| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Receptors1
*
Department of Biology, Boston College, Chestnut Hill, MA 02467; and
Institute of Cancer Biology, Danish Cancer Society, Copenhagen, Denmark
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
R) inhibits G1-to-S progression; the
mechanism by which this occurs is not completely known. We investigated
the regulation of three key cell cycle regulatory components by
BCR-FcR co-cross-linking: G1-cyclins, cyclin-dependent
kinases (Cdks), and the retinoblastoma gene product (Rb). Rb functions
to suppress G1-to-S progression in mammalian cells. Rb
undergoes cell-cycle-dependent phosphorylation, leading to its
inactivation and thereby promoting S phase entry. We demonstrate in
this paper for the first time that BCR-induced Rb phosphorylation is
abrogated by co-cross-linking with FcR. The activation of Cdk4/6-
and Cdk2-dependent Rb protein kinases is concomitantly blocked.
FcR-mediated inhibition of Cdk2 activity results in part from an
apparent failure to express Cdk2 protein. By contrast, inhibition of
Cdk4/6 activities is not due to suppression of Cdk4/6 or cyclins D2/D3
expression or inhibition of Cdk-activating kinase activity. Cdk4- and
Cdk6-immune complexes recovered from B cells following BCR-FcR
co-cross-linking are devoid of coprecipitated D-type cyclins,
indicating that inhibition of their Rb protein kinase activities is due
in part to the absence of bound D-type cyclin. Thus, BCR-derived
activation signals that up-regulate D-type cyclin and Cdk4/6 protein
expression remain intact; however, FcR-mediated signals block cyclin
D-Cdk4/6 assembly or stabilization. These results suggest that assembly
or stabilization of D-type cyclin holoenzyme complexes 1) is an
important step in the activation of Cdk4/6 by BCR signals, and 2)
suffice in providing a mechanism to account for inhibition of
BCR-stimulated Rb protein phosphorylation by
FcR. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
50–60% of activated B
cells committing to S phase and entering M phase of the cell cycle
between 31 and 48 h (1, 2). BCR cross-linking
initiates the ordered activation of signal transduction pathways
including phosphatidylinositol 3 kinase, p21ras
GTPase/mitogen-activated protein kinase, and phospholipase C-coupled
protein kinase C (reviewed in Refs. 3, 4). These
signaling pathways function at least in part to initiate cell cycle
entry by up-regulating the expression of immediate-early response genes
and G1 cyclins (5, 6, 7, 8, 9, 10). The ability
to attenuate the B cell response to Ag is a critical mechanism for
down-regulating the Ab response. This can be achieved by
co-cross-linking the BCR and Fc receptors of IgG (FcR) that produces
a dominant-negative signal which inhibits G1-to-S
progression and can lead to cell death (11, 12, 13, 14, 15, 16).
BCR-FcR coligation is thought to mimic an in vivo phenomenon in
which B cell responses undergo feedback inhibition by immune complexes
(17, 18).
The mechanism(s) by which BCR-FcR coligation inhibits
G1-to-S progression is not well understood.
Although tyrosine phosphorylation of several cellular proteins,
including Cbl, Vav, Fyn, and phospholipase C1 and the activation of
protein kinases, such as JNK1, Btk, and Lyn occurs in response to BCR
ligation alone or BCR-FcR co-cross-linking (19),
several aberrant signaling characteristics are known that may
negatively influence proliferation. These include decreased
BCR-triggered inositol 1,4,5-triphosphate production, reduced
Ca2+ entry, inhibition of BCR-induced Src
homology 2-containing phosphatase (SHP-2)-pp120 complex formation, and
Ras signaling (20, 21, 22, 23). Of interest, BCR-FcR
co-cross-linking leads to the association of Shc with SH2 (Src homology
domain) containing inositol 5-phosphatase (SHIP) (11, 24, 25). This appears to reduce the interaction between Shc and
Grb2, thereby contributing to the inhibition of the Ras/Raf-1/Erk
module (24, 26). What is not known is whether the
regulation of cell cycle regulatory components is influenced by
BCR-FcR coligation.
Growth factors and antiproliferative agents regulate cell proliferation by functioning primarily during G1 phase of the cell cycle (27). The retinoblastoma gene product (Rb) encodes a nuclear phosphoprotein that suppresses G1-to-S progression (28). Rb undergoes cell cycle-dependent phosphorylation, leading to its inactivation and thereby promoting S phase entry (29). G1-cyclin-dependent kinases (G1-Cdks) have been implicated in the phosphorylation of Rb (30, 31, 32, 33, 34). Cdk4 and Cdk6 are active during early-to-mid G1 phase and are limiting for G1 progression, whereas Cdk2 is active in late G1 phase and is required for progression through the G1/S transition (35, 36, 37, 38). The activation of Cdks is controlled in part through association with positive regulatory cyclin proteins (34). Growth factors induce the expression of D-type cyclins, which together with cyclin E, bind and activate Cdk4/6 and Cdk2, respectively (35, 36, 39, 40, 41). Recent studies indicate that cyclin D-Cdk4/6 complexes contribute only partially to the phosphorylation of Rb; cyclin E-Cdk2 complexes are also required for complete inactivation of Rb via phosphorylation (30). Importantly, Rb does not appear to be an efficient substrate for cyclin E-Cdk2 complexes in the absence of prior cyclin D-Cdk4/6 phosphorylation.
Regulation of cyclin/Cdk complexes by growth factors occurs at several levels, including phosphorylation (42). Cdks are subject to regulation by a family of stoichiometric Cip/Kip proteins (p21CIP, p27KIP1, and p57KIP2) that bind and inactivate many different cyclin/Cdk complexes (42). Recent evidence indicates that these proteins may also play a role in the assembly and nuclear targeting of cyclins and Cdks (43, 44). Ink4 proteins comprise a second family of inhibitors that specifically bind to Cdk4 and Cdk6 in the presence or absence of bound cyclin (42, 45, 46).
