* Section of Hematology, Yale University School of Medicine, New Haven, Connecticut;
Division of Oncology, Department of Medicine, Washington University School of Medicine, St. Louis, Missouri; and
University of Texas Health Science Center, Division of Hematology/Department of Medicine, San Antonio
1Correspondence: Section of Hematology, Yale University School of Medicine, New Haven, CT 06510. E-mail: arati.gupta{at}yale.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: representation difference analysis (RDA) • STAT3 • STAT5
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
32Dcl3 is an IL-3-dependent murine granulocytic precursor cell line, which can be made to undergo differentiation to mature neutrophils upon replacement of IL-3 with G-CSF. Removal of the cells from IL-3 results in apoptosis. To isolate G-CSF-responsive genes from the 32Dcl3 cell system, a RDA experiment was performed using a combination of 32Dcl3 cells grown in IL-3 and 32Dcl3 cells deprived of IL-3 for 4 h as the driver and 32Dcl3 cells treated for 4 h with G-CSF as the tester. All cells were treated with cyclohexamide to block secondary transcriptional events. The mixed driver population was used to eliminate genes expressed on IL-3 withdrawal. Two genes isolated under these conditions were identified as transglutaminase, a known immediate-early all-trans retinoic acid (ATRA)-responsive gene, and p120 nucleolar-proliferation antigen (NOL1/Nop2p). p120 is a highly conserved, 120-kDa, nucleolar-specific, RNA-binding protein expressed by cells early in the G1 phase of the cell cycle and peaking during the S phase [8 ]. A wide variety of malignant tumors expresses higher levels of the p120 gene as compared with normal, resting cells [8 ]. Overexpression of the human p120 gene in NIH3T3 cells resulted in transformation of the cells and produced rapidly growing tumors in nude mice [9 ].
In this study, we sought to identify the basis of G-CSF-induced expression of the p120 gene. We therefore screened a mouse genomic library and isolated the promoter and 5' regulatory sequences of the p120 gene and demonstrate a role for a STAT-binding site identified in the first intron in the p120 locus in mediating the G-CSF response. In addition, we show that overexpression of the p120 gene in G-CSF-induced 32D cells results in normal, morphologic maturation but defective biochemical maturation, as judged by the loss of lactoferrin (LF) gene expression. LF is a marker of normal terminal neutrophil development, and its expression is invariably defective in acute myeloid leukemias.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Approximately 1 x 107 32Dwt18 cells were transiently transfected as described previously using 10 µg of each reporter and effector plasmids and 2 µg pCMVßgal (Clontech, Palo Alto, CA), an internal control plasmid used to monitor transfection efficiency [11 ]. Transiently transfected cells were incubated at 37°C in 5% CO2 for 20 h, and luciferase activity was determined using an assay kit from Promega Biotech (Madison, WI) as per the manufacturer’s instructions. NIH3T3 cells were transfected with 2 µg reporter plasmids by lipofection using Lipofectamine as per the manufacturer’s instructions (Gibco-BRL). Transfected cells were incubated at 37°C in 5% CO2 for 48 h. Luciferase activity was then determined using an assay kit from Promega Biotech as per the manufacturer’s instructions. Cotransfection experiments included 1 µg STAT3 or STAT5 expression plasmids described previously [12 ] or the constitutively active STAT3C expression plasmid [12 ]. Luciferase expression levels were normalized to the levels of ß-gal expression [11 ].
RDA analysis of 32Dcl3 cells induced with IL-3 and G-CSF
32D cells were grown in the presence of IL-3, washed twice with 1x phosphate-buffered saline, and resuspended in medium without cytokines in the presence of cycloheximide. After a 20-min incubation at 37°C/5% CO2, the cells were incubated in no cytokines, in IL-3 (10% WEHI-conditioned medium), and in 100 ng/ml G-CSF. cDNA RDA was performed as described previously [13
]. The driver DNA was obtained by combining cDNA from 32Dcl3 cells grown in IL-3 (10% WEHI-3B-conditioned medium) with cDNA from 32Dcl3 cells grown in the absence of IL-3 for 4 h. The tester was derived from 32Dcl3 cells treated with G-CSF (100 ng/ml) for 4 h. Tester and driver cells were treated with 5 µg/ml cycloximide (Sigma Chemical Co., St. Louis, MO) for 15 min prior to G-CSF induction. Following three rounds of subtractive hybridization, the differentially expressed cDNAs obtained (DP3 products) were cloned into BamHI-digested pBlueScript (Stratagene, La Jolla, CA). Cloned DP3 products were sequenced by strandard dideoxy-sequencing technology. Clones were identified by BLAST analyses in the National Center for Biotechnology Information (NCBI; Bethesda, MD) nucleotide database.
