ICANCER RESEARCH57. 2020-2028. May 5. 19971
Increasing c-FMS (CSF-1 Receptor) Expression Decreases Retinoic Acid
Concentration Needed to Cause Cell Differentiation and
Retinoblastoma Protein Hypophosphorylation1
Andrew Yen,2 Rhonda Sturgill, and Susi Varvayanis
Cancer Cell Biology Laboratory. Department of Pathology, College of Veterinary Medicine. Cornell University, Ithaca. New York /4853
ABSTRACT
Increasing the expression of c-FMS (colony-stimulating factor 1 recep
tor) by introduction of a transgene reduced the concentration of retinoic
acid or 1,25-dihydroxy
vitamin D3 needed to cause myeloid or monocytic
cell differentiation and hypophosphorylation of the retinoblastoma tumor
suppressor protein (RB) typically associated with cell cycle G0 arrest and
differentiation of HL-60 human myelo-monoblastic precursor cells. The
data are consistent with a model in which signals originating with retinoic
acid and c-FMS integrate to cause differentiation, RB hypophosphoryla
tion,
and
G0 arrest.
Furthermore,
these
two signals
can compensate
for
each other. Three HL.60 sublines described previously (A. Yen et aL, Exp.
Cell Res., 229: 111—125,
1996) expressing low (wild-type HL-60), interme
diate, and high cell surface c-FMS were treated with various concentra
tions
of retinoic
acid.
The
lowest
concentration
tested,
10_s
M, induced
significant differentiation of only the high c.FMS-expressing cells, with no
accompanying
hypophosphorylated
RB or G0 arrest, The low and inter
mediate c-FMS expressing cells showed no induced differentiation,
hy.
pophosphorylation
concentration,
of RB,
iO'7
or G0 arrest.
M, induced
significant
A 10-fold
higher
differentiation
retinoic
of both
acid
inter
mediate and high c-FMS-expresslng cells. It induced RB hypophosphory
lationonly in high c-FMS-expressingcells but with no accompanyingG0
arrest in any of the cells. The highest retinoic acid concentration, 10_6 M,
elicited differentiation,
intermediate,
and
hypophosphorylation
high
c-FMS-expressing
of RB, and G0 arrest in low,
cells.
As the concentration
of
retinoic acid increased, cell differentiation, hypophosphorylation of RB,
and G0 arrest were progressively
elicited within this ensemble ofcells with
different c.FMS expression levels. Thus, for example, at the lowest con
centration of retinoic acid, expression of high enough c-FMS still allowed
differentiation. At higher concentrations, progressively less c-FMS was
needed for differentiation. The apparent threshold for the sum of the
retinoic acid plus
hypophosphorylation
c-FMS originated
signals to elicit differentiation,
of RB, and G0 arrest increased, in that order. Thus
factors (see Ref. 7 for review). Both RARs and RXRs must be ligand
activated to induce differentiation (8). RARs and RXRs are members
of the steroid-thyroid hormone receptor superfamily, as is the vitamin
D receptor, which can also heterodimerize with RXRs. In contrast,
c-FMS is a peptide hormone receptor of the platelet-derived growth
factor family (see Refs. 9—11 for reviews). It is the receptor for
colony-stimulating factor 1 and regulates the cell cycle and differen
tiation of myelo-monocytic hematopoietic cells. c-FMS is a trans
membrane tyrosine kinase receptor that signals through a prototypical
mitogenic ras/raf-mediated kinase cascade (12), leading to phospho
rylation and activationof MAPK and the phosphorylationof tran
scription factors such as Elk- 1. Although such kinase cascade signal
ing pathways are prototypical mitogenic signaling pathways, in this
instance they can also drive cells from proliferation to arrest. Retinoic
acid can thus determine if this signal is mitogenic or growth arresting
(1). Consistent with this, retinoic acid-induced myeloid differentiation
and G0 arrest are accelerated by ectopic expression of activated RAF,
a prototypical serine threonine kinase associated with proliferation
and transformation (13). Vitamin D3-induced monocytic differentia
tion is likewise accelerated. Whereas retinoic acid can determine if
c-FMS signaling promotes or inhibits mitogenesis, the putative
c-FMS-originated signal can enhance the cellular effect of the retinoic
acid- or vitamin D3-originated signal. Thus, the two pathways appear
to regulate each other. How these steroid and peptide hormone sig
naling pathways interact with each other is not known. One obvious
question is whether upstream events might be involved that allow the
c-FMS-originated signal to enhance cellular sensitivity to retinoic acid
or vitamin D3, allowing cells to respond to lower doses of these
ligands.
The RB tumor suppressor gene is a putative immediate regulator of
the
cell cycle and, by extension, cell differentiation (see Refs. 14—16
G0arrest havedifferent signalthresholdrequirements.1,25-Dihydroxy
for
reviews). The regulatory activity is believed to be exercised
vitamin
D3, also a ligand for a member
of the steroid thyroid
hormone
receptor superfamily, caused monocytic differentiation with a similar through its phosphorylation by cyclin-driven kinases. However, re
c-FMS dependency, indicating that these effects characterize both myeloid cently reported nonconcordances between expression of cyclins and
and monocytic differentiation.
RB phosphorylation (17, 18), as well as nonconcordances between
phosphorylation and the cell cycle phases (1, 13, 19, 20), and also
expression levels and growth arrest (21—24),indicate that the function
INTRODUCTION
of RB is still unclear. This ambiguity is furthered by nonconcordances
between apparent available free E2F, a putative principal binding
Retinoic acid provides a signal that induces the growth arrest and
partner of RB, and cell proliferation (see Refs. 20 and 25 for reviews).
myeloid differentiation of a human myeloblastic leukemia cell line,
Giventhatthe retinoicacidandFMSsignalingpathwaysbothregulate
HL-60, in a process that depends on another signal originating with
the c-FMS receptor (1). The RARs3 (2—4)and the related RXRs (5, 6) cell cycle arrest and differentiation, it becomes of potential interest to
can homo- or heterodimerize and act as ligand-activated transcription
determinetheir effecton RB phosphorylation.Manipulationof either
of these pathways provides a means of determining how RB phos
phorylation and expression changes when cell proliferation and dif
Received 12/6/96; accepted 3/25/97.
The costs of publication of this article were defrayed in part by the payment of page
ferentiation are thus altered.
charges. This article must therefore be hereby marked advertisement in accordance with
HL-60 is a human myeloblastic leukemia cell line (26). It was
18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported in part by grants from NIH (United States Public Health
originally designated as a cell line of acute promyelocytic origin, and
Service), United States Department of Agriculture, and the Children's Leukemia Fund.
later, it was reevaluated as myeloblastic (27). The cell line has
2 To whom requests for reprints should be addressed. Phone: (607) 253-3354; Fax:
amplified myc and activated ras, and it is p53 negative. It retains the
(607) 253-3317.
3 The abbreviations used are: RAR, retinoic acid receptor; RXR, retinoid X receptor;
capability to undergo myeloid or monocytic differentiation in re
RB, retinoblastoma; PKC, protein kinase C; AP-l , activator protein I; NBT, nitroblue
sponse to a variety of agents (see Ref. 28 for review). The most
tetrazolium; c-FMS, CSF- I receptor; MAPK, mitogen-activated protein kinase; VD3,
1.25-dihydroxy vitamin D3.
commonly used inducers are retinoic acid (29), a myeloid inducer, and
retinoic acid-induced cell differentiation, RB hypophosphorylation, and
@
2020
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1997 American Association for Cancer Research.
c-FMS REGULATES RB. CELL CYCLE. AND DIFFERENTIATION
l,25-dihydroxy vitamin D3 (30), a monocytic inducer. Typical treat
ments with l0_6 M retinoic acid or 0.5 X l06 M vitamin D3 induce
onset ofphenotypic conversion and also G3/G0 arrest by 48 h, a period
corresponding to approximately two division cycles in the subline
studied. Early events during the first —24 h of this cascade are
differentiation lineage independent, whereas late events during the
second —24 h specify myeloid or monocytic differentiation (31).
