|
Originally published In Press as doi:10.1074/jbc.M104093200 on June 13, 2001
J. Biol. Chem., Vol. 276, Issue 33, 31067-31073, August 17, 2001
Heterogeneous RNA-binding Protein M4 Is a Receptor for
Carcinoembryonic Antigen in Kupffer Cells*
Olga V.
Bajenova,
Regis
Zimmer,
Eugenia
Stolper,
John
Salisbury-Rowswell,
Afshan
Nanji, and
Peter
Thomas
From the Department of Surgery, Boston University School of
Medicine, Boston, Massachusetts 02118
Received for publication, May 7, 2001, and in revised form, June 12, 2001
 |
ABSTRACT |
Here we report the isolation of the recombinant
cDNA clone from rat macrophages, Kupffer cells (KC) that encodes a
protein interacting with carcinoembryonic antigen (CEA). To isolate and identify the CEA receptor gene we used two approaches: screening of a
KC cDNA library with a specific antibody and the yeast two-hybrid system for protein interaction using as a bait the N-terminal part of
the CEA encoding the binding site. Both techniques resulted in the
identification of the rat heterogeneous RNA-binding protein (hnRNP) M4 gene. The rat ortholog cDNA
sequence has not been previously described. The open reading frame for
this gene contains a 2351-base pair sequence with the
polyadenylation signal AATAAA and a termination poly(A) tail. The
mRNA shows ubiquitous tissue expression as a 2.4-kilobase
transcript. The deduced amino acid sequence comprised a 78-kDa membrane
protein with 3 putative RNA-binding domains, arginine/methionine/glutamine-rich C terminus and 3 potential membrane
spanning regions. When hnRNP M4 protein is expressed in pGEX4T-3 vector
system in Escherichia coli it binds
125I-labeled CEA in a
Ca2+-dependent fashion. Transfection of rat
hnRNP M4 cDNA into a non-CEA binding mouse macrophage
cell line p388D1 resulted in CEA binding. These data provide evidence
for a new function of hnRNP M4 protein as a CEA-binding protein in
Kupffer cells.
| |
INTRODUCTION |
The liver is a common site for metastasis from various forms of
primary malignancies. Both experimental and clinical results reveal
that the presence of carcinoembryonic antigen
(CEA)1 enhances liver
metastasis from colorectal carcinoma cells (1, 2). CEA is a highly
characterized, cell surface glycoprotein overexpressed by various tumor
cells and provides a tool for tumor tissue-specific targeting.
Increasing amounts of CEA in the serum correlates with the development
of metastatic recurrence after surgical removal of the primary tumor
(3). Earlier we have shown that CEA production of human colorectal
cancer cell lines directly correlates with the metastatic potential
(4). Poorly metastatic colon cancer cell lines become highly metastatic
when transfected with the cDNA coding for CEA (5, 6). As a member of the immunoglobulin supergene family, CEA is involved in
intercellular recognition and may facilitate attachment of colorectal
carcinoma cells to sites of metastasis. In an experimental metastasis
model of colorectal carcinoma in athymic nude mice, systemic injection of CEA enhanced experimental liver metastasis and implantation in liver
by weakly metastatic tumor cells (7, 8).
To successfully treat cancers, it is necessary to prevent the
development of metastasis after the treatment or removal of the primary
malignancy. Therefore, it is important to elucidate the mechanism by
which CEA enhances metastatic potential. The molecular basis by which
CEA can influence metastasis is only partly understood. We have shown
that CEA is rapidly cleared from the circulation of experimental
animals, accumulates in the liver, and is endocytosed in
vitro by Kupffer cells. This initiates a series of signaling
events that leads to tyrosine phosphorylation on at least two
intracellular proteins (9) and is followed by induction of IL-1 ,
IL-6, IL-10, and tumor necrosis factor- cytokines (10). CEA
uptake by Kupffer cells is independent of its carbohydrate composition
and is mediated by an 80-kDa binding protein (11). CEA is recognized by
this binding protein through a 5-amino acid sequence,
Pro-Glu-Leu-Pro-Lys (PELPK), located at the hinge region (amino acids
108-112) between the N-terminal and the first immunoglobulin loop
domain (12). Molecular modeling studies have suggested that this region
is exposed on the surface of the
molecule.2 We have recently
shown in a subset of colorectal cancer patients that mutations in the
PELPK region of CEA results in the accumulation of large amounts of CEA
in the serum (13). To further study the interaction between the peptide
sequence, PELPK, and rat Kupffer cells, the CEA-binding protein was
purified using a combination of gel filtration, preparative
polyacrylamide gel electrophoresis, and affinity chromatography on
CEA-Sepharose (14). A polyclonal antibody to the rat 80-kDa protein was
produced in mice that blocks uptake by isolated rat Kupffer cells of
both CEA and PELPK peptide conjugated to albumin. This antibody shows a
high degree of specificity for the rat 80-kDa protein by both
fluorescein isothiocyanate analysis and Western blotting (14). In this
study we used this antibody to clone the rat CEA-binding protein. We
have isolated the rat ortholog of the human hnRNP M4 protein and
present the data on the analysis and tissue distribution of this
protein. It was also shown that transfection of rat hnRNP M4
cDNA into a mouse P388D1 macrophages resulted in CEA binding.
| |
EXPERIMENTAL PROCEDURES |
Isolation of Kupffer Cells--
Kupffer cells were isolated
according to our standard laboratory protocol (15) from the livers of
male Harlan Sprague-Dawley rats. The rodents were starved overnight and
the livers perfused through the portal vein with collagenase. Suspended
cells were separated into parenchymal and non-parenchymal fractions by
differential centrifugation. Kupffer cells were purified from the
non-parenchymal cell fraction using a gradient of 17.5% metrizamide in
Gey's balanced salt solution. The interface layer containing Kupffer
cells was isolated and washed in phosphate-buffered saline. Further
purification of the cells is achieved by attachment to plastic plates
for 2 h. Generally 3-5 × 108 cells are
routinely isolated from one liver. 1 × 108 Kupffer
cells remain after the adhesion step. Kupffer cells are identified by
their ability to phagocytose 1-µm latex particles, and by staining
for endogenous peroxidase activity. The cells were >90% viable by
trypan blue staining and 85% phagocytic.
