MOLECULAR AND CELLULAR BIOLOGY, Sept. 1999, p. 6229–6239 0270-7306/99/$04.00⫹0 Copyright © 1999, American Society for Microbiology. All Rights Reserved.
Vol. 19, No. 9
Elevated Cyclin E Levels, Inactive Retinoblastoma Protein, and Suppression of the p27KIP1 Inhibitor Characterize Early Development of Promyeloid Cells into Macrophages QIANG LIU,1,2 ROGER W. VANHOY,1 J. H. ZHOU,1 ROBERT DANTZER,3 GREGORY G. FREUND,4 AND KEITH W. KELLEY1* Department of Animal Sciences, Laboratory of Immunophysiology,1 and Department of Pathology, College of Medicine,4 University of Illinois, Urbana, Illinois 61801; INRA-INSERM U394, Institute Franc¸ois Magendie, 33077 Bordeaux Cedex, France3; and Department of Hematological Oncology, Cancer Center, Sun-Yat Sen University of Medical Science, 510060 Guangzhou, People’s Republic of China2 Received 23 July 1998/Returned for modification 27 April 1999/Accepted 28 May 1999
Cyclin-dependent kinase inhibitors such as p27KIP1 have recently been shown to lead to cellular differentiation by causing cell cycle arrest, but it is unknown whether similar events occur in differentiating promyeloid cells. Hematopoietic progenitor cells undergo lineage-restricted differentiation, which is accompanied by expression of distinct maturation markers. Here we show that the classical growth factor insulin-like growth factor I (IGF-I) potently promotes vitamin D3-induced macrophage differentiation of promyeloid cells, as assessed by measurement of a coordinate increase in expression of the integrin ␣ subunit CD11b, the CD14 lipopolysaccharide receptor, and the macrophage-specific esterase, ␣-naphthyl acetate esterase, as early as 24 h following initiation of terminal differentiation. Addition of IGF-I to cells undergoing vitamin D3-induced differentiation also leads to an early increase in expression of cyclin E, phosphorylation of the retinoblastoma tumor suppressor protein, and a doubling of the cell number. Early expression of CD11b (24 h) is simultaneously accompanied by inhibition in the expression of p27KIP1. Cell cycle analysis with propidium iodide revealed that CD11b expression at 24 h following initiation of differentiation occurs at all phases of the cell cycle instead of only those cells arrested in G0/G1. Similarly, development of a novel double-labeling intra- and extracellular flow-cytometric technique demonstrated that single cells expressing the mature leukocyte differentiation antigen CD11b can also incorporate the thymidine analog bromodeoxyuridine. Likewise, expression of the intracellular DNA polymerase ␦ cofactor/proliferating-cell nuclear antigen at 24 h is also simultaneously expressed with the surface marker CD11b, indicating that these cells continue to proliferate early in their differentiation program. Finally, at 24 h following induction of differentiation, IGF-I promoted a fourfold increase in the uptake of [3H]thymidine by purified populations of CD11b-expressing cells. Taken together, these data demonstrate that the initial steps associated with terminal macrophage differentiation occur concomitantly with progression through the cell cycle and that these very early differentiation events do not require the accumulation of p27KIP1. p27KIP1 or p21CIP1 in the absence of differentiation agents can lead to terminal differentiation of promonocytic cells. The finding that overexpression of cell cycle inhibitors is associated with cellular maturation does not necessarily indicate that cell growth and differentiation cannot occur simultaneously. For example, the increase in cellular proliferation caused by stem cell factor occurs concomitantly with enhanced megakaryocytic differentiation (61). Furthermore, germ line disruption of the three major CDK inhibitors, INK4 (59), p21CIP1 (10, 14), and p27KIP1 (17, 34, 51), has now been reported, but none of these strains of knockout mice has global defects in differentiated tissues and organs. Similarly, although the E2F transcription factor is well characterized as a cell cycle progression factor, germ line disruption of E2F was recently shown to lead to the opposite phenotype of hyperplasia rather than the expected state of hypoproliferation (18, 71). Although p27KIP1 is expressed in differentiating promyeloid cells treated with vitamin D3 (23), this event occurs several days after the cells begin to differentiate (72). This finding has recently been confirmed in human primary precursor CD34⫹ cells undergoing myeloid differentiation (63). More importantly, new evidence demonstrates a differentiation-inhibiting function of p21CIP1, and this inhibitory role can be separated from the suppressive effect of this inhibitor on cell cycle progression
Mitosis and differentiation are two cellular processes that are generally viewed as mutually exclusive events (43). Cell cycle progression is controlled by the activity of cyclin-dependent kinases (CDKs), which are finely regulated by accumulation of cyclins and association with the newly discovered CDK inhibitory proteins (CKIs). G1-phase CKIs, including INK4 (inhibitor of CDK4), p21CIP1, p27KIP1, and p57KIP2, bind CDK-cyclin complexes and thus antagonize CDK activity. The CKIs maintain the retinoblastoma (Rb) tumor suppressor protein in an active, hypophosphorylated form that effectively binds and inhibits the E2F transcription factors (60). Since cellular differentiation occurs in the G0 phase of the cell cycle, emerging evidence suggests that CKIs not only restrain cellular growth but also promote differentiation (38). Indeed, expression of p21CIP1 and maintenance of the active state of the Rb protein is correlated with cell cycle arrest of muscle cells, and this is associated with their terminal differentiation (21, 54, 62). Recently, Liu et al. (41) demonstrated that overexpression of
* Corresponding author. Mailing address: Laboratory of Immunophysiology, University of Illinois, 207 Edward R. Madigan Laboratory, 1201 West Gregory Dr., Urbana, IL 61801. Phone: (217) 333-5141. Fax: (217) 244-5617. E-mail:
[email protected]. 6229
6230
LIU ET AL.
(15). Since differentiation antigens are required to induce the expression of p21CIP1 (21), these more recent data might indicate that expression of cell cycle inhibitors is more important for maintenance of differentiated cells in G0 (20) than for directly promoting the initial stages of differentiation. Insulin-like growth factor I (IGF-I) is well known as a G1 progression factor (2) that maintains the expression of the DNA polymerase ␦ cofactor/proliferating-cell nuclear antigen (PCNA) (47) and increases the growth of lymphoid and myeloid cells (35, 69). Human HL-60 myeloid precursor cells undergo differentiation toward the macrophage lineage in the presence of vitamin D3 (13). We recently demonstrated that these cells express an endogenous type I IGF receptor and that inhibition of this intrinsic tyrosine kinase receptor blocks the growth and vitamin D3-induced maturation of these cells (40). These data suggested that IGF-I promotes both the proliferation and differentiation of myeloid precursor cells. Here we establish that IGF-I is required for optimal induction of terminal macrophage differentiation induced by vitamin D3, including CD11b, CD14, and ␣-naphthyl acetate esterase (NAE). Western blot analysis revealed that this IGF-I-promoted macrophage differentiation does not lead to early induction of p27KIP1 but, rather, causes an increase in expression of cyclin E and hyperphosphorylation of Rb. By using a novel double-labeling flow-cytometric technique that couples the expression of both differentiation (CD11b) and proliferation (PCNA) antigens, we unambiguously demonstrate that 75% of the early maturing cells simultaneously express both markers. Furthermore, early in the differentiation program (24 h), enriched populations of CD11b-positive cells incorporate [3H]thymidine. It therefore appears that early induction of the p27KIP1 CDK inhibitor and cessation of mitosis are not necessarily required for the early events that lead to the terminal differentiation of myeloid cells. MATERIALS AND METHODS Antibodies and reagents. Powdered RPMI 1640 tissue culture medium (MediaTech, Herndon, Va.) was supplemented with 2 g of sodium bicarbonate per liter, 100 U of penicillin per ml, and 100 g of streptomycin per ml (all from Sigma Chemical Co., St. Louis, Mo.). Fetal bovine serum (FBS) (HyClone Laboratories Inc., Logan, Utah [containing ⬍25 pg of endotoxin/ml by the Limulus amoebocyte lysate assay; Associates of Cape Cod, Inc., Woods Hole, Mass.]) was heat inactivated at 56°C for 30 min. The human promyeloid cell line HL-60 was purchased from the American Type Culture Collection (Rockville, Md.). These cells have a cell cycle length of 23 h, and the mean duration of G1, S, G2 and M are 11, 9, 2, and 1 h, respectively (29). Recombinant human IGF-I was purchased from Intergen (Purchase, N.Y.), and the 1␣,25-(OH)2 vitamin D3 (vitamin D3) was kindly provided by Milan Uskokovic, Hoffmann-La Roche. Rat anti-human CD11b (Mac-1, immunoglobulin G2b kappa chain [IgG2b ]) monoclonal antibody (MAb) was purchased from Boehringer Mannheim (Indianapolis, Ind.), and the irrelevant isotype-matched rat IgG2b was obtained from Sigma Chemical Co. The purified mouse anti-human cyclin E MAb (IgG1 ) and mouse anti-p27KIP1 MAb (IgG1 ) were obtained from Oncogene Science, Inc. (Uniondale, Calif.) and Transduction Laboratories (Lexington, Ky.), respectively. Mouse anti-human CD14 (gp55, IgG2a ), the irrelevant isotype-matched mouse IgG2a Ab, and the mouse anti-human Rb tumor suppressor protein (pRb, IgG1 ), which recognizes both phosphorylated and dephosphorylated forms of Rb, were purchased from Pharmingen (San Diego, Calif.). Mouse anti-proliferating cell nuclear antigen (DNA polymerase ␦ cofactor/PCNA) MAb (IgG1 ) and mouse antibromodeoxyuridine (anti-BrdU) MAb (IgG1 ) were purchased from MBL International Co. (Watertown, Mass.). The F(ab⬘)2 fragment of fluorescein isothiocyanate (FITC)-conjugated goat anti-rat IgG and anti-mouse IgG Ab was obtained from Cappel (Durham, N.C.). Differentiation of HL-60 cells with vitamin D3. Human promyeloid HL-60 cells undergo differentiation toward a macrophage phenotype following addition of vitamin D3 in serum-containing medium (13). The cells were maintained at 37°C in an atmosphere of 95% air and 7% CO2. They were passaged twice weekly in 10% FBS and studied between passages 20 and 40. The cells were washed three times (400 ⫻ g at 4°C) and then incubated in serum-free medium (RPMI 1640 supplemented with 5 g of human transferrin [Sigma Chemical Co.] per ml and 30 nM sodium selenite [Sigma Chemical Co.]) for 24 h prior to each assay. Vitamin D3 (final concentration, 1 M; stock solution diluted in ethanol) was added, and the cells were cultured for the indicated times in the presence or