Progressive differentiation and arrest result in almost entirely differ
entiated populations by 96 h. Analysis of RB expression produced
unanticipated results. Unlike some other cells, RB is phosphorylated
in all cell cycle phases; i.e., all of the RB protein in proliferating cells
is phosphorylated (1, 13, 19). The degree of phosphorylation increases
as does total RB per cell as HL-60 cells advance through progressive
cell cycle phases (19). In contrast, cells that are induced to G1/G0
arrest and differentiate express less RB per cell and have essentially
only hypophosphorylated RB. However, it is possible in some cases to
induce G0 arrest and differentiation prior to detectable hypophospho
rylated RB (13). Hypophosphorylated RB thus does not necessarily
precipitate cell cycle arrest or differentiation but rather appears as a
feature of arrested, differentiated cells, much as might be anticipated
with metabolic slow down, for example. In contrast, differentiation
and arrest are consistently associated with an unanticipated reduced
expression of this putative growth/cell cycle suppressor (21, 22). The
novelty of some of these results makes HL-60 an intriguing cell line
@
for further analysis of potential RB function and its coupling to either
growth arrest or differentiation.
The retinoic acid- and c-FMS-originated signals may converge to
regulate RB expression. In HL-60 cells, both proliferation and retinoic
acid-induced myeloid differentiation or 1,25-dihydroxy vitamin D3-
induced monocytic differentiation depend on a c-FMS-originated sig
approximately 2-fold greater mean cell surface c-FMS expression
than HL-60, whereas the Sort 10 cells have approximately 8-fold
higher c-FMS expression than HL-60. As described previously (32),
Sort 10 cells represent an asymptotic limit for the extent of allowable
c-FMS overexpression in HL-60 cells.
MATERIALS
AND METHODS
Cells and Culture
cells
were cultured
Conditions,
in RPMI
HL-60 human promyelocytic
1640 supplemented
with
leukemia
10% heat-inactivated
FCS in a humidified atmosphere of 5% carbon dioxide. Stock cultures were
maintained in continuous exponential growth by reculturing every 2 or 3 days
at an initial density of 0.2 X 106 or 0.1 X 106 cells/ml, respectively. The
wild-type
strain,
designated
F8, is not atypical
of HL-60
cells
that
have
undergone sustained passage, and has a doubling time of approximately 20 h,
and proliferates
exponentially
in the range
of approximately
0.1
X 106 to
3.0 X 106 cells/mI. The stable transfectant FMS/I overexpresses c-FMS
detectable on the cell surface. Its creation and properties have been previously
reported (1). FMS/I and its derivatives generated by fluorescence-activated cell
sorting
were maintained
as stocks
in the continuous
presence
of 400
@zg/ml
hygromycin B (Boehringer Mannheim, Indianapolis, IN) as described previ
ously (1).
Assays of cell differentiation, cell cycle phase distribution, and RB protein
were performed on sequential cell samples from cultures initiated at a density
of 0.2 X 106 cells/ml with 20—30ml per 75-cm2 flask. Inducers were added to
these cultures to make the stated final concentrations of retinoic acid or
1,25-dihydroxy
vitamin
D3 as described
previously
(31).
Retinoic
acid was
added from a stock of iO@ M in ethanol. VD3 (Solvay Duphar BV, Da Weesp,
The Netherlands) was added from a stock of 10 M in ethanol. Carrier blanks
had no effect on the measured parameters in control experiments. The exper
iments shown are each a typical case among three independent repeat exper
iments.
nal (1). In exponentially proliferating cells without retinoic acid, a
c-FMS-originated signal promotes proliferation. Adding retinoic acid
or 1,25-dihydroxy vitamin D3 causes the c-FMS-originated signal to
now promote differentiation and G0 arrest. This differentiation and G@
arrest are associated with down-regulation of RB expression and
conversion from the hyper- to the hypophosphorylated form (1, 8, 21,
22). Overexpressionof the c-FMS transmembranetyrosine kinase
receptor or an activated RAF in stable HL-60 transfectants results in
accelerated differentiation and arrest in response to retinoic acid and
1,25-dihydroxy vitamin D3 (1, 13, 32). Coupled to this, down-regu
lation of RB expression anteceding overt cell cycle or cell differen
Assays for Cell Proliferation and Differentiation. Assays for cell prolif
eration and differentiation were performed as described previously (8, 3 1). Cell
density in experimental cultures was measured by repeated counts with a
hemacytometer. Viability was assessed by exclusion of 0.2% trypan blue dye
and was routinely in excess of 90% in all cultures reported. The distribution of
cells in the cell cycle was determined by flow cytometry using propidium
iodide-stained nuclei. Cells (0.5 X 106) were harvested at each indicated time
and resuspended in 0.5 ml of hypotonic propidium iodide (0.05 mg/mI pro
pidium iodide, 1 mg/liter sodium citrate, and 0.1% Triton X-l00) and stored
refrigerated
and protected
from light until analyzed.
was done with a multiparameter
Flow cytometric
analysis
dual laser fluorescence-activated
cell sorter
tiation changes is also accelerated, although conversion to the hy
(EPICS; Coulter Electronics, Hialeah, FL) using 200 mW of488-nm excitation
from a tunable argon ion laser. Functional differentiation to a mature my
pophosphorylated form is not. Thus it appears that the signals from
elomonocytic
retinoic acid or 1,25-dihydroxy vitamin D3 and c-FMS converge at
some point to regulate RB expression and ultimately cell cycle arrest
and differentiation. This motivates interest in the effects of the relative
dose of each of these signals on RB.
The present studies determine the interactive effects of the retinoic
acid- or l,25-dihydroxy vitamin D3-originated signal and the c-FMS
originated signal on RB hypophosphorylation, cell cycle arrest, and
cell differentiation. In these studies, the effect of overexpression of
c-FMS, the receptor for colony-stimulating factor 1, on the ability of
HL-60 cells to undergo myeloid or monocytic differentiation, convert
their RB protein to the hypophosphorylated form, and undergo G1/
sayed by phorbol-12-myristate-13-acetate (Sigma Chemical Co., St. Louis,
MO) inducible intracellular reduction of nitroblue tetrazolium to formazan.
Cells (0.2 x 106)were harvested at the indicated times and resuspended in 0.2
G0-specific growth arrest was analyzed using HL-60 sublines stably
transfected with a c-FMS cDNA. Three HL-60 lines were studied,
including wild-type HL-60, FMS/I, and Sort 10. The derivation and
properties of the FMS/I and Sort 10 stable transfectants have been
previously described (1, 32). FMS/I was derived by stably transfecting
HL-60 cells to overexpress ectopic c-FMS (1). Sort 10 was derived
from FMSII by successive iterations of fluorescence-activated cell
sorting for the highest FMS-expressing cells in the population and
amplification of the sorted cells in culture (32). The FMS/1 cells have
ml of 2 mg/mI
phenotype capable of inducible oxidative metabolism
nitroblue
tetrazolium
in PBS containing
200 ng/ml
was as
phorbol
12-myristate-13-acetate. The cells were incubated in a water bath for 20 mm
at 37°Cand then scored using a hemacytometer for the percentage of cells
containing
intracellular
superoxide-precipitated
formazan.
Western Analysis of RB. Western analysis of RB protein expression and
phosphorylation
in whole cell lysates harvested
at the indicated times were
performed
as described previously
(1, 33). At indicated times, 1 X 106 cells
were harvested and fixed in methanol and stored at —
20°C
until used. The cells
were resuspended in 25—50
Ml of loading buffer [6% SDS, 4 M urea, 4 mM
EDTA, 125 mMTris (pH 6.9), 0.25% bromphenol blue, and 35 @.t1/ml
(3-mer
captoethanol],
boiled,
and gel-electrophoresed
resolving gel (37.5:1, acrylamide:bis).
using a 4% stacking
gel and 6%
0.5 X 106 cells were loaded per lane for
PAGE. Proteins were electrotransferred from the gel to a nitrocellulose mem
brane, which was probed with a murine antihuman RB monoclonal antibody
[RB Gene Product (mAbl) Monoclonal Antibody, Triton Diagnostics, Ala
meda, CA; or Zymed Laboratories, South San Francisco, CA]. Detection was
done using enhanced chemiluminescence and a horseradish peroxidase-conju
gated secondary antimurine antibody (ECL kit; Amersham Ltd.). Film images
of the membrane were densitometrically scanned to graphically represent the
2021
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1997 American Association for Cancer Research.
c-FMS REGULATES RB. CELL CYCLE. AND DIFFERENTIATION
relative
amount
of faster
migrating
unphosphorylated
RB protein
versus
Control
the
slower migrating protein phosphorylated at multiple sites. The scan axis
RA 10-8M
100
100
represents 20 mm on the gel.
