Gram-negative bacteria and is a potent activator of macrophage functions (15, 16). LPS is also a potent stimulant of junB expression, resulting in the up-regulation of AP-1 element binding activity (17, 18, 19). One of the initial events in LPS-triggered macrophage activation, which was suggested to mediate some of the downstream responses to LPS (20, 21, 22), is protein tyrosine phosphorylation. Since the members of the mitogen-activated protein (MAP) kinase family (p38, p42/extracellular signal-regulated kinase-2 (ERK2), p44/ERK1, and c-Jun NH2-terminal kinase (JNK)) are predominantly tyrosine phosphorylated in LPS-stimulated macrophages (23, 24, 25, 26), a MAP kinase pathway has been implied to be involved in LPS-induced cellular responses (19, 23, 27).
Initial exposure of various monocytes, macrophages, or their cell lines to a submaximal dose of LPS resulted in a state of refractoriness to the secondary maximal dose of LPS exposure (28, 29, 30, 31, 32, 33, 34). The macrophages in an LPS-tolerant state typically respond to a secondary LPS stimulation with a much lesser level of response (such as LPS-induced cytokine production) than the level induced by an initial stimulation. In vitro models for studying endotoxin tolerance in macrophage or monocytic cell lines have been developed (28, 29, 30, 31, 32, 33, 34, 35), although none of these models completely mimicks in vivo endotoxin tolerance. However, studying such systems provides valuable information regarding the biochemical signal pathway of endotoxin response (28, 29, 30, 31, 32, 33, 34, 35). In this paper the molecular basis of in vitro LPS tolerance was examined using LPS-induced expression of the junB gene in a mouse macrophage cell line, P388D1, because the LPS-triggered junB gene expression in P388D1 cells was found to be transient (17). The results presented show that LPS-induced desensitization of junB gene expression 1) occurs at or upstream of the level of gene transcription, and 2) may be involved in the defective LPS-induced p38 MAP kinase pathway.
| Materials and Methods |
|---|
|
|
|---|
The mouse macrophage-like cell line, P388D1, which was isolated from a methylcholanthrene-induced lymphoid neoplasm of a DBA/2 mouse and has been shown to possess characteristics typical of macrophages, was obtained from the Japanese Cancer Research Resources Bank. P388D1 cells were grown in RPMI 1640 medium containing 10% heat-inactivated FBS supplemented with 100 U/ml penicillin and 100 µg/ml streptomycin. Herbimycin A and PB202190 were purchased from Calbiochem-Novabiochem (La Jolla, CA).
Northern blot analysis
Northern blot analysis was conducted as previously described
(17). P388D1 cells (1 x 107/experiment) were treated
as indicated in the text and were then washed with cold PBS, quickly
frozen, and stored at -80°C until use. Total cellular RNA was
extracted from the cells using the guanidinium thiocyanate procedure
(36). The isolated RNA (10 µg) was electrophoresed in 1% agarose
gels containing 2.2 M formaldehyde and then transferred to Nytran nylon
membranes (Schleicher and Schuell, Dassel, Germany) by capillary action
using 10x SSC. The membranes were prehybridized for at least 2 h
at 42°C in a solution containing 50% (v/v) formamide, 5x SSPE, 5x
Denhardts solution, 0.5% SDS, 0.01 M EDTA, and 100 µg/ml salmon
sperm DNA. Membranes were then incubated for 20 h at 42°C in the
same solution with 32P-labeled cDNA probes specific for
junB (3) previously labeled by the random hexamer priming
method using [
-32P]dCTP (Amersham, Aylesbury, U.K.).
Membranes were then washed twice at 25°C in 0.1% SDS and 2x SSPE
for 30 min and twice at 42°C in 0.1% SDS and 0.1x SSPE, and
autoradiographed on Kodak XAR-5 film (Eastman Kodak, Rochester, NY) at
-80°C with intensifying screens. Membranes were then stripped and
reprobed with 32P-labeled ß-actin cDNA (37).