BCR cross-linking on mature B cells leads to the synthesis and assembly
of cyclin D2-Cdk4 and cyclin E-Cdk2 holoenzyme complexes, which acquire
Rb kinase activity during early-to-mid G1 and
late G1 phases of the cell cycle, respectively
(8, 10, 47, 48). B cells receiving partial stimuli or
undergoing abortive activation, exhibit aberrant expression and
activity of G1-cyclins and Cdks (9, 49, 50). Whether FcR could block BCR-induced Rb phosphorylation
and if this is linked to aberrant expression or inhibition of
G1-cyclin/Cdk complexes was examined in this
study. We provide evidence that the block in BCR-induced
G1-to-S progression by FcR correlates with
inhibition of BCR-induced Rb phosphorylation. The accumulation of
hypophosphorylated Rb results from inhibition of Cdk4/6- and
Cdk2-dependent Rb kinase activities. Despite expression of Cdk4/6 and
cyclins D2/D3, inhibition of the Rb kinase activities results from a
failure to assemble or stabilize cyclin D-Cdk4/6 holoenzyme complexes.
By contrast, inhibition of Cdk2 activity results from an absence of
Cdk2 expression. These results provide 1) a biochemical basis for the
observed inhibition of BCR-stimulated Rb protein phosphorylation by
FcR, and 2) insight into the mechanism(s) underlying inhibition of
BCR-stimulated G1-to-S progression following FcR
co-cross-linking.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Protein A/G PLUS-agarose, rabbit IgG for mouse Cdk4 (sc-260),
anti-human p27Kip1 Ab (sc-528),
anti-cyclin E Ab (sc-481), anti-rabbit and anti-mouse
IgG-HRP Abs were purchased from Santa Cruz Biotechnology (Santa Cruz,
CA). Anti-human Cdk2 Ab (06–505) was purchased from
Upstate Biotechnology (Lake Placid, NY). The anti-FcR mAb (clone
2.4G2) and human pRb mAb (clone G3-245) and anti-mouse p21 Ab
(13436E) were obtained from PharMingen (San Diego, CA). The
anti-pRb (Ser780) Ab was from MBL
International (Watertown, MA). The production of mouse anti-human
cyclin D2 Ab (DCS-3 and DCS-5) and anti-human cyclin D3 Ab (DCS-22)
have been described (51). Anti-actin Ab (A-2066) was
obtained from Sigma (St. Louis, MO) and anti-human Cdk7 mAb
(KAP-CC010) and Cdk7/MO15 peptide (aa 328–346) were purchased from
StressGen Biotechnologies (Victoria, BC, Canada). Whole rabbit
anti-mouse IgG, F(ab')2 fragments of rabbit
anti-mouse IgG, and anti-mouse IgM was obtained from Cappel/ICN
Biomedicals (Aurora, OH). Enhanced chemiluminescent reagents were from
Kirkegaard & Perry (Gaithersburg, MD). Affinity-purified Rb protein
corresponding to aa 769–921 (sc-4112) was from Santa Cruz
Biotechnology, and truncated p56 Rb was from QED Advanced Research
Technologies (San Diego, CA). F(ab')2 fragments
of anti-mouse µ were obtained from Jackson ImmunoResearch
Laboratories (West Grove, PA). Protein A-agarose was obtained from Life
Technologies (Gaithersburg, MD). Dr. Mark Solomon (Yale University, New
Haven, CT) kindly provided the GST-cyclin A and HA-Cdk2 expression
plasmids.
B cell isolation
BALB/c mice at 8–12 wk were obtained from Taconic Laboratories (Germantown, NY) and housed at Boston College. Mice were cared for and handled at all times in accordance with National Institutes of Health and institutional guidelines. Mature B lymphocytes were isolated from spleens by depletion of T cells with anti-Thy-1.2 plus rabbit complement (Accurate Chemical and Scientific, Westbury, NY) (52); macrophages (and other adherent cells) were removed by plastic adherence. RBC and nonviable cells were removed by sedimentation on Lympholyte M (Accurate Chemical and Scientific). Small resting B cells were selected by Percoll gradient centrifugation as described by DeFranco et al. (1). The resulting B cells were cultured in RPMI 1640 medium (BioWhittaker, Walkersville, MD) supplemented with 10 mM HEPES (pH 7.5), 2 mM L-glutamine, 5 x 10-5 M 2-ME, 100 U/ml penicillin, 100 µg/ml streptomycin, 0.25 µg/ml Fungizone, and 10% heat-inactivated FCS (BioWhittaker). B cells were maintained in a 37°C humidified incubator at 5% CO2.
Proliferation assay
B cells (2 x 105/well) were cultured in 96-well flat-bottom microtiter plates in the presence and absence of stimuli as indicated in the figure legends. Before the indicated times, cells were pulsed for 3 h with 0.5 µCi [3H]uridine (39.7 Ci/mmol, New England Nuclear, Boston, MA) and 6 h with 0.5 µCi [3H]thymidine (20 Ci/mmol, New England Nuclear). Cells were harvested onto glass fiber filters and quantitated by liquid scintillation spectrophotometry.