c-FMS Analysis. Cells were immunofluorescently stained for cell surface
80
80
20
20
c-FMS expression essentially as described previously ( 1, 32). To assay for cell
surface FMS expression, 5 X 106 cells were harvested, washed once in 150 p@l
of PBS (pH 7.2) and gently resuspended, adding 15 pAof antibody against
c-FMS (AB-2, 100 @zg
of IgG2b per ml of 0.05 M sodium phosphate buffer
containing
0. 1% sodium
aside
and 0.2%
gelatin;
Oncogene
Science,
Cam
bridge, MA). The suspension was gently mixed, incubated for 20 mm on ice
in the dark, and gently washed
with 150
@l
of PBS using a table-top
centrifuge
(800 rpm, 5 mm). The cells were resuspended in 95 p3 of PBS, to which 5 @l
of fluorescein-conjugated goat anti-rat antibody, whole molecule (Cappel
Division,
Organon
Teknika
Corp.,
West
Chester,
PA; reconstituted
per the
A
0
24
manufacturer's directions in sterile distilled water and then diluted I:10 in
normal
goat serum/PBS)
was added and then incubated
48
72
96
0
24
48
72
96
72
96
Hours
Hours
on ice for 20 mm. They
were washed in 150 @zlof PBS and resuspended in 500 @.d
of PBS. Flow
cytometric analysis was performed with a multiparameter dual laser fluores
RA 10-7M
RA 1O-6M
cence-activated cell sorter (EPICS) using 200 mW of 488-nm excitation from
a tunable argon ion laser. Emitted fluorescence was collected through a
457—502-nm band laser blocking
filter and a 550-nm
long-pass
dichroic
mirror.
Green fluorescence was detected with a photomultiplier tube masked with a
525-nm band pass filter. Photomultiplier voltages ranged from approximately
1400 to 1700 V. Forward angle light scatter was used as a trigger signal and
to eliminate debris for analysis. The ADC (analogue to digital converter)
trigger was forward angle light scatter. c-FMS expression means were deter
mined
on the basis of 10,000 events
analyzed
by onboard
statistical
analysis
packages.
Retinoic Acid Uptake. Cellular uptake of trace labeled retinoic acid was
measured using tritiated retinoic acid and liquid scintillation counting. HL-60
@
or Sort 10 cells (2 x 106) were resuspended in l0-ml cultures containing either
l06 M, l0@ M,
or 0 M (control) labeled retinoic acid. A l0@ M labeled
retinoic
acid stock
was made
from 250
@zlof 4.3
@zM[3H]retinoic
Ci/mmol,
152 mCi/mg, 0.2 mCi/ml [3H]9-cis-retinoic
plus 32.6
@zlof 8.67 mM retinoic
acid in ethanol.
retinoic
acid stock were added
to 10-mI cultures
to generate
M. The concentration
humidified atmosphere of 5% CO2. The cells were centrifuged to a pellet and
washed in 5 ml of cold PBS. The washed pellet was resuspended in S ml of
ICN Radiochemicals,
Irvine,
CA),
mixed,
and
transferred to scintillation vials for counting. Cells (2 X 106)treated with lO_6
M retinoic
acid
took
up
approximately
1%
of
the
total
radioactivity
(—8.2 x l0@cpm) available in culture, indicating that availability of retinoic
acid is not rate limiting.
RESULTS
Retinoic Acid Induced Myeloid Differentiation.
Enhanced c
FMS expression enables cells to differentiate in response to lower
concentrations of retinoic acid. HL-60, FMSII, and Sort 10 cells were
cultured with 0 M (control), l0_8 M, l0@ M, or lO_6 M retinoic acid.
The lO_6 M dosage is the standard dosage most commonly used to
elicit myeloid differentiation of wild-type HL-60 cells. The percent
age of cells having differentiated was measured by the functional
myelo-monocytic differentiation marker, inducible oxidative metabo
lism.
Fig.
1 shows
the percentage
of cells
that have
0
24
48
Hours
differentiation
of HL-60 sublines expressing progres
final concentra
lO_6 M corresponded
were incubated in 15-ml conical centrifuge tubes for 1 h at 37°Cin a
fluid (Universol;
96
inducible oxidative metabolism detected by intracellular NBT reduction. Percentage of
cells which have differentiated (Y axis) during treatment (X axis, h) of HL-60 (parental
cells; 0), FMS/I (A), and Sort 10 (U) transfectants with 0 M (control) (top left), iø@@
M
(top right), l0@ M (bottom left), and l06 M (bottom right) retinoic acid.
acid
to labeling with 0.18 MCi/ml, a relatively low dose unlikely to cause unantic
ipated effects due to radiation-induced
damage during the 1 h oflabeling.
Cells
scintillation
72
sively more c-FMS. Differentiation was measured by the percentage of cells expressing
stock contained 0.38% [3H]retinoic acid. Stocks of l0@@M and l0@ M were
made by dilution of the labeled l0@ M stock in ethanol. Ten @.dof labeled
tions of l0_6 M, l0@ M, or l0'
48
Fig. 1. Retinoic acid-induced
acid (46.0
retinoic
24
Hours
acid; Amersham Ltd)
The resulting
0
differentiated
during exposure to 0 M (control), lO_8 M, iO@ M, or lO_6 M retinoic
acid. Control cells underwent no significant differentiation during the
96-h course of observation. The generation time of HL-60 cells is
approximately 20 h, and the period of observation thus corresponds to
over four cell cycles. But as reported previously (32), the generation
time of the transfectants is longer; the generation time for Sort 10 is
up to approximately 29 h, for example. (The average after prolonged
culture for multiple cryopreserved samples restored to culture can
revert to approximately 23 h.) At l0_8 Mretinoic acid, the percentage
of differentiated cells was indistinguishable from untreated controls
for HL-60 and FMS/I, but a significant fraction of Sort 10 cells
differentiated. The onset of the increase was detected at approximately
48 h. Thus, only the cells with the highest c-FMS expression exhibited
any apparent differentiation in response to lO_8 Mretinoic acid. When
the different cell lines were treated with a 10-fold higher concentra
tion of retinoic acid, l0@ M, then a significant number of HL-60 cells
differentiated with kinetics similar to that observed for Sort 10 cells at
the lower (10_8 M) retinoic acid concentration. Differentiation in
duced in FMS/I cells was enhanced relative to HL-60. The greatest
amount of induced differentiation was by Sort 10 cells. Onset was
again at approximately 48 h. Comparing these cell lines, the amount
of differentiation increased as the amount of c-FMS expression in
creased. When the cells were treated with a further 10-fold higher
concentration of retinoic acid, lO_6 M, this relationship persisted. Sort
10 cells again exhibited the greatest fraction of cells differentiated at
all times after onset of differentiation, except at the end of the
observation period when the majority of cells in all cell lines had
finally differentiated. Sort 10 cells consistently differentiated better
than HL6O, and FMS/I was consistently intermediate at different
times, as well as at different inducer concentrations.
If the amount of differentiation at a specific time, 72 h, for each cell
line is considered, then increasing the retinoic acid concentration
consistently produced increasing differentiation. Likewise for each
retinoic acid concentration, increasing FMS expression resulted in
2022
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1997 American Association for Cancer Research.
c-FMS REGULATES RB. CELL CYCLE. AND DIFFERENTIATION
Sort 10 cells as a reproducible small subpopulation that migrated
faster than (i.e., to the right of) the main peak, which is the hyper
OMRA
L&.@
Li@.
..@.
LÀ
. A. L@.
phosphorylated RB. This subpopulation increased to a prominent peak
when the retinoic acid concentration
creasing
the retinoic
acid concentration
was increased to l0_6 M. In
to 10@ M caused
hypophos
phorylated RB in all the cell lines regardless of their c-FMS expres
sion level. RB hypophosphorylation is first evident in Sort 10 cells
treated with i07 M retinoic acid, but significant differentiation was
first evident in Sort 10 cells treated with 10_8 M retinoic acid. Similar
measurements made at 48 h showed no detectable hypophosphory
lated RB for any of the cell lines (data not shown). At 96 h, all cell
lines treated with l0@ M or lO_6 M retinoic acid had hypophospho
10-SMRA
rylated
RB, but none of the cell lines showed
hypophosphorylated
RB
with 108 Mretinoic acid, indicating that differences observed at 72 h
reflect at least in part differential kinetics. Thus, elevated c-FMS
expression increases cellular sensitivity to retinoic acid-induced RB
hypophosphorylation, but the signal threshold for eliciting hypophos
1O-7MRA
phorylation is higher than that for inducing differentiation.
@
..@
Interestingly, in some cases, down-regulation of RB appeared more
responsive than hypophosphorylation to the retinoic acid- or c-FMS
originated signaling and more closely paralleled differentiation. For
.
