QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Vol. 18, Issue 10, 3941-3951, October 2007

This Article Abstract Full Text (PDF) Supplemental Material All Versions of this Article: E07-03-0245v1 18/10/3941    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Qiu, J. Articles by Dong, S. Search for Related Content PubMed PubMed Citation Articles by Qiu, J. Articles by Dong, S.

Aberrant Chromatin Remodeling by Retinoic Acid Receptor Fusion Proteins Assessed at the Single-Cell Level

Jihui Qiu^*, Ying Huang, Guoqiang Chen, Zhu Chen, David J. Tweardy^*,, and Shuo Dong^*,

*Department of Medicine, Section of Infectious Disease, and Department of Molecular and Cellular Biology, Baylor College of Medicine, Houston, TX 77030; and Shanghai Institute of Hematology, Rui-Jin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai 200025, China

Submitted March 16, 2007; Revised July 9, 2007; Accepted July 20, 2007
Monitoring Editor: Wendy Bickmore

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Acute promyelocytic leukemia (APL) is characterized by specific chromosomal translocations, which generate fusion proteins such as promyelocytic leukemia (PML)-retinoic acid receptor (RAR) and promyelocytic leukemia zinc finger (PLZF)-RAR (X-RAR). In this study, we have applied lac operator array systems to study the effects of X-RAR versus wild-type RAR on large-scale chromatin structure. The targeting of these enhanced cyan fluorescent protein-lac repressor-tagged RAR-containing proteins to the gene-amplification chromosomal region by lac operator repeats led to local chromatin condensation, recruitment of nuclear receptor corepressor, and histone deacetylase complex. The addition of retinoic acid (RA) induced large-scale chromatin decondensation in cells expressing RAR; however, cells expressing X-RAR, especially PML-RAR, demonstrated insensitive response to this effect of all-trans retinoic acid (ATRA). Although we did not reveal differences in RA-dependent colocalization of either silencing mediator for retinoid and thyroid or steroid receptor coactivator (SRC)-1 with RAR versus X-RAR, the hormone-independent association between SRC-1 and X-RAR on the array has been identified. Rather, compared with cells expressing RAR, fluorescence recovery after photobleaching of live transfected cells, demonstrated decreased mobility of SRC-1 on the X-RAR–bound chromatin. Thus, the impaired ability of APL fusion proteins to activate gene transcription in response to ATRA corresponds to their reduced ability to remodel chromatin, which may link to their ability to impair the mobility of key nuclear receptor coregulators.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The protein products that arise from chromosomal translocations are critical factors in carcinogenesis, especially in leukemia. Chromosomal translocations that occur in leukemic cells frequently involve genes for transcription factors, leading to the formation of chimeric proteins capable of interfering in cell signaling pathways that control important aspects of growth, differentiation, and survival within hematological cells (Look, 1997). Acute promyelocytic leukemia (APL) comprises 10% of acute myeloid leukemia, and it is most commonly associated with the specific reciprocal chromosomal translocation involving the retinoic acid receptor (RAR) gene on chromosome 17 and the promyelocytic leukemia (PML) gene on chromosome 15 or the promyelocytic leukemia zinc finger (PLZF) gene on chromosome 11 (Melnick and Licht, 1999; Dong et al., 2003a). These APL-specific chromosomal abnormalities generate PML-RAR or PLZF-RAR chimeric genes, which have been demonstrated to play a key role in leukemogenesis by interfering with differentiation of bone marrow myeloid cells, causing arrest at the promyelocyte stage (Zelent et al., 2001).

Numerous studies have been performed with the goal of understanding how chimeric RAR-containing proteins contribute to leukemogenesis (Melnick and Licht, 1999; Zelent et al., 2001; Lallemand-Breitenbach et al., 2005). A critical contribution to this process made by the RAR fusion partners is the ability to mediate homodimerization or oligomerization of the chimeric protein (Kwok et al., 2006; Licht, 2006). However, the precise consequences of homodimerization in terms of leukemogenesis remain uncertain. One possibility supported by substantial in vitro data is that homodimerization results in binding to retinoic acid response element (RARE) sites required for normal myeloid differentiation, thereby competitively inhibiting binding to these sites by normal RAR/retinoid X receptor (RXR) heterodimers. An alternative explanation is that homodimerized chimeric RAR-containing proteins interfere with expression of genes critical for normal myeloid differentiation by mislocalizing within the nucleus, resulting in spatiotemporal sequestration of nuclear coregulators essential for their expression (Dong et al., 2004). Either spatiotemporal sequestration alone or competitive inhibition combined with the aberrant ligand-dependent interactions of X-RAR with nuclear coactivators and corepressors may result in impaired chromatin remodeling and repression of transcription (Lin et al., 2001).

