Department of Hematology, Research Institute, International Medical Center of Japan, Tokyo
Correspondence: Akira Yuo, Department of Hematology, Research Institute, International Medical Center of Japan, 1-21-1, Toyama, Shinjuku-ku, Tokyo 162-8655, Japan. E-mail: yuoakira{at}ri.imcj.go.jp
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: CREB • C/EBP • PU.1 • adhesion • oxidative burst
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cells acquire multiple phenotypes during differentiation: Some of them are essential for differentiation, and others seem to have no physiological significance. For example, during their maturation into granulocytes or monocytes, myeloblasts acquire respiratory burst activity essential for antibacterial reactions [3 ]. Conversely, differentiation-associated, morphological changes such as nuclear convolution do not seem to have any specific, physiological significance. It is of interest to clarify whether these two differentiation phenotypes with distinct biological significances are regulated by the same controlling system or individually regulated by distinct mechanisms.
Among the various inducers of differentiation, retinoid hormones such as all-trans retinoic acid (ATRA) are unique in that they can induce differentiation in a wide variety of cells [4 ], although the precise molecular events involved in retinoid hormone-mediated differentiation are largely unknown. Thus, clarification of the downstream molecular events involved in retinoic acid receptor (RAR)-mediated differentiation would be of use in understanding the common features of the differentiation process.
In addition to retinoid hormones, cyclic adenosine monophosphate (cAMP) analogs are potent inducers of differentiation in many cells including myelocytes [5 ]. In fact, cAMP-response element (CRE) is involved in the promoter activities of myeloid differentiation-associated genes such as CCAAT/enhancer-binding protein ß (C/EBPß) [6 ], interleukin-1ß [7 ], and prostaglandin-endoperoxide synthetase 2 [8 ]. It is interesting that CRE is also involved in cell proliferation [9 ], a completely distinct phenomenon to differentiation. Thus, elucidation of the in vivo involvement of CRE in myeloid differentiation will be of help in understanding the regulatory systems at work during the proliferation and differentiation processes in hematopoietic cells.
In the present study, we established a model system in which cAMP-dependent signals were specifically blocked by the expression of a dominant-negative CRE-binding protein (dn-CREB). Using this model, we examined the molecular mechanism of retinoid hormone-induced myeloid differentiation and the involvement of CRE-dependent transcription.
We show that CRE-dependent transcriptions play two "contrasting" roles during human myeloid differentiation: namely, positively regulating cell adhesion and respiratory burst activity but negatively regulating morphological maturation such as nuclear convolution and cytoplasmic vacuolar formation. This finding indicates that a dissociation in the regulatory system exists between physiological and morphological maturation. The biological significance of these phenomena is also discussed.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Morphological examination
The cells were washed with phosphate-buffered saline (PBS) and fixed on glass slides using a cytospin apparatus (Cytospin2, Shandon, Pittsburgh, PA) before being stained with Wright-Giemsa solution (Muto Pure Chemical Co., Ltd., Tokyo, Japan).
Examination of differentiation marker proteins by flow cytometry analysis
Cells were cultured at a density of 1 x 105/ml in 24-well dishes (1 ml/well) with 1 µl ethanol or 1 µl 1 mM ATRA (final concentration, 1 µM). After a 3-day incubation at 37°C, the cells were collected and washed with PBS containing 5% FBS and 0.05% sodium azide. They were then incubated with fluorescein isothiocyanate-conjugated mouse monoclonal anti-CD11b or -CD54 antibody (Caltag Laboratories, An-Der-Grub, Austria), and the expression of each protein was analyzed by FACSCalibur (Becton Dickinson, Mountain View, CA).
