| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
The Burnham Institute, Program on Apoptosis and Cell Death Regulation, La Jolla, California 92037;
British Columbia Cancer Agency, Department of Pathology, Vancouver, British Columbia, Canada;
Science Applications International Corp., National Cancer Institute-Frederick Cancer Research and Deveopment Center, Frederick, MD 21702;
§
Medical Research Council Centre for Immune Regulation, University of Birmingham, Birmingham, United Kingdom; and
¶
Departments of Medicine, Hematology, and Oncology and
||
Pathology, University of California, San Diego, CA 92093
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
180 aa (1). The x-ray crystallographic
structure of the TRAF domain of TRAF2 reveals a trimeric assembly, with
each TRAF domain monomer containing a surface crevice responsible for
binding peptidyl motifs found in the cytosolic domains of the TNF
family receptors to which TRAF2 is known to bind (2, 3, 4).
Similarities and differences in the peptidyl specificities of
individual TRAFs account for their selective associations with
particular TNFR family members, yielding diversity, redundancy, and
competition among TRAFs with respect to ligand-inducible recruitment to
various TNFR family receptor complexes (5, 6, 7, 8, 9, 10).
While capable of associating with multiple cytokine receptor complexes,
several TRAFs can also bind to a variety of protein kinases, including
the NF-
B-inducing kinases NKB-inducing kinase, receptor
interacting protein 1, and caspase-recruiting domain (containing
IL-1ß converting enzyme-associated kinase), as well as the c-Jun
N-terminal kinase (JNK) pathway activators mitogen-activated protein
(MAP) kinase kinase kinase-1, apoptosis signal-regulating kinase-1, and
the germinal center kinase-related kinase (GCKRK; Refs.
11, 12, 13, 14, 15, 16, 17). Thus, TRAFs physically and functionally connect
TNF family cytokine receptors to intracellular protein kinases, thereby
linking these receptors to downstream signaling pathways.
The in vivo roles of TRAFs are beginning to be elucidated through the generation of transgenic mice that overexpress these proteins or dominant-negative mutants of them, as well as through targeted gene ablation (knockouts) in mice (18, 19, 20, 21, 22, 23, 24). However, to date, little is known about the normal patterns of TRAF expression in lymphoid, hemopoietic, and other types of tissues in vivo. Moreover, little effort has been made thus far to examine how the expression of TRAFs might change in association with malignant transformation, an issue of potential importance for better understanding host immune responses to tumors.
Previously, we generated specific Abs to TRAF3 and TRAF4, defining their normal patterns of expression in human tissues in vivo (25, 26). In this report, we identified Abs that selectively recognize TRAF1, TRAF2, TRAF5, or TRAF6, and used these reagents for assessing the in vivo locations of expression of these TRAF family proteins in normal human and murine tissues. We also characterized the expression of TRAF1, TRAF2, TRAF5, and TRAF6 in 83 human tumor cell lines, and found evidence of elevations of TRAF1 expression in non-Hodgkin lymphomas (NHLs) and B cell chronic lymphocytic leukemia (B-CLL). The findings lay the foundation for a more complete understanding of the diversity of cellular responses to the TNF/Toll family receptors, which rely on TRAF family proteins for their signal transduction mechanisms.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Polyclonal antisera were generated in rabbits using synthetic peptides or recombinant protein immunogens. Unless otherwise specified, peptides were synthesized with an N-terminal cysteine appended to permit conjugation to maleimide-activated carrier proteins keyhole limpet hemocyonin KLH and OVA (Pierce, Rockford, IL) as described previously (27) and with a C-terminal amide (NH2) rather than free carboxylic acid. Among the peptides used as immunogens were: MASSSGSSPRPAPDENEFPFGC, corresponding to residues 1–22 of human TRAF1; CRADNLHPVSPGSPLTQEK, representing aa 51–69 of murine TRAF1; CVTPPGSLELLQPGFSKTLLGTKLEAK, corresponding to aa 6–31 of human TRAF2; CSFLEAQASPGTLNQVGPELLQR, representing residues 250–271 of murine TRAF2; CMAHSEEQAAVPCAFIRQNSG, corresponding to aa 1–20 of murine TRAF5; and CDAKPELLAFQRPTIPRNPK, representing residues 451–469 of human and murine TRAF6. Antipeptide antiserum specific for TRAF3 and TRAF4 have been described (25, 26).
An additional anti-TRAF3 serum was produced using recombinant human TRAF3 (residues 341–568)-His6 protein. Recombinant TRAF3 341–568(341–568) was produced as a fusion protein with a C-terminal his6 tag. This protein was expressed in BL21 (DE3) cells by induction with 1 mM isopropyl ß-D-thiogalactoside. Following cell growth and lysis, the clarified cell lysate was applied to a nickel-nitrilotriacetic acid column (Qiagen, Valencia, CA) and eluted with an imidazole gradient. The pooled TRAF3 fractions were dialyzed against 50 mM Tris at pH 8.8 and applied to a 10/10 fast protein liquid chromatography mono Q column (Pharmacia, Piscataway, NJ) and eluted with an NaCl gradient.
