[CANCER RESEARCH 47, 129-134, January 1, 1987)
Control of HL-60 Cell Differentiation
after Precommitment1
Lineage Specificity, a Late Event Occurring
Andrew Yen2, Mary Forbes, Gwen DeGala, and Justin Fishbaugh
Departments of internal Medicine [A. W., M. F., G. D.], Physiology, and Biophysics [A. Y.J and The Flow Cytometry Facility [J. F.], University of Iowa,
Iowa City, Iowa 52242
ABSTRACT
Terminal cell differentiation of HL-60 promyelocytic leukemia cells
results when they are continuously exposed to retinoic acid. This process
involves an intermediate regulatory state, the precommitment memory
state, which occurs before onset of differentiation or growth arrest in GO.
The cellular processes occurring prior to onset of terminal differentiation
can be resolved into early events anteceding development of the precommitment memory state and late events subsequent to it. While it has been
suggested that retinoic acid induced early events regulate (.•,„
specific
growth arrest associated with terminal differentiation, the significance of
induced late events is not known. Exploiting the capability of HL-60
cells to undergo either myeloid or monocytic differentiation in response
to different inducers, the present studies examine the response of HL-60
cells to the sequential application of myeloid and monocytic inducers
prior to onset of terminal differentiation. The results indicate that the
precommitment state induced by retinoic acid is not differentiation lineage
specific. Sequential application first of retinoic acid, a myeloid inducer,
and then of 1,25-dihydroxyvitamin I),, a monocytic inducer, and vice
versa, show that cellular choice of a specific differentiation lineage is
regulated by late inducer driven events. The data support a two-step
model for induction of terminal differentiation where early events anteced
ing precommitment regulate growth arrest and late events subsequent to
precommitment regulate choice of a specific differentiation lineage. The
results are of potential significance to the use of differentiation-inducing
agents in chemotherapy. The potential toxicity of prolonged exposure to
a single inducer might thus be mitigated by sequential brief exposures to
different inducers.
INTRODUCTION
Studies of the regulation of terminal cell differentiation have
been greatly facilitated by the availability of cell lines which
undergo this process in vitro. One such system is the HL-60
human promyelocytic leukemia cell line (1). HL-60 is a differentiatively bipotent cell (2) capable of selectively undergoing
either myeloid or monocytic differentiation in response to dif
ferent inducers. Exponentially proliferating cells exposed to
such inducers eventually undergo growth arrest and display the
phenotype of mature myeloid or monocytic cells. RA,3 for
instance, induces myeloid differentiation (3-6) while D3 induces
monocytic differentiation (7-11). It is not known how the
induced cellular programs leading to GI/Ospecific growth arrest
and expression of the differentiated phenotype are effected.
In the case of induced terminal myeloid differentiation due
to retinoic acid, the seminal cellular signal initiating this pro
gram apparently originates at the cell membrane (12) and is Sphase specific (13, 14). Cellular execution of this program
requires that the inducer be present for two division cycles
(approximately 48 h for these cells) before any onset of d/o
growth arrest and phenotypic differentiation occur (13, 15).
During this process, cells pass through an identifiable preconiReceived 6/9/86; revised 9/9/86; accepted 9/11/86.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1Supported in part by grants from the USPHS (NIH).
2To whom requests for reprints should be addressed.
3 The abbreviations used are: RA, retinoic acid; D3, 1,25-dihydroxyvitamin D3;
PBS, phosphate buffered saline.
mitment memory state (15). The precommitment state results
from an abbreviated exposure to inducer which by itself is
insufficient to induce G0 growth arrest or phenotypic differen
tiation. Precommitment is a functionally and structurally iden
tifiable early regulatory state. Cells enter this state when ex
posed to inducer for a period corresponding to one division
cycle (approximately 24 h for these cells). Precommitment cells
have characteristic alterations in their cytosolic complement of
Ca+2 binding proteins, altered nuclear membrane structure, and
elevated expression of the c-myc protooncogene (13, 15, 16).
Upon removal from inducer, precommitment cells continue to
proliferate and do not undergo myeloid differentiation; during
reexposure to inducer, precommitment cells demonstrate a
transient cellular memory of their previous inducer exposure
and undergo Gi/0 growth arrest and phenotypic differentiation
after only an abbreviated subsequent exposure period corre
sponding to one division cycle. The amount of abbreviation
thus equals the duration of the previous exposure to inducer.
This cellular memory of the initial inducer exposure is heritable
for several generations and then lost. After a period correspond
ing to four division cycles in inducer-free medium, cells no
longer exhibit precommitment memory. Altered nuclear struc
ture and elevated c-myc expression persist as long as precom
mitment memory persists.