Rat cDNA Library Construction and Screening--
Rat KC
library was constructed on the basis of the  ZAP Express vector
system according to the manufacturers instructions (Stratagene). Total
RNA from various tissues, KC, and cell lines was isolated using the
RNAzol method (Biotex Laboratories, Inc.) and mRNA using the
mRNA isolation kit (Qiagen Inc.). To screen the cDNA library
bacterial strain XL1 was incubated with recombinant phages.
Approximately 1 × 106 plaque forming units of the
library were plated on 150-mm Luria-Bertini agar plates containing 10 mM MgSO4 and maintained at 42 °C for 2-3 h.
Polyvinylidene difluoride filters were pretreated in 10 mM
isopropyl-1-thio- -D-galactopyranoside, overlaid in the
plates, and incubated at 37 °C for 12 h. The filters were dried
and blocked in 3% milk prior to exposure to the anti-80-kDa antibody
(1:100 dilution) (14). The enhanced chemiluminescence technique was used to detect positive cDNA clones (Amersham Pharmacia Biotech). By screening 1011 phages one positive clone was identified,
and after plaque purification, the insert was subcloned in pBK-CMV
vector for DNA sequencing.
Yeast Two-hybrid System--
We also used the HybriZAP 2.1 Two-hybrid Vector system (Stratagene) to identify CEA-binding proteins.
cDNA from rat Kupffer cells were synthesized using the ZAP Express
cDNA synthesis kit (Stratagene), according to the manufacturers
recommendations, using ~5 µg of mRNA which was purified as
described before. For preparing the target (i.e. activation
domain) vector we cloned rat cDNA into the pAD-GAL4
vector. The DNA-binding domain vector construction was performed as
follows: a 438-bp EcoRI-PstI fragment surrounding
the PELPK-binding region was isolated from a CEA plasmid, pdKCR-dhfr/dN-CEA (a gift from Dr. T. Kurihara, Suntory
Institute for Biomedical Research, Japan), and subcloned in-frame into
pBD-GAL4Cam to be used as bait. Correct insertion of the
fragment was confirmed by EcoRI-PstI digestion
and DNA sequencing. This bait did not contain any motifs for
N-linked glycosylation in order to avoid nonspecific binding
of CEA with carbohydrate-binding proteins.
Yeast Interaction Screening and Clone Isolation--
Competent
YRG-2 yeast strain cells were co-transformed by heat shock with the
target and bait plasmids at 42 °C for 20 min, in a solution
containing 1 × TE, 100 mM LiAc, pH 7.5, 50% PEG 3350. Putative interaction clones were selected in SD agar plates without leucine, tryptophan, and histidine. Specific interaction between target and bait plasmids was confirmed by a filter-lift assay,
where colonies are transferred to filter paper, permeabilized in liquid
nitrogen, and assayed for -galactosidase activity to detect
LacZ expression. Verification of interaction was performed by isolating plasmid DNA from selected yeast cells. Plasmid DNA was
purified by the alkaline method and digested with
EcoRI-XhoI enzymes to verify the insert size.
DNA Sequencing--
The sequencing was carried out using dideoxy
terminator fluorescent DNA sequencing method with the Big Dye
Terminator Cycle Sequencing Kit (Applied Biosystems). Reactions were
purified by ethanol precipitation and separated on the ABI 377-96 slab
gel automated DNA sequencer. The contingency sequence was assembled and
analyzed using the Vector NTI Suite program (Informax). The sequences
were verified by sequencing of the opposite strand.
Prokaryotic Protein Expression--
For production of
CEA-binding protein in a bacterial system, the
EcoRI/XhoI fragment corresponding to hnRNP
M4 cDNA from the recombinant phage was cloned into the
EcoRI-XhoI sites of pGEX-4T-3 vector
(Amersham Pharmacia Biotech). The pGEX-4T-3 backbone was chosen from the variety of pGEX-GST vectors to maintain the
original reading frame. The plasmid was introduced into B21
Escherichia coli host bacteria. To express the protein,
transformed cells were cultured in LB to a density
A600 = 0.5 after which
isopropyl-1-thio--D-galactopyranoside (0.3 mM) was added and cultures were maintained at 37 °C for
2 h. Total bacterial proteins were prepared and fusion proteins were purified using glutathione-Sepharose 4B affinity chromatography, according to the manufacturers instructions (Amersham Pharmacia Biotech).
Western Blot--
Kupffer cells (1 × 107) were
washed in phosphate-buffered saline and then lysed in RIPA buffer
(Santa Cruz) with protease inhibitors (1 mM sodium
orthovanadate, 2 mM sodium fluoride, 10 µg/ml leupeptin, and 0.5 mM phenylmethylsulfonyl fluoride). Bacterial cells
were lysed by sonication with protease inhibitors. Cell lysates were centrifuged at 15,000 rpm for 10 min at 4 °C. Supernatants were collected and protein concentrations were measured using the BCA method
(Pierce). Equal amounts of protein were loaded on SDS-PAGE. After
electrophoresis samples were transferred into Sequi-Blot polyvinylidene
difluoride nitrocellulose membrane (Bio-Rad). To detect CEA binding the
membrane was incubated with CEA (1 µg/ml) in TBS buffer containing 10 mM CaCl2, washed, and exposed to a monoclonal
anti-CEA antibody (clone C6G9) (Sigma). Detection of bound CEA was
carried out using the ECL technique (Amersham Pharmacia Biotech).
RT-PCR--
One-step RT-PCR was performed using the Titan One
Tube RT-PCR System (Roche Molecular Biochemicals) to detect CEA-binding protein mRNA. Briefly, a 50-µl reaction was prepared containing 500 ng of template mRNA, and final concentrations of 200 mM of each dATP, dCTP, dGTP, and dTTP, 5 mM
dithiothreitol, 1.5 mM MgCl2, 400 nM of each primer (forward, 5'-GGAAGGCCACTGAAAGTCAA-3';
reverse, 5'-TCCACGACTTTTCCCATCTT-3'), 1 × RT-PCR buffer, and 1 µl of enzyme mixture containing avian myeloblastosis reverse
transcriptase, Taq, and Pwo DNA polymerases. The following
PCR cycling was used: 1 cycle of 94 °C for 2 min; 10 cycles of
94 °C for 30 s, 60 °C for 30 s, 68 °C for 45 s;
35 cycles of 94 °C for 30 s, 60 °C for 30 s, 68 °C
for 45 s, with cycle elongation of 5 s for each cycle; and a
final extension cycle of 68 °C for 5 min. The primers were designed
surrounding the deletion region encoding amino acids 187 and 226 of the
rat hnRNP M4 protein. This generated two expected PCR products of 321 and 204 bp representing the wild type hnRNP M4 and hnRNP M4 with the
deletion mutation. As control for genomic DNA contamination PCR
reactions that included each cDNA synthesis reagent except reverse
transcriptase were set up in parallel. Two independent experiments were
performed for each amplification.