MOL. CELL. BIOL. absence of IGF-I (100 ng/ml). A similar amount of ethanol (0.1%) was added to the control, non-vitamin D3-containing RPMI 1640. Cytochemical detection of intracellular esterase activity. The enzyme NAE is an intracellular nonspecific esterase that is induced in human mature macrophages but is absent in immature myeloid blast cells (5). To determine the expression of this enzyme, HL-60 cells were seeded at 4 ⫻ 105 cells per well in 24-well culture dishes. The cells were treated with vitamin D3 (1 M) in the presence or absence of IGF-I (100 ng/ml) for 2 days. At the end of the culture period, NAE activity was evaluated by using a commercially available diagnostic kit (Sigma Chemical Co.). Briefly, 2 ⫻ 104 cells (in 0.5 ml) were cytocentrifuged and fixed with citrate-acetone-formaldehyde fixative onto microscope slides. After being rinsed, the cells were incubated with NAE substrate for 30 min and counterstained for 2 min in hematoxylin. The percentage of cells on each slide that contained cytoplasmic black formazan deposits, characteristic of NAE activity, was determined by counting at least 500 cells. Western blotting of p27KIP1, cyclin E, and p110Rb. Cells were washed once with cold phosphate-buffered saline (PBS) (1.5 M NaCl, 19 mM Na2HPO4 䡠 H2O, 8.4 mM KH2PO4) and lysed on ice in cell lysis buffer containing 50 mM sodium HEPES, 150 mM NaCl, 50 mM NaF, 25 mM -glycerophosphate, 10 mM sodium pyrophosphate, 20 mM p-nitrophenyl, 1% Nonidet P-40, 10 g of leupeptin per ml, 10 g of pepstatin A per ml, 1 mM sodium vanadate, 1 mM EDTA, 1 mM benzamide, and 1 mM phenylmethylsulfonyl fluoride (all from Sigma Chemical Co.). Insoluble material was removed by centrifugation at 12,000 ⫻ g for 10 min, and the protein concentration was determined by the Bradford dye method with a protein kit (Bio-Rad Laboratories, Richmond, Calif.) with bovine serum albumin (Sigma Chemical Co.) as the standard. Equal amounts of cell extract (50 g) were subjected to electrophoresis in sodium dodecyl sulfate (SDS)–10% polyacrylamide gels and transferred to polyvinylidene difluoride nitrocellulose sheets (Bio-Rad Laboratories) on a Bio-Rad Western transfer unit. The blotted nitrocellulose was washed twice for 15 min each with distilled water and was subsequently blocked for 20 min at room temperature with freshly prepared PBS containing 3% nonfat dry milk. The membrane was then incubated with 1 g of mouse anti-cyclin E MAb, mouse anti-p27KIP1 MAb, or mouse anti-p110Rb MAb per ml at 4°C for 24 h. After two 15-min washes with distilled water, the polyvinylidene difluoride membrane was incubated for 1 h at room temperature with a goat anti-mouse IgG1 secondary Ab linked to horseradish peroxidase (Amersham Corp.). Following additional washes with PBS–0.05% Tween (Sigma Chemical Co.), Western blot analysis of the specific protein was performed with a standard enhanced chemiluminescence kit (Amersham Corp.). The specificity of the primary Ab was confirmed by the absence of detectable proteins as assessed by blotting an identical sample with an isotype-matched control Ab followed by the appropriate alkaline phosphatase-conjugated secondary Ab. The intensity of hypophosphorylated and hyperphosphorylated Rb as detected on autoradiographs was determined by laser densitometry with a Molecular Dynamics (Sunnyvale, Calif.) personal densitometer equipped with ImageQuant 3.3 software as previously described (45). Flow cytometry to detect cell surface CD11b and CD14. Macrophage development was determined by using flow cytometry to assess the increase in the percentage of cells binding to MAbs specific for the mature macrophage surface antigens, CD11b and CD14, as previously described by others (5, 9, 67). Vitamin D3-treated cells (106) were washed once in PBS supplemented with 0.5% FBS and 0.25% bovine serum albumin (BSA) (wash buffer). The cells were then incubated for 30 min at 4°C in PBS with either rat anti-human CD11b MAb (0.2 g) and its isotype-matched control (IgG2b) or mouse anti-human CD14 MAb (1 g) and its isotype-matched control (IgG2a). After two washes, the cells were incubated in PBS with the secondary FITC-conjugated goat anti-rat or antimouse F(ab⬘)2 fragment for additional 30 min at 4°C. Subsequently, the cells were washed twice and fixed in PBS containing 1% formaldehyde until analyzed by flow cytometry (EPICS V; Coulter Instruments, Miami, Fla.). For each sample, the immunofluorescence intensity of cells stained with the isotype-matched control did not exceed that of 5% of the cell population compared to cells incubated with only the secondary F(ab⬘)2 fragment. The isotype-matched control Ab coupled with the secondary F(ab⬘)2 fragment was used to establish a bitmap of at least 5,000 cells of uniform size. Cell cycle analysis in conjunction with cell surface immunofluorescence. Cell cycle analysis was performed by washing 105 cells three times with RPMI 1640 and then culturing the cells in serum-free defined medium for 24 h. Serumstarved cells were then incubated with medium alone (control) or with vitamin D3 (1 M), IGF-I (100 ng/ml), or both for 5 days. The cells were collected at the indicated times and enumerated with a cell counter (Coulter Instruments). Subsequently, the cells were washed once with PBS and then fixed with 80% ethanol at 4°C for 24 h. After three washes, the fixed cells were incubated for 1 h with a propidium iodide (PI; 20 g/ml [Sigma Chemical Co.]) solution containing 0.1 mg of RNase A (Sigma Chemical Co.) per ml. The cells were then subjected to cell cycle analysis on an EPICS V flow cytometer. Cell surface immunofluorescence was combined with cell cycle analysis by first staining 106 cells for CD11b expression with the FITC-labeled secondary antibody, using the indirect-labeling method as described above. Stained cells were then fixed with 70% ethanol at 4°C for 12 h. After being washed once with ice-cold PBS, the cells were incubated with PI as described above. Ethanol-fixed cells without any previous labeling were used to measure background autofluo-
VOL. 19, 1999
EARLY MACROPHAGE DIFFERENTIATION
rescence. Cells labeled with only FITC or PI were used to set the background gate to exclude any signal derived from a possible FITC-PI staining interaction. Flow cytometry to simultaneously detect both surface CD11b and intranuclear PCNA or BrdU incorporation. Cells (2 ⫻ 106) were first subjected to the indirectlabeling method described above, to define the surface leukocyte differentiation marker CD11b, without 1% paraformaldehyde fixation. They were subsequently fixed in 90% methanol at ⫺20°C for 30 min. After two washes with ice-cold PBS, the cells were blocked with normal sheep serum (NSS [Sigma Chemical Co.], 50% in PBS) at room temperature for 10 min. A mouse anti-PCNA MAb (50 g/ml) or the same amount of irrelevant isotype-matched mouse IgG1 Ab was then added, and the reaction mixture was incubated at 4°C for 10 min and washed twice more with NSS. After centrifugation, the cells were dissolved in PBS containing phycoerythrin (PE)-conjugated secondary sheep anti-mouse IgG1 Ab (Sigma Chemical Co.) (1:100). The reaction mixture was incubated at room temperature for 30 min. The cells were analyzed by flow cytometry after two washes with PBS buffer containing 2% FBS. In each experiment, cells labeled with the relevant primary Ab, but with only either FITC or PE secondary Ab, were used to establish background gating. To determine nuclear BrdU incorporation, cells (2 ⫻ 106) were stained with rat anti-human CD11b Ab and the subsequent rabbit anti-rat FITC-conjugated antibody, prior to intracellular labeling, as described above. After surface antibody staining, the cells were cultured for 3 h in a PBS medium (500 l) containing 0.1 mM BrdU and 0.1 mM deoxycytidine. After three washes with PBS containing 2% FBS, the cells were fixed in 70% ethanol in PBS at ⫺20°C for 30 min. Following two additional rinses with 2% FBS, 1.5 N HCl was added to denature the DNA. Samples were then incubated at room temperature for 30 min. Following another two washes with Na2B4O7 (0.1 M) to neutralize the remaining acid, the cells were blotted with 50% NSS in PBS at room temperature for 30 min. They were then subjected to standard indirect-labeling procedures, first with the mouse anti-BrdU Ab or the same amount of irrelevant isotypematched control Ab and subsequently with PE-conjugated sheep anti-mouse antiserum. Sorting of CD11b-positive and -negative cells and [3H]thymidine incorporation. Cells (8 ⫻ 106) were differentiated for 24 h with vitamin D3 and IGF-I and then labeled with an anti-CD11b antibody (without fixation), which included five washes in PBS supplemented with 0.5% FBS and 0.25% BSA. CD11b-positive and -negative cells were separated by flow cytometry (⬎98% enriched compared to the isotype-control antibody) with a modified EPICS 753 cell sorter (Coulter Instruments) by using serum-free medium as sheath fluid and the settings described above. Triplicate samples (105 cells) were sorted into wells of a 96-well plate (200-l total volume) and incubated in serum-free medium with or without IGF-I (100 ng/ml), both in the absence of vitamin D3. After 18 h, 1 Ci of [3H]thymidine (ICN Biomedical Inc., Irvine, Calif.) was added to each well, and the plate was incubated for an additional 6 h. The cells were harvested, and incorporation of [3H]thymidine was determined with an LS 6000IC scintillation counter (Beckman, Irvine, Calif.) as previously described (44). Statistical analysis. Data were analyzed by using the Statistical Analysis System (58), with Student’s t test being used to detect differences between treatments. Differences of at least P ⬍ 0.05 were considered to be significant.
