MOLECULAR REPRODUCTION AND DEVELOPMENT 46:11–18 (1997)
Regulation of Cell Cycle Entry and
G1 Progression by CSF-1
MARTINE F. ROUSSEL*
Department of Tumor Cell Biology, St. Jude Children’s Research Hospital, Memphis, Tennessee
ABSTRACT Proliferation, differentiation, and for mitogenicity. Mol Reprod Dev 4 6 :1 1 –1 8 , 1 9 9 7 .
survival of monocytes, macrophages, and their immedi- r 1997 Wiley-Liss, Inc.
ate progenitors is regulated by the macrophage colony-
stimulating factor (CSF-1). CSF-1 initiates a mitogenic Key Words: CSF-1; receptor signaling; G1 progres-
response by binding to its receptor (CSF-1R), thereby sion
activating the receptor’s intrinsic tyrosine kinase activ-
ity and initiating signaling via multiple effector-mediated
INTRODUCTION
pathways. CSF-1 is required throughout G1 to ensure
entry of bone marrow–derived macrophages into S One of the most important decisions a cell has to
phase, and persistent CSF-1R kinase activity is neces- make is whether to divide or not to divide. That decision
sary to the expression of both immediate early (e.g.,
is made during the first gap phase (G1) of the cell cycle
and is regulated by growth factors, which come from the
c-fos, c-jun, and c-myc) and delayed early (e.g., D-type
environment (Pardee, 1989). We have focused on stud-
cyclins) response genes.
ies relating to the role of growth factors during the cell
Ectopic expression of human CSF-1R in different
cycle and, more specifically, using genetics on the
mouse cell lines, including fibroblasts, IL-3-dependent
identification of the signaling pathways that directly
myeloid cells, and early pre-B cells, confers CSF-1
affect the progression of the cells through G1 and the
responsiveness by replacing the cells’ requirements for commitment to DNA synthesis, also called S phase.
other mitogenic growth factors. NIH-3T3 fibroblasts Macrophages require CSF-1 for most of the G1 phase
engineered to express a human CSF-1 receptor point (Sherr and Stanley, 1990). Once the cells commit to the
mutant (CSF-1R [Y809F]) fail to proliferate in response S phase, they are able to continue the rest of the cell
to CSF-1 and remain arrested in the early G1 phase of cycle in the absence of growth factor stimulation. This
the cell cycle. Despite CSF-1-dependent transcription of was elegantly demonstrated in experiments performed
fos and jun family members, c-myc, D-type, and E-type by Richard Stanley’s group. These experiments were
G1 cyclin mRNAs are not expressed in the latter cells in performed using the macrophage cell line BAC1.2F5.
response to growth factor stimulation. However, en- When these cells were starved from CSF-1, they would
forced expression of c-myc or D-type cyclins, but not collect at the beginning of G1. The cells were then
cyclin E, resensitizes cells bearing CSF-1R (Y809F) to restimulated with the growth factor for different times,
the mitogenic effects of CSF-1, enabling them to after which CSF-1 was removed, and the entry into S
proliferate continuously in liquid culture and to form phase examined. These experiments showed that if
colonies in agar in response to the growth factor. D-type CSF-1 is removed at any time during G1, before the cells
cyclin mutants defective in binding to the retinoblas- commit to S phase, the cells would not enter S phase
toma protein (pRB) were unable to rescue mutant and would reset at the beginning of G1. This suggested
receptor signaling, suggesting that the ability of D-type that growth factor signaling was required minimally at
cyclin-dependent kinases to cancel pRB’s growth- two steps during the cell cycle: to enter G1 and to
suppressive function is necessary for CSF-1-induced G1 commit to S phase. Experiments using the CSF-1
exit. By contrast, cyclin E must function in a different receptor (CSF-1R) were facilitated by the fact that this
pathway. receptor can be transduced into many cell lines, except
Cells rescued by c-myc were prevented from enter-
T cells, and the transfer of this receptor into cells
ing S phase by microinjection of antibodies to cyclin D1.
renders the cells dependent on CSF-1 for their growth.
Remarkably, the cells containing the newly transferred
Conversely, cyclin D1–rescued cells were inhibited
from forming CSF-1-dependent colonies in agar when
challenged with either a dominant-negative c-myc mu-
Sponsored by NIH, grant number CA-56819; Core Grant, CA21765.
tant or mad, a transcription factor which competes with
*Correspondence to: Dr. Martine F. Roussel, St. Jude Children’s
myc for max, its requisite heterodimeric partner. Thus,
Research Hospital, Department of Tumor Cell Biology, 5th Floor
although the expression of c-myc and D-type cyclins is Danny Thomas Research Tower, 332 N. Lauderdale, Memphis, TN
rate limiting for G1 phase progression, their functions 38105.
are interdependent, with both activities being required Accepted 4 March 1996.
r 1997 WILEY-LISS, INC.