Thus when exponentially proliferating HL-60 cells are ex
posed to inducer, they undergo transition to an identifiable
precommitment state. This resolves the induced cellular pro
gram leading to onset of G0 arrest and phenotypic differentia
tion into two sequential processes. These are early events an
teceding precommitment and late events occurring after induc
tion of precommitment but before onset of terminal differentia
tion. Significantly both early and late events occur while the
cells are still proliferating. This resolution of induced cellular
processes into a grouping of early and late events is also
indicated by studies on the capability of isomers of retinoic acid
to induce terminal cell differentiation (17). These data indicate
that the retinoic acid inducer signal could be resolved into two
components distinguishable by their capability to elicit early
and late events. In this case, certain cis-trans isomers of the
alkyl chain were capable of inducing early events leading to
precommitment, but not late events. They could also subse
quently induce Go growth arrest but not differentiation. Regu
lation of cell proliferation thus appears to be primarily associ
ated with early events. While early cellular events that antecede
precommitment apparently regulate cell proliferation (17, 18),
the role of late cellular events subsequent to precommitment is
not known. The significance of this period when cells are still
proliferating and are not yet committed to terminal differentia
tion is thus of interest. In this context the question also arises
as to whether the precommitment state is differentiation lineage
specific. This present communication presents data to indicate
that these late events regulate the specificity of the differentia
tion pathway, either myeloid or monocytic, which is elicited.
Furthermore the precommitment state is apparently not lineage
specific.
129
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1987 American Association for Cancer Research.
CONTROL OF DIFFERENTIATION
MATERIALS AND METHODS
Cells and Culture Conditions. 111 6(1human leukemic promyelocytes
from the laboratory of Dr. R. Gallo were generously provided by Drs.
C. P. Burns and R. Gingrich. The cells were maintained as described
previously (6, 12, 13, 15-17) in constant exponential growth in 25-cm2
tissue culture flasks (Corning Plastics) using RPMI 1640 (Flow Labo
ratories) supplemented with 10% heat-inactivated fetal calf serum (Kan
sas City Biologicals) and 2 HIMglutamine. No antibiotics were used in
stock cultures. The cells were Mycoplasma free (Mycotrim assay; Hana
Biologies). Cells were passaged at an initial density of 0.2 x 10'Yml or
0.1 x 106/ml and passaged again by resuspension in completely fresh
medium when they reached a density of approximately 2.0 x 106/ml.
All experiments were performed with the same subline. These cells
have been maintained in continuous passage from the original cells.
Their continued responsiveness to retinoic acid, dimethyl sulfoxide,
1,25-dihydroxyvitamin I),, sodium butyrate, and tetradecanoylphorbol
acetate were periodically verified. This is in contrast to various variant
sublines which we have derived with altered response to certain of these
inducers.
In the six cases studied, exponentially proliferating cells were used
to make a cell suspension of 0.25 x IO6cells/ml to which the indicated
inducer, either RA or Dj, was added. Cultures were as follows: in Case
1, cells continuously exposed to RA, 30 ml of the cell suspension
containing IO"6 M RA were aliquoted into a 75-cm2 tissue culture flask
and incubated at 37°Cin a humidified atmosphere of 5% CO2. In Case
2, cells continuously exposed to D3, 50 ml of the cell suspension
containing IO'1 M D3 were aliquoted into 100 nun bacteriological Petri
dishes at 5 ml/replicate dish. The cells were then incubated. Bacterio
logical Petri dishes were used to prevent cell adhesion and facilitate
subsequent harvesting with a rubber policeman. In Case 3, cells exposed
to RA for 24 h before being transferred to inducer-free medium, cells
were initiated in culture in a 75-cm2 flask as above for RA-treated cells.
At 24 h the cells were washed twice by resuspension in the same volume
of fresh serum-supplemented medium (approximately 0.5 x IO6 cells/
ml) before being resuspended in the same volume of the final fresh
culture medium and aliquoted into a tissue culture flask and incubated.
In Case 4, cells exposed first to RA and then to D3, washed cells were
resuspended in the same volume of fresh medium containing I)t and
then aliquoted into replicate bateriological Petri dishes as above. In
Case 5, cells exposed to D3 for 24 h before being transferred to inducerfree medium, a cell suspension containing D3 was aliquoted into 75cm2 tissue culture flasks, at SO ml/flask, and the flask was inverted and
incubated. The flask was inverted since the roof of the flask was not
treated to allow cell adhesion. At 24 h the cells were washed as described
above and, after resuspension in the same volume of final fresh medium,
aliquoted into replicate 100-mm bacteriological Petri dishes. In Case
6, cells exposed first to D3 and then to RA, washed cells were resus
pended in the same volume of fresh final medium containing RA and
aliquoted into replicate bacteriological Petri dishes as above.