Transient Transfection of HnRNP M4 Expression Plasmid into
Macrophage Cell Lines--
Transient transfection of hnRNP M4 plasmid
was carried out in P388D1 mouse lymphoid macrophage, IC21 mouse
peritoneal, and CRL2192 rat alveolar macrophage cell lines (ATCC). None
of these cells lines were able to take up CEA under standard
conditions. We used rat Kupffer cells as a positive control. The P388D1
cells were propagated in Dulbecco's modified Eagle's medium
supplemented with 10% horse serum. The CRL 2192 cells were grown in
F-12K medium (Kaighn's modification) supplemented with 15% fetal
bovine serum (ATCC). The IC21 cells were grown in RPM1 1640 ATCC-modified medium supplemented with 10% fetal bovine serum. All
cells were maintained at 37 °C in 5% CO2 atmosphere.
All tissue culture products were obtained from Life Technologies. Cells
for transfection were plated at 2-2.5 × 105 cells
per well in six-well tissue culture clusters (Nunc products). Cells
were allowed to grow for 24 h and the plasmid construct was
transfected in serum-free medium using the Gene Pulser II Electroporation system (250 microfarads, 300 V) (Bio-Rad). The cells
were washed once with phosphate-buffered saline and resuspended at a
density of 107 cells/ml in RPM1 media without fetal bovine
serum. 10 µg of pBK-CMV/hnRNP M4 expression vector DNA was
transfected per sample. 0.4 ml of the cell suspension was used per
electroporation in 0.4-cm cuvettes. The cells were maintained at room
temperature prior to and after electroporation. At 10 min
post-electroporation, the cells were seeded into growth media at a
concentration of 106 cells/ml. 48 h
post-electroporation, the cells were harvested and resuspended in a
volume of 200 µl for measuring of CEA binding activity.
In Vitro CEA Uptake Assay--
These procedures were performed
as described previously (15). Briefly CEA (5 µg/ml) labeled with
125I was incubated in triplicate with the cells at
37 °C. Samples were taken at 0, 15-, 30-, 45-, and 60-min intervals.
These were immediately centrifuged through an oil phase (3:1 ratio of
bibutylphthalate:dioctylphthalate) at 14,000 rpm using an Eppendorf
Microfuge. The cells spin through the oil and can be separated from the
supernatant for  -counting, by cutting the bottom of the plastic
centrifuge tubes.
| |
RESULTS |
Cloning of the CEA-binding Protein from a Rat Kupffer Cell cDNA
Library--
The rat Kupffer cell cDNA library was screened with
the mouse polyclonal anti-80-kDa antibody (14). Screening of
1011 plaques from the ZAP library resulted in the
identification of one positive cDNA clone with an insert size of
2.4 kilobases. Sequencing and a Blast search of human genome data bases
for close related family members revealed that this cDNA was a
novel rodent gene with 91% of cDNA homology to the human
hnRNP M4 gene (GeneBankTM accession number NM
005968). The sequence differences are mostly single substitutions of
nucleotides and a deletion of 117 nucleotides in a spacer region
between two putative N-terminal RNA-binding domains (RBD-I and RBD-2).
A similar deletion is also characteristic to Homo sapiens
hnRNP M4 protein deletion mutant cDNA (GeneBankTM
accession number AF 061832). The cDNA sequence of the rat gene was
submitted to the GeneBankTM with accession number
U32577.
Since the anti-80-kDa antiserum used to probe the rat KC library is
polyclonal (14), there is potential to recognize nonspecific protein
epitopes. To confirm that hnRNP M4 was a CEA-binding protein we
independently screened for CEA-binding proteins using the yeast two-hybrid system (HybriZAP-2.1 Two-Hybrid Predigested Vector System,
Stratagene). The bait was an engineered fragment of the full-length CEA
containing a 146-amino acid surrounding the PELPK-binding site. DNA
sequencing showed that all clones had identical sequences to the human
hnRNP M4 gene. Both screening techniques, antibody probing
and the yeast two-hybrid system, resulted in the identification of the
same gene sequence suggesting a new functional role for the hnRNP
M4 gene encoding the 80-kDa CEA-binding protein in Kupffer cells.
Analysis of Rat HnRNP M4 cDNA--
Analysis of the rat
cDNA showed an open reading frame (ORF) starting from the
initiation sequence AAAATGG and containing a 5'-end nontranslated
region of 21 bp followed by 2351 nucleotides with the polyadenylation
signal AATAAA and a termination poly(A) tail. Fig.
1 shows this sequence. The cDNA has 3 putative RNA-binding domains (RBD) that are evolutionarily conserved
domains present in pre-mRNA-, mRNA-, pre-rRNA-, and
snRNA-binding proteins, including hnRNP proteins, splicing factors, and
polyadenylation factors (16). The first and second domains (RBD-1 and
RBD-2) are arranged in tandem close to the N terminus and RBD-3 is
located near the C terminus (Fig. 1). The N-terminal RBD-1 (nucleotides
297-376) and RBD-2 (nucleotides 634-740) regions have 82 and 87%
homology with the H. sapiens myelin gene expression factor 2 (MyEF-2) cDNA (17). These sequences are underlined on
Fig. 1. The MyEF-2 protein contains two RNA-binding domains (RBD-1 and
RBD-2), previously shown to be responsible for sequence specific
binding to both RNA and single stranded DNA (17). MyEF-2 is a
transcription factor that was isolated from the mouse brain, maps to
mouse chromosome 2, and represses transcription of the myelin basic
protein gene. The hnRNP M4 RBD-1 and RBD-2 regions show a lesser degree
of homology to RBDs found in a wide variety of RNA and single strand
DNA-binding proteins (18). Sequence analysis demonstrated that
hnRNP M4 contains no other commonly recognized DNA binding
motifs, such as zinc finger, homeobox, POU, or helix-loop-helix
domains. Presumably, similar to MyEF-2, the RNA binding motifs can be
responsible for single strand DNA binding and RNA binding activity of
hnRNP M4. No significant homology with other proteins, including other
hnRNPs was found by Blast search.