Antiserum and Western blot analysis
For Western blotting, 5 x 105 cells were lysed in 20 µl 1x Laemmli’s sample buffer and were boiled before being applied to a 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel (5x104 cells/10 µl/lane). After electrophoresis, the proteins were transferred to a polyvinylidene difluoride membrane (Millipore Co., Bedford, MA). The first antibody reactions were performed using anti-CREB (New England Biolabs, Beverly, MA), anti-p67phox (Santa Cruz Biotechnology, Santa Cruz, CA), anti-p47phox (Santa Cruz Biotechnology), anti-PU.1 (Santa Cruz Biotechnology), or anti-C/EBPß (Santa Cruz Biotechnology) antibody. The second antibody reactions were performed using horseradish peroxidase (HRP)-conjugated anti-mouse and anti-rabbit monoclonal antibodies (mAb; New England Biolabs) or an HRP-conjugated anti-sheep mAb (Chemicon, Temecula, CA). The final detection procedure was performed using SuperSignal West Dura extended duration substrate (Pierce, Rockford, IL) followed by exposure to Hyperfilm (Amersham International PLC, Buckinghamshire, UK).
Reverse transcription-polymerase chain reaction (RT-PCR)
mRNA was extracted from 8 x 107 cells with a Fast Track 2.0 kit (Invitrogen, Carlsbad, CA) and was then suspended in 50 µl 0.1% diethyl pyrocarbonate in water. A 1-µl aliquot of the mRNA solution was used for the RT reaction with avian myloblastosis virus RT XL (Takara Shuzo Co. Ltd.), and the product was then suspended in 20 µl 0.1% diethyl pyrocarbonate in water. The PCR procedure was performed using a DNA Thermal Cycler PJ2000 (PerkinElmer Corp., Foster City, CA) under the following conditions: 94°C, 30 s; 55°C, 1 min; 72°C, 1 min for 30 cycles. For detection of the transgene product, the sequences 5'-CGCGTGGTACCTCTAGAGTCGACC-3' from the promoter region in the pCI-neo vector and 5'-CATTAACCCTCACTAAAGGGAAG-3' from 3' the flanking region in the pCI-neo vector were used as primers. For detection of the ß2-microglobulin message, sense primer (5'-GTGGAGCATTCAGACTTGTC-3') and antisense primer (5'-CTGCTTACATGTCTCGATCC-3') were used. The product was separated by agarose gel electrophoresis, and the DNA (product sizes: transgenes, 1.1 kbp; ß2-microglobulin, 160 bp) was visualized by ethidium bromide staining. For the molecular markers, Smart Ladder (0.2–10 kbp) was used (Wako Pure Chemical Industries).
Nitroblue tetrazolium (NBT)-reducing activity
Cells were cultured at a density of 1 x 105/ml in 24-well dishes (1 ml/well) with 1 µl ethanol or 1 µl 1 mM ATRA (final concentration, 1 µM). After a 3-day incubation at 37°C, 5 x 105 cells were collected and washed before being suspended in 1 ml RPMI-1640 medium. Next, 1 ml NBT solution (1 mg/ml) was added, and the cells were incubated for 25 min at 37°C with 100 ng 12-O-tetradecanoylphorbol 13-acetate (TPA) and were finally washed once with PBS. The numbers of NBT-positive cells were counted with a light microscope (Olympus Optical Co. Ltd., Tokyo, Japan).
Transfection and luciferase assay
The pCRE-firefly luciferase reporter (fLuc) was constructed by inserting a 36-bp CRE-enhancer fragment (CTC GGG GCG CCT CCT TGG CTGACG TCA GAG AGA GAG) from the somatostatin gene [10
] into the NheI/XhoI site of pGL2-Basic fLuc (Nippon Gene Co. Ltd., Tokyo, Japan) together with a herpes simplex virus thymidine kinase (TK) promoter fragment from pRL-TK (Promega, Madison, WI) into the BglII site. pCRE-Luc (20 µg) and 0.4 µg pRL-simian virus 40 (SV40), a SV40 enhancer-driven Renilla luciferase reporter (Promega), were suspended in 194.4 µl water to which 5.6 µl ethoxylated polyethylenimine (EPEI; Gene Tools, LLC, Philomath, OR) was added and mixed immediately. The DNA/EPEI solution was incubated at room temperature for 25 min, after which 3.8 ml serum-free medium was added. A 2-ml aliquot of the DNA/EPEI solution was mixed with 2 x 106 cells, previously washed once with serum-free medium. The cells were placed in a 24-well dish (0.5 ml/well) and incubated for 3 h at 37°C, and then the medium was replaced with 1 ml serum-containing medium. After incubation at 37°C for another 2 h, the cells were divided into two groups. They were transferred into a six-well dish (4 ml/well) to which 4 µl ethanol or 4 µl ATRA solution (1 mM) was added. After a 2-day incubation at 37°C, cell lysate was prepared, and the luciferase assay was performed with the Dual-LuciferaseTM reporter assay system (Promega). The CRE activity was assessed by normalizing the fLuc activity with that of the RL activity. There were no significant differences in SV40-enhancer activities among the three lines.