Immunoblot analysis
B lymphocytes were solubilized in 100 µl of solubilization buffer (50 mM HEPES (pH 7.4), 1.5 mM EGTA, 137 mM NaCl, 15 mM MgCl2, 0.1% Triton X-100, 75 mM ß-glycerophosphate, 1 mM Na3VO4, 1 mM PMSF, and 1 µg/ml aprotinin/leupeptin) (8, 48). Insoluble material was removed by centrifugation at 15,000 x g for 15 min and protein content was quantitated with Bradford reagents (Bio-Rad Laboratories, Hercules, CA). Cellular protein was separated through a 10–12% polyacrylamide SDS gel and transferred to Immobilon-P membrane (Millipore, Bedford, MA). The Immobilon-P membrane was incubated for 6 h at room temperature in TBST (20 mM Tris (pH 7.6), 137 mM NaCl, and 0.1% Tween 20) containing 5% nonfat dry milk. After washing the membrane with TBST (twice for 5 min), the membrane was incubated for 60 min to 18 h (4°C) with primary Ab as specified in the figure legends. The membrane was then washed several times with TBST and incubated with an anti-rabbit or anti-mouse IgG-conjugated HRP at 1:2000 in TBST for 90 min. The membrane was washed several times with TBST and developed with enhanced chemiluminescent reagent. Where indicated, autoradiograms were analyzed by densitometry using a Molecular Dynamics Personal Densitometer equipped with ImageQuant software (Sunnyvale, CA).
Immunoprecipitation
Nondenaturing extracts were prepared from 107 B cells according to the method described by Zhang et al. (44). Cells were lysed by incubation for 60 min (4°C) in ice-cold Nonidet P-40 buffer (50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 20 mM EDTA, 0.5% Nonidet P-40, 1 mM PMSF, 25 µg/ml leupeptin/aprotinin, 1 mM Na3VO4, and 10 mM ß-glycerophosphate) (8). Insoluble material was removed by centrifugation at 15,000 x g for 15 min (4°C). Cell lysates were then incubated for 5 h with 1.5 µg of the indicated Abs, followed by the addition of 50 µl of a 1:1 slurry of protein A-agarose. The resulting immune complexes were collected and separated through a 10% polyacrylamide-SDS gel. The proteins were transferred to Immobilon-P membrane and immunoblotted as described above.
Rb kinase assays
B cells were sonicated in Rb buffer (50 mM HEPES (pH 7.5), 150
mM NaCl, 1 mM EDTA, 2.5 mM EGTA, 0.1% Tween 20, and 10% glycerol)
containing 1 mM DTT, 1 mM PMSF, 1 µg/ml leupeptin/aprotinin, 1 mM
NaF, and 1 mM Na3VO4 (53). Insoluble
material was removed by centrifugation and the resulting supernatants
were incubated with 1.5 µg nonimmune rabbit IgG, 1.5 µg anti-Cdk4,
anti-Cdk6, or anti-Cdk2 Abs for 3 h at 4°C. After 3 h, the
supernatants were incubated with 50 µl of a 1:1 slurry of protein
G/A-agarose and recovered by centrifugation (15,000 x
g for 1 min). The immune complexes were washed six times
with Rb buffer, three times in a buffer containing 50 mM HEPES (pH
7.4), 5 µM ATP, and 1 mM DTT, and then resuspended in 30 µl Rb
kinase buffer (50 mM HEPES (pH 7.5), 5 mM MnCl2,
10 mM MgCl2, and 1 mM DTT) containing 5 µM ATP,
10 µCi [-32P]ATP (6000 Ci/mmol, New
England Nuclear) and 1 µg purified recombinant Rb substrate. The
kinase assays were terminated after 30 min at 30°C by the addition of
2x SDS sample buffer, and the reaction products were separated through
a 10% polyacrylamide-SDS gel. Phosphorylated Rb was detected by
autoradiography of the dried gel.
Cdk-activating kinase (CAK) assays
Production and purification of HA-Cdk2 and GST-cyclin A fusion
proteins from Escherichia coli strain BL21(DE3) were
performed as described by Connell-Crowley et al. (54). B
cells were prepared by sonication at 4°C in CAK lysis buffer (50 mM
HEPES (pH 7.5), 150 mM NaCl, 1 mM EDTA, 2.5 mM EGTA, 1 mM DTT, 0.1%
Tween 20, 1 mM PMSF, and 1 µg/ml leupeptin/aprotinin). The cell
lysates were incubated for 3 h with 2.5 µg anti-Cdk7
mAb/107 B cells, and the immune complexes were
recovered by incubation with 30 µl protein A/G PLUS agarose (90 min)
(55, 56, 57). The immune complexes were washed five times in
CAK lysis buffer, twice in Rb kinase buffer, and then resuspended in 25
µl Rb kinase buffer containing 0.5 µM ATP, 5 µCi
[-32P]ATP (6000 Ci/mmol, New England
Nuclear), 1 µg GST-cyclin A, and 3 µg of HA-Cdk2 as substrate. The
reactions were terminated after 15 min (30°C) by the addition of 2x
SDS sample buffer and separated through a 12% polyacrylamide-SDS gel.
Phosphorylated HA-Cdk2 was detected by autoradiography of the
dried gel.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
R
To evaluate the effect of BCR-FcR co-cross-linking on cell
cycle entry, splenic B lymphocytes were subjected to Percoll density
sedimentation to isolate quiescent B cells and then stimulated with 10
µg/ml F(ab')2 anti-Ig or 10 µg/ml whole anti-Ig. B cells
stimulated with F(ab')2 anti-Ig exhibited a
3.4-fold increase in [3H]uridine incorporation
compared with control B cells cultured in medium alone (Table I
). Support for
F(ab')2 anti-Ig-induced cell cycle
progression to S phase was demonstrated by increased
[3H]thymidine incorporation at 24 h and
48 h (18-fold and 49-fold above control B cells, respectively). By
contrast, B cells cultured with whole anti-Ig for 24 h
incorporated [3H]uridine and
[3H]thymidine at levels equal to control B
cells; these cells exhibited a relatively low level of
[3H]thymidine incorporation at 48 h
(3.2-fold above control B cells). Further support for inhibition of B
cell activation by whole anti-Ig was obtained from flow cytometry
in which B cells failed to undergo increased cell size, whereas B cells
stimulated with F(ab')2 anti-Ig were enlarged
at 24 h relative to control B cells, as monitored by forward
scattering (data not shown).