Cell counts before and after the washing procedures verified the lack
of any significant cell loss. Retinoic acid (Sigma Chemical Co., St.
Louis, MO) was prepared as a 10 ' M stock in ethanol and used at a
final concentration of IO"6 M. 1,25-dihydroxyvitamin D3, a generous
gift of Dr. Milan Uskokovic (Hoffman-LaRoche, Inc.), was prepared
as a 10~3M stock in propylene glycol and used at a final concentration
of IO"6 M in culture. The efficacy of the washing procedure has been
AFTER PRECOMMITMENT
was assayed by phorbol myristate acetate-induced reduction of nitroblue
tetrazolium to formazan by Superoxide as described previously (6, 12,
13, 15-17). At the indicated times, 0.2 ml of harvested cell suspension
was used to assay capability for oxidative metabolism. Cellular mor
phology was evaluated with stained slides. At the indicated times, 0.15
ml of harvested cell suspension was used to make a cytocentrifuge slide
preparation (Shandon Scientific, Ltd.). The slides were dried with a
hair dryer blowing cold air and stained with tetrachrome. Cell surface
differentiation markers Mol (19-21), Mo2 (19, 20), and My4 (21-24),
expressed by differentiated cells, were detected by immunofluorescence
using commercially available kits (Coulter Immunology Co.). Mol and
Mo2 are fluoresceinated murine monoclonal antibodies. M>4 is a
murine monoclonal antibody used with a fluoresceinated secondary
staining reagent, a goat anti-mouse immunoglobulin fluoresceinated
antibody. At the indicated times, IO6 cells were used for each assay to
prepare a wet mount for fluorescence microscopy. Briefly, IO6 cells
were centrifuged to a pellet (2000 rpm, 5 min, 4°C),the supernatant
was removed, 97.5 ¿<1
of PBS wash (containing 5% fetal calf serum and
0.2% sodium azide) plus 2.5 M' of reconstituted antibody, per the
manufacturer's instructions, was added, and the cell suspension was
agitated. The cell suspension was stored for 20-30 min on ice, protected
from light, and gently resuspended every 10 min. One ml of PBS wash
was then added and the cells were washed twice in PBS wash by
centrifugation as before and finally resuspended in 1-ml PBS wash. If
secondary staining was used, as in the case of My4, the cells were
resuspended instead in 100 n\ PBS wash containing the secondary
reagent, fluoresceinated goat antibody against mouse immunoglobulin,
per the manufacturer's instructions. The cell suspension was stored 2030 min on ice and protected from light with gentle resuspension every
10 min. One ml of PBS wash was then added and the cells were washed
twice and resuspended as above. In control experiments, induced and
uninduced HL-60 cells were assayed as above using nonspecific flu
oresceinated murine antibody (Coulter Immunology Co.) of the same
isotype as the specific reagent or nonfluoresceinated primary antibody,
in the case of staining with a secondary reagent, in lieu of the specific
reagent. These were negative in all cases. Thus, the observed reaction
of the specific antibodies used was not due to nospecific binding or
binding to an Fc receptor.
RESULTS
Distinguishing Induced Myeloid and Monocytic Differentia
tion. The capabilities of RA (3-6) to induce HL-60 myeloid
differentiation and D3 (7-11) to induce monocytic differentia
tion are well established. Progression of a cell population along
these two differentiation pathways can be ascertained and dis
tinguished by the appearance of cell surface and functional
differentiation markers.
Fig. 1 shows the response of exponentially proliferating cells
placed in culture with IO"6 M RA. The resulting progression
along the myeloid differentiation lineage is characterized by
weak expression of the Mol (19-21) cell surface antigenic
determinant, a marker for mature myeloid and monocytic cells.
There was no expression of the Mo2 (19, 20) and My4 (21-24)
cell surface antigenic determinants, markers for mature mono
demonstrated previously (15). At indicated times the growth and dif
cytic cells. Verifying induced differentiation, there was strong
ferentiation of the cells were assayed. In control experiments carrier
blanks failed to affect cell growth or induce differentiation.
expression of inducible cellular Superoxide production (25), a
Assays for Cell Growth and Differentiation. Cell density was assayed
functional marker for oxidative metabolism characteristic of
as before (6, 12, 13, 15-17) by multiple hemacytometer counts, using
mature myeloid and monocytic cells. As shown in Fig. 2A, there
at least eight counting fields. The standard deviation was routinely ± was progressive GI/O specific growth arrest indicated by the
10%. Cell viability was estimated by exclusion of 0.1% trypan blue in
enrichment in the relative number of Gr DNA cells. As we have
physiological saline as routinely in excess of 95%. At indicated times,
reported previously (13,15), onset of phenotypic differentiation
0.1 ml of harvested cell suspension was harvested to assay cell density.