RNA isolation and Northern blot analysis
Total RNA was extracted from uninduced and G-CSF-induced 32Dcl3 cells as described previously [14
], and 10 µg was analyzed by Northern blot analysis using a protocol previously described [14
]. Blots were probed with DP3 clones representing 500 base pairs (bp) of the mouse p120 nucleolar antigen cDNA probe (NM138747, nucleotide numbers 312–811) and a 600-bp fragment of the transglutaminase cDNA (M55154, nucleotide numbers 883–1450). A Northern blot of RNA isolated from uninduced and ATRA- induced MPRO cells was also carried out as described previously. The full-length human cDNA probe was provided by Dr. Harris Busch (Baylor University, Houston, TX) and has been described previously [9
].
Library screening and plasmid construction
To isolate the promoter and 5' regulatory sequences of the mouse p120 gene, we screened a mouse 129svJ genomic library (Stratagene) by methods described previously [15
]. The phage library was plated at 5x 104plaque-forming units/plate. Duplicate nitrocellulose filters were lifted and hybridized at 65°C overnight to a 32P-labeled mouse p120 cDNA isolated from the RDA screen described above. Filters were washed twice at room temperature in 0.1% sodium dodecyl sulfate (SDS) with 15 mM sodium citrate and once at 65°C, patted dry, and exposed to X-ray film with an intensifying screen at –80°C overnight. Positive plaques were purified further in secondary and tertiary screens. Three genomic clones, 13–8, 13–20, and 13–21, were isolated and purified for further analysis. A Southern blot was performed using EcoRI-digested phage clones and probed with a mouse p120 exon 1 oligonucleotide harboring the ATG and 5' sequences (5' ttgcgccccatgataactaat 3') as a probe. Only fragments in Clone 13–2 lit up, and these were purified, subcloned into pGEM7 vector (Promega Biotech), and sequenced by standard dideoxy sequencing protocols.
Subcloning promoter fragments
Using the sequence information obtained from the 13–2 phage subclones described above, a polymerase chain reaction (PCR) was performed using a 2-kb EcoRI-digested subclone of Phage 13–2 in pGEM7 as a template, the SP6 oligomer, and the oligomer in the mouse p120 gene harboring the ATG in thefirst exon detailed above. The 1-kb PCR product obtained was subcloned into the pCR11 TA vector (Invitrogen, Carlsbad, CA) and sequenced to confirm its identity. A SacI/XhoI digest of the pCR11-p120, 1-kb promoter clone was next isolated and subcloned into the promoterless pGL3B promoter digested with the same enzymes. The fragment containing the entire first intron was isolated from the 13–2 genomic clone using the following oligomers in a PCR reaction: sense 5' GGTGAGGGTCGTAGCTCCCTGATCC 3'; antisense 5' TCTTTCGAGCACGACTACAGAGCCTCTTTGG 3'.
The PCR product was first cloned into the pCR11 vector and further subcloned into the promoterless pGL2Btk vector harboring a minimal thymidine kinase promoter (a gift from Dr. Patrick Gallagher, Yale University, New Haven, CT). Mutations within the STAT site in the p120 intron were next prepared using the Quick Change kit (Stratagene), as per the manufacturer’s instructions. The oligomers used in this procedure included the following mutated sequence: p120 STAT wild-type 5' TTCCAGGAA 3' was changed to p120 STAT mutant 5' CGATATTAA 3'.