New Zealand White female rabbits were injected s.c. with a mixture of 0.25 ml KLH peptide (1 mg/ml), 0.25 ml OVA peptide (1 mg/ml), or recombinant protein (0.1–0.25 µg protein/immunization) and 0.5 ml Freund’s complete adjuvant (dose divided over 10 injection sites) and then boosted three times at weekly intervals followed by another 3–20 boostings at monthly intervals with 0.25 mg each of KLH peptide, OVA peptide, or recombinant protein immunogens in Freund’s incomplete adjuvant, before collecting blood and obtaining immune serum. Commercially available Abs for TRAF1 (H-132), TRAF2 (H-20), TRAF5 (H-257), and TRAF6 (H-274) (Santa Cruz Biotechnology, Santa Cruz, CA) were purchased for comparison and used for some experiments as indicated below.
Tissues and patient specimens
Normal tissues for immunohistochemical analysis were derived
either from human biopsy and autopsy material (n 
3)
or from adult mice of various strains. Only tissues that appeared to be
histologically free of disease were used. These tissues were fixed in
either neutral buffered formalin, zinc-buffered formalin (Z-fix;
Anatech, Battle Creek, MI), B5, or Bouin’s solution (Sigma, St. Louis,
MO), and embedded in paraffin. For part of the analysis, we constructed
a tissue microarray containing 130 specimens, representing 0.6- or 1-mm
(diameter) cylindrical cores acquired from paraffin blocks of normal
human or mouse tissues, which were sectioned at 4- to 5-µm thickness
(28). In addition to normal tissues, archival paraffin
blocks were obtained for 236 NHL specimens, representing patients’
materials from several Eastern Cooperative Oncology Group
(ECOG)-sponsored clinical trials attained through protocol E6491.
Histological diagnosis was confirmed by review by an expert
hemopathologist before immunostaining, according to the National Cancer
Institute (NCI) Working Formulation (29), then converted
to the updated Revised European-American Lymphoids neoplasm
classification where possible (30).
Immunohistochemistry
Tissue sections were immunostained using a
diaminobenzidine-based detection method as described in detail, using
either an avidin-biotin complex reagent (Vector Laboratories,
Burlingame, CA) or the Envision-Plus-HRP system (Dako, Carpinteria, CA)
using an automated immunostainer (Dako Universal Staining System)
(25, 26, 31, 32, 33). The dilutions of anti-TRAF
antiserum typically used were 1:1500 (v/v) for anti-TRAF1, 1:3000
for TRAF2, 1:500 for anti-TRAF5, and 1:1000 for anti-TRAF6.
Nuclei were counterstained with either hematoxylin or methyl green. For
all normal tissues examined, the immunostaining procedure was performed
in parallel using preimmune serum to verify the specificity of the
results. Initial confirmations of Ab specificity also included
experiments in which antiserum was preadsorbed with 5–10 µg/ml of
either synthetic peptide immunogen or recombinant protein immunogen.
The immunostaining results were arbitrarily scored according to
intensity as 0, negative; 1+, weak;
2+, moderate; and 3+,
strong. Results presented for each normal tissue were based on
immunohistochemical analysis of multiple immunostained slides
(n 3 for each tissue). For lymphomas, samples were
additionally scored for percentage of immunopositive malignant cells,
estimating the percentage in increments of 10% (0%, 10%, 20%, 30%,
etc.) in overview and from a minimum of five representative high power
fields. Comparisons were made with the germinal centers (GCs) from
non-neoplastic lymph nodes (n = 9).
In vitro translations
TRAF1, TRAF2, TRAF3, TRAF4, TRAF5, and TRAF6 cDNAs subcloned
into either pKSII-Bluescript (Stratagene, La Jolla, CA) or pcDNA-3
(Invitrogen, San Diego, CA) were transcribed and translated in vitro
using 1 µg plasmid DNA, 25 µl TNT-reticulocyte lysates and either
T3 or T7 RNA polymerase (Promega, Madison, WI) according to the
manufacturer’s protocol in the presence of
35S-labeled L-methionine (1
mCi/mmol; Amersham, Arlington Heights, IL).
Immunoblotting
Whole cell lysates were prepared from frozen human tissues obtained at autopsy, as described previously (25, 26, 33). For analysis of cultured cell lines, cells were lysed in RIPA solution (25 mM Tris, pH 7.2, 150 mM NaCl, 5 mM EDTA, 1% sodium deoxycholate, 0.1% SDS, and 1% Triton X-100) containing protease inhibitors (1 mM PMSF, 0.28 trypsin inhibitor units/ml aprotinin, 50 µg/ml leupeptin, 1 mM benzamidine, and 0.7 µg/ml pepstatin). Cell and tissue lysates were normalized for total protein content (50 µg/lane) and subjected to SDS-PAGE/immunoblot analysis using 1:500–1:2000 (v/v) dilutions of antisera and secondary HRP-conjugated goat anti-rabbit Ab (1:3000 v/v dilution; Bio-Rad, Richmond, CA), with detection accomplished using an enhanced chemiluminescence (ECL; Amersham-based) multiple Ag detection immunoblotting method that allows for multiple reprobings of blots without Ab stripping, as described previously (25, 34).
For immunoblot data qualifications, TRAF cDNAs were in vitro translated in the presence of [35S]methionine and analyzed by SDS-PAGE. The region of the gel containing each labeled TRAF protein was sliced, and the radioactivity contained in the band was quantified in a scintillation counter. The amount of TRAF protein contained in the in vitro translation mixture was determined according to the radioactivity incorporated into the protein and the number of methionines contained by each TRAF protein. Known amounts of each in vitro translated TRAF (5–50 µg) were used to create standard curves in immunoblot assays where the ECL-based Ag detection method was used. ECL data on x-ray films were quantified by scanning densitometry using the IS-1000 image analysis system (Alpha Innotech, San Leandro, CA), and the results from the in vitro translated protein standard-containing blot were used to normalize all data and thus estimate the nanograms of TRAF proteins per milligram of total cellular protein. Data from two independent standard-containing blots were within 20% agreement.