View larger version (97K):
[in this window]
[in a new window]
|
Fig. 1.
The nucleotide sequence of rat hnRNP M4
cDNA. This cDNA was isolated from rat KC as encoding the
CEA-binding protein. It is a full-length sequence that consists of a 5'
end nontranslated region of 21 bp and an ORF that starts from the
initiation sequence AAAATGG. The initiation, termination codons, and a
poly(A) tail are shown in bold. The two N-terminal
RNA-binding domains (RBD-1 and RBD-2) followed by the
methionine/arginine/glycine-rich region and C-terminal RBD-3 shown as
boxes. Inside the RBD-1 and RBD-2 the regions of homology
with the MyEF-2 transcription factor are underlined. The
linker sequence between the RBD-1 and RBD-2 containing the aa 187-226
deletion is shown in italic. The double
underlined amino acids correspond to the primers sequences
(5'-GGAAGGCCACTGAAAGTCAA, 3'-TCCACGACTTTTCCCATCTT) employed for RT-PCR
amplification. The GeneBankTM accession number for this
sequence is U32577.
|
|
The rat M4 cDNA sequence has 90-92% identity to a H. sapiens chromosome 19 clone CTD-3182G2, complete
sequence. The human gene was previously assigned to sub-bands
p13.3-p13.2 of chromosome 19 (19). The localization of the hnRNP M4 in
the rat genome is not known.
Analysis of Rat HnRNP M4 Protein--
Protein sequence analysis
based on the cDNA ORF predicts that rat hnRNP M4 encodes
a novel protein of 775 amino acids. It is consistent with its apparent
molecular mass of 78-80 kDa on SDS-polyacrylamide gel electrophoresis
and is consistent with the molecular mass previously reported for the
Kupffer cell-binding protein (14). Based on cDNA translation hnRNP
M4 is a multifunctional signaling protein that has the potential to be
modified by variety of enzymes. Structural analysis of the protein
using Swiss-Prot data base revealed that in addition to 3 putative
RNA-binding domains (aa 77-155, 171-248, and 620-696) this protein
has 3 potential membrane spanning regions (aa 263-282, 290-311, and
597-616) (Fig. 2A). It also
possesses 7 potential protein kinase C phosphorylation sites, 11 casein
kinase II phosphorylation sites, and 15 N-myristoylation motifs (Fig. 2B). The intracellular domain contains what
appears to be a tyrosine phosphorylation motif (aa 100-106) and two
glucosaminoglycan attachment sites (aa 36-39 and 379-382). The C
terminus of hnRNP M4 is unique to this molecule and contains 17 repeats
of glycine, arginine, methionine, and glutamine. Arginine is the only
known methylated amino acid residue in hnRNPs, indicating that the
methylation of these proteins is likely to have an important effect on
their functions (20). Arginine methylation modification is part of the
mechanism by which protein-RNA complexes are recognized for nuclear
export (21).
View larger version (35K):
[in this window]
[in a new window]
|
Fig. 2.
The predicted amino acid sequence of hnRNP M4
cDNA. a, the domain structure of rat hnRNP M4
protein. The protein composed of 2 N-terminal RNA-binding domains
followed by a methionine/arginine/glycine-rich region and C-terminal
RBD-3 are shown as boxes. The potential transmembrane
regions are shown. The numbers above correspond to the amino
acids. b, the post-translational modification sites of rat
hnRNP M4 protein. The predicted tyrosine kinase phosphorylation site,
KVGEVTY (aa 100-106), is shown as  . The asterisk (*)
indicates 2 potential N-linked glycosylation sites (aa
36-39 and 379-382). The plus (+) indicate 7 potential
protein kinase C phosphorylation sites. The dot (.)
indicates 11 potential casein kinase II phosphorylation sites. The
v indicates 15 myristylation sites.
|
|
The partial amino acid sequence of rat hnRNP M4 has been recently
published (22) and it is in support of our protein prediction. This
sequence is missing the N-terminal part of the protein and has several
gaps throughout the sequence (Fig. 3). It
also represents the splicing form without the aa 187-226 deletion in
the spacer region between RBD-1 and RBD-2.
View larger version (69K):
[in this window]
[in a new window]
|
Fig. 3.
Comparison of published (22) and deduced
sequences of rat hnRNP M4 protein. Amino acids are shown in the
single letter code. The sequences are indicated on the
top and the amino acid numbers on the right and
left sides. The consensus sequence shown in the
middle represents the identical amino acids in both
proteins.
|
|
In Vitro CEA Binding Studies--
The ability of the hnRNP M4
protein to interact with CEA was examined by preparing membrane lifts
from ZAP Express-hnRNP M4 phage plates and incubating them with
125I-labeled CEA or PELPK albumin conjugate (11) in the
presence of 10 mM Ca2+ or 10 mM
EDTA. Both CEA and PELPK albumin bound strongly to the phages in the
presence of Ca2+ but there was no binding when EDTA was
present. These data confirm previous observations on the
Ca2+ requirements of the CEA-binding protein using isolated
Kupffer cells (13) and adds further evidence that hnRNP M4 is the
Kupffer cell CEA-binding protein.
To further substantiate this finding and to estimate the molecular mass
of the encoded protein, the cDNA corresponding to rat hnRNP
M4 ORF was placed into a prokaryotic expression pGEX system allowing production of a GST-hnRNP M4 fusion protein. A series
of pGEX vectors were designed for inducible, high level intracellular expression of the cDNA as a fusion protein with Schistosoma japonicum GST protein (Amersham Pharmacia
Biotech). The hnRNP M4 cDNA was cloned into pGEX-4T-3
vector to maintain the proper reading frame. The fusion protein was
obtained as described under "Experimental Procedures." The Kupffer
cell lysate and pGEX-4T-3/hnRNP M4 fusion proteins (2 clones) were
subjected to SDS-PAGE and transferred to a polyvinylidene difluoride
membrane. The membrane was exposed to 1 µg/ml CEA followed by an
anti-CEA antibody. KC lysate was used as a positive control (lane
1) and the empty vector as a negative control (lane 4).
A single 80-kDa band was shown in Kupffer cells Fig.
4, lane 1, and in the fusion
proteins (lanes 2 and 3). The band was absent for
the vector alone (lane 4). This data indicated that rat
hnRNP M4 protein encoded by the cDNA is able to bind soluble CEA
similar to the 80-kDa KC receptor. The molecular weights of the hnRNP
M4 protein and an 80-kDa receptor from KC were indistinguishable.