RESULTS Kinetics of IGF-I-enhanced CD11b expression in differentiating promyeloid cells. Although generally recognized as a progression factor that acts early in the cell cycle (2), IGF-I has been reported to promote the differentiation of both primary B lymphocytes and granulocytes (12, 19, 28, 32, 33, 37). The major inducer of hepatic IGF-I synthesis, growth hormone, is synthesized by leukocytes (8, 68) and promotes the development of several lineages of hematopoietic cells (48–50, 69). Since human (65) and fetal bovine (56) serum contain an average of 150 ng of IGF-I per ml, as well as various amounts of growth hormone, we developed a defined serum-free system for differentiating HL-60 promyeloid cells into macrophages. Terminal macrophage maturation is accompanied by the sequential expression of differentiation markers (67), and induction of the ␣-subunit of the 2-integrin heterodimer (CD11b/ CD18; CR3) (16) is an early event in this process (9, 67). We therefore measured cell surface expression of CD11b over a 48-h time span, during which IGF-I enhanced macrophage differentiation in a time-dependent fashion (Fig. 1). Expression of CD11b was increased at all time points in cells treated with both IGF-I and vitamin D3 compared to those cells treated with either of these two agents separately. IGF-I acted early in the differentiation process induced by vitamin D3, as indicated by a significant increase in the percentage of CD11b-
6231
FIG. 1. Enhancement of CD11b expression by IGF-I occurs as early as 12 h following addition of vitamin D3. Cells were cultured in serum-free medium alone (Med) or with IGF-I (100 ng/ml), vitamin D3 (VD3; 1 M), or vitamin D3 plus IGF-I (VD3⫹IGF-I). During the 48-h incubation, the proportion of cells expressing CD11b was determined by flow cytometry. As early as 12 h, IGF-I increased (P ⬍ 0.05) the expression of CD11b in vitamin D3-treated cells compared to those incubated with vitamin D3 alone. Vitamin D3, in the absence of IGF-I, moderately increased (P ⬍ 0.05) CD11b expression at 24 h and later. IGF-I alone failed to increase CD11b expression above control levels. Plus signs indicate differences (P ⬍ 0.05) between medium or IGF-I alone and vitamin D3 alone, and asterisks indicate differences (P ⬍ 0.05) between vitamin D3 alone and vitamin D3 plus IGF-I. Values are expressed as means and standard errors of the mean (n ⫽ 3).
positive cells as early as 12 h (20% ⫾ 2%; P ⬍ 0.05) compared to cells in serum-free medium (4% ⫾ 1%) or treated with vitamin D3 alone (10% ⫾ 2%). This enhancement in CD11b expression caused by IGF-I in vitamin D3-treated cells persisted throughout the 48-h time span. Consistent with our previous results, IGF-I did not affect macrophage development in the absence of vitamin D3 (42). In separate experiments, we demonstrated that as little as 10 ng of IGF-I per ml plus vitamin D3 (1 M) caused a threefold increase in the proportion of CD11b-positive cells at 48 h (25% ⫾ 3% versus 74% ⫾ 4%; P ⬍ 0.01). Although receptors for both growth hormone and the closely related protein prolactin are members of the hematopoietic cytokine receptor superfamily (27, 64), neither of these hormones, at concentrations ranging from 1 to 1,000 ng/ml, increased the expression of CD11b (data not shown). These data establish that the IGF-I-induced increase in CD11b expression occurs with as little as 10 ng of IGF-I, can be detected as early as 12 h following the initiation of macrophage differentiation, and is not mimicked by either growth hormone or prolactin. IGF-I enhances the expression of CD11b, CD14, and intracellular macrophage esterase in differentiating promyeloid cells. To determine whether IGF-I increases the expression of differentiation markers other than CD11b, which is an early event in terminal macrophage differentiation (9), we measured the expression of two additional proteins, NAE and CD14, at 24 h (Fig. 2A) and 48 h (Fig. 2B). Cells were treated with vitamin D3 (1 M) in the absence or presence of IGF-I (100 ng/ml), and intracellular NAE activity was assessed by enumerating cells with intracellular black formazan deposits. Surface expression of CD14 was assessed by flow cytometry. HL-60 cells cultured in serum-free medium for either 24 or 48 h express very little CD11b, CD14, or NAE (⬍6% positive cells). However, these cells can be induced as early as 24 h to express all of these markers equally by treatment with a combination of vitamin D3 and IGF-I (42% ⫾ 5%, 42% ⫾ 4%, and 41% ⫾ 3% positive cells, respectively; P ⬍ 0.01). In the absence of IGF-I, however, vitamin D3 was significantly less effective in promoting the development of CD11b, CD14, and NAE (15% ⫾ 2%,
6232
LIU ET AL.
MOL. CELL. BIOL.
promotes cells to pass through the G1/S phase checkpoint and increases DNA synthesis in a number of cell types (2). We therefore tested whether this growth factor is also able to enhance the growth of cells cultured with the differentiating agent vitamin D3. HL-60 cells were incubated in medium, vitamin D3 (1 M), IGF-I (100 ng/ml), or both vitamin D3 and IGF-I for 5 days. The cells were harvested at the indicated times and enumerated with a cell counter (Coulter Instruments). As shown in a typical example (Fig. 3A), vitamin D3 did not affect the growth rate compared to that of control cells in medium only whereas IGF-I potently increased cell proliferation regardless of the presence of vitamin D3. A summary of three independent experiments showed that addition of IGF-I to vitamin D3-treated cells increased the cell number by 1.8 ⫾ 0.2- and 2.2 ⫾ 0.2-fold at 2 and 3 days, respectively (P ⬍ 0.05).
FIG. 2. IGF-I promotes vitamin D3-induced expression of CD11b, CD14, and NAE in vitamin D3-treated promyeloid cells in a time-dependent manner. HL-60 cells were incubated in serum-free medium alone (Med) or with IGF-I (100 ng/ml), vitamin D3 (VD3; 1 M), or vitamin D3 plus IGF-I (VD3⫹IGF-I). After 24 h (A) or 48 h (B), the proportion of cells expressing CD11b or CD14 surface antigens was determined by flow cytometry and NAE expression was determined by intracellular staining. IGF-I alone did not affect the expression of any of the differentiation markers, whereas vitamin D3 alone elicited a moderate increase in the expression of CD11b, CD14, and NAE at both 24 and 48 h. The combination of both IGF-I and vitamin D3 significantly (P ⬍ 0.01) increased the proportion of cells expressing CD11b, CD14, and NAE over that of cells incubated with only vitamin D3. Asterisks indicate that vitamin D3 increased the proportion of macrophage marker-expressing cells (P ⬍ 0.05) over that of cells cultured in medium or indicate that vitamin D3 plus IGF-I increased (P ⬍ 0.01) marker-expressing cells compared to vitamin D3 alone. Values are means and standard errors of the mean (n ⫽ 3).
9% ⫾ 2%, and 12% ⫾ 2%, respectively). IGF-I alone did not promote the expression of any macrophage differentiation marker (P ⬎ 0.10). The proportion of differentiated cells at 48 h was roughly double that detected at 24 h for all three differentiation markers, and identical trends in the expression of these macrophage differentiation markers were measured at this later time point (Fig. 2B). In serum-free medium at 48 h, addition of vitamin D3 increased (P ⬍ 0.05) the proportion of cells expressing CD11b, CD14, and NAE (25% ⫾ 4%, 23% ⫾ 2%, and 22% ⫾ 2%, respectively). Addition of IGF-I to vitamin D3-treated cells increased by approximately threefold the proportion of cells expressing these markers (76% ⫾ 5%, 78% ⫾ 3%, and 79% ⫾ 3%, respectively; P ⬍ 0.01). These results extend the findings in Fig. 1 by establishing that IGF-I also significantly enhances the expression of CD14 and NAE, both of which are expressed only by mature macrophages, as early as 24 h following induction of differentiation. Cell growth is enhanced by IGF-I in vitamin D3-treated HL-60 cells. IGF-I, acting as a classical progression factor,
FIG. 3. Cell cycle analysis during vitamin D3-induced macrophage differentiation. Cells (5 ⫻ 105) were washed three times with RPMI 1640 and cultured in serum-free defined medium for 24 h. Serum-starved cells were then incubated in medium alone (Med) or with vitamin D3 (VD3; 1 M), IGF-I (100 ng/ml), or both (VD3⫹IGF-I) for 5 days. Cell samples were collected at the indicated times and enumerated with a Coulter cell counter. Cell cycle analysis was performed by flow cytometry with PI. (A) Accumulation of HL-60 cells. While cells cultured in medium alone or vitamin D3 failed to increase in cell number, IGF-I increased cell growth, even in the presence of vitamin D3, on days 2 and 3. On days 4 and 5, cells treated with IGF-I continued to accumulate whereas cell numbers plateaued in cells treated with both IGF-I and vitamin D3. (B and C) Time courses of S (B) and G0/G1 (C) phase distribution. IGF-I increased the percentage of cells in S phase, regardless of the presence of vitamin D3, compared to that of cells cultured in medium or vitamin D3 alone on days 1 and 2. Accordingly, the percentage of cells in G0/G1 phase was reduced when cells were treated with IGF-I. On day 3 and thereafter, cells treated with both IGF-I and vitamin D3 withdrew from the cell cycle and accumulated in the G0/G1 phase whereas cells stimulated with IGF-I alone continued to replicate. The standard deviation of triplicate samples was less than 10% of the mean at each time point. Data are representative of three independent experiments (see the text for a summary of the results).