Cell isolation and culture
PBLs were isolated from the heparinized blood of normal
volunteers and cultured in the presence of anti-CD3 (OKT3) Ab, PHA,
and IL-2 as described in detail previously (35). Normal
human GC B lymphocytes were isolated from tonsils by a method involving
magnetic microbead-based immunodepletion with anti-IgD,
anti-CD39, anti-CD3, and anti-CD14 (36). B
cells were cultured in RPMI/10% FBS medium with or without 100 ng/ml
of recombinant CD40 ligand (CD40L; Immunex, Seattle, WA).
Peripheral blood B cells were isolated from peripheral blood using
anti-CD19-conjugated magnetic beads (Dynal, Lake Success, New York)
according to the manufacturer’s instructions. B-CLLs were isolated as
described (37). All CLL specimens contained 95%
leukemic cells as determined by FACS analysis using anti-CD19 and
anti-CD5 Abs. Of the 49 leukemic specimens, 21 represented
previously untreated patients and 28 were derived from B-CLL patients
with refractory disease.
Established tumor cell lines were cultured in either RPMI or DMEM with 5 or 10% FBS (38). These cell lines included the NCI panel of 60 human tumor cell lines (39) plus 23 additional human tumor lines maintained in our laboratory.
Statistical analysis
Comparisons of protein expression data derived from the NCI panel of 60 human tumor cell lines were made with other biomarker data using Spearman and Pearson test statistics for nonbinary patterns and Fisher’s exact test for pairs of binary data. In both cases, p values were calculated from two-tailed t-distributions and adjusted using a Bonferroni correction for multiple correlations (39). Correlations of the expression patterns of the different TRAF proteins with responses of these cell lines to various cytotoxic drugs and compounds can be found at http://dtp.nci.nih.gov. TRAF1 protein levels in B-CLLs were compared by an unpaired t test, using immunoblot data that had been quantified by previously described methods (37, 38, 40).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Antisera were raised in rabbits using synthetic peptides and
recombinant proteins representing fragments of TRAF1, TRAF2, TRAF5, or
TRAF6 as immunogens. To explore the specificity of these antisera,
immunoblot analysis was performed using in vitro translated TRAF1,
TRAF2, TRAF3, TRAF4, TRAF5, and TRAF6. The antisera raised against
human TRAF1 aa 1–22 (Bur32), human TRAF2 aa 6–31 (Bur 34), murine
TRAF5 aa 1–20 (1847), as well as human and murine TRAF6 aa 451–469
(Bur30) were determined to react specifically with the intended TRAF
proteins, and did not cross-react with other members of the family
(Fig. 1
). Additional commercially
available antisera for TRAF1 (H-132), TRAF2 (SC C20), TRAF5 (SC7220),
and TRAF6 (H-274) were also determined to bind selectively to their
intended TRAFs, based on immunoblot analysis of in vitro translated
TRAF1 through TRAF6 (Fig. 1). The specificity of these anti-TRAF
antisera was also confirmed through experiments where epitope-tagged
versions of TRAF1, TRAF2, TRAF5, or TRAF6 were transiently expressed in
293T cells and lysates from the transfected cells were subjected to
SDS-PAGE/immunoblot analysis (data not shown).
|
Detergent lysates were prepared from normal tissues, normalized
for total protein content, and expression of TRAFs was analyzed by
SDS-PAGE/immunoblotting using antisera with specificity for TRAF1,
TRAF2, TRAF5, or TRAF6. These experiments revealed tissue-specific
differences as well as some similarities in the relative steady-state
levels of these TRAF family members in human tissues. For example, the
most abundant amounts of the TRAF1, TRAF2, and TRAF6 proteins were
detected in thymus, testis, and epidermis (Fig. 2). By comparison, TRAF1 and TRAF6 were
far less prevalent (as a percentage of total proteins) in most other
tissues, whereas TRAF2 was more widely expressed and was found at
levels that were within a few fold of those seen in the high-expressing
tissues. Levels of TRAF5 were highest in thymus and epidermis.
|
, the expected endogenous 50- to 53-kDa TRAF1,
60- to 64-kDa TRAF2, 64-kDa TRAF5, and 66-kDa TRAF6 proteins
were the most abundant bands detected in SDS-PAGE/immunoblot assays.
However, additional minor bands of lower molecular mass were seen in
some cases, presumably representing partial degradation products of
these TRAFs, which may arise during tissue harvesting or processing
(Fig. 2). Higher molecular mass forms of TRAF2 arise from
posttranslational modifications, including ubiquitination (data not
shown). Rare isoforms of TRAFs arising from alternative mRNA splicing
cannot be excluded. Similar results were obtained in those cases where
two or more TRAF-selective Abs were available (data not shown), further
confirming the validity of these results. Immunohistochemical localization of TRAFs in tissue sections
Using antisera specific for TRAF1, TRAF2, TRAF5, or TRAF6, the in
vivo patterns of expression of these proteins were examined in normal
human or murine tissues by immunohistochemistry methods. Overall, the
cell type-specific patterns of expression of these four TRAF family
members were strikingly different, indicating that they are all
independently regulated to a large extent. Table I presents a comprehensive summary of the
immunostaining results for TRAF1, TRAF2, TRAF5, and TRAF6, and also
includes for comparison our previously published observations for TRAF3
and TRAF4 (25, 26). In the vast majority of instances,
immunostaining for TRAFs was confined to the cytosol, though
occasionally nuclear immunostaining was also found (Table I).
|
|
|
presents some examples of TRAF1,
TRAF2, TRAF5, and TRAF6 immunostaining data for the lymphoid and
hemopoietic organs of lymph node, spleen, thymus, and bone marrow.