and G,/0 specific growth arrest did not occur until 48 h of
Distribution of cells in cell cycle phases GI/O,S, G2 + M was determined
exposure to RA. Since the cell cycle duration in these cells is
by flow cytometry as described previously (6, 12, 13, 15-17), using
approximately 24 h, this period corresponds to two division
isolated nuclei stained with propidium iodide. At the indicated times,
IO6cells were harvested to assay cell cycle distribution. The capability
cycles. At 96 h most cells are morphologically indentifiable as
having matured along the myeloid pathway (Table 1). In conto undergo oxidative metabolism characteristic of differentiated cells
130
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1987 American Association for Cancer Research.
CONTROL OF DIFFERENTIATION
Mo1
trast, as we have previously shown (15), cells exposed to RA
for 24 h and subsequently washed and recultured in inducerfree medium continue to proliferate and do not undergo phenotypic differentiation or G i/o growth arrest (Table 2). This 24h exposure induces the precommitment state, as we have pre
viously reported (15).
The course of monocytic differentiation induced by D3 is
readily distinguishable from that of myeloid differentiation due
to RA. Fig. 3 shows the response of exponentially proliferating
cells placed in culture with 10~6M D3. The resulting progression
10080604020
•
9
''Mo
0'100806040200'10080604020
2---_My
4--n
oSOoo-/^"0
0
10080604020
0----oo
24
48
72
AFTER PRECOMMITMENT
96
Hours
Fig. 1. Cell surface and functional differentiation markers expressed by cells
undergoing myeloid differentiation due to continuous exposure to 10 '' M RA.
Abscissa hours in culture; ordinate, percentage of cells positive for Mol, Mo2,
My4 surface markers, and inducible cellular Superoxide (SO) production.
Fig. 2. Course of induced (;, „
growth arrest due to continuous exposure
RA, continuous exposure to D3 (O) or exposure to D3 (0-24 h) followed
washing and culture in inducer-free medium (A), exposure to RA (0-24
followed by washing and exposure to I), (24-96 h), or exposure to 1>>(0-24
followed by washing and exposure to RA (24-96 h).
to
by
h)
h)
along the monocytic differentiation lineage is characterized by
strong expression of Mol, Mo2, and My4 surface antigenic
determinants, as well as inducible Superoxide production. There
was progressive d/o specific growth arrest as evidenced in Fig.
2B by the enrichment in the relative number of GI DNA cells.
Onset of phenotypic differentiation and G1/0 growth arrest did
not occur until 48 h of exposure to D3. At 96 h the cells were
morphologically identifiable as having matured along the mon
ocytic pathway (Table 1). In contrast, cells exposed to D3 for
24 h and subsequently washed and recultured in inducer-free
medium did not differentiate (Fig. 4) and continued to prolif
erate without undergoing GI/O specific growth arrest (Fig. 2B).
Thus, while RA and D3 both induced phenotypic differentiation
and GI/O growth arrest, progression along the myeloid and
monocytic pathways was readily distinguishable. In both cases
significant expression of differentiation specific markers did
not occur until 48 h of exposure to inducer. In both cases cells
exposed to inducer for 24 h did not express any subsequent
phenotypic differentiation or GI/Ospecific growth arrest.
Lineage Specificity Determination after Precommitment. Pre
commitment cells created by a 24-h exposure to RA and then
subsequently exposed to D3 show onset of monocytic differen
tiation within 24 h. Fig. 5 shows the response of initially
exponential cells that were exposed to 10~6 M RA for 24 h,
washed, and recultured in medium containing 10~6M D3. There
was progressive expression of Mol, Mo2, My4, and inducible
Superoxide production that was similar in kinetics and extent
to the case of cells continuously exposed to D, (Fig. 3). There
was also progressive GI/O specific growth arrest similar in
kinetics and extent to the case of continuous I><exposure (Fig.
2, B, and C). In particular, onset of progressive monocytic
differentiation and GI/O specific growth arrest occurred within
24 h of exposure to D3. (Since the cell cycle duration was
approximately 24 h in these cells, this corresponded to one
division cycle.) Since this required 48 h in cells continuously
treated with D3 (Fig. 3), the RA-induced precommitment cells
were thus capable of an accelerated response to D3. At 96 h the
cells were morphologically identifiable as having differentiated
along the monocytic lineage (Table 1). The induced progressive
appearance of monocytic differentiation markers and (•,
u
growth arrest for cells continuously exposed to D3 and cells
exposed first to RA and then D3 was thus similar, indicating
that the precommitment state was without lineage specificity.