VOL. 19, 1999
FIG. 4. IGF-I inhibits p27KIP1 and enhances cyclin E expression in the early stages of macrophage development. Cells were cultured in serum-free medium for 24 h and then incubated in serum-free defined medium alone (Medium) or with vitamin D3 (VD3; 1 M), IGF-I (100 ng/ml), or vitamin D3 plus IGF-I (VD3⫹IGF-I) for 1, 2, or 3 days. Equal amounts of protein (50 g) in whole-cell lysates were resolved by electrophoresis on SDS–10% polyacrylamide gels and electrophoretically transferred onto a nitrocellulose membrane. The membrane was then probed with a MAb against p27KIP1 (A) or against the 45-kDa cyclin E protein (B). The 35-kDa protein bound to the cyclin E Ab has been previously reported and is likely to be a degraded product of cyclin E (23). Bands were visualized by enhanced chemiluminescence, and the molecular masses of protein standards are indicated on the left. A representative gel from one of three experiments is shown.
At these times, the growth of cells treated with IGF-I together with vitamin D3 was similar to that of cells treated with IGF-I alone (2.0 ⫾ 0.3- and 2.5 ⫾ 0.2-fold, respectively). After 3 days, however, cells treated with both IGF-I and vitamin D3 lost their growth potential (2.3 ⫾ 0.2-fold at 5 days) whereas cells cultured with only IGF-I continued to expand (3.1 ⫾ 0.2-fold; P ⬍ 0.05; n ⫽ 3). These data were further supported by DNA analysis which demonstrated that IGF-I promoted cells to advance into the S phase of the cell cycle, regardless of the presence of vitamin D3 (Fig. 3B). Indeed, IGF-I increased the proportion of vitamin D3-treated cells in S phase by 40% ⫾ 6% and 100% ⫾ 18% on days 1 and 2, respectively (n ⫽ 3; P ⬍ 0.05). Accordingly, the proportion of these cells in G0/G1 was reduced by 14% ⫾ 1% and 16% ⫾ 2% on days 1 and 2, respectively (Fig. 3C; P ⬍ 0.05; n ⫽ 3). At 3 days and thereafter, cells treated with IGF-I together with vitamin D3 failed to progress through S phase whereas cells treated with IGF-I alone continued to proliferate. These results establish that IGF-I increases cellular proliferation of vitamin D3-treated promyeloid cells by advancing them through the G1/S phase checkpoint and that this enhanced growth occurs in a time frame (⬍48 h) when the cells have initiated their differentiation program. IGF-I increases cyclin E and suppresses p27KIP1 expression. The recently discovered CDK inhibitors, including the G1phase p27KIP1 and p21CIP1, potently inhibit cellular proliferation and have been suggested to be required for differentiation in skeletal muscle cells (38). However, the role of these cell cycle inhibitors in differentiating hematopoietic cells has only recently begun to be elucidated (24, 41). p27KIP1 can associate with and inhibit a broad range of CDK-cyclin complexes of the G1/S transition phase (60). Indeed, the amount of this protein associated with cyclin E-CDK2 greatly increased in U937 human leukemic cells after treatment with phorbol myristate acetate for 24 to 72 h (3). We therefore tested the possibility that IGF-I promotes cell cycling by differentially regulating the expression of p27KIP1 and cyclin E, and we investigated whether this occurs early in the differentiation program of promyeloid cells. Cells were incubated in medium, vitamin D3
EARLY MACROPHAGE DIFFERENTIATION
6233
(1 M), IGF-I (100 ng/ml), or both vitamin D3 and IGF-I for 1, 2, or 3 days. As shown in Fig. 4A, p27KIP1 expression was detected in cells incubated in either serum-free medium alone or treated with vitamin D3 in serum-free medium, corresponding well to the finding that cells in both these treatments were arrested in the G0/G1 phase of the cell cycle on day 3 (Fig. 3). Addition of IGF-I almost completely suppressed the expression of p27KIP1, even in the presence of vitamin D3, at least through 72 h. However, as in the experiments by Liu et al. (41) with more mature U937 cells and by both Hengst and Reed (23) and Wang et al. (66) with promyeloid HL-60 cells, expression of p27KIP1 is significantly increased at later time points in the differentiation program. In contrast, expression of cyclin E was greatly increased by treatment with IGF-I (Fig. 4B). On day 1, cyclin E expression was increased by approximately eightfold in IGF-I-treated promyeloid cells compared to that in control cells in serum-free medium, and this enhancement was maintained throughout the 3-day experiment. Although the increase in cyclin E expression caused by IGF-I was clearly reduced by vitamin D3, cyclin E expression remained enhanced compared to the level in cells treated with vitamin D3 alone, amounting to roughly a 3-, 3-, and 20-fold increase on days 1, 2, and 3, respectively. These results establish that the IGF-I-promoted progression through the cell cycle in the initial stages of terminal macrophage differentiation occurs concomitantly with an inhibition of the G1-phase CKI p27KIP1 protein and an increase in cyclin E expression. IGF-I induces hyperphosphorylation of Rb tumor suppressor protein. Rb tumor suppressor protein is a putative target of the G1-phase cyclin-CDK complexes, including cyclin E-CDK2 and cyclin D1-CDK4 (4). The hypophosphorylated form of Rb binds to E2F transcription factors and inhibits their activity (4), whereas phosphorylation by CDKs dissociates Rb from E2F factors, which in turn activates the cell cycle machinery (36). More importantly, suppression of Rb expression in U937 promonocytic cells resulted in a diminished ability of these cells to differentiate response to vitamin D3 (7). Since IGF-I suppresses p27KIP1 and augments cyclin E expression, we wondered whether the phosphorylation of Rb tumor suppressor protein is also regulated by IGF-I in developing myeloid cells. As shown in Fig. 5, cells maintained in serum-free medium or treated with vitamin D3 alone did not express Rb protein on day 1, supporting previous findings that the Rb gene is downregulated at the onset of cell cycle arrest in HL-60 cells (72, 73). Although the Rb protein appeared on days 2 and 3, these
FIG. 5. Phosphorylation of Rb protein during IGF-I-enhanced cell differentiation. Cells were washed three times and cultured in defined medium overnight. Subsequently, these serum-deprived cells were incubated in defined medium alone (Medium) or with vitamin D3 (VD3; 1 M), IGF-I (100 ng/ml), or vitamin D3 plus IGF-I (VD3⫹IGF-I) for 1, 2, or 3 days. Equal amounts of protein (50 g) in whole-cell lysates were resolved by electrophoresis on SDS–10% polyacrylamide gels and transferred onto nitrocellulose. The blot was probed with a MAb against Rb tumor suppressor protein and visualized by enhanced chemiluminescence. Hypophosphorylated (pRb) and hyperphosphorylated (ppRb) forms of the 110-kDa Rb protein are indicated on the right, and the molecular mass of the Rb protein is shown on the left. Similar results were observed in three independent experiments.
6234
LIU ET AL.
MOL. CELL. BIOL.
FIG. 6. Cell cycle analysis of cells expressing the CD11b differentiation marker. Human HL-60 cells were washed three times and cultured in serum-free medium for 24 h. The cells (106) were then cultured in medium alone or with vitamin D3 (VD3; 1 M) or vitamin D3 plus IGF-I (VD3⫹IGF-I; 100 ng/ml) for 48 h. Cell samples were then subjected to simultaneous DNA content (bottom graph) and cell surface immunofluorescence (top graph) analysis by flow cytometry. Very few cells in medium alone expressed the CD11b marker (5% in G0/G1 and 2% in S⫹G2/M). Vitamin D3 alone induced a small increase in CD11b expression (7% in G0/G1 and 7% in S⫹G2/M). With the addition of IGF-I to vitamin D3-treated cells, a majority of cells (66%) expressed CD11b, with a significant number of cells (30%) being found in S⫹G2/M. These data are representative of three independent experiments (see the text for a summary of the results).
