| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Gastroenterology Section, Department of Medicine, The University of Chicago, Chicago, Illinois 60637
Address all correspondence and requests for reprints to: T. A. Brasitus, Gastroenterology Section, Department of Medicine, The University of Chicago, MC4076, 5841 South Maryland Avenue, Chicago Illinois 60637. E-mail: tbrasitu{at}medicine.bsd.uchicago.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,25-dihydroxyvitamin D3,
1,25-dihydroxy-16-ene-23-yne-26,27-hexafluorocholcalciferol
(F6-D3), significantly reduced the frequency of
colonic adenomas and completely abolished the development of colonic
adenocarcinomas in rats treated with azoxymethane. The mechanisms
involved in this analog’s chemopreventive actions, however, remain
unclear. In the present study, we now show that although both
1,25-dihydroxyvitamin D3 and
F6-D3 inhibited the proliferation of CaCo-2
cells, a human colonic adenocarcinoma cell line, by increasing their
doubling times, only F6-D3 caused an arrest of
these cells in the G1 phase of their cell cycle. This arrest was
accompanied by an increase in the expression of the cyclin-dependent
kinase (cdk) inhibitor proteins, p21Waf1 and
p27Kip1, which served to decrease the activity of
cyclin-dependent kinase 2 and cyclin-dependent kinase 6, whereas the
expression and phosphorylation of pRB were unchanged. In contrast to
the increased expression of these cdk inhibitors, the expression of
cyclin E was decreased, which further inhibited the activity of
cyclin-dependent kinase 2. Collectively, the inhibition of these
cyclin-dependent kinases served to arrest the CaCo-2 cells, independent
of changes in pRB. Furthermore, antibody neutralization studies suggest
that transforming growth factor-ß may mediate the coassociations
between cdk2 and p27Kip1 and cyclin E induced by
F6-D3. These data indicate that cell cycle
arrest may, at least in part, underlie the chemopreventive actions of
F6-D3 observed in the azoxymethane model of
colon cancer. Furthermore, if the antiproliferative action observed in
CaCo-2 cells also occurs in human colonic epithelium,
F6-D3 may have chemopreventive potential
against human colon cancer, as well. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,25-dihydroxyvitamin
D3
(1,25(OH)2D3), may prevent
the development of colonic cancers and other tumors (1).
In the 1,2-dimethylhydrazine model of colonic carcinogenesis, for
example, sc administration of
1,25(OH)2D3, before
carcinogenic administration, markedly reduced (
50%) the number of
colonic adenocarcinomas (2).
1,25(OH)2D3 has been shown
by our laboratory (3), and others (4, 5, 6), to
decrease cell growth and increase the differentiation of CaCo-2 and
other human cancer-derived cell lines, which may play important roles
in its ability to prevent and/or treat these malignancies. In addition,
1,25(OH)2D3 and/or several
of its analogs have been shown by our laboratory and others to
participate in several rapid events, including activation of protein
kinase C, phospholipase D, and mitogen-activated protein kinase, which
may also result in alterations in proliferation and/or differentiation
(7, 8, 9, 10, 11, 12). In human trials, however, the hypercalcemic
effects of 1,25(OH)2D3 have
precluded the achievement of adequate serum levels of this secosteroid
to induce its effect on proliferation and/or differentiation. A number
of investigators, therefore, have examined synthetic analogs of
1,25(OH)2D3 that retain
their ability to inhibit proliferation and induce differentiation but
cause significantly lower degrees of hypercalcemia
(13, 14, 15). These analogs, in general, have modifications of
the side chain and/or the D-ring of
1,25(OH)2D3, and several
have been used as anticancer agents in in vivo animal models
of colon cancer (5, 14, 15, 16, 17). Though there have been
specific synthetic analogs that failed to demonstrate antitumor
efficacy in vivo (18), our laboratory has
recently shown that one synthetic noncalcemic analog,
1,25-dihydroxy-16-ene-23-yne-26,27-hexafluorocholcalciferol
(F6-D3), when supplemented
in the diet of rats administered the colonic procarcinogen
azoxymethane, markedly reduced the incidence of colonic adenomas and
completely abolished the development of adenocarcinomas
(19). This same analog has also been shown to be effective
against tumors induced in a rat model of mammary carcinogenesis
(14). In addition, two novel 14-epi analogs of
1,25(OH)2D3 significantly
retarded tumor progression arising from MCF-7 breast cancer cells
implanted in nude mice (20). The pathways that mediate the
chemopreventive effects of
F6-D3 are, however,
unclear. Previous studies by our laboratory have shown that
F6-D3 plays a
prodifferentiating role in CaCo-2 cells [(21) and
unpublished observations] much like that observed with its parent
compound, 1,25(OH)2D3. In
the present studies using CaCo-2 cells, we characterized the
antiproliferative effects of
F6-D3 and compared them
with 1,25(OH)2D3. In
addition, we examined the possible mechanisms of action involved in
this phenomenon. The results of these studies, as well as a discussion
of their significance with respect to the potential chemopreventive
actions of these secosteroids in the colon, serve as the basis for the
present report. |
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
32P]ATP was obtained from
Perkin-Elmer Life Sciences (Boston, MA).