1O-6MRA
example,
as described
above,
lO_8 M retinoic
acid caused
differenti
ation only in Sort 10 cells. At 72 h, there was still no RB hypophos
phorylation,
@
@
..@
HL-60
.
FM.S/ I
was down-regulated
in these cells. In
rylation. In contrast, RB expression was down-regulated. This is
SORT1O
Fig. 2. Retinoic acid-induced hypophosphorylation of RB protein expressed by HL-60
sublines expressing progressively more c-FMS. Densitometric scans of Western blots of
RB protein in (left to right) HL-6O (parental), FMS/I, and Sort 10 transfectants treated
with (top to bottom), 0 M (control; top), lO_8 M, l0@ M, and 1O_6 M (bottom) retinoic acid
for 72 h. The X axis is 20 mm of the gel with the bottom of the gel oriented to the right.
The optical density representing the relative amount of protein is the Yaxis. Hypophos
phorylated RB has greater relative mobility on electrophoresis
but RB expression
another example, 10 M retinoic acid could cause detectable differ
entiation in HL-60 cells by 72 h, but no detectable RB hypophospho
than hyperphosphorylated
RB and thus appears when present as a subpopulation to the right of the hyperphospho
rylated RB peak. In the Sort 10cells treated with lO@ Mretinoic acid, approximately 15%
of the RB is hypophosphorylated, as determined by integrating the densitometric scan.
consistent with previous observations (13).
In contrast to the above, enhanced c-FMS expression did not allow
cells to G1/G0 arrest in response to lower retinoic acid concentrations.
Fig. 3 shows the percentage of cells with 01/Go DNA measured by
flow cytometry during exposure to 0 M (control), l0@ M, l0@ M, or
l0_6 M retinoic acid. Untreated control cells showed no significant
increase during the 96-h course of observation. This was paralleled by
continued increases in cell density, except at 96 h, when growth was
retarded due to depleted culture medium (data not shown). Treated
with lO_8 M retinoic acid, none of the cell lines showed any apparent
0 @/G@
growth arrest evidenced by enhancement in relative numbers of
G1/GØcells. Likewise, this was paralleled by continued increases in
cell density. Interestingly, there was no abatement at 96 h, possibly
reflecting a growth-sustaining capability of low doses of retinoic acid
more differentiation at 72 h, unless the retinoic acid dose was below
the apparent threshold for cellular response. In this respect, the reti
noic acid- and c-FMS-originated signals thus appeared to have similar
additive effects on differentiation, affecting both the rate of differen
tiation and the maximum percentage differentiated.
Enhanced c-FMS expression allows cells to hypophosphorylate RB
in response to lower concentrations of retinoic acid. It has been
established previously that the RB protein in proliferating HL-60 cells
is all in the hyperphosphorylated state (1, 13, 19, 24). Occurrence of
hypophosphorylated RB is associated with cell differentiation (13). As
shown previously, hypophosphorylated RB in HL-60 cells is distin
guishable from hyperphosphorylated
RB on a Western blot as a
subpopulation of RB molecules with greater relative mobility (1, 13,
19, 24). As shown previously also (8), as little as 4% hypophospho
rylated RB can be consistently detected this way. Fig. 2 shows the
densitometric scans of Western blot images of the RB protein for
HL-60, FMS/I, and Sort 10 cells treated with 0 M (control), l0_8 M,
io—7
M,or 10_6Mretinoicacidfor 72 h. Noneofthe untreatedcontrol
cell lines showed any hypophosphorylated RB, indicating that over
expression of c-FMS per se had no detectable effect on RB phospho
rylation. Treatment with 10_8 M retinoic acid also caused no hy
pophosphorylation, although it caused approximately 30% of the Sort
10 cells, the cells expressing the highest amount of c-FMS, to differ
entiate. Treatment with a 10-fold higher concentration of retinoic acid,
i0—@
M, caused hypophosphorylated RB in only the Sort 10 cells, but
not HL-60 or FMSII. The hypophosphorylated RB was first evident in
reported previously for HL-60 cells. Increasing the retinoic acid
concentration 10-fold to 10@ M caused no apparent arrest. Sort 10
cells showed no enrichment in G1/GØ cells distinguishable from
FMSII or HL-60. There was a slight but reproducible increase in
percentage of G1/GÃc̃ells at late times, which was similar for all the
cell lines. This was not considered an indication of arrest because
there was no significant inhibition of increasing cell density through
out the 96-h period of observation (data not shown). The slight
changes in percentages may reflect shifts in the relative durations of
cell cycle phases as the cells continue to proliferate. Increasing the
concentration of retinoic acid 10-fold further to the standard concen
tration, lO_6 M, used to induce differentiation and G1/GØarrest of wild
type HL-60 cells, caused an enrichment in G@/G@cells for all of the
cell lines, consistent with a previous report (32). Thus, enhanced
c-FMS expression does not apparently enable cells to G1/G0 growth
arrest in response to a lower concentration of retinoic acid.
The trend of the above data indicates that there is a c-FMS
dependent threshold for retinoic acid to elicit differentiation, RB
hypophosphorylation,
or G1/G0 growth
arrest.
Differentiation
can be
elicited at the lowest (lO_8 M) concentration, RB hypophosphoryla
tion at a higher (l0@ M)concentration, and G1/GØgrowth arrest at the
highest (10—6M) concentration tested. Increasing the retinoic acid
2023
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1997 American Association for Cancer Research.
c-FMS REGULATES RB. CELL CYCLE. AND DIFFERENTIATION
Control
myelo-monocytic differentiation marker, oxidative metabolism. 0.5 X
RA 1O-8M
100
100
80
80
0
lO_8 MVD3 caused a low but reproducible percentage of all cell lines
to differentiate by late times. The extent was greater for the two
transfected cells, Sort 10 and FMS/I, compared to the HL-60 parent.
Using 0.5 X iO@7 M VD3 increased the percentage of differentiated
cells for all cell lines, but Sort 10 showed the greatest percentage of
differentiated cells. Using 0.5 X lO_6 M VD3, a concentration that is
typically used to elicit monocytic differentiation, all cell lines differ
@60
060
0
0
#40@
@t40
20
20
entiated, consistent with a previous report (32). The FMS/I cells
0
24
48
72
Hours
96
0
24
48
72
96
72
96
differentiated to a greater extent than HL-60 parents at any time, and
the Sort 10 cells differentiated likewise to a greater extent than the
FMS/I cells. As for retinoic acid, at lower VD3 concentrations,
enhanced c-FMS expression increased differentiation. As with reti
noic acid, increasing c-FMS expression could compensate to some
extent for a low VD3 concentration and vice versa.
Enhanced c-FMS expression also caused RB hypophosphorylation
in response to lower concentrations of VD3. Fig. 5 shows the densi
tometric scans of RB protein Western blots for HL-60, FMS/I, and
Sort 10 cells treated with 0 M (control), 0.5 X 10_8 M, 0.5 X iO@ M,
or 0.5 x l0_6 M VD3 for 48 h. Cells treated with 0 M (control) or
0.5 X lO_8 M VD3 showed no hypophosphorylated RB. A higher
concentration of 0.5 X iO@ M induced hypophosphorylated RB in
Sort 10 cells, but not HL-60 or FMS/I. This is analogous to the case
of retinoic acid-induced RB hypophosphorylation at 72 h. VD3, at
0.5 X 10_6 M, a concentration typically used to elicit HL-60 mono
cytic differentiation and arrest, caused all the cell lines to express
hypophosphorylated RB. At 24 h, none of the cell lines treated with
any of these VD3 concentrations exhibited hypophosphorylated RB
(data not shown). At 72 h, none of the cells treated with 0.5 X l0_8
Hours
RA 1O-7M
RA 1O-6M
80
0
@60
0
40
20
0
24
48
72
Hours
96
(1
24
48
Hours
Fig. 3. Percentage of cells in G1/G0for HL-60 sublines expressing progressively more
c-FMS. The percentage of cells with G1/G0 DNA was determined by flow cytometry.
Percentage of cells in G1/G0 (Y axis) during treatment (X axis, h) of HL-60 (parental cells;
0), FMSII (A), and Sort 10 (U) transfectants with 0 M (control, top left), l0@ M(top
right), 10@ M (bottom left), and 106 M (bottom right) retinoic acid.
M showed
detectable
expression
RB
(data
when cells were treated
not
shown).
induced hypophos
phorylated RB in each of the cell lines (data not shown). For each cell
VD3-induced Monocytic Differentiation. The above findings for
line, the percentage of hypophosphorylated RB increased with in
creasing VD3 concentration, and at each concentration, the percentage
of hypophosphorylated RB increased with increasing c-FMS expres
retinoic acid motivate the question of whether these effects extend to
VD3, which induces monocytic differentiation through a related re
ceptor. The effects of c-FMS overexpression are qualitatively similar
for VD3 and retinoic acid, but in VD3 they are not as pronounced.