cells expressed Rb protein in a form that was approximately 90% hypophosphorylated (pRb) compared to the hyperphosphorylated form (ppRb). This finding is consistent with the concept that most of these cells are in the G0/G1 phase of cell cycle. However, IGF-I induced the expression of the Rb protein as well as its phosphorylation, with more than 90% of Rb in the hyperphosphorylated state on days 1 and 2, regardless of the presence of vitamin D3. Even on day 3, approximately 80% of the Rb protein was in the hyperphosphorylated form following treatment with IGF-I, which was similar to that in cells treated with both vitamin D3 and IGF-I (66% hyperphosphorylated). These data show that the IGF-I-promoted enhancement of vitamin D3-induced macrophage differentiation and promotion through the cell cycle is associated with phosphorylation of the Rb tumor suppressor protein. CD11b is expressed in differentiating macrophages at all cell cycle phases. Since IGF-I enhanced differentiation in vitamin D3-treated promyeloid cells as early as 12 to 24 h (Fig. 1), a time point when these cells are progressing through the cell cycle (Fig. 3), these data suggested that differentiating cells might continue to divide early in their differentiation program. These data are in accord with the results of Liu et al. (41), who showed that differentiation of U937 cells, as assessed by expression of mRNA for CD14, occurs long before their exit from the cell cycle. Similar results with HL-60 cells have been found at the protein level for both CD14 and nonspecific esterase by Wang et al. (66). We therefore determined whether CD11b protein is expressed throughout the cell cycle instead of only on cells that are arrested in the G0/G1 phase. As shown in a representative example in Fig. 6, cells incubated in medium alone and remaining in the G0/G1 phase were mostly negative for CD11b expression (70% of cells), with another 5% of the G0/G1-phase cells expressing this leukocyte surface antigen. Averaged over three independent experiments, 76% ⫾ 2% of control cells were in G0/G1 and only 5% ⫾ 2% expressed CD11b. Treatment with vitamin D3 alone induced CD11b ex-
pression in a small proportion of both G0/G1 (7%) and S-plusG2/M (7%) cells. When combined over three separate experiments, the proportion of CD11b-positive cells averaged 17% ⫾ 4%, which was greater (P ⬍ 0.05) than the 5% CD11bpositive cells cultured in medium alone. Addition of IGF-I to vitamin D3-treated cells greatly enhanced macrophage differentiation in both G0/G1 (36%) and S-plus-G2/M (30%) cells, for a total of 66% CD11b-positive cells (Fig. 6). This increase in CD11b-positive cells averaged 69% ⫾ 3% in three independent experiments, which was greater (P ⬍ 0.05) than in cells treated with vitamin D3 alone. More importantly, CD11b-positive cells appeared throughout each phase of cell cycle. For example, 35% ⫾ 2% (n ⫽ 3) of cells in G0/G1 expressed CD11b whereas 26% ⫾ 2% (n ⫽ 3) of G0/G1-phase cells did not. In the S-plus-G2/M phases, 32% ⫾ 2% (n ⫽ 3) of the cells expressed CD11b compared to only 7% ⫾ 1% that did not (n ⫽ 3; P ⬍ 0.05). These data demonstrate that at the onset of terminal macrophage differentiation, all cells are not arrested in G0/G1 and that early-differentiated promyeloid cells are capable of progressing through cell cycle. Expression of CD11b occurs concomitantly with expression of proliferation antigens. Since CD11b-expressing cells are able to progress through the cell cycle, we wondered whether IGF-I simultaneously promotes both cell growth and differentiation at the initial stages of terminal macrophage differentiation. We tested this possibility at the single-cell level by developing a novel flow-cytometric technique to simultaneously detect a cell surface differentiation marker and one of two definitive nuclear proliferation antigens, PCNA (DNA polymerase ␦ cofactor) and the thymidine analog BrdU. As shown in Fig. 7A, cells incubated in medium alone that were PCNA positive (48%) remained mostly CD11b negative (46%). Averaged over three independent experiments, PCNA-positive, CD11b-negative cells amounted to 45% ⫾ 3% of the total population. Vitamin D3 alone induced a small proportion of both PCNA-positive (7%) and -negative (6%) cells to undergo
VOL. 19, 1999
EARLY MACROPHAGE DIFFERENTIATION
6235
FIG. 7. Simultaneous expression of CD11b and nuclear proliferation antigens after treatment with vitamin D3 and IGF-I in HL-60 cells. Cells were cultured in medium alone or with vitamin D3 (VD3; 1 M) or vitamin D3 plus IGF-I (VD3⫹IGF-I; 100 ng/ml) for 24 h, followed by double staining for flow cytometry. (A) Labeling of surface CD11b differentiation antigen and intracellular PCNA. Only 5% of the cells expressed CD11b in medium alone (3% were PCNA negative, and 2% were PCNA positive). Vitamin D3 induced CD11b expression in 13% of the cells (6% were PCNA negative, and 7% were PCNA positive), while the addition of IGF-I further increased the expression of CD11b by nearly threefold (34%). More importantly, 74% of the CD11b-positive cells (25% of the entire cell population) also expressed PCNA. Results summarized over three independent experiments are given in the text. (B) Double staining of CD11b and the nuclear BrdU antigen. Only 5% of the control cells expressed CD11b (3% were BrdU negative, and 2% were BrdU positive). A small proportion of the cells (9%) underwent differentiation after treatment with vitamin D3 (4% were BrdU negative, and 5% were BrdU positive). However, the addition of IGF-I to vitamin D3-treated cells increased the proportion of CD11b-positive cells to 36%. Among the CD11b-expressing cells, 72% (26% of the entire cell population) were in a proliferative state, as indicated by BrdU incorporation. These data are representative of three independent experiments (see the text for a summary of the results).
macrophage differentiation by 24 h. Averaged over three independent experiments, 12% ⫾ 2% of the cells incubated with vitamin D3 expressed CD11b. Interestingly, half of these differentiating cells (6% ⫾ 1% of the entire cell population [n ⫽ 3]) also expressed the PCNA marker. More importantly, addition of IGF-I to the vitamin D3-treated cells substantially enhanced macrophage differentiation, as shown by the proportion of CD11b-positive cells increasing from 12% ⫾ 2% to 35% ⫾ 2% (n ⫽ 3; P ⬍ 0.05). Moreover, 70% of these CD11bpositive cells (24% ⫾ 2% of the entire cell population) coexpressed PCNA. Similar results were observed when the thymidine analog BrdU, which is incorporated into nuclear DNA during replication, was used as an indicator of cell proliferation (Fig. 7B). Indeed, 21% of the vitamin D3-treated cells, composed of 5% expressing CD11b and 16% remaining negative for CD11b, incorporated BrdU. Vitamin D3 alone induced CD11b expression in 9% of the cells (10% ⫾ 2% when averaged over three experiments). Addition of IGF-I to vitamin D3-treated promyeloid cells doubled the proportion of the BrdU-labeled cells from 21 to 40% (38% ⫾ 2% in three experiments [P ⬍ 0.05]) and increased the CD11b-positive cells from 9 to 36% (37% ⫾ 3% in three experiments [P ⬍ 0.05]). Among the CD11b-expressing cells, 73% (27% ⫾ 2% of the entire cell population) were also positive for BrdU, indicating active DNA synthesis in these early-differentiated cells. This result corresponds closely with the 70% CD11b-positive cells that express PCNA. Collectively, these data provide definitive evidence that individual cells undergoing IGF-I-enhanced macrophage differentiation simulta-
neously express the CD11b differentiation marker concomitantly with both PCNA and a marker of DNA replication, BrdU. CD11b-positive cells incorporate [3H]thymidine early in the terminal differentiation program. Flow cytometry was used to separate HL-60 cells that had been incubated for 24 h with vitamin D3 and IGF-I into populations that expressed CD11b and those that did not. As expected from our earlier results (40), incubation of CD11b-negative cells with IGF-I in serumfree medium for an additional 24 h significantly (P ⬍ 0.01) increased their incorporation of [3H]thymidine by approximately fivefold (Fig. 8). CD11b-positive cells cultured in medium alone incorporated an amount of [3H]thymidine that was not significantly different from the amount incorporated by CD11b-negative cells cultured in medium alone. More importantly, even in CD11b-positive cells, IGF-I increased the incorporation of [3H]thymidine by 4-fold (P ⬍ 0.05) (Fig. 8). These data are consistent with the idea that very early during development of macrophages, cell differentiation and cell proliferation can occur simultaneously. DISCUSSION The recently discovered CKIs, including the G1-phase p27KIP1 and p21CIP1, not only inhibit cell cycle progression but have been suggested to be required for the differentiation of a variety of cells (43). Unlike skeletal muscle cells, where differentiation programs are inversely related to cell cycle progression (21, 62), we establish here that proliferation and the early
6236
LIU ET AL.
FIG. 8. Enriched populations of CD11b-positive and -negative cells incorporate [3H]thymidine. Following a 24-h treatment with vitamin D3 (1 M) and IGF-I (100 ng/ml), the cells were sorted by flow cytometry on the basis of expression of CD11b, and both CD11b-negative (⬎98% negative) and -positive (⬎98% positive) cells were cultured for another 24 h in serum-free medium alone (Med) or in medium plus IGF-I. Incorporation of [3H]thymidine was increased by IGF-I in both CD11b-negative (P ⬍ 0.01; n ⫽ 4) and CD11bpositive (P ⬍ 0.05; n ⫽ 5) cells. Asterisks indicate statistical significance, and values are means and standard errors of the mean.