1,25(OH)2D3 was purchased
from Steroids Limited (Chicago, IL).
F6-D3 was a generous gift
from Dr. M. R. Uskokovic (Hoffman-LaRoche Inc.,
Nutley, NJ). Cell culture products were purchased from Life Technologies (Bethesda, MD). Tissue culture plates and flasks
were purchased from Corning (Corning, NY). The
cytotoxicity detection kit was purchased from Roche Molecular Biochemicals (Indianapolis, IN). Polyclonal antibodies specific
for cdk2, cdk4, cdk6, cyclin A, cyclins D1–D3, cyclin E, and
p16INK, p27Kip1, and
p57Kip2, as well as monoclonal antibodies
specific for pRb and p21Waf1, were purchased from
Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).
Antibodies specific for the hyperphosphorylated forms of pRb were
purchased from PharMingen (San Diego, CA). Pan-specific
neutralizing antibodies to transforming growth factor (TGF)-ß were
obtained from R & D Systems (Minneapolis, MN).
Porcine TGF-ß, sharing complete sequence identity with human TGF-ß,
was obtained from R & D systems. The chemiluminescent system was
obtained from Amersham Pharmacia Biotech Products
(Arlington Heights, IL). All other chemicals, unless otherwise
indicated, were purchased from Sigma (St. Louis, MO) and
were of the highest grade available.
Cell culture
CaCo-2 cells, derived from a human colonic adenocarcinoma
(22), were cultured at 37 C in an atmosphere of 5%
CO2. Cells were maintained in DMEM with 4.5
g/liter glucose, 2 mM L-glutamine, 50 U/ml
penicillin, 50 µg/ml streptomycin, 2 µg/ml gentamicin, 10
mM HEPES, 1% essential and nonessential amino acids, and
20% FBS. Fresh media was supplemented daily with the indicated
concentrations of
1,25(OH)2D3 or
F6-D3 in ethanol. The final
concentration of ethanol in all experiments was 0.05%.
Cell number
Cells were grown in 6-well multiwell plates, protected from
fluorescent light, in media supplemented daily with
1,25(OH)2D3,
F6-D3, or vehicle (ethanol)
at the indicated concentrations. Cellular proliferation was assessed,
3–4 days post plating, by measuring the total cell number. Cells were
removed using 0.2% trypsin-EDTA in HBSS. Trypsin activity was quenched
by addition of an equal volume of media containing serum. An aliquot
was then counted using a Coulter Counter (Coulter Electronics, Miami,
FL), and data were reported as the number of cells per well.
Cell toxicity assay
The effect of
1,25(OH)2D3 or
F6-D3 on cell death was
assessed by measuring release of lactate dehydrogenase (LDH) from the
cytosol of damaged cells using the Roche Molecular Biochemicals Cytotoxicity Detection kit, according to
manufacturer’s instructions. Briefly, membrane damage results in an
increase in the LDH released into the culture supernatant, as assessed
by the LDH catalyzed conversion of tetrazolium to formazan. Activity
units were calculated based on absorbance of formazan salt at 490 nm
obtained from cell supernatants. Lactate dehydrogenase (% total)
= [(agonist treated - vehicle treated) ÷ (triton
treated - vehicle treated)] x 100; triton causes maximum LDH
release.