These tissues are emphasized because many of the TNF family cytokine
receptors to which TRAFs bind are prominently involved in the
physiology of these organs. Note that although several TRAFs were
expressed in the GC lymphocytes of lymph nodes (all except TRAF4),
small resting lymphocytes in the interfollicular regions of nodes were
typically immunonegative for all TRAFs (Fig. 3; Table I). Thus,
expression of TRAFs in lymphocytes may be dynamically regulated in vivo
with changes in lymphocyte activation. Interestingly, TRAF6 was the
only TRAF family member expressed at immunodetectable levels in
cortical thymocytes, whereas TRAF3 was the only member detected in
medullary thymocytes (Fig. 3; Table I). Thus, TRAF expression may be
modulated during thymocyte differentiation. Differences in the
intensity of TRAF immunoreactivities in myeloid progenitor cells and
mature granulocytes in bone marrow similarly suggest fluctuations in
the expression of these proteins during hemopoietic differentiation.
Additional details are provided in Table I. It should be noted that
because of the nonquantitative nature of immunohistochemistry,
these data should be interpreted as only approximations of the
relative amounts of these TRAF family proteins in various types of
cells in vivo.
|
To explore whether changes in TRAF family protein might occur
during lymphocyte activation, PBLs (representing mostly T cells) were
cultured for various times with or without PHA and IL-2 or with OKT3
monoclonal anti-CD3 Ab. As shown in Fig. 4A, unstimulated PBLs
contained little or no detectable TRAF1, TRAF2, TRAF5, or TRAF6 when
initially placed into culture and during up to 6 days of culture in the
absence of mitogens or lymphokines (with the exception of TRAF6, where
a slight increase in protein levels was also detected after several
days of culture without stimulus). In contrast, marked increases in all
of these TRAF family proteins were induced by anti-CD3 Ab and by
the combination of PHA and IL-2. However, note that because elevations
in TRAFs occurred relatively late in these cultures, it is unlikely
that PHA, anti-CD3, or IL-2 are directly responsible.
|
B). Probing the same blots with Abs specific for
other proteins, such as caspase-10 and poly(ADP-ribose) polymerase,
demonstrated the specificity of these CD40L-induced changes in TRAF
expression and also confirmed loading of approximately equal amounts of
total protein in all lanes (data not shown). Analysis of TRAF protein levels in human tumor cell lines
Lysates were prepared from 83 human tumor cell lines, representing
a wide range of malignancies, including the entire NCI 60 cell line
anticancer drug screening panel (38). After normalization
for total protein content (50 µg/lane), samples were subjected to
immunoblot analysis using various anti-TRAF Abs. Fig. 5 shows a representative example of some
immunoblot data obtained for TRAF1, TRAF2, TRAF3, TRAF5, and TRAF6.
Though the most abundant bands seen in immunoblot analyses corresponded
to the anticipated molecular masses for each of these TRAF family
members, in some cases additional bands were also observed in lesser
amounts, possibly representing posttranslationally modified
versions of TRAFs, low abundance isoforms arising from alternative
mRNA splicing, or partially degraded proteins.
|
summarizes the results, expressing the
data as nanograms of TRAF protein per milligram of total protein.
TRAF2, TRAF3, TRAF5, and TRAF6 were widely expressed among various
types of cancer cell lines, although the levels of TRAF2 and TRAF3
tended to be higher by 2- to 3-fold overall compared with TRAF5 and
TRAF6. In contrast, expression of TRAF1 was far more restricted. The
most consistent and highest levels of expression of TRAF1 were found in
B cell lines. Indeed, all B cell lines tested contained detectable
levels of TRAF1, including the B cell lymphoma lines RL, Raji, DND39,
MuTu, SU-DHL.1, and RS11846, the lymphoblastoid B cell line BJAB, and
the pre-B cell leukemia 380, which arose from an antecedent non-Hodgkin
B cell lymphoma (41). The pre-B cell line 697 and all T
cell lines tested failed to express TRAF1, as did all myeloid leukemia
lines tested except the SR line. However, the AML2 cell line did
contain a smaller molecular mass anti-TRAF1 reactive protein, which
remains to be characterized.
|
TRAF1 expression in NHLs
The abundant expression of TRAF1 in B cell lines prompted us to
explore the expression of this TRAF family member in NHLs. Patient
specimens used for these studies represented archival paraffin blocks
obtained from a variety of ECOG trials. A total of 232 lymphoma
specimens were immunostained using anti-TRAF1 Ab, and the
immunostaining results were scored with respect to the percentage of
TRAF1-immunopositive neoplastic cells (0–100%). Based on comparisons
with non-neoplastic nodes, the presence of 30% immunopositive
malignant lymphocytes was chosen as a cut-off for dichotomizing
immunostaining data into positive vs negative groups. Using this
method, 112 of 232 (48%) lymphomas were determined to be TRAF1
positive, having percentages of TRAF1 positive exceeding those found in
non-neoplastic reactive lymph node biopsies (n = 6).