|
R mAb (2.4G2), an
immunologic reagent that binds FcR and blocks IgG-FcR binding
(14), exhibited increased
[3H]thymidine incorporation that was comparable
to F(ab')2 anti-Ig, indicating that
inhibition of proliferation by whole anti-Ig was mediated by FcR
(data not shown). Collectively, these findings are consistent with
several studies demonstrating inhibition of
G1-to-S progression by co-cross-linking the BCR
to FcR (12, 13, 14, 15). Of note, in subsequent experiments the
analyses of B cells treated with whole anti-Ig was not extended
beyond 31 h due to decreased cell viability and detection of
apoptotic cells; B cells stimulated with F(ab')2
anti-Ig exhibited >90% viability through 48 h.
BCR-induced Rb protein phosphorylation is blocked by coengagement
of FcR
The mechanism underlying inhibition of BCR-stimulated
G1-to-S progression by FcR remains
incompletely defined. Rb functions to suppress cell growth and is
expressed in a hypophosphorylated state in quiescent mammalian cells
(28). Hyperphosphorylation of Rb during
G1 phase, a modification that results in its
functional inactivation, leads to S phase entry (28, 29, 30).
We therefore investigated the regulation of Rb phosphorylation by
F(ab')2 anti-Ig and whole anti-Ig. To
monitor Rb phosphorylation, whole cell lysates prepared from B cells
were separated by SDS-PAGE and immunoblotted with an anti-human pRb
mAb. Previous studies have established the specificity of this assay
for monitoring Rb phosphorylation (56). The existence of
multiple Rb forms during SDS-PAGE is indicative of Rb phosphorylation
with hyperphosphorylated Rb exhibiting a greater apparent m.w. in
comparison to hypo- or unphosphorylated Rb. Rb was expressed as a
single band of 110 kDa in quiescent B cells (Fig. 1A). Stimulation of resting B
cells with F(ab')2 anti-Ig led to the
appearance of a second slower migrating Rb form at the 12-h time point
(Fig. 1A, indicated by arrow/pRb); this band was more
prominent at 24 h and 31 h. The appearance of the second
Mr Rb form parallels the timing of
G1-to-S progression (1, 2). In the
same experiment, Rb in lysates from B cells treated with whole
anti-Ig was expressed as a single band on SDS-polyacrylamide gels
and exhibited a Mr similar to resting
B cells. We conclude that quiescent B cells stimulated to progress to S
phase by F(ab')2 anti-Ig exhibit a
time-dependent increase in Rb phosphorylation, whereas signals derived
from FcR prevent BCR-induced Rb phosphorylation.
|
R co-cross-linking blocks
phosphorylation of Rb by cyclin D holoenzyme complexes was obtained
using an anti-pRb Ser780 Ab that recognizes
the Ser780 phosphorylated form of Rb. Cyclin
D-Cdk4/6 complexes in vitro phosphorylate Ser780
(33). In vivo this site is phosphorylated in early
G1 phase, concomitant with the expression and
activation of Cdk4 (33). Although cyclin D-Cdk4/6
complexes target additional sites on Rb, Ser780
phosphorylation of endogenous Rb should be indicative of the presence
of in vivo activated cyclin D-Cdk4 or -Cdk6 complexes.
Ser780 phosphorylation was initially detected in
lysates from F(ab')2 anti-Ig stimulated B
cells at the 12-h time point and increased to maximal levels at 31
h, declining thereafter. By comparison, extracts from quiescent and
whole anti-Ig treated B cells were devoid of detectable Rb
Ser780 phosphorylation at each time point
examined (Fig. 1
B). G1-Cdks from B lymphocytes treated with whole anti-Ig do not support in vitro Rb protein phosphorylation
Abundant evidence supports the role of Cdk4/6 and Cdk2 in
attenuating the growth inhibitory function of Rb via phosphorylation
(30, 31, 32, 33, 34). In mature B cells, Cdk4 and Cdk6 are rapidly
up-regulated in response to F(ab')2 anti-Ig
stimulation, whereas Cdk2 exhibits a delayed expression profile,
occurring near the G1/S transition (8, 9, 47, 48). Because cyclin D-Cdk4/6 and cyclin E-Cdk2 complexes
contribute to the phosphorylation of Rb following B cell stimulation,
we considered the possibility that they might be targets of biochemical
signals derived from BCR-FcR co-cross-linking. We therefore measured
Cdk4/6- and Cdk2-dependent Rb kinase activities in lysates prepared
from B cells under growth inhibitory and stimulatory conditions.
Control B cells were essentially devoid of Cdk4/6 and Cdk2 activities,
as measured by in vitro immune complex kinase assays with recombinant
Rb protein as substrate (Fig. 2,
lanes M). F(ab')2 anti-Ig
stimulation of B cells led to activation of Cdk4 that was detected at
12 h, increased at 24 h and 31 h, and then declined at
48 h (Fig. 2, F(ab')2 anti-Ig,
lanes Cdk4). Cdk6 immune complexes from BCR-stimulated B
cells also supported Rb phosphorylation with a similar activation
profile (Fig. 2, F(ab')2 anti-Ig, lanes
Cdk6). Cdk2 exhibited a delayed Rb phosphorylation profile
following BCR cross-linking, in keeping with previous reports (Fig. 2, F(ab')2 anti-Ig, lanes Cdk2)
(8, 9, 48). By contrast, in several experiments we did not
detect significant levels of Cdk4- or Cdk6-dependent Rb kinase activity
from whole anti-Ig-treated B cells at the time points examined
(Fig. 2, whole anti-Ig). Moreover, Cdk2-immune complexes from
BCR-FcR co-cross-linked B cells were also devoid of Rb kinase
activity. It should be noted that B cells precipitated with nonimmune
rabbit serum were devoid of measurable Rb kinase activity (Fig. 2, NRS). Collectively, these data suggest that maintenance of endogenous
Rb in a hypo- or unphosphorylated state following BCR-FcR coligation
results from a failure to activate the cellular pool of known
G1-Cdks.