View larger version (62K):
[in this window]
[in a new window]
|
Fig. 4.
Western blot of rat hnRNP M4 protein
expressed in E. coli showing CEA binding. The rat
cDNA corresponding to hnRNP M4 ORF was placed into prokaryotic
expression vector pGEX4T-3 allowing production of a
GST-hnRNP M4 fusion protein. The Kupffer cell lysate and
pGEX-4T-3/hnRNP M4 fusion proteins (2 clones) were subjected to
SDS-PAGE and transferred into the polyvinylidene difluoride membrane.
This membrane was exposed to the soluble 1 µM CEA
followed by the anti-CEA antibody. Similar to KC (lane 1),
hnRNP M4 protein (lanes 2 and 3) binds to the
soluble CEA and has a molecular mass 80 kDa. Lane 4 corresponds to a negative control that is a pGEX4T-3 vector
without the hnRNP M4 cDNA.
|
|
Tissue Distribution of HnRNP M4--
Both rat and human tissues
were examined for the presence of the binding protein mRNA using
Northern blots. A nick translated probe with [32P]
corresponding to hnRNP M4 cDNA was used. The results from rat and
human tissues were similar. Northern blot analysis revealed that the
mRNA encoding the hnRNP M4 is abundantly expressed as a single
transcript of 2.4 kilobases in the liver, heart, lung, skeletal muscle,
kidney, stomach, and thyroid. The RNA seems to have a ubiquitous
distribution but the Northern blot analysis (not shown) was unable to
distinguish between the full-length and deletion mutant. To determine
whether the form with the deletion in a spacer region between the RBD-1
and RBD-2 is expressed in Kupffer cells we synthesized PCR primers that
incorporated the deleted region as part of their product. Amplification
results in a larger PCR product for the full-length form (321 bp) and a
shorter product (204 bp) for the deletion mutant. The results show that
the deletion form encoding aa 187-226 is distributed throughout all
rat tissues examined. The larger form is generally also present but in
what seems to be lower copy numbers. Both mRNA forms (with and
without the deletion) were characteristic of the rat KC. Only the form
with the deletion was detected in human KC (Fig.
5). These data indicate that there are
multiple subclasses of hnRNP M4 mRNA that may arise by alternative
pre-mRNA splicing and may carry different functions in
vivo.
View larger version (43K):
[in this window]
[in a new window]
|
Fig. 5.
Distribution of different splicing form of
hnRNP M4 protein in rat tissues. To determine whether the form
with the deletion (aa 187-226) in a spacer region between the RBD-1
and RBD-2 is expressed in Kupffer cells we synthesized PCR primers that
incorporated the deleted region as part of their product. Amplification
will result in a larger PCR product for the full-length form (321 bp)
and a shorter product (204 bp) for the deletion mutant. We examined a
number of rat tissues. The results show that the deletion form of hnRNP
M4 is distributed throughout all tissues examined. Both mRNA forms
(with and without the deletion) were characteristic of the rat KC. Only
the short form was determined in human KC. These data suggest that
multiple forms of the hnRNP M4 protein may exist, possibly with
different functions in vivo.
|
|
Transfection of HnRNP M4 in Macrophage Cell Lines--
To
elucidate whether macrophage cell lines can take up soluble CEA,
in vitro experiments with 125I-labeled CEA were
performed. We tested 3 macrophage cell lines: CRL2192, P388D, and IC21.
CEA uptake by isolated rat KC was used as a positive control. Cells
were incubated with 125I-CEA (5 µg/ml) for 15, 30, 45, and 60 min. Fig. 6 shows that none of the
macrophage cell lines was able to take up the labeled CEA in
vitro. Earlier it was also determined that freshly isolated lung
alveolar macrophages express the 80-kDa protein and rapidly endocytose
CEA in a similar manner to Kupffer cells (23). These data suggest that
the ability of CEA uptake is cell specific and perhaps is tissue
restricted to Kupffer cells and alveolar macrophages.
View larger version (9K):
[in this window]
[in a new window]
|
Fig. 6.
The ability to take up CEA depends on the
tissue examined and is restricted to KC and alveolar macrophages.
To elucidate whether the macrophage cell lines CRL2192, P388D, and IC21
can take up soluble CEA, in vitro experiments with
125I-labeled CEA were performed. Cells were incubated with
the iodine-labeled CEA (5 µg/ml) for 15, 30, 45, and 60 min. In
contrast to KC and peritoneal macrophages none of the cell lines is
able to take up CEA. CRL 2192,  ; U937, ; IC21,  ; KC,  .
|
|
To further confirm the role of hnRNP M4 as a CEA-binding protein and to
begin to elucidate the mechanism underlying the intracellular signaling
associated with Kupffer cell activation, the hnRNP M4 cDNA in the pBK-CMV expression vector was transfected into
macrophage cell lines. The mouse alveolar-derived P388D1 and mouse
peritoneal-derived IC21 macrophages were transfected with the
hnRNP M4/PBK-CMV (10 g of hnRNP M4/pBK-CMV
plasmid) expression vector. Fig. 7 shows that after 48 h 125I-CEA binding in P388D1 cells
increased with time in the transfected as compared with the parental
cell line. A transfectant containing a matching concentration of the
empty vector (not shown) did not take up CEA. This data shows that
transient transfection of hnRNP M4 cDNA results in CEA
binding by P388D1 cells and implies expression of the hnRNP M4 protein
on the cell surface. Interestingly, only P388D1, but not IC21 cells
were able to take up CEA. At present, the reasons for lack of response
by IC21 cells are not known.
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 7.