Preparation of whole cell extracts and Western blot analysis
Whole cell extracts were prepared from uninduced 32Dcl3 and 30 min, G-CSF-induced 32Dcl3 cells essentially as described previously [16
]. Total protein concentration in the whole cell extract preparations was assayed using the Bradford assay (BioRad kit, BioRad, Hercules, CA), as per the manufacturer’s instructions. Approximately 40 µg whole cell extracts prepared from uninduced and 15-min, 30-min, 60-min, and 4-h, G-CSF-induced 32Dcl3 cells were transferred to 2x Laemelli’s loading buffer, boiled for 5 min, and loaded onto a 4–20% Tris/Glycine gel (Novex, San Deigo, CA). Electrophoresis was carried out at 150 V for 2 h at room temperature. Electrophoresed proteins were transferred to a polyvinylidene difluoride membrane (BioRad) and blocked with 5% nonfat dry milk in Tris-buffered saline containing 0.1% Tween-20 (TBS-T) for 1 h at room temperature. The blocked membrane was next incubated with phosphor-STAT3 (Y705) antibody (Cell Signaling Technology, Beverly, MA) at a 1:1000 dilution in TBS-T at 4°C overnight. The membrane was washed three times with TBS-T+ 5% nonfat dry milk and incubated with an anti-rabbit horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology, CA) for 1 h at room temperature. The membrane was washed five times in TBS-T, and chemiluminescent detection was performed as per the manufacturer’s instructions (Amersham, Piscataway, NJ). The membrane was next stripped and reprobed as described with a STAT3 antibody (Cell Signaling)
Oligonucleotide pull-down assay
Uninduced and 30-min, G-CSF-induced 32Dcl3 cells were lysed in Nonidet P-40 (NP-40) lysis buffer [50 mM Tris-Hcl, pH 8.0, 0.5% NP-40, 150 mM NaCl, 0.1 mM EDTA, 1 mM dithiothreitol (DTT), 10% glycerol, and the complete protease inhibitor cocktail (Roche Molecular Systems, Alameda, CA)]. Cell lysates or nuclear extracts (200 µg; prepared as above) were incubated with 1 µg biotinylated p120 STAT probe (see above) or a known STAT-binding site m67 [12
] in the presence of 1 µg poly:d(I)/d(C) in binding buffer (10 mM HEPES, pH 7.9, 75 mM NaCl, 1 mM DTT, 1 mM EDTA, 10 mM MgCl2, 10% glycerol). The lysate-oligomer mix was allowed to incubate for 1 h at 4°C. Streptavidin-agarose beads (40 µl; Invitrogen) were next added, and incubation continued overnight at 4°C. The DNA–protein complexes were then collected by low-speed centrifugation and washed four times in binding buffer, heated to 95°C for 5 min in 2x sample loading buffer (Invitrogen), and resolved in a 4–12% gradient NuPage gel (Invitrogen). Western blotting was performed as above.
Reverse transcriptase (RT)-PCR analysis of p120 expression in bone marrow-expressing, mutant G-CSF-Rs
Mice harboring knock-in mutations, d715 and d715F, in the G-CSF-R have been described previously [12
]. Bone marrow cells from wild-type, d715, and d715F mice were harvested by flushing both femoral bones with 
-minimal essential medium with 10% fetal bovine serum. A total of 5–10 x 104 mononuclear bone marrow cells in each case was suspended in Myelocult 5300 (Stem Cell Technologies, Vancouver, CA) containing human G-CSF (100 ng/ml, Amgen) and 100 ng/ml stem cell factor (R&D Systems, Minneapolis, MN) and incubated for up to 8 days at 37°C in a 5% CO2 incubator. Cells were harvested on Days 0, 3, 5, and 7 or 8 post-induction, and total RNA was isolated by the Trizol method according to the manufacturer’s protocol (Invitrogen).
The total RNA samples so isolated were subjected to RT analysis using standard protocols [15 ]. The cDNA from the RT reaction was used as a templete in a PCR reaction designed to amplify mouse p120, and the mouse p120 oligos used were as follows: p120 mouse (sense): 5' gaggatgaggaggaaaagctg 3'; p120 mouse (antisense): 5' gcgcccatcatagtggctgat 3'. Previously described ß-actin oligos were used to determine the efficiency of the RT reaction and to determine equal loading in each sample.
Real-time PCR analysis
Total RNA (0.1 µg) from wild-type and d715F bone marrow from Days 3 and 5 post-G-CSF treatment was used to generate first-strand cDNA using random hexamer primers with the Superscript II RT kit (Invitrogen). Each cDNA (5 ng) was then analyzed by real-time PCR in duplicate for each primer pair, and the corresponding threshold cycle values were determined. All reactions were performed in a 25-ul vol using 50 ng each primer and the platinum SYBR Green quantitative PCR (Q-PCR) Supermix UGD kit from Invitrogen as per the manufacturer’s instructions using default cycling parameters in an ABI Prism 7200 Sequence detector. Transcript levels of each mRNA were normalized to that of ß-actin and then expressed as a percentage of the signal observed in the normal sample using the above primer pairs.