|
R blocks up-regulation of cyclin E and
Cdk2 protein levels, but does not affect BCR-induced D-type cyclin and
Cdk4/6 protein expression
Given the above data, we postulated that FcR may abrogate
G1-Cdks and/or cyclin expression and in this way
achieve its inhibitory effect on Cdk4/6 and Cdk2 activities. To test
this possibility, B cells were evaluated for expression of
G1-Cdks and their associated cyclin subunits.
Stimulation of B cells with F(ab')2 anti-Ig
led to an increased cellular level of Cdk4 at 12 h, which was
maintained at each subsequent time point examined (Fig. 3A). Similarly, whole
anti-Ig treatment of B cells led to a time-dependent increase in
Cdk4 levels. Cdk4 in lysates from F(ab')2
anti-Ig stimulated B cells migrated in SDS-polyacrylamide gels as
two distinct bands. Both bands were judged specific insofar as
preincubation of the anti-Cdk4 Ab with Cdk4 antigenic peptide
resulted in a significant reduction in immunoreactivity. The nature of
the faster migrating Mr form is
presently unknown but may represent a BCR-dependent posttranslational
modification of Cdk4, which is prevented by FcR
co-cross-linking.
|
A). The level of
Cdk6 in whole anti-Ig-treated B cells at 12 h was equal to
quiescent B cells and exhibited a modest increase at 24 h and
31 h. To ensure that equal amount of protein was immunoblotted in
each lane, the blot was stripped and actin levels measured by
immunoblotting with an anti-actin Ab (Fig. 3A).
Interestingly, antiserum to p33Cdk2 did not detect expression of Cdk2
in extracts from control and whole anti-Ig-treated B cells at the
time points examined, whereas F(ab')2 anti-Ig
led to a time-dependent increase in Cdk2 protein levels, initially
detected at 24 h (Fig. 3B).
Low levels of cyclins D2 and D3 expression were detected in lysates
from control B cells (Fig. 4A); whole anti-Ig
treatment of resting B cells led to a time-dependent increase in
cyclins D2 and D3 levels with maximal expression at 24 h and
31 h, respectively. As expected, the levels of cyclins D2 and D3
were up-regulated in BCR-stimulated B cells at 24 h, in agreement
with a more detailed time course analyses previously reported (8, 9, 47). Using antisera specific for cyclin E, we detected two
relatively faint bands at approximate m.w. of 55 kDa in control B cells
(Fig. 4B, denoted by arrows). Both bands were blocked by
preincubation of the anti-cyclin E Ab with immunizing peptide
before immunoblotting and thus, determined to be specific (Fig. 4B, lanes pcE). The level of the immunoreactive
bands detected in control B cells did not appear to significantly
change in whole anti-Ig-treated B cells. By contrast, expression of
both cyclin E bands in F(ab')2 anti-Ig
stimulated B cells were elevated in a time-dependent manner. The level
of actin in each lane of Fig. 4B was equal as determined by
stripping the blot and immunoblotting with anti-actin Ab (data not
shown).
|
R co-cross-linking suffices in explaining the
inability of Cdk2-immune complexes to support in vitro Rb
phosphorylation. Of note, it is not known if the level of cyclin E
expression in whole anti-Ig-treated B cells is below a threshold
level necessary to otherwise activate Cdk2. In contrast, individual
components comprising cyclin D-Cdk4/6 complexes are up-regulated in
response to BCR-FcR co-cross-linking, suggesting that inhibition of
Cdk4/6 kinase activities by FcR is mediated at a posttranslational
level.
CAK activity is not affected by cross-linking BCR-FcR
Apart from binding regulatory cyclin, posttranslational
modification of Cdks is a focal point for regulating their kinase
activity. In particular, phosphorylation on a single threonine residue
(Thr160 in Cdk2 and Thr172
in Cdk4) by CAK is essential for activity (55, 57). CAK
consists of Cdk7, cyclin H, and the p36 MAT1 assembly factor
(58). It is possible that CAK might be inhibited as a
consequence of BCR-FcR coligation, thereby preventing activation of
Cdk4 and Cdk6. Therefore, we sought to develop an in vitro assay to
measure CAK activity in primary B lymphocytes. B cells were solubilized
in a 0.1% Tween 20 buffer to maintain CAK and CAK-like activities, and
then Cdk7 was immunoprecipitated using an anti-human Cdk7 mAb
(54, 55, 57). The resulting immune complexes were assayed
in vitro for phosphorylation of HA-Cdk2 substrate based on the method
of Connell-Crowley et al. (54). In preliminary
experiments, we found that both Cdk7 (Fig. 5A) and cyclin H (Fig. 5B) were expressed in control, whole anti-Ig, and
F(ab')2 anti-Ig stimulated B cells. (Note
that only a single time point is shown in Fig. 5, A and
B; however, the levels of Cdk7 and cyclin H did not change
over time through 31 h and 48 h following BCR-FcR
coligation and BCR cross-linking, respectively (data not shown).