Nuclear run-on assay
P388D1 cells (
5 x 107/experiment) were
treated with or without LPS (1 µg/ml) for 1 h. After washing
twice with LPS-free medium, cells were incubated for 2 h in
LPS-free medium, then stimulated with or without LPS (1 µg/ml) for
1 h at 37°C. Isolation of nuclei and nuclear run-on
transcription assays were performed as previously described (38). The
plasmids used in these experiments included pGEM-2 vector alone;
pGEM-2, a vector containing a 1.0-kb cDNA insert of mouse
junB (3), pHA4.1, a vector containing a 3.5-kb cDNA insert
of human ß-actin (37); pG3PDH, a vector containing a 1.2-kb cDNA
insert of human glyceraldehyde-3-phosphate dehydrogenase (39); and
pMuTNF, a vector containing a 1.0-kb cDNA insert of mouse TNF-
.
mRNA stability analysis
P388D1 cells (1 x 107/experiment) were treated with or without LPS (1 µg/ml) for 1 h. After washing twice with LPS-free medium, cells were first incubated for 2 h in LPS-free medium, then stimulated with LPS (1 µg/ml) for 1 h at 37°C. At the end of the incubation period, actinomycin D was added at a final concentration of 5 µg/ml. At various times after the addition of actinomycin D, total RNA was isolated, and 10 µg of each sample was subjected to Northern blot analysis.
Western blot analysis
The cells treated with LPS were rapidly chilled on ice, washed twice with ice-cold PBS containing 2 mM sodium orthovanadate (Na3VO4), and then lysed in 1 ml of lysis buffer (1% Triton X-100, 20 mM Tris-HCl (pH 8.0), 137 mM NaCl, 10% glycerol, 2 mM Na3VO4, 10 mM NaF, 2 mM PMSF, 100 µg/ml leupeptin, and 100 µg/ml aprotinin) for 15 min on ice. The lysate was centrifuged (10,000 x g, 15 min, 4°C), and the clear supernatant was subjected to electrophoresis in 5 to 20% SDS-polyacrylamide gels. The proteins separated by SDS-PAGE were transferred to a nitrocellulose membrane (Schleicher and Schuell) using an electrophoretic transfer cell. Subsequently, the nitrocellulose membrane was blocked with TBST (10 mM Tris, 150 mM NaCl, and 0.05% Tween-20, pH 8.0) containing 2% BSA overnight at room temperature. The membrane was then incubated with TBST containing the relevant Abs for 3 h. The primary Abs used were the monoclonal anti-phosphotyrosine Ab (clone 4G10, Upstate Biotechnology, Lake Placid, NY), the monoclonal anti-p42/44 MAP kinase (New England Biolabs, Beverly, MA), and the monoclonal anti-p38 MAP kinase (New England Biolabs) at a 1/1000 dilution. The membrane was then rinsed with three changes of TBST, once for 15 min and twice for 5 min, incubated with horseradish peroxidase-conjugated goat anti-mouse IgG at a 1/4000 dilution for 1 h at room temperature, washed three times, and developed using an enhanced chemiluminescence system (Amersham). Membranes were stripped of the primary Ab-secondary Ab complex by incubating them in stripping buffer (100 mM 2-ME, 2% SDS, and 62.5 mM Tris-HCl, pH 6.7) for 30 min at 50°C. To detect activated forms of p42/44 MAP kinase and p38 MAP kinase, Abs specifically recognized p42/44 MAP kinase phosphorylated on Thr202 and Tyr204 or Abs that specifically recognize Thr180 and Tyr182 on p38 MAP kinase were purchased from New England Biolabs and used in immunoblot analyses. Dual phosphorylation on Thr202 and Tyr204 is diagnostic for activated p42/44 MAP kinase, and dual phosphorylation on Thr180 and Tyr182 is diagnostic for activated p38 MAP kinase. Detection of Ab binding was conducted as described in the manufacturers instructions using the Photope chemiluminescent detection system (New England Biolabs).
| Results |
|---|
|
|
|---|
The kinetics and dose-response characteristics of junB mRNA
expression in P388D1 cells in response to LPS treatment were first
examined by Northern analysis. P388D1 cells (1 x 107)
were incubated with LPS (1 µg/ml) for varying periods of time (06
h). Northern blot analysis of the total RNAs extracted from the cells
(Fig. 1
A) showed that the
expression of junB mRNA was increased as early as 30 min
after LPS treatment, peaked at 60 min, and had declined substantially
by 3 h. The degree of junB mRNA expression 1 h
after LPS exposure was also dependent on the dose of LPS; the minimum
LPS concentration necessary to induce the detectable junB
mRNA level was about 1 ng/ml (Fig. 1
B). The exposure of
cells to PMA also markedly increased the expression of junB
mRNA by 30 min, further increased expression by 1 h, then
decreased expression almost to the constitutive level by 3 h
(Fig. 1
C).