HL-60, FMS/I, and Sort 10 cells were cultured with 0 M (control),
0.5 X i0@8 M, 0.5 x l0@ M, or 0.5 X 106 M VD3 in lieu of retinoic
acid, and cell differentiation was assayed as above at sequential times.
Fig. 4 shows the percentage of cells that expressed the functional
Table
RBhypophosphorylation,
I Lowest retinoic acid concentration (M)needed to elicit differentiation,
or
10expressing
G0 arrest for HL-60 sublines HL-60, FMS/J. and Sort
progressively more
c-FMSRBSubline
arrestSort
10-8M
Differentiation
Hypophosphorylation
l0_6FMSII
10
l0'
iO7
l0_6HL-60
iO@
106
10@
10_6
VD3
10-7M
VD3
G0
lO_6
10-6M
Fig. 4. Vitamin D3-induced differentiation of
HL-60 sublines expressing progressively more
c-FMS. Differentiation was measured by the per
centage of cells expressing inducible oxidative
metabolism detected by intracellular NBT reduc
tion. Percentage of cells that have differentiated
(Y axis) during treatment (X axis, h) of HL-60
(parental cells; 0), FMS/I (A), and Sort 10 (U)
transfectants with 0.5
x
l0'
M (left),
0.5 x l0@ M (middle), and 0.5 X l0_6 M (right)
1,25-dihydroxy vitamin D3.
0
24
48
72
But
with 0.5 X 10_6 M of VD3 (Fig.
6). At 96 h, all of the tested VD3 concentrations
concentration allows cells expressing lower c-FMS to respond.
Table 1 summarizes the results. Thus, to some extent it appears that
the retinoic acid- and c-FMS-originated signals can compensate for
each other as if it is their sum that determines response.
VD3
hypophosphorylated
treatment with 0.5 X l0@ M induced hypophosphorylated RB in all
of the cell lines. Comparing HL-60, FMS/I, and Sort 10, the relative
amount of hypophosphorylated RB increased with increasing c-FMS
96
Hours
0
24
48
Hours
72
96
0
24487296
Hours
2024
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1997 American Association for Cancer Research.
c-FMS REGULATES RB, CELL CYCLE. AND DIFFERENTIATION
OMD3
inhibited cell density increases. Like retinoic acid, the intermediate
VD3 concentration induced a modest G1/G0 enrichment (by up to
—20%in some cases) but no associated abatement in increasing cell
density. This was consequently not taken to indicate 01/G0-specific
growth arrest, as occurred at the highest VD3 concentration, but rather
a redistribution of cell cycle phases as reported previously for these
cells (32).
As for retinoic acid, the lowest concentration of VD3 could elicit
cell differentiation if sufficient c-FMS receptors were overexpressed.
An intermediate concentration was needed to elicit RB hypophospho
rylation, and this occurred first in only the highest c-FMS-expressing
cells. The highest concentration was needed to elicit G1/G0-specific
growth arrest. Thus, as for retinoic acid, the signal threshold for
response was lowest for cell differentiation, higher for RB hypophos
phorylation, and highest for G1/G0-specific growth arrest. The effects
of c-FMS overexpression on VD3-induced monocytic differentiation
are thus similar to those for retinoic acid-induced myeloid differenti
ation. In conclusion, c-FMS enhanced the sensitivity to both these
inducers, indicating that the effects with retinoic acid induced myeloid
differentiation extend to the case of VD3-induced monocytic differ
entiation.
LA@
E.L@.
LA.@..
LA.@.
LA...
L.A.
@LL@.. [.@
.
@L&
.
__
10-SM
D3
1O-7M D3
1O-6M D3
Cellular Retinoic Acid Uptake. Overexpressionof c-FMS had no
significant effect on uptake of exogenous retinoic acid (Table 2).
Exponentially proliferating HL-60 or Sort 10 cells were cultured for
HL-60
FMS/I
5ORT1O
Fig. 5. VD3-induced hypophosphorylation of RB protein expressed by HL-60 sublines
expressing
progressively
more
c-FMS.
Densitometric
scans
of Western
blots
of RB
proteinin (leftto right)HL-60(parental),FMS/I,and Sort 10transfectantstreatedwith
(top to bottom) 0 M (top), 0.5 X 10@ M, 0.5 X 10@ M, 0.5 X l06 M (bottom) VD3 for
48 h. The X axis represents 20 mm of the gel with the bottom of the gel oriented to the
right.The opticaldensityrepresentingthe relativeamountof proteinis the ordinate.
sion. Increasing c-FMS or VD3 had similar downstream effects on RB
hypophosphorylation.
Thus, as for retinoic acid, elevated c-FMS
expression increased cellular sensitivity to VD3 induced RB hy
pophosphorylation, but the threshold for response is higher than that
for inducing differentiation.
In contrast to the above, enhanced c-FMS expression did not cause
cells to G@/G@arrest in response to lower VD3 concentrations. As
above, HL-60, FMSII, and Sort 10 cells were cultured with 0 M
(control), 0.5 X 10_s M, 0.5 X l0@ M, or 0.5 X lO_6 M VD3. The
percentage of cells having G1/G0 DNA (Fig. 7), measured by flow
cytometry,
and the cell density
(data
not shown)
were measured
at
subsequent times until 96 h. As with retinoic acid, there was no arrest
apparent by both G@/G@enrichment and inhibition of increases in cell
density for any of the cell lines treated with the lower concentrations
of VD3, but 0.5 X lO_6 M VD3 induced G1/G0 enrichment and
VD3
1 h with 0 M(control), lO8 M, l0@ M,or lO_6 Mtrace tritium-labeled
retinoic acid. An equal number of labeled cells were harvested in each
case, washed, and solubilized for radioactive liquid scintillation
counting. Fig. 8 shows the results. Cellular uptake of retinoic acid was
approximately proportional to the exogenous retinoic acid concentra
tion. Successive reductions in exogenous concentrations by 10-fold
caused comparable 10-fold reductions in uptake. Uptake by HL-60
cells, which expressed low c-FMS, and by Sort 10 cells, which
expressed high c-FMS, was comparable for each concentration of
exogenous retinoic acid tested.
LD3
LA
..,,.,,.
j
HL.60
FMS/1
SORT1O
Fig. 6. VD3-induced hypophosphorylation of RB for HL-6O sublines after 72 h for (left
to right) HL-.60, FMS/I, and Sort 10 cells treated with 0.5 X 10_6 M VD3.
10-8M
VD3
1O-7M
VD3
10-6M
100
80
Fig. 7. Percentage of cells in G@/G@for HL-60
sublines expressing progressively more c-FMS.
HL-60 cells (0), FMS/I (A), and Sort 10 (U) were
treated with 0.5 x 10_8 M (left), 0.5 x 10@ M
80
0
¶@60
0
C,
C,
@60
@40
@4o
20
20
(middle) and 0.5 x l0_6 M (right) VD3.
0
0
0
24
48
72
96
0
24
48
Hours
Hours
72
96
0
24
48
72
Hours
2025
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1997 American Association for Cancer Research.
96
@
@
51
0
0
c-FMS REGULATES RB. CELL CYCLE. AND DIFFERENTIATION
DISCUSSION
@
@
@
12000@
The results of the present study indicate that increasing c-FMS
expression increases the sensitivity of HL-60 promyelocytic leukemia
cells to the myeloid differentiation inducer, retinoic acid. The cellular
responses analyzed were cell differentiation, cell cycle arrest, and RB
hypophosphorylation,
a putative regulator of such processes. Cells
with progressively higher c-FMS expression could respond to pro
gressively lower concentrations of retinoic acid. Cell differentiation
could be elicited by the lowest concentration tested, without accom
panying RB hypophosphorylation or G1/G0-specific growth arrest, if
c-FMS expression was high enough. A higher concentration of reti
noic acid elicited cell differentiation, even for cells expressing less
c-FMS, and also RB hypophosphorylation, without G1/G0-specific
growth arrest, if c-FMS expression was sufficiently high. The highest
concentration tested corresponded to that routinely used to elicit
HL-60 myeloid differentiation and caused cell differentiation, RB
hypophosphorylation, and G1/G0-specific growth arrest, even for cells
expressing low c-FMS levels. Similar although not as pronounced
results were observed with the monocytic differentiation inducer,
1,25-dihydroxy vitamin D3, indicating that c-FMS had similar regu
latory effects on both of the differentiation lineages available to
HL-60 promyelocytic leukemia cells.