Cdk7-mediated phosphorylation of HA-Cdk2 substrate was observed in
lysates from control, whole anti-Ig, and
F(ab')2 anti-Ig stimulated B cells (Fig. 5C). The specificity of this assay was confirmed in several
control experiments. 1) HA-Cdk2 was not efficiently phosphorylated by
nonimmune IgG immunoprecipitates in comparison to anti-Cdk7
immunoprecipitates (compare Fig. 5D, lanes A and
B, respectively). 2) Preincubating the anti-Cdk7 mAb
with immunizing Cdk7/MO15 peptide before immunoprecipitation
significantly reduced phosphorylation of HA-Cdk2 substrate by immune
complexes (Fig. 5D, lane C). 3) No anti-Cdk7 immune
complex-mediated phosphorylation was observed from in vitro kinase
assays devoid of added HA-Cdk2 substrate (Fig. 5D, lane D)
or in assays containing HA-Cdk2 substrate devoid of GST-cyclin A (Fig. 5D, lane E). It is noteworthy that the autoradiogram was
exposed for an extended period of time to detect any low level
phosphorylation. We conclude that FcR-derived signals do not
significantly affect CAK activity.
|
R blocks stable cyclin D-Cdk4/6 holoenzyme complex formation
We next sought to examine the possibility that inhibition of
Cdk4/6 Rb kinase activities in the presence of FcR-derived signals
might be explained by a failure to form stable complexes with the
regulatory D-type cyclin. To evaluate this, nondenaturing detergent
lysates were prepared from B cells and incubated with either
anti-Cdk4 or anti-Cdk6 Abs. The resulting immunoprecipitates
were analyzed by SDS-PAGE and immunoblotted with anti-cyclin D2 or
anti-cyclin D3 Abs. As a positive control for these experiments, we
found that both cyclins D2 and D3 coprecipitated with Cdk4 and Cdk6
from lysates of F(ab')2 anti-Ig-stimulated B
cells (Fig. 6A). In control
experiments for the specificity of the immunoprecipitating Abs,
nonimmune rabbit IgG bound to protein A-agarose did not coprecipitate
cyclins D2 or D3 (Fig. 6A). Most significantly, cyclins D2
and D3 proteins did not coprecipitate with anti-Cdk4 or
anti-Cdk6 immune complexes recovered from B cells treated with
whole anti-Ig at the time points examined. We interpret these data
to mean that D-type cyclins and Cdk4/6 are not expressed as stable
holoenzyme complexes in response to BCR-FcR co-cross-linking.
|
B). These observations are
consistent with the lack of detectable Cdk2 protein expression in whole
anti-Ig-treated B cells. As a control for the immunoprecipitations,
Cdk2 protein coprecipitated with cyclin E immune complexes recovered
from parallel BCR-stimulated B lymphocytes at 24 h. The timing of
cyclin E/Cdk2 complex assembly in BCR-stimulated B cells is consistent
with its role in late G1 phase of the cell cycle
(34, 36). We also monitored cyclin A/Cdk2 complex assembly
in BCR-stimulated B cells, which was expressed at a later time in
comparison to cyclin E/Cdk2 complexes (Fig. 6B). In
mammalian cells, cyclin A/Cdk2 complexes assemble during S phase of the
cell cycle, and therefore their assembly serves as an endogenous
readout for BCR-stimulated S phase entry. p27 displays differential expression in B cells stimulated with F(ab')2 anti-Ig and whole anti-Ig
We also investigated whether the ability of whole anti-Ig to
block G1-to-S progression was associated with
changes in the expression of several known Cdk inhibitors, including
Cip1 (p21) and Kip1 (p27) (42). As shown in Fig. 7, p21 levels in
F(ab')2 anti-Ig-stimulated B cells exhibited
a modest increase at 31 h. Expression of p21 in whole
anti-Ig-treated B cells was similar to that of
F(ab')2 anti-Ig-stimulated B cells.
Interestingly, the level of p27 in control B cells was relatively high;
B cells stimulated with F(ab')2 anti-Ig
exhibited a decrease in p27 levels at 12 h that was 50% of
control B cells. The level of p27 declined further at each subsequent
time point to a level that was 25% of control B cells at 31 h.
These results are consistent with studies by Snow and coworkers
(10) and others (47), demonstrating decreased
p27 levels in response to mitogenic F(ab')2 anti-Ig, LPS and
CD40 stimulation. Importantly, the BCR-induced decline in p27 levels
was blocked in response to co-cross-linking with FcR.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
R on the activation of B cells, the
current study investigated the regulation of Rb protein phosphorylation
by G1 Cdks. BCR cross-linking caused a
time-dependent increase in the phosphorylation of endogenous Rb that
correlated with G1-to-S progression. In the
presence of signal(s) derived by co-cross-linking the FcR,
BCR-stimulated phosphorylation of Rb did not occur under the
experimental parameters used in this study. The maintenance of
hypophosphorylated Rb correlated with inhibition of BCR-induced de novo
RNA and DNA synthesis. On the basis of these findings, together with
the previously established inhibitory role of Rb on
G1-to-S phase progression (28), we
consider it likely that Rb plays an important role in growth arrest
induced by FcR in mature B cells. Hitherto, Rb phosphorylation has
not been considered as a contributing factor in the inhibition of B
cell activation by FcR.
We specifically examined whether alterations in BCR-induced expression
and/or activities of known Rb kinases occurred following
co-cross-linking with FcR. BCR cross-linking led to a time-dependent
heightened expression of Cdk4 and Cdk6, consistent with previous
reports (8, 9, 47). Cdk4- and Cdk6-dependent Rb kinase
activities increased concomitant with G1-to-S
progression; the activation of Cdk2-dependent Rb kinase activity was
delayed compared with Cdk4/6. Surprisingly, we found that BCR-FcR
co-cross-linked B cells exhibited increased levels of both Cdk4/6
proteins; however, immunoprecipitated Cdk4 and Cdk6 from parallel B
cell cultures did not support in vitro phosphorylation of Rb protein
substrate at the time points examined in this study. We conclude that
inhibition of Cdk4- and Cdk6-dependent Rb protein kinase activities by
co-cross-linking FcR explains the observed failure of the BCR to
induced endogenous Rb phosphorylation in mature B cells.