Population doubling time
The time taken for cell populations to double was calculated
using four separate time points in the exponential (log phase) of cell
growth, and the population doubling levels (PDL) were determined
according to published methods (23). The population
doubling time was calculated by dividing the number of hours in culture
by the PDL. Briefly, the PDL was calculated using the equation for
biological samples derived in Ref. 23 : PDL = [log (final cell
number) - log (initial cell number)] x 3.322.
Cell cycle analysis by flow cytometry
The effects of daily supplementation of
1,25(OH)2D3 (100
nM) or F6-D3
(100 nM) on the distribution of cells in the cell cycle
were assessed in preconfluent cells, at 3–4 days post plating,
according to the method of Darzynkiewicz (24). Briefly,
cells were fixed in 70% ethanol and stained for 30 min at room
temperature with 10 µg/ml propidium iodide in PBS containing 20 U/ml
deoxyribonuclease-free ribonuclease A. Cells (20,000 per sample) were
analyzed using CellQuest software in a FacScan flow cytometer
(Becton Dickinson and Co. Immunocytometry Systems, San
Jose, CA). The distribution of DNA in the cell cycle was determined
using Modfit LT software (Verity Software House, Topsham, ME).
Western blotting of cell cycle regulatory proteins
CaCo-2 cells, treated daily, for 3–4 days post plating, with
F6-D3 (100 nM),
1,25(OH)2D3 (100
nM), or vehicle, were lysed by boiling in PBS/1% SDS
containing 100 µg/ml AEBSF and 1 mM sodium orthovanadate.
Proteins were measured using the bicinchoninic acid method
(25). Proteins were separated by SDS-PAGE and
electroblotted to polyvinylidene difluoride membranes (Millipore Corp., Bedford MA) (26). Blots were stained with
0.1% India ink in 50 mM Tris-HCl (pH 7.4), 150
mM NaCl with 0.05% Tween-20 (TBST) to assess equal
loading. Nonspecific binding of antibodies was blocked by incubating
the blots in 5% nonfat dry milk in TBST for 2 h. After blocking,
the blots were incubated overnight at 4 C with the indicated antibodies
to p16INK, p21Waf1,
p27Kip1, p57Kip2, cyclin A,
cyclin D1, cyclin D2,
cyclin D3, cyclin E, pRB, or pRb specific for the
hyperphosphorylated form, at final concentrations of 0.5 µg/ml in 1%
BSA/TBST. After four washes (10 ml each) with TBST, blots were
incubated with 1:3000 final dilutions of appropriate peroxidase-coupled
secondary antibody. The blots were washed four times in TBST, and
proteins were detected using an enhanced chemiluminescent system,
according to manufacturer’s directions. The xerograms were digitized
using a flat bed scanner, and the band densities were quantified using
the IP Lab Gel P, version 2.0a software.
Immunoprecipitation kinase assays for cdk2, cdk4, and cdk6
CaCo-2 cells, treated for 4 days with daily additions of 100
nM F6-D3 or
vehicle (ethanol), were lysed in 50 mM HEPES (pH 7.5), 150
mM NaCl, 2.5 mM EGTA, 1 mM
dithiothreitol (DTT), 0.1% Tween-20, 10% glycerol, 0.1 mM
AEBSF, 10 µg/ml leupeptin, 20 U/ml aprotinin, 10 mM
ß-glycerophosphate, 1 mM NaF, and 0.1 mM
sodium orthovanadate. Lysates (50 µg) were precleared with 1 µg of
normal rabbit serum, then immunoprecipitated with the indicated
antibodies and protein A Sepharose beads for 3 h. After four
washes in lysis buffer, immunoprecipitated proteins were assayed for
kinase activity at 37 C in 30 µl of 50-mM HEPES (pH 7.5),
1 mM DTT, 2.5 mM EGTA, 10 mM
ß-glycerophosphate, 0.1 mM sodium orthovanadate, 1
mM NaF, 10 µM ATP, 1 µCi
[32P] ATP, and either 2 µg histone H1 for
cdk2 and cdk6, or 2 µg GST-pRB for cdk4. Kinase reactions were run
under linear conditions and stopped, after 30 min of incubation, by
addition of an equal volume of 2x Laemmli buffer and boiling for 5
min. Proteins were separated by SDS-PAGE, on a 12% resolving gel, and
electroblotted to polyvinylidene difluoride membranes
(27). [32P]-histone H1 and
[32P]-GST-pRB were detected by autoradiography.