events associated with terminal differentiation of promyeloid cells into macrophages are not mutually exclusive processes. Indeed, IGF-I clearly promotes cell differentiation toward the macrophage lineage in vitamin D3-treated promyeloid cells, as indicated by increasing the expression of the ␣-integrin adhesion molecule CD11b (Fig. 1), the CD14 lipopolysaccharide receptor (Fig. 2), and the mature macrophage-specific esterase enzyme, NAE (Fig. 2). This increase in expression of differentiation markers, which occurs within 24 h (Fig. 2A), is accompanied by an increase in the percentage of cells in the S phase of the cell cycle, a reduction in the percentage of cells in G0/G1 (Fig. 3B and C), and a subsequent doubling of the number of cells (Fig. 3A). More importantly, expression of the CKI p27KIP1 is not induced in the presence of IGF-I as promyeloid cells begin to terminally differentiate towards the macrophage lineage, as determined by acquisition of the differentiation antigen CD11b (Fig. 1 and 4A). Accordingly, IGF-I increases cyclin E expression (Fig. 4B) and maintains the Rb tumor suppressor protein in a hyperphosphorylated state in differentiating macrophages (Fig. 5). Simultaneous analysis of DNA content and cell surface CD11b established that IGF-I enhances macrophage differentiation in all phases of the cell cycle, instead of only those arrested in G0/G1 (Fig. 6). A novel double-labeling flow-cytometric analysis at the single-cell level revealed simultaneous expression of CD11b and two different nuclear proliferation markers, PCNA and BrdU, in nearly 75% of myeloid precursor cells as they initiated terminal macrophage differentiation (Fig. 7). Finally, the finding that sorted populations of CD11b-positive cells incorporate significant amounts of [3H]thymidine is consistent with the idea that these cells synthesize DNA (Fig. 8). These data provide clear evidence that complete cell cycle arrest is not required at the onset of myeloid-cell differentiation and establish that the initial stages of terminal macrophage development are characterized by (i) expression of three different markers of macrophage differentiation, (ii) proliferation, (iii) expression of low levels of p27KIP1, and (iv) maintenance of Rb in a state of hyperphosphorylation. Collectively, the results indicate that p27KIP1 and hypophosphorylated Rb are not necessary to initiate macrophage differentiation. These data are consistent with recent results (23) that established a negative relationship between the expression of p27KIP1 and cyclin E protein during promyeloid proliferation,
MOL. CELL. BIOL.
but they do not support a role for this CKI in the initial events leading from promyeloid cells to macrophages. For example, IL-2 promotes the G1/S transition by inhibiting p27KIP1 in T lymphocytes (52). Similarly, induction of p27KIP1 leads to G1 cell cycle arrest in cyclic AMP-treated macrophages cultured with CSF-1 (30). Although these data suggest an important role for p27KIP1 in inhibiting cell cycle entry, they do not directly address the possibility that p27KIP1 is involved in early maturation events of macrophages. New findings that ectopic overexpression of p27KIP1 leads to monocyte differentiation in U937 cells in the absence of a priming differentiation signal (41) and that p27KIP1 is expressed in differentiating HL-60 cells (23) also support a role for p27KIP1 in monocyte differentiation. The fact that the expression of p27KIP1 is increased only late in the differentiation program of U937 cells (⬎48 h) strongly supports a role for this protein in the final stages of macrophage differentiation (41, 66). These data showing a direct role of p27KIP1 in leading to cell cycle arrest in both T cells and macrophages is consistent with the notion that this CKI is somehow involved in promoting the differentiation of hematopoietic cells. In contrast, our experiments examined the potential contribution of p27KIP1 at much earlier stages of myeloid-cell differentiation in a progenitor myeloid cell rather than the more differentiated monocytic U937 cells. Our rationale for this objective was that although vitamin D3 has been reported to induce the expression of p27KIP1, this protein is not detectable at significant levels until 4 days after initiation of cellular differentiation in HL-60 cells (23, 66). Recent studies on expression of the CKI p21 in normal myeloid-cell differentiation found a similar result for human primary CD34⫹ progenitors. Indeed, expression of p21 is not apparent until 9 to 12 days after initiation of differentiation process whereas expression of a characteristic leukocyte maturation marker, the granulocyte colony-stimulating factor receptor, starts on days 3 to 6 (63). Our data extend these results by showing that in the presence of IGF-I, p27KIP1 is not induced during the first 48 h of macrophage differentiation (Fig. 4). However, during this initial 24- to 48-h period, the cells had clearly initiated their terminal differentiation program, as assessed by expression of CD11b, CD14, and NAE (Fig. 1 and 2). However, in more differentiated human U937 cells, expression of p21CIP1 at both the mRNA and protein levels is increased by 4 h following addition of vitamin D3 (41). It is therefore possible that this CDK inhibitor, rather than p27KIP1, regulates the development of more mature macrophages. Recent evidence suggests that induction of CKIs such as p21CIP1 and p27KIP1 may be more important for maintenance of terminally differentiated cells in the G0 phase of the cell cycle than for the initiation of differentiation, and our results are not inconsistent with this hypothesis. Expression of the muscle-specific protein MyoD transactivates the p21CIP1 promoter and induces the expression of its transcripts (21, 46). Moreover, induction of p21CIP1 is correlated with terminal cell cycle arrest in several fully differentiated cell lineages, including muscle, skin, cartilage, and nasal epithelial cells, in a p53independent fashion (54). Accordingly, induction of p21CIP1 occurs as a consequence of a terminal differentiation event in normal epithelial cells, and one major consequence of this CDK inhibition is an increase in the growth-inhibitory activity of hypophosphorylated Rb protein. However, new data have now demonstrated an inhibitory role of p21CIP1 in the differentiation of primary mouse keratinocytes, suggesting a distinct function for this CKI in cell development that can be separated from its effects on cell cycle control (15). Indeed, cell cycle regulators have been shown to expand their role in cell differentiation. For example, Rb protein appears to direct the dif-
VOL. 19, 1999
ferentiation program in myoblasts (53), adipocytes (11), and hematopoietic cells (7) because these cells fail to differentiate in the absence of Rb protein. Thus, it is possible that the CKI p27KIP1, which is induced long after the majority of promyeloid cells have initiated their differentiation program (Fig. 1 and 4A) (23), plays a role in maintaining the postmitotic state of differentiated cells by preventing them from reentering the cell cycle. Differentiation antigens expressed later in the development of hematopoietic cells may function to induce the expression of p27KIP1 and therefore maintain mature, terminally differentiated macrophages in the G0/G1 phase of the cell cycle. In contrast to previous reports (5, 9), we used a defined culture system, which excluded FBS, to differentiate vitamin D3-treated HL-60 promyeloid cells into macrophages. Although IGF-I alone failed to induce terminal macrophage differentiation, addition of IGF-I to vitamin D3-treated cells potently increased the expression of three different mature macrophage antigens, including CD11b. This differentiation antigen is induced by a novel nuclear transcription factor, MS-2, during monocyte differentiation (16). Similarly, vitamin D3 induces macrophage differentiation and, after several days, causes cell cycle arrest. This late exit from the cell cycle was recently shown to be associated with a reduction in the expression of cyclin E (23), a decline in the expression of DNA polymerase ␦ cofactor/PCNA (26) and hypophosphorylation of the Rb tumor suppressor protein (73). In contrast, we investigated some of the earliest events associated with macrophage differentiation and separated the effects of vitamin D3 from those of FBS, which contains abundant amounts of IGF-I. This 70-amino-acid peptide is well known to be a mitogenic proliferation factor for many types of hematopoietic cells (35, 69). IGF-I is well known to exert its proliferative actions as a G1 progression factor very early in the cell cycle near the G0-G1 interface (2). Since by 24 h most of the CD11b-positive cells had already passed completely through their S phase, it is likely that at least some of the CD11b-positive cells reentered the cell cycle and incorporated [3H]thymidine in their second round of DNA synthesis. Our results also confirm that IGF-I increases PCNA expression in HL-60 cells (55). Indeed, an intact IGF-I receptor is required for simian virus 40 T-antigeninduced transformation in 3T3-like cells (6) and for the expression of PCNA in platelet-derived growth factor-induced fibroblasts (47). Although IGF-I is likely to act largely by maintaining the hyperphosphorylated state of Rb tumor suppressor protein (57) and by increasing cyclin E expression (70), the regulation of Rb phosphorylation and cyclin E by IGF-I in hematopoietic cells has remained unknown. Here we clearly establish that IGF-I increases the expression of cyclin E and the phosphorylation of Rb tumor suppressor protein in promyeloid cells. More importantly, these events occur early (⬍24 h) and at the same time as these cells are differentiating from promyeloid cells into the macrophage lineage. The newly identified c-Mpl ligand thrombopoietin promotes megakaryocyte progenitor proliferation at the same time as expression of the differentiation marker gpIb, a component of the platelet von Willebrand factor receptor (31). Simultaneous expression of proliferation and differentiation markers in these cells suggests a critical role of c-Mpl ligand in both events and supports our results that hematopoietic cells can simultaneously differentiate and continue to divide. Similarly, c-kit ligand has now been reported to stimulate the proliferation of human megakaryocytes by increasing the expression of cyclin A and the ratio of hyperphosphorylated to hypophosphorylated Rb protein (61). Incubation with c-kit ligand simultaneously leads to increased expression of IIb/IIIa platelet-related glycoprotein (gpIIb, IIIa), indicating enhanced megakaryocytic differentia-
EARLY MACROPHAGE DIFFERENTIATION
6237
tion. Likewise, cytokine-regulated proliferation of human hematopoietic stem cells occurs concomitantly with induction of the 1-integrin’s very late antigens 4 and 5 (VLA-4 and VLA-5 [39]), which suggests that maturation of these pluripotent blood stem cells is functionally linked to cell growth. In addition, during lymphocyte differentiation, T-cell-dependent B-cell activation induces isotype switching, and this event has recently been shown to be positively related to the division cycle number (24). The correlation between the percentage of IgG1-positive B cells and division number was independent of time after stimulation, arguing for a requirement of cell division during B-cell maturation. Interestingly, rapamycin, which abrogates the activation of p34cdc2 and p33CDK2 necessary for entry into S phase, also inhibits the differentiation of normal B lymphocytes into Ig-secreting plasma cells (1). Taken together, these reports support our findings with myeloid progenitors that cell growth and early differentiation events occur simultaneously. Collectively, these data clearly show that IGF-I promotes macrophage differentiation and that this process occurs concomitantly with progression through the cell cycle. The finding that a single cell can express both differentiation and proliferation antigens early in the development of myeloid cells unequivocally argues that these two cellular processes are not exclusive events. Indeed, the IGF-I-induced elevation of cyclin E expression, hyperphosphorylation of Rb, and inhibition of CKI p27KIP1 show that permanent withdrawal from the cell cycle does not necessarily precede cell differentiation. Indeed, nearly 75% of early-differentiated promyeloid cells express PCNA and incorporate BrdU, and purified populations of CD11b-positive cells incorporate substantial amounts of [3H]thymidine. Thus, both p27KIP1 and hypophosphorylation of Rb are not induced at the initiation of the terminal differentiation program in promyeloid cells. Since the cells acquire p27KIP1 and Rb hypophosphorylation later in their differentiation program, the expression of these proteins is likely to be of clinical relevance because disruption of either Rb (25) or CKI (22) function results in terminally differentiated cells reentering the cell cycle during carcinogenesis. ACKNOWLEDGMENTS We gratefully acknowledge the instrumentation and expert technical assistance provided by Gary Durack in the University of Illinois Urbana-Champaign Biotechnology Center Flow Cytometry Facility. This research was supported by grants to K.W.K from the National Institutes of Health (AG-06246, DK-49311, and MH-51569) and the Pioneering Research Project in Biotechnology financed by the Japanese Ministry of Agriculture, Forestry and Fisheries and to G.G.F. from the National Institutes of Health (CA 61931). The UIUC Biotechnology Center Flow Cytometry Facility was supported by NIH grant PHS 1S10 RR02277. REFERENCES 1. Aagaard-Tillery, K. M., and D. F. Jelinek. 1994. Inhibition of human B lymphocyte cell cycle progression and differentiation by rapamycin. Cell. Immunol. 156:493–507. 2. Aaronson, S. A. 1991. Growth factors and cancer. Science 254:1146–1152. 3. Asiedu, C., J. Biggs, and A. S. Kraft. 1997. Complex regulation of CDK2 during phorbol ester-induced hematopoietic differentiation. Blood 90:3430– 3437. 4. Bartek, J., J. Bartkova, and J. Lukas. 1996. The retinoblastoma protein pathway and the restriction point. Curr. Opin. Cell Biol. 8:805–814. 5. Barton, A. E., C. M. Bunce, R. A. Stockley, P. Harrison, and G. Brown. 1994. 1␣,25-Dihydroxyvitamin D3 promotes monocytopoiesis and suppresses granulocytopoiesis in cultures of normal human myeloid blast cells. J. Leukoc. Biol. 56:124–132. 6. Baserga, R. 1994. Oncogenes and the strategy of growth factors. Cell 79: 927–930. 7. Bergh, G., M. Ehinger, T. Olofsson, B. Baldetorp, E. Johnsson, H. Brycke, G. Lindgren, I. Olsson, and U. Gullberg. 1997. Altered expression of the retinoblastoma tumor-suppressor gene in leukemic cell lines inhibits induction
6238
LIU ET AL.