|
Next, the question of how repeated exposure of P388D1 cells to LPS
would affect junB gene expression was investigated. To this
end, P388D1 cells were exposed for 1 h at 37°C to varying doses
of LPS (10 pg/ml to 1 µg/ml). The cells were then washed, incubated
for an additional 2 h in LPS-free medium, and restimulated with 1
µg/ml of LPS for 1 h. Northern analysis of total RNAs extracted
from the cells (Fig. 2
A)
showed that the cells first exposed to >1 ng/ml LPS did not respond to
the secondary LPS stimulation by junB gene activation,
whereas the initial exposure of cells to <1 ng/ml LPS did not
substantially alter junB mRNA expression in response to the
secondary exposure. Next, we examine how long after washing is required
for recovery of junB expression by restimulation. Figure 2
B
shows that responsiveness was almost completely recovered by 36 h
after the initial stimulus. To examine whether this observed refractory
response is restricted to LPS, the capacities of PMA and lipid A as
secondary inducers of junB mRNA were tested by exposing the
control and LPS-primed cells to LPS, PMA, or lipid A. Northern analysis
of total RNA extracted from these cells (Fig. 3
) showed that the cells exposed to a
primary LPS treatment displayed the expected hyporesponsiveness to
secondary LPS challenge (compare lane 3 to lane
4). The initial exposure of cells to LPS did not change the
response of the junB gene to PMA (lane 6
vs lane 5). In contrast, LPS-tolerant cells showed marked
reduction in response to treatment with lipid A, as observed for LPS
(lane 8 vs lane 7).
|
|
We speculated that the decreased steady state level of
junB mRNA in LPS-tolerant cells may have been due to an
increased rate of degradation of junB mRNA. To test this
possibility, LPS-tolerant and control cells were first incubated for
1 h with 1 µg/ml LPS. At the end of the incubation period,
actinomycin D was added at 5 µg/ml. Total RNA was extracted from the
cells at 0, 15, 30, 60, and 90 min after the addition of actinomycin D.
As shown in Figure 4
, junB
mRNA expression was reduced 30 min after the addition of actinomycin D
in both LPS-tolerant and control cells, suggesting no significant
difference in junB mRNA half-life between control and
LPS-tolerant cells.
|
-32P]UTP. Figure 5
. In fact, in
this cell line, TNF-
expression occurred in response to LPS with a
peak at 1 h, and then quickly declined to the basal level. A
similar desensitization phenomenon was detected for TNF-
(data not
shown). The transcriptional levels of housekeeping genes such as
glyceraldehyde-3-phosphate dehydrogenase and ß-actin were almost
unchanged in response to LPS. These results suggest that the decreased
levels of LPS-induced steady state junB mRNA in these cells
were primarily due to decreased transcription.
|
The tyrosine phosphorylation of several proteins catalyzed by
tyrosine kinases has been shown to be involved in LPS-induced cytokine
gene activation and cytokine production (20, 21, 22). It follows that one
of the upstream events associated with LPS-induced junB gene
transcription may require tyrosine phosphorylation through tyrosine
kinases. To test this possibility, the effect of herbimycin A, a potent
tyrosine kinase inhibitor (40), on the LPS-induced expression of
junB mRNA in P388D1 cells was examined, as follows. P388D1
cells were incubated for 4 h with varying concentrations (0.15
µg/ml) of herbimycin A and then were incubated for 1 h with or
without LPS. The total RNA extracted from the cells was analyzed by
Northern blotting. As shown in Figure 6
,
pretreatment of P388D1 cells with herbimycin for 4 h resulted in a
significant inhibition of junB gene expression in a
dose-dependent manner, suggesting the involvement of tyrosine
phosphorylation through tyrosine kinases during LPS-triggered
junB mRNA expression. In addition to junB,
herbimycin A inhibited TNF-
expression in response to LPS. Thus, a
tyrosine kinase-sensitive step(s) is not a junB-specific
phenomenon.
|
The p38 MAP kinase, p42/p44 MAP kinase, and JNK are the most
prominent tyrosine-phosphorylated proteins in response to LPS (20, 21, 22).