Statistical analysis
The Student’s t-test was used to determine statistical significance. A P value of <0.01 was considered significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
According to the results of the Western blot analysis, we selected three clones for each line: clone-1, clone-2, and clone-3 for CTL (neo); clone-1, clone-2, and clone-5 for dn-CREB with a high-protein expression as a result of transgene-derived dn-CREB (Fig. 1A ); and clone-1, clone-2, and clone-9 for CTL (anti) with reduced CREB as a result of the antisense message expression (Fig. 1B) . All these clones showed equivalent levels of ß-tublin expression (Fig. 1C) , confirming that the difference in protein-expression level was specific to CREB. The expression of the transgene-derived message was further confirmed by RT-PCR using transgene-specific probes (Fig. 1D) . We also checked the growth of each line, as cAMP-mediated signals are involved in cell proliferation [9 ] as well as cell viability [12 ]. We found that the viable cell number of dn-CREB lines was significantly smaller than that of CTL (neo) lines (Fig. 2 ). Conversely, the viable cell number of CTL (anti) lines was almost identical to that of CTL (neo) lines (Fig. 2) , probably as a result of compensation by other CREB/ATF family members. These findings further support that dn-CREB protein is in fact expressed in the three clones in dn-CREB lines. Thus, we performed the following experiment using these nine clones.
 View larger version (29K): [in a new window] |
Figure 1. The construction and evaluation of U937 transformant lines. (A–C) Western blot analysis. Cell lysates were prepared from neomycin-selected clones of U937 cells after transfection with an expression vector for the (A, B and C) CTL (neo), an expression vector for (A and C) dn-CREB, and an expression vector for the (B and C) CTL (anti). An aliquot of each lysate was used for Western blot analysis (5x104 cells/lane) with the anti-CREB antibody (A and B) or anti-ß-tublin antibody (C). (D) RT-PCR analysis. The mRNA was extracted from three clones of the CTL (neo) lines (clone-1, clone-2, and clone-3); three CTL (anti) lines (clone-1, clone-2, and clone-9); and three dn-CREB lines (clone-1, clone-2, and clone-5); 1 µg was used for RT-PCR with primers derived from the promoter region and the 3' noncoding region of the pCI-neo vector (upper) or ß2-microglobulin (lower). The 1177-bp PCR products derived from the transgene were detected only in the CTL (anti) and dn-CREB lines (upper), and the 160-bp PCR products derived from ß2-microglobulin were detected in all lines (lower). The transgene product in CTL (neo) lines was only 56 bp and was not visible in this gel electrophoresis.
|
 View larger version (36K): [in a new window] |
Figure 2. The involvement of CRE in cell growth. Three clones from the CTL (neo) lines (closed square, diamond, and rectangle), CTL (anti) lines (open square, diamond, and rectangle), and dn-CREB lines (dotted square, diamond, and rectangle) were cultured at a density of 2 x 104/ml in 24-well dishes. The number of viable cells was then counted over time. The experiment was performed in triplicate, and the average cell number is presented. Note that the number of viable cells in dn-CREB lines was significantly low compared with that of CTL (neo) lines.