To explore the possibility in vivo that chimeric RAR-containing fusion proteins contribute to leukemogenesis by impairing chromatin remodeling at the single-cell level, we used lac operator-lac repressor tethering systems, A03_1 and RRE_B1 cells, which contain multiple integrated copies of a high-affinity-binding site for the lac repressor, to directly visualize the effect of wild-type RAR compared with PML-RAR and PLZF-RAR (X-RAR) on large-scale chromatin dynamics. In A03_1 cell line system, multiple copies of the lac operator were engineered into the genome of Chinese hamster ovary (CHO) cells, and, together with the surrounding genomic sequences, they were amplified to produce a 90-Mb heterochromatic region or array (Robinett et al., 1996; Li et al., 1998). The expression of a chimeric protein containing the lac repressor and the activation domain of viral transcription factor VP16 in these A03_1 cells resulted in recruitment of the protein to the lac operator array and large-scale chromatin decondensation (Belmont et al., 1999a; Tumbar et al., 1999; Belmont, 2001).

Interrogation of the effects of wild-type RAR versus X-RAR gene products on chromatin structure by using this lac repressor-tethering system revealed that expression of either cyan fluorescent protein (CFP)-LacR–tagged wild-type RAR or X-RAR protein results in 60% condensation of the array. However, compared with wild-type RAR, cells expressing X-RAR proteins, in particular PML-RAR, were insensitive to normalization of the array volume when exposed to all-trans retinoic acid (ATRA). This aberrant response to ATRA did not correlate with abnormalities in the ability of X-RAR to modulate their association with nuclear coactivators or corepressors on the array; rather, it correlated with the reduced mobility of coactivators on the array compared with expression of wild-type RAR. Thus, the findings of this study support the hypothesis that expression of X-RAR results in aberrant ATRA-mediated chromatin remodeling, possibly due to reduced mobility of critical nuclear coactivators.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Reagents and Cell Lines
ATRA was obtained from Sigma-Aldrich (St. Louis, MO). A03_1 cell and RRE_B1 cells, which were kindly provided by Dr. Belmont (Robinett et al., 1996; Li et al., 1998), were maintained at 37°C with 5% CO₂ in customized F-12 Ham's media with 10% dialyzed fetal bovine serum (FBS; HyClone Labs) and methotrexate (0.3 µM for A03_1 cells and 10 µM for RRE_B1 cells), without phenol red, hypoxanthine and thymidine (Li et al., 1998; Stenoien et al., 2001). 293T cells were cultured in DMEM (Invitrogen, Carlsbad, CA) with 10% fetal bovine serum. HeLa cells were maintained in Opti-MEM I media (Invitrogen) with 4% FBS (Dong and Tweardy, 2002; Dong et al., 2004).

Vector Constructs
The CFP-LacR construct was made by inserting the LacR (lac repressor) sequence, amplified from polymerase chain reaction (PCR)-reaction, into the BglII-digested pECFP-C1 (Clontech, Mountain View, CA). The CFP-LacR-RAR, CFP-LacR-PML-RAR, and CFP-LacR-PLZF-RAR constructs were made by inserting the LacR fragment into constructs CFP-RAR, CFP-PML-RAR, and CFP-PLZF-RAR, respectively, described previously (Dong et al., 2004). The silencing mediator for retinoid and thyroid (SMRT) and histone deacetylase 3 (HDAC3) cDNA fragments removed from a pCMX expression vector (Chen and Evans, 1995) and pcDNA3 expression vector (Invitrogen) (Yang et al., 1996) were subcloned into pEYFP-C1 vector (Clontech) to make yellow fluorescent protein (YFP)-tagged SMRT and YFP-tagged HDAC3 expression vector, respectively. The YFP-tagged steroid receptor coactivator (SRC)-1 and RXR expression vectors (in pEYFP-C1; Clontech) have been reported previously (Stenoien et al., 2001; Dong et al., 2004). The YFP-tagged RAR, PML-RAR, and PLZF-RAR expression vectors were made from CFP-RAR, CFP-PML-RAR, and CFP-PLZF-RAR (Dong et al., 2004), respectively, by a swap involving placement of the cDNA for RAR, PML-RAR, and PLZF-RAR into the pEYFP-C1 expression vector (Clontech). The YFP sequence within YFP-tagged SMRT vector was replaced with mCherry sequence from mCherry expression vector (in pRSET-B; Invitrogen) (Shaner et al., 2005), to make cherry fluorescent protein (ChFP)-tagged SMRT expression vector. The fragment consisting of eight copies of lac operator (LacO) sequence, which was cut from vector p3216PECMS2 (Prasanth et al., 2005), was placed into pGL3-promoter vector (Promega, Madison, WI) with appropriate restriction enzymes, to make 8xLacO luciferase reporter construct. All plasmid constructs were confirmed by DNA sequencing and immunoblotting (Figure 1A).