The mechanism underlying inhibition of Cdk4/6 Rb protein kinase
activities cannot be attributed to suppression of CAK activity, because
MO15/Cdk7 immune complexes isolated from resting, BCR-stimulated, or
BCR-FcR co-cross-linked B cells supported in vitro phosphorylation
of HA-Cdk2 substrate. In addition, expression of cyclins D2 and D3
increased in the presence of BCR-FcR signals, suggesting that the
inhibition of Cdk4/6-dependent Rb phosphorylation is not due to an
absence of D-type cyclin proteins. It may be significant, however, that
the relative amount of cyclin D3 in growth arrested B cells was reduced
at 24 h compared with BCR-stimulated B cells. The possibility that
the level of cyclin D3 may be below a threshold level necessary for
activation of Cdk4/6 cannot be completely ruled out as a contributing
factor in suppression of Rb kinase activity. It has been postulated
that the levels of D-type cyclins may influence the ability of putative
assembly factors to facilitate joining or stabilization of cyclin
D-Cdk4/6 complexes (59).
Coggeshall and coworkers (23) recently reported inhibition
of p21/Ras activation by BCR-FcR co-cross-linking. p21/Ras is an
immediate upstream activator of the Raf-1/mitogen-activated protein
kinase kinase/Erk module (60, 61). These observations may
be relevant to the findings herein, particularly in light of studies
demonstrating that the transcriptional activation of the cyclin D1 gene
is dependent on upstream activation of Erk (62, 63). In
addition, expression of a dominant-negative p21/Ras has been shown to
inhibit cyclin D1 gene expression (62). We do not know if
increased expression of cyclins D2/D3 is controlled at the
transcriptional or posttranscriptional levels; however, our results
together with those of Coggeshall and coworkers (23),
raise the possibility that expression of D-type cyclins in mature B
lymphocytes by BCR-FcR signals may occur via a p21/Ras-independent
pathway.
We suspected that the inhibition of Cdk4/6 activation might reflect an
inability to efficiently assemble or stabilize cyclin D-Cdk4/6
complexes. This conclusion is based on several independent experiments
in which cyclins D2 and D3 were not detected in anti-Cdk4 and
anti-Cdk6 immune complexes recovered from B cells following
BCR-FcR coengagement. In control experiments, D2- and D3-type
cyclins coprecipitated with both Cdk4 and Cdk6 immune complexes from
BCR-stimulated B cells. This latter observation indicates that B cell
Cdk4/6 and D-type cyclins are tightly associated and amenable to the
combined immunoprecipitation-immunoblot procedure for the detection of
Cdk4 and Cdk6 binding partners. These data strongly support the notion
that assembly or stabilization of cyclin D-Cdk4/6
holoenzyme complexes, under negative signaling conditions, may be
rate-limiting for acquisition of Cdk4/6-dependent Rb kinase activity.
It is important to emphasize that we do not mean to imply the absence
of a bound regulatory D-type cyclin is the only mechanism contributing
to the inactivation of Cdk4/6 activities. It is possible that
phosphorylation on inhibitory tyrosine and threonine residues of
the Cdk subunit may concomitantly contribute to suppression of
Cdk4/6 activity (38). Nonetheless, the assembly or
stabilization of cyclin D-Cdk4/6 complexes, rather than
simply regulating the levels of D-type cyclins and/or Cdk4/6 proteins,
suggests a new posttranslational mechanism for the regulation of their
kinase activities in mature B lymphocytes. Our data are analogous to a
recent report suggesting that assembly might be an important step in
Cdk4 activation in some specialized cell types. In quiescent thyrocytes
expression of cyclin D3 and Cdk4 is not sufficient for Cdk4 assembly,
but rather requires an additional signal supplied by
thyrotropin/cAMP-dependent pathway (64). Moreover,
assembly of active cyclin D3-Cdk4 complexes in fibroblasts remains
dependent upon mitogenic stimulation, despite constitutive
overexpression of recombinant cyclin D3 and Cdk4 (53).
Efficient assembly of cyclin D-Cdk4/6 complexes may require an
additional signal, not provided by BCR-FcR co-cross-linking, that
removes or inactivates a bound Cdk inhibitor protein. Candidate
inhibitors include the Ink4 family, which currently comprise four gene
products (p15, p16, p18, and p19) and inhibit kinase activity by
interacting directly with Cdk4 and Cdk6 either free or in complex with
D-type cyclin (42, 45, 46). Several Ink4 proteins have
been shown to be expressed in B cells, including p16, p18, and p19
(47, 65). We found that comparable amounts of Cdk4
coprecipitated with p18 and p19 immune complexes recovered from lysates
of BCR-FcR co-cross-linked B cells as well as from BCR-stimulated B
cells; Cdk4 was not found associated with p16 under any of the
aforementioned conditions (data not shown). We interpret these findings
to mean that these Ink4 proteins are not likely to play a direct role
in preventing Cdk4 binding to D-type cyclins. It has been reported that
the Cdk inhibitor, p21, can promote assembly and nuclear import
of cyclin D-Cdk4 complexes (43). BCR cross-linking alone
or BCR-FcR coligation led to a modest increase in the cellular
levels of p21 compared with resting B cells, in agreement with previous
reports (10, 47). We do not know at present whether p21
functions as an assembly factor for cyclin D-Cdk4/6 complexes in mature
B cells; however, in data not included herein, p21 was not detected in
D-type cyclin or Cdk4/6 immunoprecipitates under growth stimulatory or
inhibitory conditions.