Transfection of hnRNP M4 cDNA can
initiate CEA binding in P388D1 macrophages. To determine the
mechanism underlying the intracellular signaling associated with KC
activation and CEA binding, the macrophage cell lines CRL2192, P388D,
and IC21 were transfected with the hnRNP M4 cDNA in the
pBK-CMV expression vector in a transient assay. As shown, after 48 h, the hnRNP M4 cDNA expression induced CEA binding in
P388D1 cells. An ~4 times increase in CEA uptake was observed on
transfecting 10 µg of hnRNP M4/pBK-CMV plasmid. Control,
; transfected line, .
|
|
| |
DISCUSSION |
There is significant experimental and clinical data that suggest
an important role for CEA in enhancement of hepatic metastasis from
colorectal cancer (2). A precise mechanism for this function requires
the identification of the proteins involved in metastatic signaling
through KC. In this work, we employed two independent methods to
identify the gene responsible for CEA binding to Kupffer cells. The
expression cloning technique using an antiserum specific to the 80-kDa
CEA-binding protein and the yeast two-hybrid system utilized
protein-protein interactions to identify and isolate the cDNA
encoding the CEA receptor. Both techniques resulted in the isolation of
a cDNA with high homology (~90%) to the hnRNP M4
gene. To further study the CEA-KC interaction we developed an in
vitro model that will reconstitute the signaling events that occur
in vivo in KC. Identifying the nature of biochemical differences between KC and macrophage cell lines, which allow hnRNP M4
function more efficiently in some cell lines than others, will help to
understand the complex regulatory mechanisms involved in the CEA
receptor expression and binding.
To elucidate whether the hnRNP M4 cDNA can initiate CEA
uptake, the macrophage cell lines, CRL2192, P388D, and IC21, were transfected with the expression pBK-CMV/hnRNP M4 vector. In
support to our previous findings, it was shown that the ability to take up CEA depends on the tissue examined and is restricted to
Kupffer cells (Fig. 6) and alveolar macrophages (22). Transient
transfection assays using the mouse P388D1 macrophages resulted in CEA
binding (Fig. 7), suggesting that the type of cell is very important
for this function to occur. Thus, the identified rat hnRNP M4 protein (derived from the open reading frame of the cDNA sequence) displays features similar to that of 80-kDa KC CEA receptor by three independent criteria. 1) By SDS-PAGE the molecular masses of both the recombinant and the cellular proteins were found to be similar corresponding to
78-80 kDa. 2) The rat hnRNP M4 fusion protein expressed in E. coli binds CEA and this interaction is a
Ca2+-dependent process (Fig. 5). 3) Transient
transfection of the hnRNP M4 cDNA into a non-CEA binding
macrophage cell line resulted in the ability to take up CEA in the
presence of Ca2+.
The hnRNPs are a large group of 20 proteins (hnRNP A-hnRNP U) that
associate with pre-mRNAs in the eukaryotic cell (16). The majority
of hnRNPs are extremely abundant nuclear components, their amounts
approximately the same as those of the core histones (24). They are
involved in mRNA processing, maturation (capping, splicing, and
polyadenylation) (25), and mRNA stability (26). Some hnRNPs also
bind single and double stranded DNA and act as transcription factors
(27). Most hnRNPs are expressed only in the nucleus. However, several,
among them, hnRNP A, D, E, I, and K, have been found to shuttle between
the nucleus and the cytoplasm, and are believed to promote mRNA
export by acting as adapters between mRNA and the transport
machinery (27). The hnRNP M4 protein belongs to the hnRNP M subfamily
that consists of 4 splicing forms with the molecular mass 68-80
kDa (22). The human hnRNP M4 protein has been shown to participate in
RNA splicing and processing, as well as to changes related with heat
shock (19, 29, 30). Evidence that this protein can be expressed on the
cell surface comes from the fact that the close homologue of human
hnRNP M4 was also described as a monomer of
N-acetylglucosamine-specific receptor for the thyroid
hormone NAGR1 (31). Later studies showed that this receptor is
identical to the hnRNP M4 (32). Thus in both these instances (CEA and
NAGR1) hnRNP M4 is capable of acting as a receptor.
We hypothesize (Fig. 8) that the hnRNP M4
is involved in a multiprotein complex that can recognize CEA on the
surface of KC and initiate a series of signaling events that lead to
the induction of IL-1, IL-6, IL-10, and tumor necrosis factor-
cytokines (10) and tyrosine phosphorylation (9). Based on this study
three mechanisms of cytokine regulation by CEA can be envisioned.
First, enhancement in tyrosine phosphorylation after receptor
activation is an important signaling event leading to cellular
responses. Similar to other family members, hnRNP M4 has a tyrosine
internalization signal, KVGEVTY, aa 100-106, 7 potential protein
kinase C, and 11 casein kinase II phosphorylation sites and can undergo
post-translational modifications. Particularly, the hnRNP K protein is
tyrosine phosphorylated in vitro by Src and Lck that
regulates in vivo the K-protein-protein and K-protein-RNA
interactions (33). HnRNP A2 and hnRNP C1 proteins in the liver are
phosphorylated and this process is modulated by calmodulin (34). With
phosphorylation being the hallmark, several types of modifications may
participate in signal transduction. Many proteins have been identified
that carry complex N- and O-linked post-translational glycosylation including transcription factors, cytoskeletal, and nuclear pore proteins, oncogene products, and tumor
suppressors (35). The hnRNP M4 protein possesses two potential N-glucosaminoglycan attachment sites and these
N-glucosaminoglycan chains could serve to orient the
receptor proteins on the cell surface by correct insertion into the
plasma membrane and to stabilize the protein (36).
View larger version (45K):
[in this window]
[in a new window]
|
Fig. 8.
A hypothesis on the role of the hnRNP M4
protein in CEA metabolism. We suggest that CEA interacts
with hnRNP M4 on the surface of Kupffer cells through the PELPK-binding
site. The CEA-hnRNP complex is endocytosed by KC and initiates the
signaling cascade that stimulates production of IL-1, IL-6, IL-10,
and tumor necrosis factor- cytokines. Upon stimulation the CEA-hnRNP
M4 complex dissociates, CEA is modified by removal of sialic acid
residues and is recycled by the hepatocytes asialo-glycoprotein
receptor. The further pathway for the hnRNP M4 is unknown.
|
|
Multiple protein interactions characteristic to hnRNPs can also be part
of signaling. For example, HnRNP K has a diverse repertoire of
molecular partners including G protein, tyrosine, and serine/threonine kinases as well as the proto-oncoprotein Vav (37).