Parallel immunoprecipitations were probed for expression of the cdks,
to confirm comparable kinase abundance in each assay.
Coprecipitation assays
To analyze proteins coassociating with cdk2, cdk4, and cdk6 (by
Western blotting), cells treated with vehicle or
F6-D3 were broken in
modified DTT-free lysis buffer and immunoprecipitated with antibodies
to cdk2, cdk4, or cdk6. After 3 h, immunoprecipitates were washed
in modified lysis buffer and suspended in nonreducing Laemmli buffer.
Where indicated, cdk2, cdk4, and cdk6 immunoprecipitates were probed
for coassociating proteins, by Western blotting, as described above,
with antibodies to p21Waf1,
p27Kip1, or cyclin E. Parallel cdk-specific
immunoprecipitates were probed by Western blotting with anti-cdk
antibodies to confirm comparable amounts of kinase in each sample. To
analyze the role of TGF-ß in the
F6-D3-induced alterations
in p27Kip1 and cyclin E bound to cdk2, cells were
treated for 3 days with vehicle or
F6-D3 in the presence of
neutralizing concentrations of a pan-specific antibody to TGF-ß (20
µg/ml), control rabbit nonimmune serum (20 µg/ml), or exogenous
porcine TGF-ß (2.5 ng/ml); and lysates were then immunoprecipitated
with cdk2 antibody.
Statistical analysis of data
Numerical data are expressed as means ± SEM.
One-way ANOVAs were performed using MINITAB statistical analysis
software (MINITAB Inc., State College, PA) (28), and
differences with P 
0.05 were considered
statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, both
1,25(OH)2D3 and
F6-D3 caused significant
dose-dependent decreases in cell numbers. The former finding is in
agreement with results previously reported by our laboratory for
1,25(OH)2D3 in these cells
(3). Furthermore, as shown in this figure, compared with
1,25(OH)2D3,
F6-D3 caused a
significantly greater inhibition of cellular proliferation at the doses
indicated. Measurement of the cytotoxicity of these compounds by LDH
release (Fig. 2) revealed no cytotoxicity
for 1,25(OH)2D3 at
concentrations as high as 300 nM. Because
F6-D3 induced significant
cell death only at 300 nM, the highest concentration
tested, we chose nontoxic concentrations of 100 nM each for
1,25(OH)2D3 and
F6-D3 for all subsequent
studies.
|
|
), 100
nM F6-D3 caused
a significant increase (P 0.05), from 36% to 44%,
in the percentage of cells in the G0/G1 phase. This G1 arrest was
accompanied by a reciprocal decrease (P 0.05), from
44% to 37%, in the percentage of cells observed in the S-phase.
F6-D3 treatment, however,
did not alter the percentage of cells in the G2/M phase (Fig. 3).
|
shows that
1,25(OH)2D3 (100
nM) caused a statistically significant increase in the
number of hours required for CaCo-2 cells to double, compared with
vehicle. There was, numerically, an even more pronounced increase in
the doubling time in cells treated with
F6-D3 (100 nM).
The mechanism for the increase in cell doubling time in
1,25(OH)2D3-treated cells
is unclear; but, because the cell cycle distribution did not change,
this suggests that this secosteroid may cause a uniform slowing of cell
cycling, as has been previously described for other agents
(29). In addition, in preliminary studies,
1,25(OH)2D3 and
F6-D3 were each found to
induce apoptosis in CaCo-2 cells (30), which, if
uncompensated, would also be expected to prolong the doubling times of
these cells (manuscript in preparation).
|
, F6-D3 caused a significant
(P 0.05) increase in the expression of both of the
cyclin-dependent kinase inhibitors, p21Waf1
(8-fold) and p27Kip1 (2-fold), whereas the
expression levels of p16INK and
p57Kip2 were unchanged (Fig. 4, and data not
shown). As noted earlier,
F6-D3 induced a G1 arrest.