Therefore, we tested whether the tyrosine phosphorylation of these
proteins would be suppressed in LPS-tolerant cells. First, to examine
the effect of LPS on cellular tyrosine phosphorylation in P388D1 cells,
the total cell lysates were prepared from P388D1 cells exposed to LPS
for 0, 7, 15, or 30 min or to PMA for 15 min. Then, cell lysates were
subjected to SDS-PAGE. The proteins separated were electrophoretically
transferred to a nitrocellulose membrane and blotted with an
anti-phosphotyrosine Ab (4G10). As shown in Figure 7
A (left
panel), LPS treatment of cells (lanes 14)
resulted in a time-dependent enhancement of tyrosine phosphorylation of
two protein bands of approximately 38 to 42 kDa. LPS-induced tyrosine
phosphorylation of these bands was significantly inhibited after
pretreatment of P388D1 cells with herbimycin A at doses (0.55
µg/ml) that effectively inhibited LPS-induced junB gene
expression (Fig. 7
B). To establish whether the 38- to 42-kDa
tyrosine-phosphorylated polypeptide was a member of the MAP kinase
family, the membrane shown in the left panel of Figure 7
A was reprobed after being stripped of 4G10 Ab using Abs
that recognize 42/44 MAP kinase (independent of phosphorylation) or p38
MAP kinase (independent of phosphorylation). The middle and
right panels of Figure 7
A show that the
electrophoretic mobility of the tyrosine-phosphorylated 38- to 42-kDa
protein appeared to coincide with those of p38 MAP kinase and p42 MAP
kinase, respectively. Although the p42/44 MAP kinase Ab used in this
study recognize both p42 and p44 MAP kinases, the expression of p44 MAP
kinase was below the detection level in this cell line. In contrast, we
observed the tyrosine phosphorylation of three distinct bands of p38,
p42, and p44 MAPKs in response to LPS in the mouse macrophage cell line
RAW 264.7 (data not shown). The levels of p38 MAP kinase and p42 MAP
kinase were relatively unaffected by exposure to LPS or PMA.
|
|
The above findings indicated that the down-regulated
junB expression in response to LPS may be associated with
reduced activation of p38 MAP kinase in LPS-tolerant cells. Since the
role of p38 MAP kinase in the up-regulation of the junB gene
in response to LPS had not been established, we evaluated whether
PB202190, a potent p38 MAP kinase inhibitor (19), inhibits the
LPS-triggered junB mRNA expression in P388D1 cells. P388D1
cells were incubated for 1 h with varying concentrations (0.110
µM) of PB202190 and then incubated for 1 h with or without LPS.
As shown in Figure 9
, inhibition of p38
MAP kinase by PB202190 markedly suppressed LPS-induced junB
mRNA expression, suggesting that the p38 MAP kinase pathway is required
for LPS-triggered junB mRNA expression. These results
indicate that the reduced response to LPS in the junB gene
expression may be due at least in part to the diminished p38 MAP kinase
activity in LPS-tolerant cells.
|
| Discussion |
|---|
|
|
|---|
, IL-1
IL-1ß, IL-6, and IFN-ß, as well as proto-oncogenes, such as
c-fos, c-jun, and junB (17, 41, 42, 43).
LPS is also known to cause desensitization of the expression of
LPS-inducible genes, such as TNF-
and IL-1ß (28, 29, 30, 31, 32, 33, 34, 35), and
proto-oncogenes, such as c-fos (44). Our present data
demonstrate that junB gene expression in P388D1 cells is
also desensitized after exposure of cells to >1 ng/ml LPS (Figs. 1
The data shown in Figure 3
suggest a probable difference between
signal transduction mechanisms of junB gene expression
mediated by LPS and PMA, since LPS-tolerant cells were able to respond
to PMA with full expression of the junB gene, but responded
to secondary stimulation with LPS with a substantially reduced level of
junB gene expression. Indeed, we have previously
demonstrated that the activation of protein kinase C was important in
PMA-induced junB gene expression, but was not required for
LPS-triggered junB expression in a mouse macrophage cell
line, J774 (17). The noted reduced response of LPS-tolerant cells to
lipid A is expected because lipid A is a biologically active moiety of
LPS.