Second, possible mechanisms of cytokine regulation by the hnRNP M4-CEA
interaction emerges from the sequence homology of N-terminal RBD-1 and
RBD-2 domains with the MyEF-2 transcription factor (17). It is known
that RBD motifs of the RNA-binding proteins can provide RNA and DNA
binding activity (38). A previous study (39) indicated that the hnRNP K
protein trans-activates the c-myc promoter by increasing the
level of transcription. Several hnRNP proteins, including hnRNPA/B
(40), hnRNP DOB (27), and hnRNP U (41), can act as transcription
factors and by regulating promoter(s) function(s) of the target genes
can directly or indirectly (through other genes) modulate cytokine gene
expression. Additionally, this group of proteins can effect
transcription by binding with the cell cycle regulators. It was shown
that P2P-R RNA interacts with RB1 through the RB pocket domain and this
interaction has an effect on co-transcriptional or post-transcriptional
regulation of mRNA expression (42). HnRNPs are also involved in the
multiprotein complexes that include nuclear receptors (40) and RNA
polymerase II (43).
The third mechanism of how hnRNP M4 protein can trigger the IL-1,
IL-6, IL-10, and tumor necrosis factor- cytokines production is in
the control of mRNA stability. For many proto-oncogenes, lymphokines, and cytokines, a common feature is the existence of A + U-rich elements (25). Uridylate stretches also found in regulatory
regions of RNAs such as the 3' splice site of introns appear to be
common targets for RNA-binding proteins (44) and it was shown that the
family of hnRNP M proteins bind to poly(U) stretches in high salt
conditions (1 M NaCl) (17, 28, 29).
This report on isolation of CEA-binding protein thus provides both a
direction for further analysis of the mechanism of KC activation and
cytokine regulation, and the clinical application to determine whether
abnormal expression of hnRNP M4 is involved in metastatic cancers. The
in vitro model using overexpression of hnRNP M4 in
macrophage cell lines may be used to reconstitute the signaling events
that occur in vivo in KC. It will also allow us to examine
downstream signaling events and the protein domains needed to initiate
receptor-mediated endocytosis and to induce the cytokine secretion in
macrophages by CEA. These studies have a potential to not only increase
our knowledge of how metastases occur but to give insight into new
therapeutic approaches.
| |
ACKNOWLEDGEMENTS |
We thank Dr. Joseph Ozer for helpful
discussion and critical reading of this manuscript. We also thank Wendy
Thomas for technical assistance and help with the preparation of the manuscript.
| |
FOOTNOTES |
*
This work was supported by National Institutes of Health
Grant CA 74941 (to P. T.).The costs of publication of this
article were defrayed in part by the
payment of page charges. The article must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
To whom correspondence should be addressed: Boston University
School of Medicine, Laboratory of Surgical Biology, 801 Albany St., Rm.
310, Boston, MA 02118. Tel.: 617-414-8066; Fax: 617-414-8078; pthomas{at}bu.edu.
Published, JBC Papers in Press, June 13, 2001, DOI 10.1074/jbc.M104093200
2
P. A. Bates, personal communication.
| |
ABBREVIATIONS |
The abbreviations used are:
CEA, carcinoembryonic antigen;
KC, Kupffer cell;
hnRNP, heterogeneous
RNA-binding protein;
RBD, RNA-binding domain;
aa, amino acid;
IL, interleukin;
PAGE, polyacrylamide gel electrophoresis;
bp, base pair(s);
RT-PCR, reverse transcriptase-polymerase chain reaction;
ORF, open reading frame;
MyEF-2, myelin gene expression factor 2;
GST, glutathione S-transferase.
| |
REFERENCES |
| 1.
|
Kim, J. C.,
Gong, G.,
Roh, S. A.,
and Park, K. C.
(1999)
Mol. Cells
9,
133-137
|
| 2.
|
Jessup, J. M.,
and Thomas, P.
(1998)
in
Cell Adhesion and Communication Mediated by the CEA Family: Basic and Clinical Perspectives
(Stanners, C. P., ed)
, pp. 195-222, Harwood Academic Publishers, Amsterdam, Netherlands
|
| 3.
|
Keesee, S. K.,
Briggman, J. V.,
Thill, G.,
and Wu, Y.-J.
(1996)
Crit. Rev. Eukaryotic Gene Exp.
6,
189-214
|
| 4.
|
Wagner, H. E.,
Toth, C. A.,
Steele, G. D.,
and Thomas, P.
(1992)
Clin. Exp. Metastasis
10,
25-31
|
| 5.
|
Hashino, J.,
Fukada, Y.,
Oikawa, S.,
Nakazato, H.,
and Nakanishi, T.
(1994)
Clin. Exp. Metastasis
12,
324-328
|
| 6.
|
Thomas, P.,
Gangopadhyay, A.,
Steele, G., Jr.,
Andrews, C.,
Nakazato, H.,
Oikawa, S.,
and Jessup, J. M.
(1995)
Cancer Lett.
25,
59-66
|
| 7.
|
Leconte, A.,
Garambois, V.,
Ychou, M.,
Robert, B.,
Pourquier, D.,
Terskikh, A.,
Mach, J. P.,
and Pelegrin, A.
(1999)
Br. J. Cancer.
9,
1373-1379
|
| 8.
|
Hostetter, R. B.,
Augustus, L. B.,
Mankarious, R.,
Chi, K.,
Fan, D.,
Toth, C. A.,
Thomas, P.,
and Jessup, J. M.
(1990)
J. Natl. Cancer Inst.
82,
380-385
|
| 9.
|
Gangopadhyay, A.,
Lazure, D. A.,
and Thomas, P.
(1997)
Cancer Lett.
118,
1-6
|
| 10.
|
Gangopadhyay, A.,
Bajenova, O.,
Kelly, T. M.,
and Thomas, P.
(1996)
Cancer Res.
56,
4805-4810
|
| 11.
|
Toth, C. A.,
Thomas, P.,
Broitman, S. A.,
and Zamcheck, N.
(1982)
Biochem. J.
204,
377-381
|
| 12.
|
Gangopadhyay, A.,
and Thomas, P.
(1996)
Arch. Biochem. Biophys.
334,
151-157
|
| 13.
|
Zimmer, R.,
and Thomas, P.
(2001)
Cancer Res.
61,
2822-2826
|
| 14.
|
Gangopadhyay, A.,
Lazure, D. A.,
Kelly, T. M.,
and Thomas, P.
(1996)
Arch. Biochem. Biophys.
328,
151-157
|
| 15.
|
Toth, C. A.,
Thomas, P.,
Broitman, S. A.,
and Zamcheck, N.
(1985)
Cancer Res.
45,
392-397
|
| 16.
|
Kenan, D. J.,
Query, C. C.,
and Keene, J. D.
(1991)
Trends Biochem. Sci.