An alternative interpretation for the absence of coprecipitated D-type
cyclin in Cdk4/6 immune complexes is the notion that an assembly factor
required to facilitate the formation of stabilize cyclin D-Cdk4/6
subunits is either absent or inactive under negative signaling
conditions. Proteins that facilitate assembly of cyclin D-Cdk4/6 in
mammalian cells have remained elusive; however, it has been postulated
that candidate proteins might share similarity to the mammalian
homologue of the Saccharomyces cerevisiae Cdc37 protein,
which is required for assembly of Cln2 and Cdc28 (66, 67).
Recent studies indicate that activation of the Ras/Raf-1/Erk pathway
may be required for assembly of cyclin D1 with Cdk4 (23, 59, 68). It has been postulated that this occurs via an Erk-mediated
phosphorylation of the D-type cyclin, Cdk subunits, or an
assembly factor/chaperone. Given the reported inhibition of p21/Ras by
FcR (23), it is possible that an Erk-mediated
phosphorylation of cyclins D2/D3 or Cdk4/6 is blocked by
FcR-derived signals, thereby preventing assembly of the
holoenzyme complex. Experiments are presently underway to test this
hypothesis.
Our experiments also revealed that BCR-FcR co-cross-linking blocked
BCR-induced cyclin E and Cdk2 protein accumulation. We interpret these
results as an explanation for the lack of measurable Cdk2-dependent Rb
kinase activity under these conditions. It is interesting that Monroe
and coworkers (50) identified aberrant expression of
cyclin E and Cdk2 in primary immature B cells undergoing BCR-induced
growth arrest and apoptosis. The lack of cyclin E expression in
immature B cells may be linked to an observed inhibition of BCR-induced
c-myc expression, given previous reports that constitutive
expression of Myc or activation of conditional MycER chimeras led to
elevated levels of cyclin E mRNA (69). It is noteworthy
that accumulating evidence suggests that B cells receiving partial
stimuli exhibit aberrant expression of cyclins and Cdks. Mature B cells
stimulated to exit G0 and arrest in
G1 phase of the cell cycle by IL-4 or phorbol
diester expressed Cdk4, but lacked cyclin D2 and cyclin E
(9). B cells from xid mice, which exhibit
aborted activation in response to BCR cross-linking, do not up-regulate
Cdk4, cyclin A and cyclin D2 proteins (49). Scott and
coworkers (70) demonstrated that BCR-induced growth arrest
of WEHI-231 B cells correlated with inhibition of cyclin E- and cyclin
A-Cdk2 activity. Interestingly, Cdk2 inhibition was mediated by the
Kip1 inhibitor, p27. In mature B cells stimulated via the BCR, p27
levels rapidly decline (9, 47), whereas in immature B cell
lymphomas which undergo BCR-induced growth arrest, p27 levels increase
(70). We found herein that F(ab')2
anti-Ig stimulation of mature B cells led to decreased p27 levels,
whereas BCR-FcR co-cross-linking blocked this decline. These
observations suggest that p27 may contribute to the growth arrest
phenotype induced by BCR-FcR coligation. Determining whether or not
the stabilization of p27 levels by FcR contributes to the growth
arrest phenotype will be an important area of future investigation.
In summary, the data herein demonstrate that BCR-FcR
co-cross-linking prevents BCR-induced Rb protein phosphorylation. Our
results indicate that this is due largely to inhibition of Cdk4- and
Cdk6-dependent Rb kinase activities. The inhibition of Cdk4/6 activity
does not result from a global suppression of BCR-induced D-type cyclin
and Cdk4/6 expression or decreased CAK activity, but rather from a
failure to express stable cyclin D-Cdk4/6 holoenzyme complexes. Thus,
although BCR-derived activation signals that function to up-regulate
D-type cyclin and Cdk4/6 protein expression remain intact, an
additional signal is required to facilitate joining or stabilization of
cyclin D-Cdk4/6 complexes, which is not provided by BCR-FcR-derived
signals. These data suggest a novel mechanism for the regulation of
Cdk4/6 kinase activation in mature B lymphocytes. Elucidating the
mechanism(s) responsible for the block in cyclin D-Cdk4/6 complex
assembly or stabilization will be an important next step in
understanding the molecular basis underlying abortive
G1-to-S progression.
|
Footnotes |
|---|
2 Current address: Boston University Medical School, Department of Microbiology, 80 East Concord Street, Boston, MA 02118.
3 Address correspondence and reprint requests to Dr. Thomas C. Chiles, 411 Higgins Hall, Department of Biology, Boston College, Chestnut Hill, MA 02467. E-mail address:
4 Abbreviations used in this paper: BCR, B cell Ag receptor; FcR, Fc receptor for IgG; Cdk, cyclin-dependent kinase; Rb, retinoblastoma gene product; pRb, phosphorylated Rb; CAK, Cdk-activating kinase; Erk, extracellular signal-regulated kinase.
Received for publication April 29, 1999. Accepted for publication July 7, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Rb. Curr. Opin. Immunol. 10:306.[Medline]
receptors and membrane immunoglobulin inhibits anti-immunoglobulin-induced blastogenesis. J. Immunol. 132:627.[Abstract]
receptor to surface immunoglobulin on B cells provides an inhibitory signal that closes the plasma membrane calcium channel. J. Biol. Chem. 269:11409. ITIM peptide motif. Mol. Cell. Biol. 17:4305.[Abstract]
RIIB1 and is tyrosine phosphorylated during "negative" signaling. Immunol. Lett. 54:77.[Medline]
RIIb in B cells. J. Immunol. 162:1408.