This arrest is consistent with the roles of
p21Waf1 and p27Kip1 in
maintaining cells in the G0/G1 phase of the cell cycle
(31). As shown in Fig. 4A, F6-D3
did not alter the expression of cyclins A, D1, or D3. Cyclin D2 was not
detectable in CaCo-2 cells (data not shown). Whereas the pRb expression
(Fig. 4A) and phosphorylation of pRb (data not shown) were unchanged in
cells treated with F6-D3,
this secosteroid inhibited cyclin E expression by 55%
(P 0.05) (Fig. 4). We next sought to investigate why
1,25(OH)2D3, while altering
the doubling times, did not seem to arrest CaCo-2 cells in G1. As shown
in Fig. 5, there were no alterations in
the expression of p21Waf1,
p27Kip1, or cyclin E by
1,25(OH)2D3, in contrast to
the changes in these proteins observed when cells were treated with
F6-D3.
|
|
). In contrast, there was no detectable
change in the phosphorylation of GST-pRB by cdk4 in cells treated with
F6-D3, compared with
vehicle-treated cells (Fig. 6). The total amount of each of the
immunoprecipitated cdks from
F6-D3 and vehicle-treated
cells were comparable (data not shown). These data indicate that
inhibition of both cdk2 and cdk6, but not cdk4, are likely involved in
the G1 arrest induced by
F6-D3.
|
). In
addition, F6-D3 increased
binding of p27Kip1 to cdk6 (Fig. 7). In contrast,
whereas there were decreased levels of cyclin E and increased levels of
p27Kip1 detected in lysates from cells treated
with F6-D3, there were no
detectable alterations in the binding of these proteins to cdk4 (data
not shown).
|
). Addition of
exogenous TGF-ß, however, did not seem to enhance the binding of
p27Kip1 to cdk2 or further inhibit cyclin E
binding (data not shown). This suggests that the
F6-D3-induced changes in
these associations are maximally achieved by the endogenous TGF-ß
secreted from these cells. These results, moreover, suggest that
TGF-ß may mediate, at least in part, the alterations in
p27Kip1 and cyclin E induced by
F6-D3.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this study, compared with 1,25(OH)2D3, F6-D3 was found to be a more potent antiproliferative agent of colon cancer-derived CaCo-2 cells. Furthermore, growth inhibition of CaCo-2 cells by this synthetic secosteroid occurred via a G1 arrest. Specifically, for cells in the log phase of growth, F6-D3 caused a significant increase in the percentage of cells in the G1 phase, compared with vehicle-treated cells. This increase in the G1 phase was accompanied by a reciprocal decrease in the number of cells in the S phase, with no change in the G2/M fraction. These findings are the first demonstration in CaCo-2 cells of growth inhibiting effects by F6-D3, a potential colon cancer chemopreventive agent, and this is in agreement with prior studies in which 1,25(OH)2D3 or other analogs of this secosteroid inhibited proliferation via a G1 arrest in several noncolonic cell lines (5, 37, 40, 41). In contrast to phase-specific cell cycle inhibition by F6-D3, 1,25(OH)2D3 seemed to uniformly slow the cell cycle progression of CaCo-2 cells and caused no detectable alterations in the expression of several cell cycle regulatory proteins.
Control of the cell cycle involves changes in the expression and/or activity of three major classes of regulators: cyclin-dependent kinases (cdk), cyclins, and cyclin-dependent kinase inhibitors (31, 42). Cell cycle progression is regulated by sequential formation, activation, and subsequent inactivation of a series of cyclin-cdk complexes (43, 44). The major cyclin-cdk complexes involved in cell cycle progression from the G1 to the S phase are cyclin D-cdk4/cdk6 and cyclin E-cdk2, respectively (43, 44). In the present studies in CaCo-2 cells, we have partially characterized the growth inhibitory mechanisms of a fluorinated analog of 1,25(OH)2D3 that involve a number of these specific regulators of the cell cycle. In the G1 arrest of CaCo-2 cells by F6-D3, although cyclin A, cyclin D1, and cyclin D3 were not involved, we have demonstrated that decreases in cyclin E expression seemed to play a role. The functional consequence of this finding was the inhibition of cdk2 kinase activity in cells treated with this secosteroid.