The reduction in the steady state junB level in
LPS-tolerant cells could have been due to accelerated degradation of
the junB transcript. However, the data shown in Figure 4
rule out this possibility, because the half-life of the junB
transcript of LPS-tolerant cells did not differ from that in control
cells. The data in Figure 5
suggest, on the other hand, that the steady
state junB gene expression in response to secondary LPS
stimulation is down-regulated primarily at or upstream of the
transcriptional level. A possibility is that a negatively acting
factor(s) that is synthesized as part of the response to initial LPS
treatment may repress junB mRNA expression at or upstream of
junB transcription. A potential role for such a negatively
acting factor in IFN-
-induced desensitization in fibroblasts (45)
and in LPS-induced tolerance for IL-1ß gene expression in a human
promonocytic cell line, THP-1 (35), has been reported. Lae and McCall
suggested that a labile protein inhibits the activity of transcription
in a repressor manner, since CHX treatment of the LPS-primed cells
recovered not only the steady state level of IL-1ß but also the
transcription of IL-1ß (35). These authors further suggested that
I
B, which is an inhibitor protein of NF-
B, may be a candidate for
the labile repressor in their system (35). Studies of nonlymphoid cells
suggested that the transcriptional activation of junB is
regulated by a variety of stimuli, including IL-6 (46, 47, 48, 49), tumor
promoter (50), cAMP-elevating agent (50, 51), platelet-derived growth
factor, and fibroblast growth factor (52, 53). Moreover, negative
regulation of cAMP response element binding protein-targeted protein
phosphatase-1 or -2A resulted in junB transcriptional
activation in B lymphocytes in response to membrane Ig (54). Each of
these stimuli was shown to use a unique signal transduction pathway
that apparently targets distinct cis-acting DNA sequences
and transcriptional factors (46, 47, 48, 49, 50, 51, 52, 53, 54). The identification of
cis-acting DNA sequence and a transcriptional factor(s)
important for LPS-triggered junB activation would assist in
understanding the mechanism(s) by which LPS causes desensitization of
junB gene expression.
The reduced junB gene response in LPS-tolerant cells
could be due to an aberration in the signal transduction pathway from
LPS-LPS receptor interaction to transcriptional activation. Protein
tyrosine phosphorylation has been shown to be one of the initial events
in macrophage activation, which may mediate some of the downstream
responses to LPS (20, 21, 22). The data presented in Figures 6
and 7
support this idea, because herbimycin A, which is a potent inhibitor of
tyrosine kinase, but neither protein kinase C nor protein kinase A,
simultaneously inhibited the tyrosine phosphorylation of the probable
p38 MAP kinase and p42 MAP kinase and junB expression during
LPS stimulation.
Among proteins that are tyrosine phosphorylated in response to
LPS stimulation, MAPKs, including p38, p42 (ERK-2)/p44 (ERK-1), and
JNK, are known to play important roles in the signal transduction by
LPS (23, 24, 25, 26). Of interest is the differential effect of LPS tolerance
on p38 MAP kinase and p42 MAP kinase activation (Fig. 8
). The
activation of p42 MAP kinase by a secondary exposure of LPS occurred in
LPS-tolerant cells, whereas the activation of p38 MAP kinase was much
less than that in control cells. The junB promoter was
positively regulated by p21ras and ERK in
association with an ETS transcription factor (55). Therefore, our
results suggest that the site of LPS tolerance in junB
expression may be downstream to and independent of p42 MAP kinase. In
this regard, our results differ from Ag-dependent T cell anergy
characterized by the inability to produce IL-2, another example of
adaptation to continuous or repetitive stimulation. Clonal anergy in
Th1 cells appeared to be associated with a block in signal transduction
to ERK and JNK, resulting in defective JunB and/or c-Fos and FosB
expression (56, 57, 58). In contrast to p42 MAP kinase, it is tempting to
speculate that the diminished activation of p38 MAP kinase in response
to LPS may contribute to the down-regulated junB gene
expression in the LPS-tolerant state. This is supported by the finding
that pretreatment with SB202190 inhibited the up-regulation of
junB gene expression in response to LPS, indicating that the
activation of p38 MAP kinase is required for LPS-induced
junB gene expression (Fig. 9
). In vitro studies have shown
that p38 MAP kinase directly phosphorylates activation transcriptional
factor-2 in response to LPS (59). The activation transcriptional factor
family is known to bind the cAMP response element (CRE) motif. Indeed,
the junB gene contains a CRE-like site at a region of the 5'
flanking sequence as well as a CRE site at the 3' noncoding sequence
(46, 49, 51, 52). Identification of downstream substrates may be
helpful for a better understanding of the possible association between
the defective LPS-induced p38 MAP kinase pathway and the
down-regulation of junB gene expression. It also would be
interesting to define the upstream element of p38 MAP kinase at which
LPS-tolerant state causes aberration. Further study is needed to answer
these questions.