Determination of lineage specificity was thus a late event that
occurred after precommitment since the cells did not distin
guish between an initial exposure to RA or D3 in their subse
quent responses.
Determination of lineage specificity as a late event can be
verified in the reciprocal experiment where cells are first ex
posed to D3 and then to RA. Fig. 6 shows the response of
initially exponential cells exposed to 10~6 M D3 for 24 h,
washed, and recultured in medium containing 10~6 M RA. The
response with respect to Mol, Mo2, My4, and inducible superoxide production was similar to the case of cells continuously
131
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1987 American Association for Cancer Research.
CONTROL
OF DIFFERENTIATION
AFTER PRECOMMITMENT
Table I Differentiation ofHL-60 cells after treatment with different ¡nducers
(%)Monocyte
counts at 96 h
Indurar'Differential
BandRA
Promonocyte
Monoblast
Promyelocyte
Myelocyte
11
3
20
16
Metamyelocyte
43
26
Dj/RA
51
30
D3
9
49
42
52* RA/D,
7
41
RA, cells continuously treated with RA; D3/RA, cells treated with D3 (0-24 h) and RA (24-96 h); D3, cells continuously treated with D3; RA/D3, cells
with RA (0-24 h)
h).Table
and D3 (24-96
pRelativi2 Response of cells exposed to RAfor 24 h
8°ability
of cells expressing
Time(%)0
(h)
SO(+)° Mol(+)
cells%
1.0
24
1.8
48
3.0
72
6.7
0" 96
7.7
SO, inducible supi100
99
99
96
97
98
0
4
2
2
0
0
0
0
0
0
100
Mo2(+)
My4(+)
0
0
0
0
0
0
0
0
0
2°
0«
2My
Mo 1
/O'
8040
-_
o
0
20
100
8040
•—
"
4SOtreated24
200<
100
Mo 2
8040
20
0<
o
100__o/
jT
8°^__^_o-—
600
20Oc
'
(-o)
100
8040
My 4
°'^^^^
4020
J>¿r
48
20 j__
0'
— ç^^
i
i
72
96
Hours•
Fig. 4. Cell surfac«and functional differentiation markers expressed by cells
exposed
to 10"* M D3Mo1Mo
inducer-freemedium.
(0-24 h) followed by washing and culture in
Abscissa and ordinate as in Fig. 1
i
.SO
100
80m
6°4020
mon-o-^^
^^°
respond as if they had been continuously exposed to the
ocytic inducer. Conversely, cells initially exposed to the monocytic inducer and then to the myeloid inducer respond as if
continuously exposed to the myeloid inducer. Thus in effecting
^^^
^^"
n:roxide.40
24
48
72
96
Hours
Fig. 3. Cell surface and functional differentiation markers expressed by cells
undergoing monocytic differentiation due to continuous exposure to IO"6 M D3.
Abscissa and ordinate as in Fig. 1.
exposed to RA (Fig. 1). The occurrence of G)/0 specific growth
arrest was also similar to the case of continuous RA exposure
(Fig. 2, A and D). In particular, within 24 h of exposure to RA
there was onset of expression of inducible Superoxide produc
tion, Gi/o growth arrest, and weak Mol expression character
istic of myeloid differentiation. At 96 h the cells were morpho
logically identifiable as having progressed along the myeloid
lineage (Table 1).
DISCUSSION
The results presented here show that proliferating HL-60
cells driven to precommitment by a brief exposure to a myeloid
inducer (RA) and then exposed to a monocytic inducer (D3)
(RA) and monocytic (D3) inducers apparently induced similar
necessary early functions which were independent of lineage.
In contrast, late cellular events induced by these inducers specify
the differentiation lineage. Thus, for example, in the case of an
initial exposure to RA, the induced precommitment state is
without lineage specificity.
Collectively the data from this and previous studies suggest
a possible model of the induction of terminal differentiation in
HL-60 cells. In this hypothesis, the process by which prolifer
ating cells ultimately yield G0 arrested and phenotypically dif
ferentiated descendents involves an identifiable intermediate
state, i.e.,, precommitment. Accordingly the cellular processes
occurring while the cells are still proliferating can be resolved
in two steps: (a) early events leading to precommitment; and
(b) late events subsequent to precommitment but prior to onset
of Go arrest and phenotypic differentiation. Precommitment
cells have a heritable but labile memory of their initial exposure
to inducer which confers upon them the capability to undergo
terminal differentiation with only an abbreviated subsequent
132
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1987 American Association for Cancer Research.