of differentiation but not G1-accumulation. Blood 89:2938–2950. 8. Blalock, J. E. 1994. The syntax of immune neuroendocrine communication. Immunol. Today 15:504–511. 9. Brackman, D., F. Lund-Johansen, and D. Aarskog. 1995. Expression of leukocyte differentiation antigens during the differentiation of HL-60 cells induced by 1,25-dihydroxyvitamin D3: comparison with the maturation of normal monocytic and granulocytic bone marrow cells. J. Leukoc. Biol. 58:547–555. 10. Brugarolas, J., C. Chandrasekaran, J. I. Gordon, D. Beach, T. Jacks, and G. J. Hannon. 1995. Radiation-induced cell cycle arrest compromised by p21 deficiency. Nature 377:552–556. 11. Chen, P. L., D. J. Riley, Y. Chen, and W. H. Lee. 1996. Retinoblastoma protein positively regulates terminal adipocyte differentiation through direct interaction with C/EBPs. Genes Dev. 10:2794–2804. 12. Clark, R., J. Strasser, S. McCabe, K. Robbins, and P. Jardieu. 1993. Insulinlike growth factor-I stimulation of lymphopoiesis. J. Clin. Investig. 92:540– 548. 13. Collins, S. J. 1987. The HL-60 promyelocytic leukemia cell line: proliferation, differentiation, and cellular oncogene expression. Blood 70:1233–1244. 14. Deng, C., P. Zhang, J. W. Harper, S. J. Elledge, and P. Leder. 1995. Mice lacking p21CIP1/WAF1 undergo normal development, but are defective in G1 checkpoint control. Cell 82:675–684. 15. Di Cunto, F., G. Topley, E. Calautti, J. Hsiao, L. Ong, P. K. Seth, G. P. Dotto. 1998. Inhibitory function of p21Cip/WAF1 in differentiation of primary mouse keratinocytes independent of cell cycle control. Science 280:1069– 1071. 16. Farokhzad, O. C., C. M. Shelley, and M. A. Arnaout. 1996. Induction of the CD11b gene during activation of the monocytic cell line U937 requires a novel nuclear factor MS-2. J. Immunol. 157:5597–5605. 17. Fero, M. L., M. Rivkin, M. Tasch, P. Porter, C. E. Carow, E. Firpo, K. Polyak, L. H. Tsai, V. Broudy, R. M. Perlmutter, K. Kaushansky, and J. M. Roberts. 1996. A syndrome of multiorgan hyperplasia with features of gigantism, tumorigenesis, and female sterility in p27kip1-deficient mice. Cell 85:733–744. 18. Field, S. J., F.-Y. Tsia, F. Kuo, A. M. Zubiaga, W. G. Kaelin, D. M. Livingstone, S. H. Orkin, and M. E. Greenberg. 1996. E2F-1 functions in mice to promote apoptosis and suppress proliferation. Cell 85:549–561. 19. Gibson, L. F., D. Piktel, and K. S. Landreth. 1993. Insulin-like growth factor-I potentiates expansion of interleukin-7-dependent pro-B cells. Blood 82:3005–3011. 20. Gu, W., J. W. Schneider, G. Condorelli, S. Kaushal, V. Mahdavi, and B. Nadal-Ginard. 1993. Interaction of myogenic factors and the retinoblastoma protein mediates muscle cell commitment and differentiation. Cell 72:309– 324. 21. Halevy, O., B. G. Novitch, D. B. Spicer, S. X. Skapek, J. Rhee, G. J. Hannon, D. Beach, and A. B. Lassar. 1995. Correlation of terminal cell cycle arrest of skeletal muscle with induction of p21 by MyoD. Science 267:1081–1021. 22. Hall, M., and G. Peters. 1996. Genetic alterations of cyclins, cyclin-dependent kinases, and Cdk inhibitors in human cancer. Adv. Cancer Res. 68:67– 108. 23. Hengst, L., and S. I. Reed. 1996. Translational control of p27Kip1 accumulation during the cell cycle. Science 271:1861–1864. 24. Hodgkin, P. D., J. H. Lee, and A. B. Lyons. 1996. B cell differentiation and isotype switching is related to division cycle number. J. Exp. Med. 184:277– 281. 25. Horowitz, J. M., S.-H. Park, E. Bogenmann, J.-C. Cheng, D. W. Yandell, F. J. Kaye, J. D. Minna, T. P. Dryja, and R. A. Weinberg. 1990. Frequent inactivation of the retinoblastoma antioncogene is restricted to a subset of human tumor cells. Proc. Natl. Acad. Sci. USA 87:2775–2779. 26. Hsieh, T. Y., C. Y. Ng, C. Mallouh, H. Tazaki, and J. M. Wu. 1996. Regulation of growth, PSA/PAP and androgen receptor expression by 1 alpha, 25-dihydroxyvitamin D3 in the androgen-dependent LNCaP cells. Biochem. Biophys. Res. Commun. 223:141–146. 27. Ihle, J. N. 1995. Cytokine receptor signalling. Nature 377:591–594. 28. Jardieu, P., R. Clark, D. Mortensen, and K. Dorshkind. 1994. In vivo administration of insulin-like growth factor-I stimulates primary B lymphopoiesis and enhances lymphocyte recovery after bone marrow transplantation. J. Immunol. 152:4320–4327. 29. Juan, G., X. Li, and Z. Darzynkiewicz. 1998. Phosphorylation of retinoblastoma protein assayed in individual HL-60 cells during their proliferation and differentiation. Exp. Cell Res. 244:83–92. 30. Kato, J.-Y., M. Matsuoka, K. Polyak, J. Massague, and C. J. Sherr. 1994. Cyclic AMP-induced G1 phase arrest mediated by an inhibitor (p27kip1) of cyclin-dependent kinase 4 activation. Cell 79:487–496. 31. Kaushansky, K., S. Lok, R. D. Holly, V. Broudy, N. Lin, M. C. Balley, J. W. Forstrom, M. M. Buddle, P. J. Oort, F. S. Hagen, G. J. Roth, T. Papayannopoulou, and D. C. Foster. 1994. Promotion of megakaryocyte progenitor expression and differentiation by the c-Mpl ligand thrombopoietin. Nature 369:568–571. 32. Kimata, H., and M. Fujimoto. 1994. Growth hormone and insulin-like growth factor-I induce immunoglobin (Ig) E and IgG4 production by human B cells. J. Exp. Med. 180:727–732.