Overexpression of p120 in 32D cells
The full-length cDNA for human p120 (a gift from Dr. Harris Busch) was subcloned into the mammalian expression vector pIRES2-enhanced green fluorescent protein (EGFP; Clontech), digested with SalI and SacI. The resultant p120 expression plasmid was stably transfected by electroporation into 32Dcl3 cells as described previously [11
]. Stable transfectants were isolated following growth in G-418 (Gibco-BRL, 600 µg/ml). Single-cell clones of vector alone or plasmid EGFP-p120 were isolated by limiting dilution and analyzed for expression of p120 by Northern blot analysis as described above. Clones A and C (expressing high levels of p120, see Fig. 7A
) were used for further analysis. G-CSF inductions of the clones were carried out as described above.
 View larger version (54K): [in a new window] |
Figure 7. Overexpression of p120 in 32 cells. (A) p120 was overexpressed in 32D cells, and stable transfectants were isolated (see Materials and Methods). A Northern blot of 6 p120-overexpressing clones and three vector-alone clones was performed and probed sequentially for p120 and actin. Clones A, C, and E express high levels of p120. (B) A Northern blot of G-CSF induced 32D cells of a vector clone (Lanes 1–4), and p120-overexpressing Clone C (Lanes 5–8) was performed. The blot was probed sequentially with the neutrophil maturation marker LF and subsequently with actin. Twenty-four-hour ATRA-induced MPRO cell RNA was used as a positive control (+; Lane 9).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (26K): [in a new window] |
Figure 1. Northern blot analysis of DP3 clones from 32Dcl3/G-CSF RDA, which was carried out in 32Dcl3 cells using a combination of cDNAs from 32Dcl3 cells in IL-3 with cDNA from 32Dcl3 cells grown in the absence of IL-3 for 4 h as the mixed driver and 32Dcl3 cells treated with G-CSF (100 ng/ml) for 4 h as the tester. Tester and driver cells were treated with cycloheximide (CHx) from 15 min prior to and 4 h after G-CSF induction to identify only direct targets of G-CSF signaling. Following three rounds of subtractive hybridization, the differentially expressed cDNAs obtained (DP3 products) were cloned, and sequenced (see Materials and Methods) included p120 and transglutaminase. RNA (10 µg), obtained from 32Dcl3 cells in the absence of IL-3 (–IL-3, Lane 1), in the presence of IL-3 (+IL-3, Lane 2), and in G-CSF (–IL-3+G-CSF, Lane 3), was subjected to Northern blot analysis and sequentially probed for p120 and transglutaminase expression. Equal loading of RNA in each lane was ensured by reprobing the blot with a cDNA probe for actin.
|
 View larger version (39K): [in a new window] |
Figure 2. Expression of p120 in myeloid cells. (A) Northern blot analysis of 32D cells induced with G-CSF for 0 and 4 h and Days 1–3 and 8 was performed. The blot was probed sequentially with p120 and actin cDNA probes. (B) A Northern blot of ATRA-induced MPRO cells treated for 0, 24 h, 48 h, and 72 h (Lanes 1–4, respectively) was performed. Normal mouse brain RNA was used as a positive control (A, Lane 5). The blot was probed sequentially with p120 and actin cDNAs, the latter to ensure equal loading of RNA in each lane.
|
gene. These cells are more mature than 32D cells and can be induced to undergo complete neutrophil maturation upon the addition of pharmacological levels of ATRA within 3 days [13
, 18
]. A Northern blot of ATRA-induced MPRO cells treated for 0, 24 h, 48 h, and 72 h was performed. Normal mouse brain RNA was used as a positive control (Fig. 2B
, Lane 5). The blot was probed sequentially with p120 and actin. p120 mRNA is abundant in uninduced MPRO promyelocytes (Fig. 2B
, Lane 1) but diminishes to undetectable levels within 48 h post-ATRA treatment (Fig. 2B
, Lanes 2–4), suggesting that p120 expression is inversely proportional to the differentiation stage of MPRO as well as 32D cells. Thus, the expression of p120 appears to decrease upon neutrophil maturation. A similar pattern of p120 expression has been reported recently in microarray analyses of human bone marrow populations highly enriched for promyelocytes, myelocytes/metamyelocytes, and neutrophils [19
].