|
 View larger version (19K): [in a new window] |
Figure 3. CRE-dependent transcription. (A) Luciferase assay. Approximately 2 x 106 cells of CTL (neo) clone-1, CTL (anti) clone-1, and dn-CREB clone-1 were cotransfected with 20 µg CRE-enhancer-containing fLuc reporter together with 0.4 µg SV40 enhancer-containing RL reporter. Two hours after transfection, cells were treated with ethanol (EtOH) or 1 µM ATRA. After a 2-day incubation, the luciferase assay was performed. The CRE-enhancer activity was normalized by SV40-enhancer activity; the latter was used as an internal control for transfection efficiency. Open bars indicate the CRE-enhancer activities of ethanol-treated samples, and hatched bars indicate those of ATRA-treated ones. The data are expressed as mean ± SD of three independent experiments. Note that CRE activities in dn-CREB lines were significantly reduced compared with those in CTL (neo) lines in ethanol- and ATRA-treated cases. The results of the other two clones for each line were similar (data not shown). (B) C/EBPß induction. Approximately 1 x 105 cells of the three clones from the CTL (neo) lines, CTL (anti) lines, and dn-CREB lines were treated with or without ATRA. After a 3-day incubation, cell lysates were prepared, and Western blot analysis was performed using the anti-C/EBPß antibody.
|
Phenotype analysis of U937 transformants during functional maturation
We investigated the differentiation parameters regarding functional maturation. First, we examined the differentiation-associated cell adhesion and aggregation by light microscopic observation 3 days after ATRA treatment. As shown in Figure 4A
, ATRA-induced cell adhesion and aggregation were significantly attenuated in dn-CREB lines. Moreover, differentiation-dependent induction of an adhesion molecule, CD11b, was significantly attenuated in dn-CREB lines compared with CTL (neo) and CTL (anti) lines (Fig. 4B
, left). Conversely, the basal expression of CD54 was severely blocked in dn-CREB lines, and ATRA treatment up-regulated the surface CD54 expression in dn-CREB lines to an almost equivalent level to that observed in CTL (neo) and CTL (anti) lines (Fig. 4B , right).
 View larger version (49K): [in a new window] |
Figure 4. ATRA-induced cell adhesion. (A) Cell aggregation. Aliquots of the CTL (neo) line (clone-3; left), CTL (anti) line (clone-9; center), and dn-CREB line (clone-5; right) were cultured at a density of 1 x 105 cells/ml in 24-well dishes before being treated with 1 µl ethanol (EtOH) or 1 µM ATRA. Cell aggregation was examined using an inverted light microscope after a 3-day incubation (original magnification, x40). Note the loss of cell aggregation in the dn-CREB line (lower right). The other two clones of each transformant, CTL (neo), CTL (anti), or dn-CREB, showed similar results (data not shown). (B) CD11b and CD54 expression. Aliquots of the CTL (neo) line (clone-3; top), CTL (anti) line (clone-9; middle), and dn-CREB line (clone-5; bottom) were cultured at a density of 1 x 105 cells/ml in 24-well dishes before being treated with 1 µl ethanol or 1 µM ATRA. After a 3-day incubation, the surface expression of CD11b (left) or CD54 (right) was determined by flow cytometry analysis. Thin lines indicate the results of isotype-control immunoglobulin G staining. Bold lines represent anti-CD11b (left) or anti-CD54 (right) antibody staining of ethanol-treated samples, and dotted lines represent ATRA-treated samples. The other two clones of each transformant line showed similar results (data not shown).
|
 View larger version (30K): [in a new window] |
Figure 5. ATRA-induced respiratory burst activity. (A) NBT reduction activity. The three CTL (neo) lines, CTL (anti) lines, and dn-CREB lines were treated with 1 µM ATRA. After a 3-day incubation, the cells were collected and incubated with TPA for 25 min. The NBT reduction-positive cells were then counted under an inverted light microscopy. The data are expressed as mean ± SD of three independent experiments. Note that NBT-reducing activities in dn-CREB lines were significantly reduced compared with those in CTL (neo) lines. Nontreated cells and cells treated with ethanol alone did not show any detectable NBT reduction (data not shown). (B and C) The expression of p67phox, p47phox, and PU.1. The three CTL (neo) lines, CTL (anti) lines, and dn-CREB lines were treated with ethanol or ATRA. After a 3-day incubation, the cell lysates were prepared, and Western blot analysis was performed using anti-p67phox and anti-p47phox antibodies (B) or an anti-PU.1 antibody (C).
|
Thus, CRE-dependent transcriptions contribute to the induction of respiratory burst activity by up-regulating p67phox and p47phox expression through PU.1 gene induction.