Cyclin-dependent kinase inhibitors block the activation of cyclin-dependent kinases and, thereby, restrain cells in a particular phase of the cell cycle. There are two major classes of cyclin-dependent kinase inhibitors. The first group includes the p21Waf1 family of inhibitors (p21Waf1, p27Kip1, and p57Kip2), which are able to inhibit a number of the cdk/cyclin complexes and have diverse regulatory roles in many phases of the cell cycle (42). For example, p21Waf1 can inhibit the activation of cdk2 by either cyclin E in the G1 phase of the cell cycle or by cyclin A in the S-phase of the cell cycle (42). In our system, cyclin E (but not cyclin D) seemed to play a role in the G1 arrest, whereas cyclin A was not implicated in the S-phase changes. The second group includes the Ink4 family of inhibitors, which have been shown to be more specific for their interactions with cdk4 and its accompanying cyclins (45). In the present studies, alterations in cdk4 activity did not seem to be involved in the G1 cell cycle arrest induced by F6-D3 in CaCo-2 cells, and this was reflected in the observation that alterations in p16INK were not observed.
Previous studies have identified p21Waf1 and
p27Kip1 as mediators of a G1 arrest by
1,25(OH)2D3 and/or one of
its analogs (40, 46, 47). 1,25
(OH)2D3 has been shown to
increase the expression of both p27Kip1 and (more
transiently) p21Waf1 in HL-60 cells
(40). In contrast, this secosteroid did not alter these
cdk inhibitors in CaCo-2 cells.
F6-D3, however, increased
the expression of both p21Waf1 and
p27Kip1, whereas the expression of
p57Kip2 was unchanged. The increase in
p21Waf1 expression, as well as the increase in
p27Kip1 and the decrease in cyclin E bound to
cdk2, are likely to be responsible for the decrease in cdk2 activity
and, thereby, to have contributed to the
F6-D3-induced inhibition of
the G1
S transition.
A vitamin D response element in the p21Waf1 promotor has been shown to mediate the secosteroid-induced up-regulation of this cyclin-cdk inhibitor in U937 cells (46). In future studies, it will be of interest to determine whether F6-D3 mediates its increase in the expression of p21Waf1 in CaCo-2 cells by a similar mechanism. The mechanism(s) of up-regulation of the expression of p27Kip1 by F6-D3 is also unclear. In other cell types, 1,25(OH)2D3 has been shown to increase levels of p27Kip1 by a TGF-ß-dependent, vitamin D response element-independent mechanism (35, 43, 48, 49). Interestingly, this was accompanied by an enhanced association of p27Kip1 to cyclin E-cdk2. In this study, F6-D3 similarly increased p27Kip1 binding to cdk2 in CaCo-2 cells, and this effect was blocked by TGF-ß antibodies, suggesting that TGF-ß may play a role in these alterations, as well as in the G1 arrest by F6-D3. In contrast to HL-60 cells treated with EB1089, another analog of 1,25(OH)2D3 (50), addition of exogenous TGF-ß, did not further augment the F6-D3-induced binding of p27Kip1 to cdk2 in CaCo-2 cells. G1 arrest of the cell cycle by TGF-ß in rat intestinal cells has been associated with phosphorylation changes in pRb that occur as a result of decreased cyclin D1 (49, 51). Because, in our studies involving a human colon carcinoma cell line, there were no changes in cyclin D or pRb, the mechanism(s) by which TGF-ß mediates the actions of F6-D3 on p27Kip1, cyclin E, and cdk2 seems to be species- and tissue-specific.
In the present studies, the activity of cdk6 was also inhibited by treatment of CaCo-2 cells with F6-D3. Furthermore, this synthetic secosteroid caused an increased association of p27Kip1 with cdk6, similar to the finding observed in HL60 cells with increased p27Kip1 binding to cdk6 in response to 1,25(OH)2D3 (40). Increasing levels of cyclin D bind to cdk6 to drive cells from G1 to S phase. Though there was no decrease in the expression of cyclin D1 or D3 to account for the G1 arrest by F6-D3, the enhanced binding of p27Kip1 to cdk6 induced by this secosteroid seems, at least in part, to have inhibited cdk6 activity.