Isolation and characterization of the mouse p120 promoter and regulatory sequences
To identify promoter elements responsible for the p120 expression response to G-CSF, we isolated genomic sequences of the mouse p120 gene. A mouse 129sv genomic library was screened using the p120 fragment isolated by RDA as a probe (see Materials and Methods). Three positive clones were identified, one of which was analyzed further. Sequence analysis of the genomic locus of mouse p120 is shown in Figure 3A
. The TAATA box is indicated in gray, and three previously identified transcription start sites (indicated by asterisks) [20
] are indicated 5' to the first ATG represented in bold (+1). The first 500 bp of the promoter region and the first exon of the mouse p120 gene are nearly 70% identical to the human p120 gene [21
, 22
] (data not shown). In studies of the human p120 promoter, Haidar et al. [20
] identified an Sp-1 site (Fig. 3A)
, which is conserved in the mouse and human p120 genes and appears to be indispensable to promoter activity in NIH3T3 cells. They also indicated the importance of an adjacent site in the human p120 promoter, which they refer to as the PRE. They demonstrated that the PRE was capable of up-regulating human p120 promoter activity 2.5-fold in NIH3T3 cells. We identified several putative transcription factor-binding sites by using the TRANSFAC database, including sites for ATF/CREB, Ets, AP-1, and a STAT site (underlined in black), all of which are indicated in Figure 3A
. Of these putative transcription factor-binding sites, the STAT site located in the first intron of the p120 gene was most noteworthy with respect to the G-CSF response.
 View larger version (40K): [in a new window] |
Figure 3. Genomic organization of the mouse p120 gene. (A) The DNA sequence of the genomic locus of the mouse p120 gene is indicated. The TAATA box and translation start site (ATG) encompass the three transcription start sites (indicated by asterisks). The conserved and functionally relevant specificity protein 1 (Sp-1) and positive regulatory element (PRE) are indicated in the promoter. In addition, binding sites for activating transcription factor (ATF)/cyclic adenosine monophosphate response element-binding protein (CREB), Ets, and activated protein 1 (AP-1) are indicated. A STAT (underlined in black) is indicated in the first intron. (B) Schematic of the mouse p120 locus indicating the location of the oligonucleotide pairs used to amplify and subclone the 1-kb promoter and 5' regulatory region (1S and 1AS) and the first intron of the p120 gene (2S and 2AS). The location and sequence of the STAT site are indicated.
|
Transient cotransfection analyses were first performed in the myeloid 32Dwt18 cell line, a subline of 32Dcl3 cells described previously [11 ]. In transient cotransfection analysis of ptkluc1.0 harboring a 1-kb fragment of the p120 promoter and 5' regulatory sequences, no change in luciferase reporter gene activity was observed in the presence of a constitutively activated STAT3 expression plasmid (Fig. 4 ). However, a 3.7-fold increase in luciferase activity was observed upon cotransfection of the p120 intron 1-containing reporter plasmid (ptklucin; see Fig. 3B ) with the constitutively activated STAT3 plasmid. The increased reporter gene activity was eliminated following cotransfection of a plasmid bearing a STAT site mutation in the p120 intron (ptklucinmut; Fig. 4A ). Based on these observations, the presence of a functional STAT site is indicated in the first intron of the p120 gene.
 View larger version (14K): [in a new window] |
Figure 4. Transient transfection analysis of the p120 intronic STAT site in myeloid and nonmyeloid cells. (A) Reporter plasmids (10 µg) ptkluc1.0, harboring a 1-kb fragment of the p120 promoter and 5' regulatory sequences (see Fig. 3B
); p120 intron 1, containing reporter plasmid (ptklucin; see Fig. 3B
); a plasmid bearing a STAT site mutation in the p120 intron (ptklucinmut); and the empty ptkluc vector were transiently cotransfected into 32Dwt18 cells, with and without a constitutively active STAT3 expression plasmid (STAT3C) and with 2 µg pCMVßGal plasmid. Cells were harvested 24 h post-transfection, and luciferase reporter gene activity was measured. Reporter gene activity has been represented as fold increase in luciferase activity with STAT3C over luciferase activity without STAT3C. The figure represents normalized mean ± SE obtained from three independent experiments, each performed in duplicate. (B) Transient cotransfection analyses were carried out in the nonmyeloid NIH3T3 cell line. The STAT site (2 µg) containing p120 intron reporter plasmid (ptklucin) was cotransfected with 2 µg expression plasmids for STAT3 and STAT5 along with 0.5 µg pCMVßGal plasmid. Luciferase activity was measured 48 h post-transfection and represented as fold increase in luciferase activity over empty vector (ptkluc). The figure represents normalized mean ± SE obtained from three independent experiments, each performed in duplicate.