Morphological findings of U937 transformants during differentiation
We then evaluated differentiation-dependent morphological alterations. In CTL (neo) lines, no apparent changes in nuclear or cytoplasmic morphologies were observed, although there was a reduction in total cell size (Fig. 6A
, left). In CTL (anti) lines, mild nuclear convolution was detected as well as a reduction in cell size, but no apparent cytoplasmic vacuolar formation was detected (Fig. 6A
, center). In contrast, dn-CREB lines showed remarkable changes, including nuclear convolution and multiple cytoplasmic vacuole formation, as well as a reduction in cell size (Fig. 6A , lower right). It is interesting that nuclear convolution was detected even without ATRA treatment in dn-CREB lines (Fig. 6A , upper right). Thus, the inhibition of CRE-dependent transcription resulted in the acceleration of morphological maturation with or without ATRA treatment.
 View larger version (29K): [in a new window] |
Figure 6. ATRA-induced morphological changes. (A) Wright-Giemsa staining. Aliquots of the CTL (neo) line (clone-3; left), CTL (anti) line (clone-9; center), and dn-CREB line (clone-5; right) were cultured at a density of 1 x 105 cells/ml in a 24-well dish before being treated with 1 µl ethanol (EtOH) or 1 µM ATRA. After a 3-day incubation, the cells were fixed on slide glasses and stained with Wright-Giemsa solution. Cell morphology was observed using a light microscope (original magnification, x1000). The other two clones of each transformant line showed similar results (data not shown). (B) Flow cytometry analysis. Aliquots of the CTL (neo) clone-3 (top), CTL (anti) clone-9 (middle), and dn-CREB clone-5 (bottom) were cultured at a density of 1 x 105 cells/ml in 24-well dishes before being treated with 1 µl ethanol or 1 µM ATRA. After a 3-day incubation, the cells were fixed with 1% paraformaldehyde, and two-dimensional flow cytometry analysis [x-axis, forward-scattered light (FSC); y-axis, side-scattered light (SSC)] was performed. The numbers indicate the percentage of cells located in the FSC-low/SSC-high area. Similar results were obtained from the other two clones of each transformant line (data not shown).
|
Thus, CRE enhancer is required for the inhibition of morphological maturation during myeloid differentiation.
Differentiation-dependent growth arrest in U937 transformants
As ATRA has antiproliferative activity on myeloid cells [3
, 4
], we finally examined the involvement of CRE-dependent transcription in differentiation-associated cell-growth arrest. Time-course analysis of the numbers of viable cells was conducted with ATRA treatment in CTL (neo), CTL (anti), and dn-CREB lines. We found that the growth rate was reduced by approximately 20% after ATRA treatment in all three lines and could not detect any significant differences among them (Fig. 7
).
 View larger version (17K): [in a new window] |
Figure 7. Differentiation-associated growth inhibition. Aliquots of the CTL (neo) clone-3 (left), CTL (anti) clone-9 (center), and dn-CREB clone-5 (right) were cultured at a density of 1 x 105 cells/ml in 24-well dishes before being treated with 1 µl ethanol (EtOH) or 1 µM ATRA. Viable cell numbers were counted over a time course. The data are expressed as mean ± SD of three independent experiments. The other two clones of each transformant line showed similar results (data not shown).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The dn-CREB lines showed insufficient inductions of p67phox, p47phox, and PU.1 by ATRA treatment. Although PU.1 is also an activator of the gp91phox promoter, we could not find any enhancement of the gp91phox protein expression in ATRA-treated CTL (neo), CTL (anti), and dn-CREB lines. We could also not detect any significant differences in the levels of gp91phox protein expression even after ATRA treatment among the three lines. Moreover, p67phox is the dose-limiting component of phagocyte NADPH oxidase [14 ]. Thus, we speculate that the molecular basis for reduced respiratory burst activity in dn-CREB lines may be the insufficient induction of p67phox as a result of failed PU.1 induction [15 , 16 ]. The defects in PU.1 induction may also be the reason for the insufficient CD11b induction in ATRA-treated dn-CREB lines [17 ]. At present, however, we do not know the precise molecular mechanism by which CRE-dependent transcriptions contribute to PU.1 gene induction. There is no CRE consensus sequence in the PU.1 promoter region, at least up to -30 bp upstream from the transcriptional initiation site [18 ]. Thus, the CRE enhancer may function at a more upstream region in the PU.1 promoter, or alternatively, CRE-dependent transcriptions may induce a specific transcription factor responsible for activation of the PU.1 promoter. Further investigations are required to determine the exact mechanism.