| Footnotes |
|---|
2 Current address: Laboratory of Molecular Biology, Basic Research Laboratories, Central Research Laboratories, Ajinomoto Co., Inc., Kawasaki 210, Japan. ![]()
3 Current address: University of Tokyo, Department of Cell Processing, Institute of Medical Sciences, Tokyo 108, Japan. ![]()
4 Abbreviations used in this paper: AP-1, transcription factor activating protein-1; TRE, 12-O-tetradecanoylphorbol 13-acetate response element; MAP, mitogen-activated protein; ERK, extracellular signal-regulated kinase; JNK, c-Jun NH2-terminal kinase; TPA, 12-O-tetradecanoylphorbol 13-acetate; CRE, cyclic adenosine 3',5'-monophosphate response element. ![]()
Received for publication March 17, 1997. Accepted for publication May 27, 1998.
| References |
|---|
|
|
|---|
gene. J. Leukocyte Biol. 56:27.[Abstract]
and IL-1ß production by human monocytes. J. Immunol. 153:1818.[Abstract]
mRNA in LPS-primed macrophages occurs at the level of nuclear factor-
B activation, but not at the level of protein kinase C or CD14 expression. J. Immunol. 154:4803.[Abstract]
and lipopolysaccharide. J. Biol. Chem. 268:1908.
and lipopolysaccharide. J. Leukocyte Biol. 48:412.[Abstract]
B site downstream of the junB gene that mediates responsiveness to interleukin-6 in a murine plasmacytoma. J. Biol. Chem. 270:31129.This article has been cited by other articles:
![]() |
A. E. Medvedev, I. Sabroe, J. D. Hasday, and S. N. Vogel Invited review: Tolerance to microbial TLR ligands: molecular mechanisms and relevance to disease Innate Immunity, June 1, 2006; 12(3): 133 - 150. [Abstract] [PDF] |
||||
![]() |
Hongkuan Fan and J. A. Cook Review: Molecular mechanisms of endotoxin tolerance Innate Immunity, April 1, 2004; 10(2): 71 - 84. [Abstract] [PDF] |
||||
![]() |
S.-J. Yeo, J.-G. Yoon, S.-C. Hong, and A.-K. Yi CpG DNA Induces Self and Cross-Hyporesponsiveness of RAW264.7 Cells in Response to CpG DNA and Lipopolysaccharide: Alterations in IL-1 Receptor-Associated Kinase Expression J. Immunol., January 15, 2003; 170(2): 1052 - 1061. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Fujihara, S. Wakamoto, T. Ito, M. Muroi, T. Suzuki, H. Ikeda, and K. Ikebuchi Lipopolysaccharide-triggered desensitization of TNF-{alpha} mRNA expression involves lack of phosphorylation of I{kappa}B{alpha} in a murine macrophage-like cell line, P388D1 J. Leukoc. Biol., August 1, 2000; 68(2): 267 - 276. [Abstract] [Full Text] |
||||
![]() |
A. E. Medvedev, K. M. Kopydlowski, and S. N. Vogel Inhibition of Lipopolysaccharide-Induced Signal Transduction in Endotoxin-Tolerized Mouse Macrophages: Dysregulation of Cytokine, Chemokine, and Toll-Like Receptor 2 and 4 Gene Expression J. Immunol., June 1, 2000; 164(11): 5564 - 5574. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||