|
Oligonucleotide pull-down assay (OPD) confirms binding of phospho-STAT3 and demonstrates binding of phospho-STAT5 to the p120 STAT site in G-CSF-induced 32Dcl3 cells
Having demonstrated the presence of a functional STAT site in the first intron of the p120 gene, we undertook to determine the identity of the STAT factors involved with binding to this site in 32D cells. We initially performed a Western blot using uninduced and G-CSF-induced 32Dcl3 cells to demonstrate activation of STAT3. As evidenced in Figure 5A
, G-CSF treatment of 32Dcl3 cells results in rapid phosphorylation of the STAT3 protein within 15 min. This activation of STAT3 appears to be sustained over the course of the experiment (4 h; Fig. 5A
, Lane 5). Previous studies have demonstrated that STAT3 remains phosphorylated and therefore activated in G-CSF-treated 32Dcl3 cells for up to 5 days [26
].
 View larger version (29K): [in a new window] |
Figure 5. STAT3 binds the STAT site in the first intron of the p120 gene. (A) Western blot analysis was performed using whole cell extracts prepared from uninduced (0, Lane 1) and G-CSF-induced 32Dcl3 cells for 15 min–4 h (Lanes 2–5). The blot was probed sequentially with phospho-STAT3 and STAT3 antibodies. (B) Oligonucleotide pull-down assay. Cell extracts (200 µg), prepared from uninduced and G-CSF-induced 32Dcl3 cells, were incubated with a biotinylated, double-stranded p120 STAT probe. The DNA–protein complexes were recovered using streptavidin-agrose beads, and the bound proteins were resolved by SDS-polyacrylamide gel electrophoresis and Western blot analysis. The blot was first probed with STAT3 antibodies (top panel). The blot was stripped and reprobed first with a phospho-STAT3 antibody (STAT3-P; middle panel) and then with a phospho-STAT5 antibody (bottom panel). Twenty percent of the input cell extract sample probed with the STAT3 antibody has been shown (bottom panel of B). The arrows point to the phospho-STAT3 and phospho-STAT5 proteins. (C) Oligonucleotide pull-down assay was performed as described in B using a known STAT-binding probe m67. The blot was first probe with phospho-STAT3, stripped, and reprobed with phospho-STAT 5. Twenty percent of the input cell extract sample was probed with the STAT5 antibody.
|
p120 expression is abrogated in the bone marrow of mutant G-CSF-R knock-in mice defective in STAT3 signaling
Thus far, we have demonstrated that p120 nucleolar antigen is a direct target of G-CSF signaling in 32Dcl3 cells. We have also shown the presence of a functional STAT site in the first intron of the p120 gene. To demonstrate that p120 expression is dependent on STAT signaling, we examined the expression of p120 in mice lacking normal G-CSF-induced STAT3 signaling. We studied transgenic mice carrying a mutant G-CSF-R (designated d715F) in which the distal 98 amino acids are deleted, and the remaining cytoplasmic tyrosine residue mutated to a phenylalanine (Fig. 6A
). G-CSF-mediated STAT3 and STAT5 activation by the d715F mutant is severely impaired [12
]. Transgenic mice, carrying an identical, truncated G-CSF-R mutant but with Y704 intact (designated d715, Fig. 6A
), which demonstrate normal STAT3 activation in response to G-CSF, were used as a control [12
].
 View larger version (18K): [in a new window] |
Figure 6. Expression of p120 in the bone marrow of mutant G-CSF-R knock-in mice defective in STAT3 signaling. (A) A schematic representing wild-type (WT) and mutant G-CSF-Rs. The extracellular (EC), transmembrane (TM), and cytoplasmic domains (CD) are indicated. The conserved tyrosine residues (Y) are indicated as are the conserved Boxes 1 and 2. Mutant d715 lacks 98 amino acids of the cytoplasmic domain and retains only one tyrosine residue (Y704), which is mutated to a phenylalanine in mutant d715F (F704). (B) Myeloid progenitors were isolated from mice homozygous for the d715 G-CSF-R mutation and cultured in the presence of G-CSF for 3–8 days. Cells were harvested on Days 3, 5, and 8, and RT-PCR analysis was performed using p120 and actin-specific oligonucleotides. The p120 transcript is expressed at Days 3 and 5 but not at Day 8 of G-CSF induction in the d715 G-CSF-R sample. (C) Real-time Q-PCR analysis of G-CSF-induced wild-type (wt) and d715F G-CSF-R mutant was performed using oligomers specific for p120 and for actin. Transcript levels of p120 mRNA were normalized to that of B-actin and expressed as a percentage of the signal observed in the wild-type-Day 3 sample (100%). The arrows indicate lack of measurable p120 transcript.