CONTROL
OF DIFFERENTIATION
and dimethyl sulfoxide-induced (18) differentiation as well as
the present studies. It has also been suggested that growth
regulation is primarily associated with early events (17, 18),
while, according to the present studies, control of differentiation
lineage specificity is primarily associated with late events. In
similar studies using sodium butyrate as a monocytic inducer
in lieu of D3, we have found similar results as with the present
studies.4 As has been shown here, the precommitment state is
Mo1
apparently without lineage specificity.
Previous findings that growth arrest can be uncoupled from
differentiation in these cells support this two-step model (17,
18, 26). Observation of such an uncoupling is not limited to
HL-60 cells. It has also been observed in the pokeweed mitogeninduced proliferation of normal B-lymphocytes and their dif
ferentiation into plasmacytes (27). Interestingly, in this case,
occurrence of terminal GI/O specific growth arrest was also
anteceded by two cell division cycles (28).
The molecular basis of these processes is still obscure. How
ever, certain details are apparent. The initial signal appears to
be S-phase specific. This was found in mathematical modeling
of the behavior of asynchronous cells (13) as well as in synchro
nized cells (14). Furthermore, a pulse exposure of subcytotoxic
hydroxyurea, an S-phase specific agent, could induce precommitm it (29). These data point to the involvement of DNA
replication in initiating the program of terminal differentiation.
Considered in the light of previous studies of hydroxyureainduced dihydrofolate reducÃ-ase gene amplification (30-33),
the hydroxyurea studies suggest that this process may involve
selective gene amplification.
Precommitment cells have characteristically altered abun
dances of cytosolic calcium binding proteins (15). However, the
derivation of precommitment and subsequent differentiation
appears to be relatively insensitive to inhibitors of calcium flux
or calmodulin antagonists (34). Precommitment cells induced
by retinoic acid do have elevated levels of c-myc RNA (16). The
elevated c-myc RNA levels are transient and decrease in ter
minally differentiated cells. Strikingly, onset of terminal differ
entiation precedes the decrease, suggesting that the decrease is
a result, but not a cause, of terminal differentiation. Further
more, induction of precommitment by a brief exposure to
hydroxyurea also causes elevation in c-myc RNA levels.5 Fi
Fig. 5. Cell surface and functional differentiation markers expressed by cells
exposed to 10"* M RA (0-24 h) followed by washing and exposure to I0~* M D3
(24-96 h). Abscissa and ordinate as in Fig.l.
10080604020
°^^\-Mo
—~°
OÃ-10080
2---My4--SO°--~~~~~a**O¿_
604020
nally, pulse exposure of cells to retinoic acid for one division
cycle causes elevation of c-myc levels such that the level remains
transiently elevated despite removal of inducer. The kinetics of
elevation and subsequent decrease parallels the duration of
cellular precommitment memory.5 The c-myc protooncogene
010080
604020
may thus have a regulatory role in inducing and sustaining the
precommitment state.
In this regard it is of significance to note that c-myc RNA
levels decrease early during D3-induced monocytic differentia
tion (8). This is in contrast to the transient early increase (16)
for RA-induced myeloid differentiation. Thus modulation of
the c-myc RNA levels is different for different differentiation
pathways. As we have suggested previously (16), c-myc may
have a regulatory role in controlling cellular choice of a specific
differentiation lineage.
The present results are of potential significance to the appli
cation of differentiation inducing agents in chemotherapy. They
show that there is an additive effect of short doses of different
inducers. Thus, this may be a way to minimize any potential
toxicity incurred by prolonged exposure to a single inducer.
0100806040200
l)
I-Mol--„ 24
AFTER PRECOMMITMENT
'
96Hours
48
l
72
Fig. 6. Cell surface and functional differentiation markers expressed by cells
exposed to 10"* M D3 (0-24 h) followed by washing and exposure to 10~* M RA
(24-96 h). Abscissa and ordinate as in Fig. 1.
exposure to inducer (15, 17). This two-step character of the
induced cellular programs leading to terminal differentiation
has been indicated in previous studies of RA-induced (15, 17)
4 A. Yen et al., unpublished observation.
5 A. Yen, unpublished observation.
133
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1987 American Association for Cancer Research.
CONTROL
OF DIFFERENTIATION
AFTER PRECOMMITMENT
ment memory associated with altered nuclear structure. J. Cell. Physiol.,
118: 277-286, 1984.
16. Yen, A., and Guernsey, D. L. Increased c-Myc RNA levels associated with
the pre-commitment state during HL-60 myeloid differentiation. Cancer
Res., 46:4156-4161, 1986.
17. Yen, A., Powers, V., and Fishbaugh, J. Retinoic acid induced HL-60 myeloid
differentiation: dependence of early and late events on isomerie structure.