There was an apparent threshold in cumulative retinoic acid plus
c-FMS-originated signal needed to elicit cellular response. The thresh
old to elicit cell differentiation, RB hypophosphorylation, and cell
cycle arrest progressively increased. Within the range of ligand (ret
inoic acid or I ,25-dihydroxy vitamin D3) concentrations or c-FMS
expression levels tested, cell differentiation could be elicited without
occurrence
of detectable
RB hypophosphorylation
or cell cycle arrest.
RB hypophosphorylation could be elicited without cell cycle arrest,
but not without differentiation. Cell cycle arrest could only be elicited
with cell differentiation and RB hypophosphorylation. This progres
sion of effects from cell differentiation to RB hypophosphorylation to
cell cycle specific arrest paralleled increasing concentrations of reti
noic acid or VD3. In the case ofcell differentiation and RB hypophos
phorylation,
it was apparent
that within
the range of ligand
concen
tration or c-FMS expression tested, for any given concentration of
retinoic acid or VD3, increasing c-FMS expression paralleled increas
ing extent of response. Thus, it appears as if the retinoic acid
originated
and c-FMS-originated
cell differentiation
signal thresholds
signals
integrate,
or RB hypophosphorylation.
for causing
cell differentiation,
lation, and cell cycle-specific
This rules out simple
models
summing
to elicit
Thus, the apparent
by retinoic
Table 2 Uptake of retinoic acid lxvilL-dO and c-FMS-transfected HL-60
(cpm)SubLine
Sort10
Control
8
9
6000@
4000@
2000@
U
10'M
159 ±5
122 ±23
107M
1,320 ±62
1,196±213
co@
0
(9
(0.
..J
-:@
@.
to
0
0
@O
0
,@
0
(0
@.
0
(9
0
-
-:@
Fig. 8. Uptake of retinoic acid by HL-6O and c-FMS-transfected
HL-60 cells. HL-60
or Sort 10 HL-60 cells transfected with c-FMS were labeled for I h in trace (—0.4%)
labeled (3H) retinoic acid and harvested. Cellular radioactivity measuring uptake of
retinoic acid is shown for (left to right) HL-60 cells treated with 0 M (control), l0'
M,
l0@ M, and lO_6 M trace-labeled retinoic acid, and Sort 10 cells treated with 0 M
(control), 1O' M, l0@@M,and lO'
Mtrace-labeled retinoic acid. Columns, means of
triplicate repeats; bars. SD. c-FMS overexpression did not affect cellular retinoic acid
uptake. which was approximately proportional to the exogenous concentration.
Recent literature suggests a variety of potential mechanisms by which
the retinoic acid/vitamin D3 and c-FMS signal transduction pathways
may intersect. Such an intersection may occur at the retinoic acid/
vitamin D steroid receptor, at intermediate signal transduction mole
cules downstream of c-FMS, or at promoters targeted by the two
signal transduction pathways.
Phosphorylation of the steroid receptor may be one intersection of
steroid hormone signaling with a kinase cascade. The RARa is
phosphorylated, although in embryonal carcinoma cells, the phospho
rylation
is not affected
by retinoic
acid (35). Likewise,
in HL-60 cells,
RARa is also phosphorylated, but the relative amount of phosphoryl
ated RARa appears unchanged by retinoic acid, based on detecting a
phosphorylation induced gel electrophoretic mobility shift of RARa
by Western analysis.4 RAR(3- and RAR'y are also phosphorylated in
acid, whereby
Uptake of trace (0.4%)-Iabeled retinoic acid at the indicated concentrations. c-FMS
overexpression did not affect cellular retinoic acid uptake, which was approximately
proportional to the exogenous concentration. repeats.Uptake
SDs are for triplicate
106MHL-6O
8000
RB hypophosphory
the ligand-activated transcription factor induces expression of a cen
tral regulatory molecule causing pleiotropic effects, specifically dif
ferentiation, RB hypophosphorylation, and G1/G0-specific arrest.
The present results indicate that there is an intersection between the
retinoic acid- and vitamin D3-originated signals and the c-FMS
originated signals. The identity of this mechanism is not known.
Because both retinoic acid and vitamin D3 are similarly involved and
both putatively induce common essential early events (31, 34), the
mechanism likely involves early events common to both inducers.
I
C-)
growth arrest increase, in that order.
of regulation
I
10000@
embryonal
carcinoma
cells (36, 37). Using ectopic
expression,
RARa
can be phosphorylated by protein kinase A, causing enhanced tran
scriptional activation of a reporter by ligand-activated RARa (38).
Likewise, the vitamin D receptor can also be phosphorylated by PKC,
a kinase implicated in signal transduction by typical transmembrane
tyrosine kinase receptors for mitogenic growth factors (39). The
vitamin D receptor can also be phosphorylated by casein kinase (40),
a kinase known to target nuclear proto-oncogenes myc, myb, and max
and RNA polymerases I and II, thereby also enhancing transcriptional
activation. In general, the phosphorylation of steroid hormone recep
tors has been implicated
in their ligand
activation,
nuclear
transport,
and DNA binding (41). Thus, signaling molecules downstream of
c-FMS may affect such processes for the retinoic acid or vitamin D
receptors.
The intersection of retinoic acid/vitamin D3 and c-FMS signaling
10,092 ±1,102
8,914±1,640
4 5.
C.
Brooks
III
and
A.
Yen,
unpublished
work.
2026
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1997 American Association for Cancer Research.
c-FMS REGULATES RB. CELL CYCLE. AND DIFFERENTIATION
may also involve retinoic acid or vitamin D3 control of signaling
molecules typically downstream of transmembrane tyrosine kinase
receptors, such as c-FMS. Ligand-activated c-FMS binds the adapter
molecules Grb2 and 5051, associated with RAS activation (42),
implicating the signaling molecules typically associated with RAS/
RAF mediated signaling. And RAF has demonstrated regulatory ef
fects on retinoic acid- or vitamin D3-induced HL-60 differentiation
(13). Retinoic acid can specifically increase PKCa levels in murine
melanoma cells (43). In U937 monoblastoid leukemia cells, low doses
and may be a regulator thereof. Several recent findings suggest this
possibility. RB can be found concentrated in nucleoli (56) and can
interact with the UBF transcription factor (56) and inhibit RNA
polymerases (56, 57, 58). Thus, RB may function as a general regu
lator of transcriptional activity. In the particular cases of E2F, Spl,
AP-1, or p53 dependent transcriptional activation, a RB-GAL4 fusion
protein can repress transcriptional activation by this range of tran
scription factors (58). Thus, RB may have a more global regulatory
function than just G1/G0 control by regulating cellular anabolic levels.
of <1 flMretinoic acid enhanced protein phosphorylationand cell
differentiation by l2-O-tetradecanoylphorbol-l3-acetate,
an activator
of PKC (44). Furthermore, in HL-60 cells, retinoic acid increased
expression of the src-like kinase fgr, whereas vitamin D3 increased
the expression of the src-like kinases fyn, fgr, and lyn (45). Thus,
retinoic acid can have a direct effect on signaling molecules. Like
wise, vitamin D3 can also affect downstream effectors of transmem
brane tyrosine kinases. Vitamin D3 can activate RAF within 1 mm of
stimulation in a l2-O-tetradecanoylphorbol-13-acetate
or staurospo
rifle-sensitive process suggesting a PKC dependence (46). Further
downstream, vitamin D3 can, for example, enhance DNA binding
activity of AP-l (47).
Finally, the intersection of the retinoic acid/vitamin D3- and c
FMS-originated signals may occur at the promoter(s)-targeted by
these signals. The Mr 265,000 CREB-binding protein required for
AP-1 activation is part of an integrator complex that also mediates
transcriptional activation by RAR (48). This provides a potential
mechanism for the observed ability of retinoic acid to act as a negative
regulator of AP-l (49). In this case, the RAR may deprive AP-l sites
of the needed CBP. One can speculate, however, that if the RAR binds
in proximity of an AP-l, then a factor such as CBP, which is attracted
to AP-1, for example, might also positively regulate transcriptional
activation by RAR. Thus, the AP-1 activation downstream of some
prototypical transmembrane tyrosine kinase receptor might affect
retinoic acid signaling, and potentially vice versa. Such a process
might, for example, be the basis of the integration of the retinoic
acid/vitamin D- and c-FMS-originated signals observed here.