|
Overexpression of p120 in 32D cells results in partial neutrophil maturation
Previous studies have demonstrated that p120 nucleolar-proliferating protein is expressed at higher levels in malignant tumors as compared with normal tissues [27
]. In fact, expression levels of p120 have been used as a prognostic marker for several malignancies including adenocarcinomas of the lung and in prostrate cancer [28
, 29
]. We have demonstrated increased expression of p120 in a number of human myeloid leukemic cell lines (data not shown).
In an attempt to investigate the role of p120 in neutrophil development, we overexpressed this gene in 32D cells and evaluated the ability of the resulting transgenic cells to undergo G-CSF-induced neutrophil maturation. The p120 gene was constitutively expressed in 32D cells, and single-cell clones were selected for further analysis. Figure 7A represents a Northern blot demonstrating varying levels of expression of p120 in clones A–F. In comparison, empty vector-transfected 32D cells expressed little or no endogenous p120 (Fig. 7A , Vector alone 1–3).
p120-overexpressing clones (A and C were used for further analysis) were incapable of growing in the absence of IL-3, confirming that constitutive expression of p120 in 32D cells does not render the cells factor-independent (data not shown). Although morphologic maturation of p120-overexpressing clones appeared to be normal (data not shown), a block in late neutrophil gene expression was observed. The expression of LF, a marker of late neutrophil maturation, was diminished greatly (Fig. 7B , p120clc, Lanes 5–8) in p120-overexpressing 32D cells induced with G-CSF as compared with the empty vector expressing 32D cells (Fig. 7B , Lanes 1-4) as measured by Northern blot analysis. These observations suggest that the temporal expression of p120 during neutrophil development is critical and that aberrant expression can contribute to the disregulation of normal neutrophil development, which is a hallmark of acute myeloid leukemias.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
G-CSF mediates neutrophil maturation via G-CSF-R binding, which activates the JAK-STAT, phosphatidylinositol-3 kinase, and the RAS-MAPK pathways (reviewed in ref. [1
]). In the JAK-STAT pathway, G-CSF stimulates members of the JAK family (JAK1 and -2 and Tyk2), which in turn activate STAT1, -3, and -5. A significant role for G-CSF in mediating myeloid gene regulation via the modulation of key transcription factors such as CCAAT/enhancer-binding protein C/EBP and C/EBP
has been well established [31
, 32
]. Using the 32D/G-CSF cell model of neutrophil differentiation, Numata et al. [33
] recently demonstrated that the expression of C/EBP was up-regulated in G-CSF-induced 32D cells in a STAT3-dependent manner, but no STAT-binding site has been identified in the C/EBP promoter. Additional studies have identified other G-CSF-regulated genes including the G-CSF-R, SOCS3, and several known immediate-early genes such as Erg-1, c-fos, c-jun and junB, Tis-7, Tis-8, and Tis-11, which are up-regulated in G-CSF-stimulated myeloid precursor cells [4
5
6
]. Again, the mechanism involved in the up-reglation of these genes as a result of G-CSF signaling remains unclear. Our data suggest that G-CSF induction of 32Dcl3 cells results in the phosphorylation of STAT3 and STAT5, which in turn bind to the STAT site in the p120 intron, resulting in expression of the p120 gene. Although the other cis-acting elements in the p120 promoter may be involved in up-regulation of the p120 gene, they are unlikely to contribute to the up-regulation of this gene in response to G-CSF. Our study, thus, adds p120 to a growing list of immediate-early genes such as SOCS3, JunB, and p19INK/4D, which are direct targets of STAT3 and do not require de novo protein synthesis for their expression upon cytokine induction.