Fig. 7 shows some representative examples
of TRAF1 immunostaining in a reactive node compared with lymphomas,
including a TRAF1-negative follicular small cleaved-cell lymphoma
(FSCCL), a TRAF1-positive follicular mixed small/large cell lymphomas,
and a TRAF1-positive diffuse large cell lymphoma (DLCL). In normal and
reactive nodes, note that although follicular dendritic cells and
extrafollicular histiocytes were stained intensely for TRAF1, only
occasional large (activated) GC lymphocytes were TRAF1 immunopositive.
Similarly, only occasional malignant B cells were stained for TRAF1 in
the FSCCL specimen shown in Fig. 7, though TRAF1-positive dendritic
cells and histiocytes were prevalent in the same tissue section. In
contrast, a substantial proportion of the malignant lymphocytes was
strongly stained for TRAF1 in the follicular mixed small/large cell
lymphomas and DLCL specimens presented (Fig. 7).
|
summarizes the TRAF1
immunostaining results, correlating the TRAF1 data with lymphoma
histology. Because of the archival nature of these specimens, outdated
methods for histopathological categorization were used (mostly NCI
working formulation) at the time these specimens were collected and
entered into ECOG databases. Thus, only tentative conclusions can be
drawn with respect to histology. However, approximately equivalent
proportions of lymphomas with nodular (follicular) (51/107; 48%) vs
diffuse (61/125; 49%) cytoarchitecture were TRAF1 immunopositive.
Comparisons of lymphomas according to cell size (small cells vs large
cells) suggested a greater tendency of large cell lymphomas to be TRAF1
immunopositive than small cell lymphomas (73/133 (55%) large cell vs
19/59 (32%) small cell) (p = 0.004). However,
using the histopathological diagnoses to correlate TRAF1 immunostaining
with lymphoma cell size may have underestimated these differences,
because in 31 of 40 (78%) mixed small and large cell lymphomas,
TRAF1-immunopositive cells nevertheless were predominantly larger
malignant cells (Fig. 7). Altogether, TRAF1 expression appears to be
up-regulated compared with normal nodes in roughly half of
NHLs.
|
TRAF1 expression in B-CLL specimens was also evaluated using
immunoblotting methods (37), including 21 previously
untreated and 28 treatment-refractory patients. Comparisons were made
with purified normal peripheral blood B lymphocytes (n
= 4). Though a limited comparison, TRAF1 protein levels appeared to be
higher in many B-CLLs than in normal B cells (Fig. 8). Moreover, treatment-refractory B-CLLs
contained significantly higher levels of TRAF1 compared with previously
untreated B-CLLs (p = 0.03 by unpaired
t test) or normal peripheral blood B lymphocytes
(p = 0.01). Thus, TRAF1 protein levels may be
pathologically elevated in some B-CLLs, with higher TRAF1 possibly
representing a progression event or adaptation associated with
resistance to chemotherapy.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One of the reasons why differences in the expression of combinations of
TRAF family proteins can have important implications relates to
hetero-oligomerization among TRAFs. In particular, interactions of some
TRAF family members with certain TNF family receptors are dependent on
hetero-oligomerization with other TRAFs. For example, although TRAF1
can directly interact with the cytosolic domain of CD30, it does not
directly bind the cytosolic domains of TNFR1, TNFR2, and CD40,
depending instead on recruitment to these receptors through association
with TRAF2 (5, 6, 8, 9, 42). Thus, TRAF1 depends on TRAF2
for its association with certain TNF family receptors. Moreover,
TRAF1/TRAF2 heterodimers have been shown to recruit the antiapoptotic
proteins cIAP1 and cIAP2 to TNFR1 and TNFR2 receptors
(43). The simultaneous combination of TRAF1, TRAF2, cIAP1,
and cIAP2 protein expression has been reported to be required for
optimal suppression of caspase-8 activation induced by TNFR1, thus
preventing TNFR1-induced apoptosis (44). Thus, it may be
of relevance to mechanisms of apoptosis suppression in the context of
responses to TNF-
that several types of cells appear to express
TRAF1 in the absence of TRAF2 (based on immunohistochemical analysis of
normal tissues), including dendritic cells in nodes, thymus, and
spleen, as well as megakaryocytes, smooth muscle cells, neurons in
peripheral autonomic ganglia, and selected types of epithelial cells
(Table I). Similarly, TRAF5 directly binds the cytosolic domain LTßR
(45), but relies on hetero-oligomerization with TRAF3 for
its association with CD40 (8). The interdependence of
TRAF3 and TRAF5 coexpression is underscored by the observation that
TRAF5 can activate kinases involved in NF-B induction and JNK
activation, whereas TRAF3 does not (45, 46). Thus,
CD40-expressing cells that contain TRAF5 in the absence of TRAF3
presumably cannot use TRAF5 for CD40 signal transduction responses.
Examples of TRAF5 expression in the absence of immunodetectable TRAF3
were found by immunohistochemical analysis in interdigitating reticulum
cells of nodes (paracortex), thymus, and spleen; follicular dendritic
cells of GCs; as well as in macrophages and histiocytes, all cell types
that are known to express CD40 and respond to CD40L. Thus, one presumes
that these types of cells rely upon other NF-B- and JNK-activating
TRAF family members, besides TRAF5, which are known to bind the
cytosolic domain of CD40.