The present results motivate a re-evaluation of current ideas on the
role of RB in regulating the cell cycle and, putatively, differentiation.
REFERENCES
1. Yen, A., Forbes, M. E., Varvayanis, S., Tykocinski, M. L., Groger, R. K., and Platko,
J. D. c-FMS-dependent HL-6O cell differentation and regulation of RB gene expres
sion. J. Cell. Physiol., 157: 379—391, 1993.
2. Petkovich, M., Brand, N. J., Krust, A., and Chambon, P. A human retinoic acid
receptor which belongs to the family of nuclear receptors. Nature (Lond.), 330:
444—450,1987.
3. Brand, N., Petkovich, M., Krust, A., Chambon. P.. de The, H., Marchio, A., Tiollais,
P., and Dejean, A. Identification of a second human retinoic acid receptor. Nature
(Lond.), 332: 850—853, 1988.
4. Krust, A., Kastner, P., Petkovich, M., Zelent, A., and Chambon, P. A third human
retinoic acid receptor. hRAR-gamma. Proc. NatI. Acad. Sci. USA, 86: 5310—5314,
1989.
5.
@
D.
J.,
Ong,
E.
S.,
Dyck.
J.
A.,
and
Evans,
R.
M.
Nuclear
receptor
that
identifiesa novel retinoic acid response pathway. Nature (Lond.), 345: 224—229,1990.
6. Mangelsdorf, D. J., Borgmeyer, U., Heyman. R. A.. Thou, J. Y., Ong, E. S., Oro,
A. E., Kakizuka, A., and Evans, R. M. Characterization of three RXR genes that
mediate the action of 9-cis-retinoic acid. Genes Dcv., 6: 329—344,1992.
7. Mangelsdorf, D. J., Kazuhiko, U., and Evans, R. M. The retinoid receptors. In: M. B.
Spom, A. Roberts, and D. S. Goodman (eds.), The Retinoids, 2nd Ed., pp. 319—349.
New York: Raven Press, 1994.
8. Brooks, S. C., III, Kazmer, S., Levin, A. A., and Yen, A. Myeloid differentiation and
retinoblastoma phosphorylation changes in HL-60 cells induced by retinoic acid
receptor and retinoid X receptor-selective retinoic acid analogs. Blood, 87: 227—237,
1996.
9. Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, 3. D. Renewal by
pluripotent stem cells: blood cell formation. In: Molecular Biology of The Cell, 2nd
Ed., pp. 973—983.New York: Garland Publishing Inc., 1989.
10. Quesenberry, P. J. Hemopoietic stem cells. progenitor cells, and growth factors. In:
w. j. Williams,E.Beutler,A.1.Erslev,M.A.Lichtman,(eds.),Hematology,4thEd.,
p. 129. New York: McGraw-Hill, Inc., 1990.
I 1. Sherr, C. J. The colony-stimulating factor 1 receptors: pleiotropy of signal-response
coupling. Lymphokine Res., 9: 543—548,1990.
12. Choudhury. G. G., Sylvia. V. L., Pfeifer, A., Wang. L-M., Smith, E. A., and
Sakaguchi. A. Y. Human colony stimulating factor-I receptor activates the c-RAF- 1
proto-oncogene kinase. Biochem. Biophys. Res. Commun. 172: 154—159,1990.
13. Yen, A., Williams, M., Platko, J. D., Der, C., and Hisaka, M. Expression of activated
RAF accelerates cell differentiation and RB protein down regulation but not hy
pophosphorylation. Eur. J. Cell Biol., 65: 103—1
13, 1994.
14. Zacksenhaus, E., Bremner, R., Jiang, Z., Gill, R. M., Muncaster. M., Sopta, M.,
Phillips, R. A., and Gallie, B. L. Unraveling the function of the retinoblastoma gene.
RB is thoughtto regulatethe cell cycle and, putatively,cell differen
tiation through its phosphoiylation. In this present paradigm, the
phosphorylation of RB, possibly at the G@restriction point (50, 51) by
cyclin D-, E-, or A-dependent kinases (52, 53), is believed to cause
progression to S phase, due in large part to the release of transcription
factors binding only the unphosphorylated RB (see Refs. 54 and 55 for
review). Along with other recent data not entirely consistent with the
anticipations of this paradigm (1, 13, 17—25),two pieces of the present
data specifically militate against this paradigm. One is that, although
HL-60 cells have a substantial G@phase, with approximately 45% of
exponentially proliferating cells in G@,there is no evident hypophos
phorylated RB; i.e., all of the RB is hyperphosphorylated. The other
is that it is possible to cause appearance of significant, —15% hy
pophosphorylated RB without precipitating growth arrest. Thus, RB
phosphorylation or dephosphorylation per se does not immediately
regulate
to S cell cycle progression or G@arrest. Furthermore,
because differentiation can occur without observable hypophospho
rylated RB, phosphosylation shift also does not appear to be causal for
cell differentiation. Interestingly, in proliferating HL-60 cells, the
amount and degree of phosphorylation of RB progressively increase
as cells advance through the cell cycle (19, 21). Furthermore, when
HL-60 cells differentiate, the amount of RB protein per cell decreases
as it shifts to the unphosphorylated state, as cells G@ arrest and
differentiate (1, 13, 19). In these instances, the amount of RB and its
phosphorylation appear to generally reflect cellular anabolic activity,
Mangelsdorf,
Adv.CancerRes.,61:
115—141,1993.
15. Levine, A. J. The tumor suppressorgenes. Annu. Rev. Biochem., 62: 623—651, 1993.
16. Wanberg, R. A. Tumor suppressor genes. Science (Washington DC), 254: 1138—
1146, 1991.
17. Zukerberg, L. R., Benedict, W. F., Arnold, A., Dyson, N., Harlow, E., and Harris,
N. L. Expression of the retinoblastoma protein in low-grade B-cell lymphoma:
relationship to cyclin Dl . Blood, 88: 268—276.1996.
18. Pier Paolo, C., Dc Luca, A., Howard, C. M., Baldi, A., Firpo, E. J., Koff, A., Paggi,
M. G., and Giordano, A. Functional analysis of pRb2/p130 interactions with cyclins.
Cancer Rca., 56: 2003—2008,1996.
19. Yen, A., and Varvayanis, S. Late dephosphorylation of the RB protein in G2 during
the process of induced cell differentiation. Exp. Cell Res., 2/4: 250—257, 1994.
20. Ikeda, M-A., Jakoi, L., and Nevins, J. R. A unique role for the RB protein in
controlling E2F accumulation during cell growth and differentiation. Proc. NatI.
Acad. Sd. USA, 93: 3215—3220,1996.
21. Yen, A., Chandler, S., Forbes, M., Fung, Y-K., T'Ang, A., and Pearson, R. Coupled
down-regulation of the RB retinoblastoma and c-myc genes antecedes cell differen
tiation: possible role of RB as a ‘statusquo―gene. Eur. J. Cell Biol., 57: 2 10—221,
1992.
22. Yen, A., and Chandler, S. Inducers of leukemic cell differentiation
cause down
regulation of RB gene expression. Proc. Soc. Exp. Biol. Med., 199: 291—297,1992.
23. Conddorelli, G. L, Testa, U., VaItieri, M., Vitelli, L, De Luca, A., Barbed, T., Monte
sam, E, Campisi, S., Giordano, A., and Peschle, C. Modulation of retinoblastoma gene
in normal adult hematopoiesis: peak expression and functional role in advanced erythroid
differentiation. Proc. Nail. Acad. Sci. USA, 92: 4808—4812, 1995.
24. Yen, A., and Soong, S. Retinoic acid induced RB (retinoblastoma) hypophosphory
lation enhanced by CGP 5241 1 (4,5-dianilinophthalimide),
an EGF family tyrosine
kinase receptor inhibitor. Eur. J. Cell Biol, 69: 327—334,1996.
25. Weinberg, R. A. E2F and cell proliferation: a world turned upside down. Cell, 85:
457—459,1996.
2027
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1997 American Association for Cancer Research.
c-FM5 REGULATES
RB. CELL CYCLE. AND DIFFERENTIATION
26. Collins, S. J., Gallo, R. C., and Gallagher. R. E. Continuous growth and differenti
ation of human myeloid leukemia cells in suspension culture. Nature (Lond.), 270:
347—349.1977.
27. Dalton, W. T., Ahearn, M. J., McCredie, K. B., Freireich, E. J., Stass, S. A., and
Turjillo, J. M. HL-60 cell line was derived from a patient with FAB-M2 and not
FAB-M3. Blood. 71: 242—247,1988.