The cytoplasmic domain of the G-CSF-R harbors four highly conserved tyrosine residues (Y704, Y729, Y744, and Y764), of which Y704 and Y744 have been implicated in modulating myeloid maturation [33 ]. Studies from several groups have demonstrated that phosphorylation of Y704 and Y744 recruits STAT3 and to a lesser extent, STAT1 and -5 [1 , 34 , 35 ]. Previous studies in transgenic mice carrying a mutant G-CSF-R lacking the C-terminal 98 amino acids with a mutation in the remaining tyrosine residue (Y704) referred to as d715F have demonstrated a lack of STAT3 activation and attenuation of STAT5 activation. These mice demonstrated severe neutropenia with accumulation of immature myeloid precursors in the bone marrow [12 ]. This study in particular lends credence to the vital role played by STAT3 in myeloid proliferation and maturation. Abrogation of p120 expression in G-CSF-treated bone marrow of G-CSF-R mutant (d715F) mice, which lack the ability to activate STAT3 and STAT5, but not in that of the wild-type or d715 G-CSF-R mice (in which activation of STAT3 and STAT5 is not affected), provides strong evidence of the role of STAT3 as well as STAT5 in mediating the expression of the p120 gene. The expression of p120 in other cell types may, however, be mediated by a STAT-independent mechanism, which may involve one or more of the putative cis-elements in the p120 promoter. In this context, we have demonstrated up-regulation of p120 expression in the monocyte-macrophage lineage in 12-O-tetradecanoylphorbol 13-acetate (TPA)-induced KG-1 cells (data not shown). Phorbol esters (TPA) are known to mediate their effects via protein kinase C and the AP-1 transcription factor.
As the G-CSF-induced expression of p120 appears to correlate with the induced binding of phospho-STAT3 and phospho-STAT5, it is to be noted that p120 is up-regulated marginally in the presence of IL-3 in 32D cells (Fig. 1) . IL-3 is also known to be associated with phosphorylation of the same STAT family members [36 ]. Our data therefore suggest that the levels of phospho-STAT3 and -5, which are increased markedly upon G-CSF induction in 32D cells, are responsible for the up-regulation of p120 expression.
p120, also known as NOL1, is the mammalian homologue of a yeast protein known as Nop2p, which is required for the methylation of 27S RNA and ultimately for the processing of 27S to mature 25S ribosomal RNA (rRNA) during ribosome biogenesis [37 ]. Mutations in the yeast Nop2p protein disrupt rRNA processing by abolishing methylation [37 ], and the mammalian homologue p120 has therefore been proposed to be a rRNA methyltransferase [38 ]. Human p120 has been shown to bind to 28S rRNA in vitro via its arginine-rich domain. It has also been shown to associate with 60S and 80S preribosomal particles [39 ]. p120 expression therefore appears to be a hallmark of rapidly growing cells. In accordance with our data, the expression of p120 has recently been shown to be down-regulated in highly enriched populations of normal human bone marrow representing the entire spectrum of the neutrophil development pathway [19 ].
The expression of p120 has been shown to be induced rapidly following growth stimulation [40 ], making the overexpression of p120 in tumor cells potentially significant. In fact, the levels of p120 have been correlated with tumor stage and prognosis [41 ]. Apoptosis of HeLa cells in the presence of antisense p120 suggests that p120 is required for growth of mammalian cells [42 ]. Our study demonstrates that overexpression of p120 in 32D cells results in partial neutrophil differentiation as judged by abrogation of LF gene expression. Although the mechanism underlying this observation remains speculative, it has been demonstrated previously that overexpressed p120 in NIH3T3 fibroblasts results in transformation of these cells [20 ]. We hypothesize that the early up-regulation of p120 following G-CSF induction of 32Dcl3 cells aids, in part, in boosting ribosome biogenesis to assist in the proliferation and differentiation aspects of the myeloid program elicited by G-CSF and that sustained expression of p120 in myeloid cells may contribute to the transformed phenotype by an undefined mechanism.
Received February 2, 2005; revised January 25, 2006; accepted January 31, 2006.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and (C/EBP and C/EBP) and CDP/cut in myeloid maturation induced lactoferrin gene expression Blood 101,3460-3468 ratio and granulocyte colony-stimulating factor Nat. Immunol. 4,1029-1036[CrossRef][Medline] Blood 98,897-905 in granulocyte-colony stimulating factor signaling pathway J. Biol. Chem. 280,12621-12629 and Stat3 ß in immature normal and leukemic human myeloid cells Blood 88,2442-2449
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