Leuk. Res., 10:619-629, 1986.
18. Yen, A. Control of HL-60 myeloid differentiation: evidence of uncoupled
growth and differentiation control, S-phase specificity, and two-step regula
tion. Exp. Cell. Res., 156: 198-212, 1985.
19. Todd, R. F., Griffin, J. D., Ritz, J., Nadler, L. M., Abrams, T., and
Schlossman, S. F. Expression of normal monocyte-macrophage differentia
tion antigens on HL-60 promyelocytes undergoing differentiation induced
by leukocyte-conditioned medium or phorbol diester. Leuk. Res., 5: 491495, 1981.
20. Todd, R. F., Nadler, L. M., and Schlossman, S. F. Antigens on human
monnaies identified by monoclonal antibodies. J. Immunol., 126: 14351444, 1981.
21. Kufe, D. W., Griffin, J., Mitchell, T., and Shafman, T. Polyamine require
ments for induction of HL-60 promyelocyte differentiation by leukocyteconditioned medium and phorbol ester. Cancer Res., 44:4281-4284, 1984.
22. Griffin, J. D., Ritz, J., Nadler, L. M., and Schlossman, S. F. Expression of
myeloid differentiation antigens on normal and malignant myeloid cells. J.
Clin. Invest., 68:932-941, 1981.
23. Nadler, L. M., Ritz, J., Griffin, J. D., Todd, R. F., Reinherz, E. L., and
Schlossman, S. F. Diagnosis and treatment of human leukemias and lymphomas utilizing monoclonal antibodies. Prog. Hematol., 12:187-225,1981.
24. Griffin, J. D., Mayer, R. J., Weinstein, H. J., Rosenthal, D. S., Coral, F. S.,
Beveridge, R. P., and Schlossman, S. F. Surface marker analysis of acute
myeloblastic leukemia: identification of differentiation-associated phenotypes. Blood, 62: 557-563, 1983.
25. Groopman, J. E., and Golde, D. W. Biochemistry and function of monocytes
and macrophages. In: W. J. Williams, E. Beutler, A. J. Erslev, and M. A.
Lichtman (eds.), Hematology, p. 848, Ed. 3. New York: McGraw-Hill Book
Co., 1983.
26. Wickstrom, E. L., Wickstrom, E., Lyman, G. H., and Freeman, D. L. HL60 cell proliferation inhibited by an anti-c-myc pentadecadeoxynucleotide.
Federation Proceedings, 45: 1330, 1986.
27. Yen, A., and Lewin, D. Uncoupling lymphocyte proliferation from differen
tiation: dissimilar dose-response relations for pokeweed mitogen-induced
proliferation and differentiation of normal human lymphocytes. Cell Immu
nol., 61: 332-342, 1981.
28. Yen, A., and Stein, L. S. Polyclonal mitogenesis of human lymphocytes by
PWM: two preprogrammed division cycles resulting in cells refractile to
further mitogenesis. Cell Immunol., 57:440-454, 1981.
29. Yen, A., Freeman, L., and Fishbaugh, J. Hydroxyurea induces precommitment during retinoic induced HL-60 terminal myeloid differentiation: pos
sible involvement of gene amplification. Leuk. Res., in press, 1986.
30. Schimke, R. T. Gene amplification in cultured animal cells. Cell, 37: 705713, 1984.
31. Hill, A. B., and Schimke, R. T. Increased gene amplification in L5178Y
mouse lymphoma cells with hydroxyurea-induced chromosomal aberrations.
Cancer Res., 45: 5050-5057, 1985.
32. Brown, P. C., Tlsty, T. D., and Schimke, R. T. Enhancement of methotrexate
resistance and dihydrofolate reducÃ-asegene amplification by treatment of
mouse 3T6 cells with hydroxyurea. Mol. Cell. Biol., 3: 1097-1107, 1983.
33. Tlsty, T., Brown, P. C., Johnston, R., and Schimke, R. T. Enhanced frequency
of generation of methotrexate resistance and gene amplification in cultured
mouse and hamster cell lines. In: R. T. Schimke (ed.), Gene Amplification,
pp. 231-238. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory
Press, 1982.
34. Yen, A., Freeman, L., Powers, V., Van Sant, R., and Fishbaugh, J. Cell cycle
dependence of calmodulin levels during HL-60 proliferation and myeloid
differentiation: no changes during pre-commitment. Exp. Cell Res., 165:
139-151, 1986.
ACKNOWLEDGMENTS
We are grateful to Joyce .lamini for skillful secretarial assistance in
the preparation of this manuscript. We thank Dr. Duane Guernsey for
critically reading this manuscript.