, 12 M.F. ROUSSEL
CSF-1R can dispose of their previous growth factor
requirements and replace them with CSF-1 (Roussel
and Sherr, 1993). Therefore, if the CSF-1R is trans-
ferred into fibroblasts, which normally require PDGF
and EGF or IGF for proliferation, they can proliferate
in chemically defined cultured conditions with CSF-1 as
the sole growth factor.
In all the experiments I will present on the regulation
of G1 progression during the cell cycle, the human
CSF-1R has been transferred into mouse fibroblasts.
This is a very useful system because the human CSF-1R
cannot be activated by the murine growth factor, and
therefore one can use human CSF-1 to study the effect
of mutations in the human receptor on the proliferation
of mouse fibroblasts. In this system, human CSF-1 is
the only growth factor affecting these cells. All of our
experiments were done by this system, but they could
also be done in myeloid cells, and some of the experi-
ments I present were also done using IL3-dependent
cells.
CSF-1 RECEPTOR STRUCTURE
When CSF-1 binds to its receptor, the receptor dimer-
izes, the intrinsic tyrosine kinase activity of the recep-
tor is activated, and various tyrosines are phosphory-
lated (Fig. 1). The tyrosine residues on the receptor that
become phosphorylated have been mapped. These are
tyr 561, tyr 571, three tyrosines in the kinase insert
Fig. 1. The human CSF-1 receptor: structure and the immediate
domain, tyr 699, 708, 723, and 809 in the core kinase. early response to CSF-1. The CSF-1R is a transmembrane glycopro-
Tyrosine 561 is the binding site for the Src family tein composed of an extracellular ligand binding domain, a short
kinases, and activation of Src is important for signal hydrophobic transmembrane domain, and an intracellular kinase
transduction by the receptor (Roche et al., 1995; Alonso domain split in two by a short kinase insert. Upon CSF-1 binding, the
et al., 1995). The tyrosine at position 571 has not been receptor dimerizes, its tyrosine kinase is activated, and specific
tyrosine residues are phosphorylated in trans. Phosphorylated resi-
shown, so far, to bind specific proteins. The first of three dues serve as anchors for SH2-containing cellular effectors that relay
tyrosines in the kinase insert domain of the receptor, receptor proliferative signals. Y, tyrosine residue; P, phosphorylated;
tyr 699, binds to Grb2 (Van der Geer and Hunter, 1993). P13K, phosphoinositol 3 kinase.
Recently, tyr 708 was shown to be required for Stat 1
activation (Novak et al., 1995). Tyrosine 723 is the in
vitro binding site for the PI-3 kinase (Shurtleff et al., activity between the wild-type and the Y809F mutant
1990; Reedijk et al., 1992), and the tyrosine at position receptor. The cellular proteins that were phosphory-
809 is one of the major autophosphorylation sites of the lated on tyrosine in response to CSF-1 activation were
receptor. compared in the cells expressing the Y809F mutant
receptor and cells expressing the wild-type receptor.
EFFECT OF MUTATIONS IN THE CSF-1 There was slightly less tyrosine phosphorylation in the
RECEPTOR ON CELL PROLIFERATION Y809F cells compared with the wild type, but all of the
Mutations in the receptor were made to evaluate the bands in the wild type could be accounted for in the
roles of various tyrosine residues in receptor function Y809F cells (Roussel et al., 1990). Therefore, these
(Roussel and Sherr, 1993; Roussel, 1994). Remarkably, experiments suggested there was nothing intrinsically
the deletion of the kinase insert did not have any wrong with the activity of the mutant receptor, but that
significant effect on the ability of the cell to respond to the problem was downstream of the receptor.
CSF-1, suggesting that the binding of all of the partners
to the phosphorylated tyrosines in that domain was not EFFECT OF THE Y809F CSF-1 RECEPTOR
essential for CSF-1-dependent proliferation. However, MUTATION ON THE IMMEDIATE EARLY
the substitution of phenylalanine for tyrosine at posi- GENE RESPONSE
tion 809 gave a very striking result. The cells contain- We decided to look at the effect of the Y809F mutant
ing this mutant CSF-1R survived in the presence of the receptor on the transcription of genes associated with
growth factor, but were unable to proliferate. One the immediate early response to growth factor activa-
explanation for these results was that the receptor was tion (Roussel et al., 1991). RNA was associated from
defective in its overall kinase activity. However, we cells containing either the wild-type or the Y809F
found no significant difference in the in vitro kinase mutant CSF-1 receptor, which had been stimulated