Active cyclin D-cdk complexes cause hyperphosphorylation of the retinoblastoma gene product (pRb), leading to liberation of the E2F transcription factor that transactivates genes involved in S-phase progression (52, 53). Cyclin E, which is expressed in late G1, forms an active kinase complex with cdk2 (54) and contributes to the G1/S transition by regulating pRb-dependent and -independent pathways (52, 55, 56), downstream of cyclin D1-cdk4/cdk6 (57). In the present study, we found that the expression of cyclin D and pRb, and the phosphorylation state of pRb, were unaffected by F6-D3 treatment of CaCo-2 cells. Cyclin E-cdk2 activity was, however, decreased by F6-D3, indicating that inhibition of this kinase contributes to a pRb-independent G1 arrest induced by this secosteroid in these cells. In this regard, it has been shown that cyclin E regulates the transition of pRb-null fibroblasts to the S-phase (56). The potential mechanism by which cyclin E expression may be inhibited by this Vitamin D3 analog will require further investigation. In other cells, cyclin E gene expression has been shown to be inhibited by activation of Sp1 transcription factors (58). Furthermore, Sp1 has been shown to be activated by 1,25(OH)2D3 in leukemic cells, where it regulates the transcription of p21Waf1 (59). Additional studies to examine the role, if any, of Sp1 in inhibiting the expression of cyclin E will, therefore, be of interest. Our data also suggests that F6-D3 may inhibit cyclin E expression by a TGF-ß-dependent mechanism.
Finally, it is worth noting that although the antiproliferative
effects of F6-D3 were
statistically significant, but relatively small, their impact on tumor
progression may still be considerable. Tumorigenesis is characterized
by a net increase in proliferation over programmed cell death.
Mathematical models of tumorigenicity have suggested that even modest
decreases in this proliferative rate, as observed in the present study,
could result in significant inhibition of tumor formation (60, 61). Likewise, even a modest increase in the apoptotic rate, as
has been found in preliminary studies in our laboratory with
F6-D3 (30),
and in HT-29 intestinal cells with a closely related compound
(17), could result in significant inhibition of tumor
formation (60, 61). Furthermore, it bears emphasis that
CaCo-2 cells are very resistant to antiproliferative interventions. For
example, in previous studies in our laboratory, we have found that
these cells, as assessed by fluorescence-activated cell sorting, could
not be synchronized by any of several strategies, including serum
deprivation, or treatment with hydroxyurea and nocadazole (unpublished
observations). Indeed, this cell line has been shown to be dysregulated
in specific mitotic cell cycle checkpoints (62). Hence,
the observation that F6-D3
inhibits the G1S transition in these autonomous cells further
emphasizes the potential antiproliferative efficacy of this agent.
Furthermore, at earlier stages in the malignant transformation process,
F6-D3 may be an even more
potent antiproliferative agent than observed in these colon
cancer-derived cells.
In summary, we have shown that
F6-D3, a fluorinated analog
of 1,25(OH)2D3, decreases
CaCo-2 cell proliferation by an arrest of cells in the G1 phase. This
arrest is induced, at least in part, by increased expression of
p21Waf1 and p27Kip1, and
decreased expression of cyclin E. The former
(
p21Waf1 and p27Kip1)
serve to inhibit the kinase activities of cdk2 and cdk6; and the latter
(
cyclin E), to inhibit cdk2. The alterations in
p27Kip1 and cyclin E binding to cdk2 may be
mediated by TGF-ß. Furthermore, these cell cycle changes may underlie
the antiproliferative and chemopreventive actions of
F6-D3 observed in the
azoxymethane model of colonic carcinogenesis (19).
Finally, these results, if applicable to human colonic epithelium,
suggest that F6-D3 may be
useful as a chemopreventive agent in human colon cancer, as well.