REFERENCES
1. Collins, S. J., Gallo, R. C., and Gallagher, R. E. Continuous growth and
differentiation of human myeloid leukemia cells in suspension culture. Nature
(Lond.), 270: 347-349, 1977.
2. Fontana, J. A., Colbert, D. A., and Deisseroth, A. B. Identification of a
population of bipotent stem cells in the 111 6(1human promyelocytic leuke
mia cell line. Proc. Nati. Acad. Sci. USA, 78: 3863-3866, 1981.
3. Breitman, T. R., Selonick, S. E., and Collins, S. J. Induction of differentiation
of human promyelocytic leukemia cells (HL-60) by retinole acid. Proc. Nati.
Acad. Sci. USA, 77: 2936-2940, 1980.
4. Breitman, T. R. Induction of terminal differentiation of HL-60 and fresh
leukemic cells by retinoic acid. In: R. P. Revoltella, C. Basilico, R. C. Gallo,
G. M. Pontieri, G. Rovera, and J. H. Subak-Sharpe (eds.). Expression of
Differentiated Functions in Cancer Cells, pp. 257-273. New York: Raven
Press, 1982.
5. Perussia, B., Lebman, D., Ip, S. H., Rovera, G., and Trinchieri, G. Terminal
differentiation surface antigens of myelomonocytic cells are expressed in
human promyelocytic leukemia cells (HL-60) treated with chemical inducers.
Blood, 5«:836-843, 1981.
6. Yen. A., Van Sant, R., Harvey, J., and Fishbaugh, J. Myeloid differentiation
inducing factors produced by PWM treated normal G,/0 lymphocytes but not
CLL cells. Cancer Res., 45: 4060-4066, 1985.
7. Mangelsdorf, D. J., Koeffler, H. P., Donaldson, C. A., Pike, J. W., and
Haussler, M. R. 1,25-Dihydroxyvitamin I><indiimi differentiation in a
human promyelocytic leukemia cell line (HL-60): receptor-mediated matura
tion to macrophage-like cells. J. Cell Biol., 98: 391-398, 1984.
8. Reitsma, P. H., Rothberg, P. G., Astrin, S. M., Trial, J., Bar-Shavit, Z., Hall,
A., Teitelbaum, S. I... and Kahn, A. J. Regulation ofMyc gene expression in
HL-60 leukemia cells by a vitamin D metabolite. Nature (Lond.), 306: 492494, 1983.
9. McCarthy, D. M., San Miguel, J. F., Freake, H. C., Green, P. M., Zola, H.,
Catovsky, D., and Goldman, J. M. 1,25-Dihydroxyvitamin D3 inhibits pro
liferation of human promyelocytic leukemia (HL-60) cells and induces monocyte-macrophage differentiation in HL-60 and normal human bone marrow
cells. Leuk. Res., 7: 51-55, 1983.
10. Tanaka, H., Abe, E., Miyaura, C., Shiina, Y., and Suda, T. la,25-Dihydroxyvitamin I >, induces differentiation of human promyelocytic leukemia cells
(HL-60) into monocyte-macrophages, but not into granulocytes. Biochem.
Biophys. Res. Commun., 117: 86-92, 1983.
11. Polli, N., O'Brien, M., De Castro, J. T., RodrÃ-guez,B., McCarthy, D., and
Catovsky, D. Monocytic differentiation induced by 1,25-dihydroxyvitamin
I), in myeloid cells: an ultrastructural immunocytochemical study. Leuk.
Res., 9: 259-270, 1985.
12. Yen, A., Reece, S. L., and Albright, K. Membrane origin for a signal eliciting
a program of cell differentiation. Exp. Cell. Res., 152:493-499, 1984.
13. Yen, A., Reece, S. L., and Albright, K. L. Control of cell differentiation
during proliferation. II. Myeloid differentiation and cell cycle arrest of HL60 promyelocytes preceded by nuclear structural changes. Leuk. Res., 9: 5171, 1985.
14. Yen, A., and Albright, K. Evidence for cell cycle phase specific initiation of
a program of HL-60 cell myeloid differentiation mediated by inducer uptake.
Cancer Res., 44: 2511-2515. 1984.
15. Yen, A., Reece, S. L., and Albright. K. L. Dependence of HL-60 myeloid cell
differentiation on continuous and split retinoic acid exposures: pre-commit-
134
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1987 American Association for Cancer Research.
Control of HL-60 Cell Differentiation Lineage Specificity, a Late
Event Occurring after Precommitment
Andrew Yen, Mary Forbes, Gwen DeGala, et al.
Cancer Res 1987;47:129-134.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/47/1/129
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 June 17, 2017. © 1987 American Association for Cancer Research.