16,
214-220
|
| 17.
|
Haas, S.,
Steplewski, A.,
Siracusa, L. D.,
Amini, S.,
and Khalili, K.
(1995)
J. Biol. Chem.
270,
12503-12510
|
| 18.
|
Gorlach, M.,
Burd, C. G.,
Portman, D. S.,
and Greyfuss, G.
(1993)
Mol. Biol. Rep.
18,
73-78
|
| 19.
|
Gattoni, R.,
Mahe, D.,
Mahl, P.,
Fischer, N.,
Mattei, M. G.,
Stevenin, J.,
and Fuchs, J. P.
(1996)
Nucleic Acids Res.
24,
2535-2542
|
| 20.
|
Liu, Q.,
and Dreyfuss, G.
(1995)
Mol. Cell. Biol.
15,
2800-2808
|
| 21.
|
Shen, E. C.,
Henry, M. F.,
Weiss, V. H.,
Valentini, S. R.,
Silver, P. A.,
and Lee, M. S.
(1998)
Genes Dev.
12,
679-691
|
| 22.
|
Kafasla, P.,
Patrinou-Georgoula, M.,
and Guialis, A.
(2000)
Biochem. J.
350,
495-503
|
| 23.
|
Toth, C. A.,
Rapoza, A.,
Zamcheck, N.,
Steele, G.,
and Thomas, P.
(1989)
J. Leukocyte Biol.
45,
370-376
|
| 24.
|
Huang, S.
(2000)
J. Struct. Biol.
129,
233-240
|
| 25.
|
Krecic, A. M.,
and Swanson, M. S.
(1999)
Curr. Opin. Cell Biol.
11,
363-371
|
| 26.
|
Loflin, P.,
Chen, C. Y.,
and Shyu, A. B.
(1999)
Genes Dev.
13,
1884-1897
|
| 27.
|
Tolnay, M.,
Vereshchagina, L. A.,
and Tsokos, G. C.
(1999)
Biochem. J.
338,
417-425
|
| 28.
|
M, Jang, S. K.,
Shyu, A. B.,
and Wilkinson, M. F.
(2000)
Cell
102,
135-138
|
| 29.
|
Datar, K. V.,
Dreyfuss, G.,
and Swanson, M. S.
(1993)
Nucleic Acids Res.
21,
439-446
|
| 30.
|
Mahe, D.,
Fischer, N.,
Decimo, D.,
and Fuchs, J.
(2000)
Biochim. Biophys. Acta
1492,
414-424
|
| 31.
|
Blank, O.,
Perrin, C.,
Mziaut, H.,
Darbon, H.,
Mattei, M. G.,
and Miquelis, R.
(1994)
Genomics
21,
18-26
|
| 32.
|
Blank, O.,
Perrin, C.,
Mziaut, H.,
Darbon, H.,
Mattei, M. G.,
and Miquelis, R.
(1995)
Genomics
27,
561-562
|
| 33.
|
Schullery, D. S.,
Ostrowski, J.,
Denisenko, O. N.,
Stempka, L.,
Shnyreva, M.,
Suzuki, H.,
Gschwendt, M.,
and Bomsztyk, K.
(1999)
J. Biol. Chem.
274,
15101-15109
|
| 34.
|
Bosser, R.,
Faura, M.,
Serratosa, J.,
Renau-Piqueras, J.,
Pruschy, M.,
and Bachs, O.
(1995)
Mol. Cell. Biol.
15,
661-670
|
| 35.
|
Rudd, P. M.,
Wormald, M. R.,
and Dwek, R. A.
(1999)
J. Mol. Biol.
239,
351-366
|
| 36.
|
Boer, U.,
Neuschafer-Rube, F.,
Moller, U.,
and Puschel, G. P.
(2000)
Biochem. J.
15,
839-847
|
| 37.
|
Bomsztyk, K.,
Van Seuningen, I.,
Suzuki, H.,
Denisenko, O.,
and Ostrowski, J.
(1997)
FEBS Lett.
403,
113-125
|
| 38.
|
Fierro-Monti, I.,
and Mathews, M. B.
(2000)
Trends Biochem. Sci.
25,
241-246
|
| 39.
|
Lee, M. H.,
Mori, S.,
and Raychaudhuri, P.
(1996)
J. Biol. Chem.
271,
3420-3427
|
| 40.
|
Leverrier, S.,
Cinato, E.,
Paul, C.,
Derancourt, J.,
Bemark, M.,
Leanderson, T.,
and Legraverend, C.
(2000)
J. Biol. Chem.
381,
1031-1040
|
| 41.
|
Kim, M. K.,
and Nikodem, V. M.
(1999)
Mol. Cell. Biol.
19,
6833-6844
|
| 42.
|
Witte, M. M.,
and Scott, R. E.
(1997)
Proc. Natl. Acad. Sci. U. S. A.
94,
1212-1217
|
| 43.
|
Eggert, M.,
Michel, J.,
Schneider, S.,
Bornfleth, H.,
Baniahmad, A.,
Fackelmayer, F. O.,
Schmidt, S.,
and Renkawitz, R.
(1997)
J. Biol. Chem.
272,
28471-28478
|
| 44.
|
Gallouzi, I. E.,
Brennan, C. M.,
Stenberg, M. G.,
Swanson, M. S.,
Eversole, A.,
Maizels, N.,
and Steitz, J. A.
(2000)
Proc. Natl. Acad. Sci. U. S. A.
97,
3073-3078
|
Copyright © 2001 by The American Society for Biochemistry and Molecular Biology, Inc.
 CiteULike  Complore  Connotea  Del.icio.us  Digg  Reddit  Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
J. B. Kruskal, A. Azouz, H. Korideck, M. El-Hallak, S. C. Robson, P. Thomas, and S. N. Goldberg
Hepatic Colorectal Cancer Metastases: Imaging Initial Steps of Formation in Mice
Radiology,
June 1, 2007;
243(3):
703 - 711.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. N. Samara, L. M. Laguinge, and J. M. Jessup
Carcinoembryonic Antigen Inhibits Anoikis in Colorectal Carcinoma Cells by Interfering with Trail-R2 (DR5) Signaling
Cancer Res.,
May 15, 2007;
67(10):
4774 - 4782.
[Abstract]
[Full Text]
[PDF]
|
|
|
Copyright © 2001 by the American Society for Biochemistry and Molecular Biology.
|
Advertisement
Advertisement
|
TOP
ABSTRACT
INTRODUCTION