|
Footnotes |
|---|
Received March 29, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in rat
colonocytes: role of c-src in PLC- activation. J Clin Invest 99:1831–1841[Medline]
,25-dihydroxyvitaminD3 in promelocytic NB4
leukemia cells: a structure function study. Endocrinology 139:457–465,25-Dihydroxy-16-ene-23-yne-26,27-hexafluorocholecalciferol
(RO24–5531), a new deltanoid (vitamin D analog) for prevention of
breast cancer in the rat. Cancer Res 54:1653–1656,25-dihydroxy-16-ene-23yne-26,27 hexafluoro-19-nor-cholecalciferol
(RO-25–6760), on a human colon cancer xenograft. Clin Cancer Res 4:2869–2876[Abstract]
,25-dihydroxy-16-ene-23-yne-26,27-hexafluorocholecalciferol, a
noncalcemic analog of 1,25-dihydroxyvitamin
D3, inhibits azoxymethane-induced colonic
tumorigenesis. Cancer Res 55:3050–3054,25-dihydroxyvitamin D3 synergism
toward transforming growth factor-ß1-induced AP-1 transcriptional
activity in mouse osteoblastic cells via its nuclear receptor. J
Biol Chem 273:14738–14744This article has been cited by other articles:
 |
|
 |
|
  B. NURNBERG, S. GRABER, B. GARTNER, J. GEISEL, C. PFOHLER, D. SCHADENDORF, W. TILGEN, and J. REICHRATH Reduced Serum 25-Hydroxyvitamin D Levels in Stage IV Melanoma Patients Anticancer Res, September 1, 2009; 29(9): 3669 - 3674. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ng, J. A. Meyerhardt, K. Wu, D. Feskanich, B. W. Hollis, E. L. Giovannucci, and C. S. Fuchs Circulating 25-Hydroxyvitamin D Levels and Survival in Patients With Colorectal Cancer J. Clin. Oncol., June 20, 2008; 26(18): 2984 - 2991. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. G. Fakih, D. L. Trump, J. R. Muindi, J. D. Black, R. J. Bernardi, P. J. Creaven, J. Schwartz, M. G. Brattain, A. Hutson, R. French, et al. A Phase I Pharmacokinetic and Pharmacodynamic Study of Intravenous Calcitriol in Combination with Oral Gefitinib in Patients with Advanced Solid Tumors Clin. Cancer Res., February 15, 2007; 13(4): 1216 - 1223. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Taghizadeh, M. J. Tang, and I. T. Tai Synergism between vitamin D and secreted protein acidic and rich in cysteine-induced apoptosis and growth inhibition results in increased susceptibility of therapy-resistant colorectal cancer cells to chemotherapy Mol. Cancer Ther., January 1, 2007; 6(1): 309 - 317. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Martin-Venegas, S. Roig-Perez, R. Ferrer, and J. J. Moreno Arachidonic acid cascade and epithelial barrier function during Caco-2 cell differentiation J. Lipid Res., July 1, 2006; 47(7): 1416 - 1423. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. T. Jacobs, M. R. Haussler, and M. E. Martinez Vitamin D Activity and Colorectal Neoplasia: A Pathway Approach to Epidemiologic Studies Cancer Epidemiol. Biomarkers Prev., September 1, 2005; 14(9): 2061 - 2063. [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Harris and V. L. W. Go Vitamin D and Colon Carcinogenesis J. Nutr., December 1, 2004; 134(12): 3463S - 3471S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Lucas, J. Domenico, and E. W. Gelfand Cyclin-Dependent Kinase 6 Inhibits Proliferation of Human Mammary Epithelial Cells Mol. Cancer Res., February 1, 2004; 2(2): 105 - 114. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. K. Wali, S. Khare, M. Tretiakova, G. Cohen, L. Nguyen, J. Hart, J. Wang, M. Wen, A. Ramaswamy, L. Joseph, et al. Ursodeoxycholic Acid and F6-D3 Inhibit Aberrant Crypt Proliferation in the Rat Azoxymethane Model of Colon Cancer: Roles of Cyclin D1 and E-Cadherin Cancer Epidemiol. Biomarkers Prev., December 1, 2002; 11(12): 1653 - 1662. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Galbiati, L. Polastri, S. Gregori, M. Freschi, M. Casorati, U. Cavallaro, P. Fiorina, F. Bertuzzi, A. Zerbi, G. Pozza, et al. Antitumorigenic and Antiinsulinogenic Effects of Calcitriol on Insulinoma Cells and Solid {beta}-Cell Tumors Endocrinology, October 1, 2002; 143(10): 4018 - 4030. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Chen, B. H. Davis, M. D. Sitrin, T. A. Brasitus, and M. Bissonnette Transforming growth factor-beta 1 signaling contributes to Caco-2 cell growth inhibition induced by 1,25(OH)2D3 Am J Physiol Gastrointest Liver Physiol, October 1, 2002; 283(4): G864 - G874. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
). Data are representative of three
independent experiments. *, 
,
,
P 