Our studies with isolated PBLs and GC B cells suggest that expression of TRAFs can be dynamically regulated in response to various stimuli that induce lymphocyte activation, proliferation, or differentiation. Though little or no TRAF1, TRAF2, TRAF3, TRAF5, or TRAF6 protein was found in circulating resting PBLs, expression of all of these TRAF family members was induced by TCR complex ligation using anti-CD3 Ab or other T cell mitogens. Because TRAFs participate in signaling by a variety of TNF family receptors (CD27, CD30, and OX4B), which provide costimulatory functions in T cell responses (47, 48, 49, 50), the inducibility of TRAF expression suggests a mechanism for ensuring that only Ag-stimulated T cells are competent to respond to costimulatory signals for TNF family cytokines. In this regard, in vitro studies of mature T cells isolated from TRAF knockout mice indicate that TRAF3 is essential for proliferative responses of T cells, whereas TRAF6 is not (20, 23). Moreover, CD28-independent costimulation of resting T cells by the TNFR family member 4-1BB has been demonstrated to require TRAF2 (51). Thus, a critical role exists for TCR-inducible expression of certain TRAF family members in T cell-proliferative responses. Furthermore, poststimulatory degradation of TRAFs may help to ensure that T cells are competent to respond to costimulatory TNF family ligands during only a brief window, as has been documented recently for the case of CD30-dependent degradation of TRAF2 (52).
In contrast to peripheral blood T cells, GC B cells constitutively
expressed TRAF2, TRAF3, TRAF5, and TRAF6. Previous studies have shown
that GC B cells do not express TRAF4 (26). Thus, these
cells, which respond to CD40, possess a diversity of TRAFs, suggesting
that they are poised to rapidly respond to CD40L within the
microenvironment of the GC. Among the five TRAF family proteins
surveyed, only TRAF1 was induced by CD40L. Recent studies have
demonstrated that TNF- and CD40L induce TRAF1 mRNA expression
(53), consistent with our results.
Many TNF family receptors are expressed on malignant cells, providing
opportunities for modulating tumor cell growth, Ag-presenting capacity,
and sensitivity to apoptotic agents such as anticancer drugs. TNF
derives its name from early investigations of this cytokine that
suggested that it might be selectively cytotoxic to tumor cells,
sparing normal cells. Differences in sensitivity to TNF--induced
cell death have been linked to TRAF-dependent signal transduction
responses, including NF-B-dependent and -independent mechanisms
(54, 55, 56). Thus, alterations in the levels and ratios of
various TRAF family proteins can translate into differences in tumor
cell sensitivity to TNF-mediated cytotoxicity. For example, TRAF1 and
TRAF2, in combination, reportedly promote resistance to TNF-induced
apoptosis at a proximal step, squelching activation of the apical
caspase in the death receptor pathway, namely, pro-caspase-8
(44). Thus, elevated levels of TRAF1 or TRAF2 might
promote tumor cell resistance to TNF- and related death ligands such
as Apo3L (DR3-Ligand), whose receptors use very similar signaling
mechanisms (57). Conversely, overexpression of TRAF3 can
interfere with NF-B induction by TNFR1, CD40, and TRAF2, in at least
some circumstances (58), suggesting that reductions in
TRAF3 levels might provide tumor cells with improved resistance to
TNF-mediated cytotoxicity. Similarly, TRAF3 has been implicated as an
important mediator of the growth-suppressing and apoptosis-sensitizing
effects of LTßR and CD40 on epithelial cancer cells
(59, 60, 61). However, because NF-B and other
TRAF-activated signaling pathways also can up-regulate expression of
molecules involved in Ag presentation, promoting NF-B induction may
have disadvantages in terms of tumor avoidance of immune recognition.
The immunoblot analysis of TRAF protein levels in 83 tumor cell lines,
when compared with the patterns of TRAF expression observed in normal
tissues by immunohistochemical methods, suggests some possible
tumor-specific alterations in TRAF family protein expression. For
instance, though TRAF1 was not present at immunodetectable levels in
normal mammary epithelium, some breast cancers contained abundant
amounts of TRAF1 protein. Similarly, although TRAF2 was not detectable
by immunostaining in normal prostatic epithelium, most prostate cancer
cell lines contained high levels of TRAF2 protein. Conversely, although
normal colonic epithelial cells were uniformly TRAF3 immunopositive,
most colon cancer cell lines lacked expression of TRAF3, suggesting
cancer-associated down-regulation of TRAF3 expression.