The accelerated, morphological maturation in dn-CREB lines under ATRA-treated and nontreated conditions indicates that CRE-dependent transcriptions play a role in inhibiting the basal as well as differentiation-associated, morphological alterations. Although the molecular mechanism involved in morphological maturation is still elusive, our findings at least show that CRE-dependent transcriptions negatively regulate morphological maturation. The reason CRE-dependent transcription inhibits morphological maturation may be a result of the fact that CRE-dependent transcription works in cell proliferation [9 ] where morphological alterations must not occur. Thus, other molecular events, which work more strongly than CRE-dependent transcription, would transmit signals for morphological alterations during differentiation. Further investigations will elucidate the precise molecular mechanism of morphological maturation.
As we showed in Figure 7 , CRE-dependent transcriptions are not required for differentiation-associated growth arrest. This is in a clear contrast to the fact that CRE-dependent transcriptions play some roles in baseline cell proliferation (Fig. 2) , probably by inducing the genes for cyclin D [19 ] and cyclin A [20 ]. Thus, distinct mechanisms are at work in growth regulation under undifferentiated and differentiated conditions. It has been reported that interferon regulatory factor (IRF) is involved in differentiation-associated growth arrest [21 ]. Thus, the function of IRF-dependent transcriptions is superior to CRE-dependent transcriptions regarding growth regulation, the former playing a dominant role in ATRA-induced growth arrest.
Our data concerning the involvement of CRE-dependent transcriptions in myeloid differentiation and proliferation are summarized in Figure 8 . As we showed, ATRA-mediated signals might be divided into three categories: CRE-activated, CRE-inhibited, and CRE-independent pathways. Among these, the signal regulating CD54 expression was rather unique in that CRE-dependent transcriptions are essential for its basal expression, but retinoic acid can transmit signals by bypassing CRE-dependent signaling during differentiation. This finding indicates that there is a redundancy in the regulation of CD54 expression.
 View larger version (20K): [in a new window] |
Figure 8. Model of molecular signaling during ATRA-induced differentiation. RXR, Retinoid-X receptor.
|
Using in vitro cultured cells, we have shown the involvement of CRE-dependent transcription in myeloid differentiation, although the in vivo significance of our findings remains unclear. The involvement of CREB-binding factors, such as CREB-binding protein [22 ] and Tax [23 ], in the development of acute leukemia suggests a possible involvement of CRE-dependent transcription in in vivo hematopoietic differentiation. Recently, it was reported that CREB-/-/ATF-1-/- double-knockout mice are embryonically lethal at the morula stage [24 ]. Thus, it may be somewhat difficult to understand the requirement of CRE-dependent transcriptions during hematopoiesis using this murine system. In this sense, our system would be very useful, providing valuable information concerning the functions of CRE-dependent transcriptions in human hematopoiesis. Further investigations to clarify the in vivo importance of CRE-dependent transcriptions during myeloid development will be of help in establishing a novel therapy for hematopoietic disorders.
Received October 29, 2002; revised January 21, 2003; accepted January 24, 2003.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  D. Sawka-Verhelle, L. Escoubet-Lozach, A. L. Fong, K. D. Hester, S. Herzig, P. Lebrun, and C. K. Glass PE-1/METS, an Antiproliferative Ets Repressor Factor, Is Induced by CREB-1/CREM-1 during Macrophage Differentiation J. Biol. Chem., April 23, 2004; 279(17): 17772 - 17784. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