28. Yen, A. HL-60 cells as a model of growth control and differentiation: the significance
of variant cells. Hematol. Rev., 4: 5—46,1990.
29. Breitman, T. R., Selonick, S. E.. and Collins, S. J. Induction of differentiation of the
human promyelocytic leukemia cell lines (HL-60) by retinoic acid. Proc. NatI. Acad.
Sci.USA, 77:2936—2940,
1980.
30. Miyaura. C., Abe, E., Kuribayashi, T., Tanaka, H., Konno, K., Nishii, Y., and Suda,
1. 1-a, 25-Dihydroxyvitamin D3 induces differentiation of human myeloid leukemia
cells. Biochem. Biophys. Res. Commun., 102: 937—943,1981.
31 . Yen, A., Forbes, M., deGala, G., and Fishbaugh, J. Control of HL-6O cell differen
tiation lineage specificity: a late event occurring after precommitment. Cancer Res.,
47: 129—134,
1987.
32. Yen, A., Sturgill, R., Varvayanis. S.. and Chern, R. FMS (CSF-I receptor) prolongs
cell cycle, promotes retinoic acid induced hypophosphorylation of RB protein, G1
arrest and cell differentiation. Exp. Cell Res., 229: 111—125,
1996.
33. Yen, A., and Varvayanis, S. RB phosphorylation in sodium butyrate resistant HL-6O
cells: cross-resistance to retinoic acid but not vitamin D3. J. Cell. Physiol., 163:
502—509,1995.
34. Yen, A., and Forbes, M. E. c-myc down-regulation and precommitment in HL-60
cells due to bromodeoxyuridine. Cancer Rca., 50: 141 1—1420,1990.
35. Gaub, M. P., Rochette-Egly, C., Lutz, Y., Ali, S., Matthes, H., Scheuer, I., and
Chambon, P. Immunodetection of multiple species of retinoic acid receptor: evidence
for phosphorylation. Exp. Cell Res., 201: 335—346,1992.
36. Rochette-Egly, C., Guab, M-P., Lutz, Y., Ali, S., Scheuer, I., and Chambon, P.
Retinoic acid receptor-@3immunodetection and phosphorylation on tyrosine residues.
Mol. Endocrinol., 12: 2197—2209,1992.
37. Rochette-Egly, C., Lutz, Y., Saunders, M., Scheuer, I., Gaub, M. P., and Chambon,
P. Retinoic acid receptor -y specific immunodetection and phosphorylation. J. Cell
Biol., 115: 535—546.1991.
38. Rochette-Egly, C., Oulad-Abdelghani,
M., Staub, A., Pfister, V., Scheuer, I.,
Chambon,P., and Gaub,M-P. Phosphorylation
of the retinoicacid receptor-aby
protein kinase. A. Mol. Endocrinol., 9: 860—871, 1995.
39. Hsieh, J-C., Jurutka, P. W., Nakajima, S.. Galligan, M. A., Haussler, C. A., Shimizu, Y.,
Shimizu, N., Kerr Whitfield, G., and Haussler M. R. Phosphorylation of the human
43. Gruber, J. R., Desai, S., Blusztajn, I. K., and Niles, R. M. Retinoic acid specifically
increases nuclear PKCa and stimulates AP-1 transcriptional activity in B16 mouse
melanoma cells. Exp. Cell Res., 221: 377—384,1995.
44. Ways, D. K., Dodd, R. C., Bennett. T. E., Hooker, J. L., and Earp, H. S. Effect of
retinoic acid on phorbol ester-stimulated differentiation and protein kinase C-depend
ent phosphorylation in the U937 human monoblastoid cell. Cancer Res., 48: 5779—
5787,1988.
45. Katagiri, K., Katagiri, T., Koyama, Y., Morikawa, M., Yamamoto, T., and Yoshida,
T. Expression of src family genes during monocytic differentiation of HL-60 cells.
I. Immunol., 146: 701—707,1991.
46. Lissoos, T. W., Beno, D. W. A., and Davis, B. H. 1,25-Dihydroxyvitamin
D3
activates raf kinase and raf perinuclear translocation via a protein kinase C-dependent
pathway. I. Biol. Chem., 33: 25132—25138, 1993.
47. Lasky, S. R., Iwata, K., Rosmarin, A. G., Caprio, D. G., and Maizel, A. L. Differential
regulation of JunD by dihydroxycholecalciferol in human chronic myelogenous
leukemia cells. J. Biol. Chem., 270: 19676—19679,1995.
48. Kamei, Y., Xu, L., Heinzel, T., Torchia, I., Kurokawa, R., Gloss, B., Lin, S-C.,
Heyman, R. A., Rose, D. W., Glass, C. K., and Rosenfeld, M. G. A CBP integrator
complex
mediates
transcriptional
activation
and AP-l
inhibition
by nuclear receptors.
Cell, 85: 403—414,1996.
49. SchUle, R., Rangarajan. P., Yang, N., Kliewer, S., Ransone, L. I., Bolado, J., Verma,
I. M., and Evans, R. M. Retinoic acid is a negative regulator of AP-l-responsive
genes. Proc. Natl. Acad. Sci. USA, 88: 6092—6096, 1991.
50. Yen, A., and Pardee, A. B. Exponential 3T3 cells escape in mid-G, from their high
serum requirement. Exp. Cell Res., 116: 103—1
13, 1978.
51. Dou, Q. P., Levin, A. H., Thao, S., and Pardee, A. B. Cyclin E and cyclin A as
candidates for the restriction point protein. Cancer Rca., 53: 1493—1497,1993.
52. Dowdy, S. F., Hinds, P. W., Louie, K., Reed, S. I., Arnold, A., and Weinberg, R. A.
Physical interaction of the retinoblastoma protein with human D cyclins. Cell, 73:
499—511, 1993.
53. Hinds, P. W., Mittnacht, S., Dulic, V., Arnold, A., Reed, S. I., and Weinberg, R. A.
Regulation of retinoblastoma protein functions by ectopic expression of human
cyclins. Cell, 70: 993—1006, 1992.
54. Shirodkar, S., Ewen, M., DeCaprio, I. A., Morgan, J., Livingston, D. M., and
Chittenden, T. The transcription factor E2F interacts with the retinoblastoma product
and a p107-cyclin A complex in a cell cycle-regulated manner. Cell, 68: 157—166,
1992.
55. Nevins, i. R. E2F: a link between
the RB tumor suppressor protein and viral
oncoproteins. Science (Washington DC), 258: 424—429,1992.
vitamin D receptor by protein kinase. C. J. Biol. Chem., 268: 15118—15126,
1993.
40. Jurutka, P. W., Hsieh, i-C., Nakajima. S., Haussler, C. A., Whitfield, G. K., and
Haussler, M. R. Human vitamin D receptor phosphorylation by casein kinase II at
Ser-208 potentiates transcriptional activation. Proc. Nail. Acad. Sci. USA, 93: 3519—
3524,1996.
41 . Kuiper, G. G. J. M., and Brinkman, A. 0. Steroid hormone receptor phosphorylation:
is there a physiological role? Mol. Cell. Endocrinol, It%t: 103—107,1994.
42. Lioubin, M. N., Myles, G. M., Carlberg, K., Bowtell, D., and Rohrschneider, L. R.
Shc, Grb2, SosI, and a 150-kilodalton tyrosine-phosphorylated protein form com
plexes with fms in hematopoietic cells. Mol. Cell. Biol., 14: 5682—5691, 1994.
56. Cavanaugh, A. H., Hempel, W. M., Taylor, L. J., Rogaisky, V., Todorov, G., and
Rothblum, L. I. Activity of RNA polymerase I transcription factor UBF blocked by
Rb gene product. Nature (Lond.), 374: 177—180,1995.
57. White, R. i., Trouche, D., Martin, K., iackson, S. P., and Kouzarides, T. Repression
of RNA polymerase III transcription by the retinoblastoma
protein. Nature (Lond.),
382: 88—90,1995.
58. Adnane, I., Shao, Z., and Robbins, P. D. The retinoblastoma susceptibility gene
product represses transcription when directly bound to the promoter. I. Biol. Chem.,
270: 8837—8843, 1995.
2028
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1997 American Association for Cancer Research.
Increasing c-FMS (CSF-1 Receptor) Expression Decreases
Retinoic Acid Concentration Needed to Cause Cell
Differentiation and Retinoblastoma Protein
Hypophosphorylation
Andrew Yen, Rhonda Sturgill and Susi Varvayanis
Cancer Res 1997;57:2020-2028.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/57/10/2020
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1997 American Association for Cancer Research.