Examination of TRAF1 expression in lymphomas suggests that striking
up-regulation occurs commonly in these neoplasms, particularly during
transformation of the malignant cells to large cell disease. Although
normal lymphocytes can be induced to express TRAF1, immunohistochemical
analysis of normal lymph nodes and other peripheral lymphoid organs
suggests that relatively small percentages of normal (presumably
activated) lymphocytes express this particular member of the TRAF
family in vivo. In contrast, much higher percentages of malignant
lymphoma cells commonly expressed TRAF1, implying a deregulation in
TRAF1 expression. B cell lymphoma cell lines were also the most
consistent expressers of TRAF1 among the 83 human tumor cell lines
examined in vitro, suggesting that TRAF1 overexpression does not
require unique microenvironments found in lymphoid organs in vivo and
implying autonomous deregulated expression of this TRAF family member
in these malignancies. Furthermore, circulating B-CLLs also commonly
contained high levels of TRAF1, and recently it has been reported that
Reed-Sternberg cells of Hodgkin’s disease and EBV-transformed lymphoid
cells also express high levels of TRAF1 (62). The
molecular explanation for the aberrant expression of TRAF1 in lymphomas
and B-CLLs remains to be revealed. The chromosomal locus where the
TRAF1 gene resides in humans is not frequently involved in
chromosomal translocations or other cytogenetic abnormalities typically
seen in lymphomas or leukemias (63). The promoter of the
TRAF1 gene contains several NF-B binding sites and is
highly inducible by NF-B (53). Thus, deregulation of
signal transduction pathways that control NF-B represents at least
one hypothetical mechanism by which TRAF1 overexpression could occur in
lymphomas and B-CLLs. Though the functional significance of
constitutively elevated levels of TRAF1 remains to be fully elucidated,
T cells of transgenic mice engineered to overexpress TRAF1 display
resistance to Ag-induced apoptosis in vitro and in vivo
(18). Thus, elevated TRAF1 levels may endow malignant
lymphocytes with enhanced protection from apoptosis induced either
intrinsically by Ag receptor-mediated induction of TNF family death
receptors and ligands (18) or extrinsically by other
immune effector cells. Future studies of TRAF1 expression in cohorts of
lymphoma and B-CLL patients receiving uniform therapy are needed to
establish whether expression of this TRAF family member is of
prognostic significance with respect to clinical outcome.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. John C. Reed, The Burnham Institute, 10901 North Torrey Pines Road, La Jolla, CA 92037.
3 Abbreviations used in this paper: TRAF, TNFR-associated factor; DLCL, diffuse large cell lymphoma; NHL, non-Hodgkin lymphoma; B-CLL, B cell chronic lymphocytic leukemia; CD40L, CD40 ligand; FSCCL, follicular small cleaved-cell lymphoma; LTßR, lymphotoxin-ß receptor; JNK, c-Jun N-terminal kinase; ECOG, Eastern Cooperative Oncology Group; ECL, enhanced chemiluminescence; GC, germinal center; NCI, National Cancer Institute; KLH, keyhole limpet hemocyonin.
Received for publication March 13, 2000. Accepted for publication August 9, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B and JNK activation. J. Biol. Chem. 274:22414.B activation in CD27 signaling: involvement of TNF receptor-associated factors in its signaling and identification of functional region of CD27. J. Immunol. 161:4753.B and c-jun N-terminal kinase (JNK/SAPK) pathways at TNF receptor-associated factor 2. Proc. Natl. Acad. Sci. USA 94:9792.B activation and regulates lymphocyte proliferation and survival. Immunity 7:703.[Medline]
B activation, and increased sensitivity to TNF-induced cell death in TRAF2-deficient mice. Immunity 7:715.[Medline]
B antiapoptosis: induction of TRAF1 and TRAF2 and c-IAP1 and c-IAP2 to suppress capase-8 activation. Science 281:1680.B and putative signal transducer for the lymphotoxin-ß receptor. J. Biol. Chem. 271:14661.B activation. J. Biol. Chem. 272:2042.B and c-Jun N-terminal kinase. J. Biol. Chem. 274:19368.B pathways. J. Exp. Med. 188:1381.B, by the tumor necrosis factor (TNF) receptor 1 through a TNF receptor-associated factor 2- and germinal center kinase related-dependent pathway. J. Biol. Chem. 272:32102.B by TNF receptor 2 and CD40. Science 269:32767.
B. Proc. Natl. Acad. Sci. USA 94:2460.This article has been cited by other articles:
|
|
 |
|
  S. Kitada, C. L. Kress, M. Krajewska, L. Jia, M. Pellecchia, and J. C. Reed Bcl-2 antagonist apogossypol (NSC736630) displays single-agent activity in Bcl-2-transgenic mice and has superior efficacy with less toxicity compared with gossypol (NSC19048) Blood, March 15, 2008; 111(6): 3211 - 3219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. K. Oyoshi, P. Bryce, S. Goya, M. Pichavant, D. T. Umetsu, H. C. Oettgen, and E. N. Tsitsikov TNF Receptor-Associated Factor 1 Expressed in Resident Lung Cells Is Required for the Development of Allergic Lung Inflammation J. Immunol., February 1, 2008; 180(3): 1878 - 1885. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Cerhan, S. M. Ansell, Z. S. Fredericksen, N. E. Kay, M. Liebow, T. G. Call, A. Dogan, J. M. Cunningham, A. H. Wang, W. Liu-Mares, et al. Genetic variation in 1253 immune and inflammation genes and risk of non-Hodgkin lymphoma Blood, December 15, 2007; 110(13): 4455 - 4463. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Kuwahara, E. P. Lillehoj, T. Koga, Y. Isohama, T. Miyata, and K. C. Kim The Signaling Pathway Involved in Neutrophil Elastase Stimulated MUC1 Transcription Am. J. Respir. Cell Mol. Biol., December 1, 2007; 37(6): 691 - 698. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Zhang, Z. Wang, T. Li, E. N. Tsitsikov, and H.-F. Ding NF-{kappa}B2 mutation targets TRAF1 to induce lymphomagenesis Blood, July 15, 2007; 110(2): 743 - 751. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Zhou, H. Ye, J. I. Martin-Subero, S. Gesk, R. Hamoudi, Y.-J. Lu, R. Wang, J. Shipley, R. Siebert, P. G. Isaacson, et al. The pattern of genomic gains in salivary gland MALT lymphomas Haematologica, July 1, 2007; 92(7): 921 - 927. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Wullaert, G. van Loo, K. Heyninck, and R. Beyaert Hepatic Tumor Necrosis Factor Signaling and Nuclear Factor-{kappa}B: Effects on Liver Homeostasis and Beyond Endocr. Rev., June 1, 2007; 28(4): 365 - 386. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Sabbagh, C. C. Srokowski, G. Pulle, L. M. Snell, B. J. Sedgmen, Y. Liu, E. N. Tsitsikov, and T. H. Watts A critical role for TNF receptor-associated factor 1 and Bim down-regulation in CD8 memory T cell survival PNAS, December 5, 2006; 103(49): 18703 - 18708. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. C. Davies, T. W. Mak, L. S. Young, and A. G. Eliopoulos TRAF6 Is Required for TRAF2-Dependent CD40 Signal Transduction in Nonhemopoietic Cells Mol. Cell. Biol., November 15, 2005; 25(22): 9806 - 9819. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Yang, J. L. Wiser, J. J. Peairs, J. N. Ebright, Z. J. Zavodni, C. Bowes Rickman, and G. J. Jaffe Human RPE Expression of Cell Survival Factors Invest. Ophthalmol. Vis. Sci., May 1, 2005; 46(5): 1755 - 1764. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. S. Gricks, D. Zahrieh, A. J. Zauls, G. Gorgun, D. Drandi, K. Mauerer, D. Neuberg, and J. G. Gribben Differential regulation of gene expression following CD40 activation of leukemic compared to healthy B cells Blood, December 15, 2004; 104(13): 4002 - 4009. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Zapata, M. Krajewska, H. C. Morse III, Y. Choi, and J. C. Reed TNF receptor-associated factor (TRAF) domain and Bcl-2 cooperate to induce small B cell lymphoma/chronic lymphocytic leukemia in transgenic mice PNAS, November 23, 2004; 101(47): 16600 - 16605. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. J. Weston, B. Austen, W. Wei, E. Marston, A. Alvi, S. Lawson, P. J. Darbyshire, M. Griffiths, F. Hill, J. R. Mann, et al. Apoptotic resistance to ionizing radiation in pediatric B-precursor acute lymphoblastic leukemia frequently involves increased NF-{kappa}B survival pathway signaling Blood, September 1, 2004; 104(5): 1465 - 1473. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. So, S. Salek-Ardakani, H. Nakano, C. F. Ware, and M. Croft TNF Receptor-Associated Factor 5 Limits the Induction of Th2 Immune Responses J. Immunol., April 1, 2004; 172(7): 4292 - 4297. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Tracey, R. Villuendas, A. M. Dotor, I. Spiteri, P. Ortiz, J. F. Garcia, J. L. R. Peralto, M. Lawler, and M. A. Piris Mycosis fungoides shows concurrent deregulation of multiple genes involved in the TNF signaling pathway: an expression profile study Blood, August 1, 2003; 102(3): 1042 - 1050. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G Siegler, E Kremmer, R Gonnella, and G Niedobitek Epstein-Barr virus encoded latent membrane protein 1 (LMP1) and TNF receptor associated factors (TRAF): colocalisation of LMP1 and TRAF1 in primary EBV infection and in EBV associated Hodgkin lymphoma Mol. Pathol., June 1, 2003; 56(3): 156 - 161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. G. Eliopoulos, E. R. Waites, S. M. S. Blake, C. Davies, P. Murray, and L. S. Young TRAF1 Is a Critical Regulator of JNK Signaling by the TRAF-Binding Domain of the Epstein-Barr Virus-Encoded Latent Infection Membrane Protein 1 but Not CD40 J. Virol., December 20, 2002; 77(2): 1316 - 1328. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Munzert, D. Kirchner, H. Stobbe, L. Bergmann, R. M. Schmid, H. Dohner, and H. Heimpel Tumor necrosis factor receptor-associated factor 1 gene overexpression in B-cell chronic lymphocytic leukemia: analysis of NF-kappa B/Rel-regulated inhibitors of apoptosis Blood, November 15, 2002; 100(10): 3749 - 3756. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Arron, Y. Pewzner-Jung, M. C. Walsh, T. Kobayashi, and Y. Choi Regulation of the Subcellular Localization of Tumor Necrosis Factor Receptor-associated Factor (TRAF)2 by TRAF1 Reveals Mechanisms of TRAF2 Signaling J. Exp. Med., October 7, 2002; 196(7): 923 - 934. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Challa, A. G. Eliopoulos, M. J. Holder, A. S. Burguete, J. D. Pound, A. Chamba, G. Grafton, R. J. Armitage, C. D. Gregory, H. Martinez-Valdez, et al. Population depletion activates autonomous CD154-dependent survival in biopsylike Burkitt lymphoma cells Blood, May 1, 2002; 99(9): 3411 - 3418. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Y. Chung, Y. C. Park, H. Ye, and H. Wu All TRAFs are not created equal: common and distinct molecular mechanisms of TRAF-mediated signal transduction J. Cell Sci., February 15, 2002; 115(4): 679 - 688. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