MOL. CELL. BIOL. 33. Kimata, H., and A. Yoshida. 1994. Differential effect of growth hormone and insulin-like growth factor-I, insulin-like growth factor-II, and insulin on Ig production and growth in human plasma cells. Blood 83:1569–1574. 34. Kiyokawa, H., R. D. Kineman, K. O. Manova-Todorova, V. C. Soares, E. S. Hoffman, M. Ono, D. Khanam, A. C. Hayday, L. A. Frohman, and A. Koff. 1996. Enhanced growth of mice lacking the cyclin-dependent kinase inhibitor function of p27kip1. Cell 85:721–732. 35. Kooijman, R., E. L. Hooghe-Peters, and R. Hooghe. 1996. Prolactin, growth hormone and insulin-like growth factor-I in the immune system. Adv. Immunol. 63:377–454. 36. Lam, E. W., and N. B. La Thanggue. 1996. DP and E2F proteins: coordinating transcription with cell cycle progression. Curr. Opin. Cell Biol. 6:859– 866. 37. Landreth, K. S., R. Narayanan, and K. Dorshkind. 1992. Insulin-like growth factor-I regulates pro-B cell differentiation. Blood 80:1207–1212. 38. Lassar, A. B., S. X. Skapek, and B. Novitch. 1994. Regulation mechanisms that coordinate skeletal muscle differentiation and cell cycle withdrawal. Curr. Opin. Cell Biol. 6:788–794. 39. Levesque, J. P., D. N. Haylock, and P. J. Simmons. 1996. Cytokine regulation of proliferation and cell adhesion are correlated events in human CD34⫹ hemapoietic progenitors. Blood 88:1168–1176. 40. Li, Y. M., D. H. Schacher, Q. Liu, S. Arkins, N. Rebeiz, R. H. McCusker, R. Dantzer, and K. W. Kelley. 1997. Regulation of myeloid growth and differentiation by the insulin-like growth factor-I receptor. Endocrinology 138: 362–368. 41. Liu, M., M. H. Lee, M. Cohen, M. Bommakanti, and L. P. Freedman. 1996. Transcriptional activation of the Cdk inhibitor p21 by vitamin D3 leads to the induced differentiation of the myelomonocytic cell line U937. Genes Dev. 10:142–153. 42. Liu, Q., W. Ning, R. Dantzer, G. G. Freund, and K. W. Kelley. 1998. Activation of protein kinase C- and phosphatidylinositol 3⬘-kinase and promotion of macrophage differentiation by insulin-like growth factor-I. J. Immunol. 160:1393–1401. 43. Marx, J. 1995. Cell biology. Cell cycle inhibitors may help brake growth as cells develop. Science 267:963–964. 44. Minshall, C., S. Arkins, G. G. Freund, and K. W. Kelley. 1996. Requirement for phosphatidylinositol 3⬘-kinase to protect hemopoietic progenitors against apoptosis depends upon the extracellular survival factor. J. Immunol. 156: 939–947. 45. Minshall, C., S. Arkins, J. Straza, J. Conners, R. Dantzer, G. G. Freund, and K. W. Kelley. 1997. IL-4 and insulin-like growth factor-I inhibit the decline in Bcl-2 and promote the survival of IL-3-deprived myeloid progenitors. J. Immunol. 159:1225–1232. 46. Missero, C., and G. P. Dotto. 1996. p21 WAF1/CIP1 and terminal differentiation control of normal epithelia. Mol. Cell. Differ. 4:1–16. 47. Miura, M., S. W. Li, G. Dumenil, and R. Baserga. 1994. Platelet-derived growth factor-induced expression of messenger RNA for the proliferating cell nuclear antigen requires a functional receptor for the insulin-like growth factor I. Cancer Res. 54:2472–2477. 48. Murphy, W. J., S. K. Durum, M. R. Anver, and D. L. Longo. 1992. Immunologic and hematologic effects of neuroendocrine hormones: studies on DW/J dwarf mice. J. Immunol. 148:3799–3805. 49. Murphy, W. J., S. K. Durum, and D. L. Longo. 1992. Role of neuroendocrine hormones in murine T cell development. Growth hormone exerts thymopoietic effects in vivo. J. Immunol. 149:3851–3857. 50. Murphy, W. J., G. Tsarfaty, and D. L. Longo. 1992. Growth hormone exerts hematopoietic growth-promoting effects in vivo and partially counteracts the myelosuppressive effects of azidothymidine. Blood 80:1443–1447. 51. Nakayama, K., N. Ishida, M. Shirane, A. Inomata, T. Inoue, N. Shishido, I. Horii, D. Y. Loh, and K.-I. Nakayama. 1996. Mice lacking p27kip1 display increased body size, multiple organ hyperplasia, retinal displasia, and pituitary tumors. Cell 85:707–720. 52. Nourse, J., E. Firpo, W. M. Flanagan, S. Coats, K. Polyak, M. H. Lee, J. Massague, G. R. Crabtree, and J. M. Roberts. 1994. Interleukin-2-mediated elimination of the p27kip1 cyclin-dependent kinase inhibitor prevented by rapamycin. Nature 372:570–573. 53. Novitch, B. G., G. J. Mulligan, T. Jacks, A. B. Lassar. 1996. Skeletal muscle cells lacking the retinoblastoma protein display defects in muscle gene expression and accumulate in S and G2 phases of the cell cycle. J. Cell Biol. 135:441–456. 54. Parker, S. B., G. Eichele, P. Zhang, A. Rawls, T. Sands, A. Bradley, E. N. Olson, J. W. Harper, and S. J. Elledge. 1995. p53-independent expression of p21cip1 in muscle and other terminally differentiating cells. Science 267: 1024–1027. 55. Reiss, K., P. Porcu, C. Sell, Z. Pietrzkowski, and R. Baserga. 1992. The IGF-I receptor is required for the proliferation of hematopoietic cells. Oncogene 7:2243–2248. 56. Roldan, A., E. Charreau, and R. Schillaci. 1989. Insulin-like growth factor-I increases the mitogenic response of human peripheral blood lymphocytes to phytohemagglutinin. Immunol. Lett. 20:5–8. 57. Rosenthal, S. M., and Z. Q. Cheng. 1995. Opposing early and late effects of insulin-like growth factor I on differentiation and the cell cycle regulatory
VOL. 19, 1999
58. 59. 60. 61.
62. 63. 64. 65. 66.
retinoblastoma protein in skeletal myoblasts. Proc. Natl. Acad. Sci. USA 92:10307–10311. SAS Institute, Inc. 1989. SAS/STAT user’s guide, version 6, vol. 2, 4th ed. SAS Institute, Inc., Cary, N.C. Serrano, M., H.-W. Lee, L. Chin, C. Cordon-Cardo, D. Beach, and R. A. Depinho. 1996. Role of the INK4a locus in tumor suppression and cell mortality. Cell 85:27–37. Sherr, C. J. 1996. Cancer cell cycles. Science 274:1672–1677. Shimizu, E., Y. Takahashi, A. Shinohara, Y. Yamamoto, S. Yoshida, A. Yamamoto, S. Sone, Y. Takaue, and T. Ogura. 1996. Stem-cell factor regulates the expression of cyclin A and retinoblastoma gene products in the growth and differentiation pathway of human megakaryocytic cells. J. Cancer Res. Clin. Oncol. 122:445–452. Skapek, S. X., J. Rhee, D. B. Spicer, and A. B. Lassar. 1995. Inhibition of myogenic differentiation in proliferating myoblasts by cyclin D-1-dependent kinase. Science 267:1022–1024. Steinman, R. A., J. Huang, B. Yaroslavskiy, J. P. Goff, E. D. Ball, and A. Nguyen. 1998. Regulation of p21(WAF1) expression during normal myeloid differentiation. Blood 91:4531–4542. Taniguchi, T. 1995. Cytokine signaling through nonreceptor protein tyrosine kinases. Science 268:251–255. Thissen, J.-P., J.-M. Ketelslegers, and L. E. Underwood. 1994. Nutritional regulation of insulin-like growth factors. Endocr. Rev. 15:80–94. Wang, Q. M., J. B. Jones, and G. P. Studzinski. 1996. Cyclin-dependent
EARLY MACROPHAGE DIFFERENTIATION
67.
68. 69. 70.
71. 72. 73.
6239
kinase inhibitor p27 as a mediator of the G1-S phase block induced by 1,25-dihydroxyvitamin D3 in HL60 cells. Cancer Res. 56:264–267. Watanabe, M., Y. Shimamoto, S. Yoshida, K. Suga, Y. Mizuguchi, O. Kohashi, and M. Yamaguchi. 1993. Intracellular multiplication of Legionella pneumophila in HL-60 cells differentiated by 1,25-dihydroxyvitamin D3 and the effect of interferon ␥. J. Leukoc. Biol. 54:40–46. Weigent, D. A., and J. E. Blalock. 1991. The production of growth hormone by subpopulations of rat mononuclear leukocytes. Cell. Immunol. 135:55–65. Weigent, D. A., and J. E. Blalock. 1995. Associations between the neuroendocrine and immune systems. J. Leukoc. Biol. 58:137–150. Yamamoto, K., A. Hirai, T. Ban, J. Saito, K. Tahara, T. Terano, Y. Tamura, Y. Saito, and M. Kitagawa. 1996. Thyrotropin induces G1 cyclin expression and accelerates G1 phase after insulin-like growth factor I stimulation in FRTL-5 cells. Endocrinology 137:2036–2042. Yamasaki, L., T. Jacks, R. Bronson, E. Goillot, E. Harlow, and N. J. Dyson. 1996. Tumor induction and tissue atrophy in mice lacking E2F-1. Cell 85: 537–548. Yen, A., M. Williams, J. D. Platko, C. Der, and M. Hisaka. 1994. Expression of activated RAF accelerates cell differentiation and RB protein downregulation but not hypophosphorylation. Eur. J. Cell Biol. 65:103–113. Yen, A., and S. Soong. 1996. Retinoic acid-induced RB hypophosphorylation enhanced by CGP 52411, an EGF family tyrosine kinase receptor inhibitor. Eur. J. Cell Biol. 69:327–334.