Cell Transfection, Immunoblotting, Luciferase Assay, and Gel-Shift DNA-Binding Assays
For transient transfections, HeLa cells, 293T cells, A03_1 cells, and RRE_B1 cells were grown in six-well plates to 60–70% confluence. The cells were transiently transfected with the indicated expression vectors, reporter genes, or a combination by using GeneJuice (Novagen, Madison, WI) as reported previously (Dong and Tweardy, 2002). Twenty-four to 48 h after transfection, cells were lysed in lysis buffer for either immunoblotting or luciferase assay as described previously (Dong et al., 1996; Dong and Tweardy, 2002; Dong et al., 2003b, 2004). Equivalent amounts of protein were electrophoresed on 7.5% SDS polyacrylamide gels and transferred to polyvinylidene difluoride membrane developed using antibody against RAR (C-20; Santa Cruz Biotechnology, Santa Cruz, CA) or green fluorescent protein (GFP) (Dong et al., 2003b).

For the gel-shift DNA-binding assays, CFP-tagged or CFP-LacR–tagged RAR, PML-RAR, and PLZF-RAR proteins were expressed in 293T cells. Gel shift assays were performed as described previously (Dong et al., 1996; Dong and Tweardy, 2002) using 20 µg of whole cell extracts of 293T cells transfected with the indicated constructs. Briefly, cell extracts were preincubated for 10 min at room temperature in the 20 µl of reaction buffer, which contained 20 mM HEPES, pH 7.9, 0.5 mM EDTA, 2 mM dithiothreitol, 10% glycerol, 50 mM KCl, and 1 µg of poly(dI-dC). Radiolabeled lac operator or RARE (DR5G) duplex oligonucleotide was added and incubated for 50 min. The lac operator double-stranded oligonucleotide contains the sequence (top strand) 5'-TGTGGAATTGTGAGCGGATAACAATT-3' (Sasmor and Betz, 1990). The sequences of the RARE is as follows: 5'-GGGTAGGGGTCACCGAAAGGTCACTCG-3' (Dong et al., 1996). One microgram of antibody against RAR was included in the reaction for supershift experiments (Figure 1, C and D). Protein–DNA complexes were separated on 5% polyacrylamide gels equilibrated in 0.25x Tris borate-EDTA. Gels were dried, exposed, and analyzed on the PhosphorImager (GE Healthcare, Little Chalfont, Buckinghamshire, United Kingdom).

Fluorescence Microscopy and Fluorescence Recovery after Photobleaching (FRAP)
For fixed cell experiments, A03_1 and RRE_B1 cells were plated onto poly-D-lysine–coated coverslips in 24-wells plate and cultured in customized F-12 Ham's media with 10% charcoal-dextran treated FBS (Hyclone Laboratories). After 24 h, cells were transfected with the indicated CFP-, YFP-, or ChFP-labeled constructs using GeneJuice (Novagen) as described previously (Dong et al., 2004). Forty-eight hours after transfection, cells were fixed in 4% formaldehyde in PEM buffer [80 mM piperazine-N,N'-bis(2-ethanesulfonic acid), pH 6.8, 5 mM EGTA, and 2 mM MgCl₂) and quenched in 1 mg/ml NaBH₄ (Sigma-Aldrich) in PEM buffer. The A03_1 and/or RRE_B1 cells were stained with 1 µg/ml 4',6'-diamidino-2-phenylindole (DAPI; Sigma-Aldrich) for 5 min before mounting in Slow Fade reagent (Invitrogen). Deconvolution microscopy was performed on a Zeiss Axiovert S100 TV microscope (Carl Zeiss, Thornwood, NY) by using DeltaVision restoration and microscopy system (Applied Precision, Issaquah, WA) as described previously (Stenoien et al., 2000). A series of Z-sections were imaged and deconvolved with the DeltaVision constrained iterative algorithm. For quantitative measurement, the DeltaVision software was used to measure the volume of the three-dimensional images of the LacO arrays of A03_1 cells, which were built by the software using the series of Z-sections of each array. The lac operator arrays within RRE_B1 cells were much more extended, and instead of measuring the volume of the array, the array occupied chromatin area was measured using the Z-section containing the largest array area. Both the volume and area measurements were exported, analyzed, and graphed in Excel (Microsoft, Redmond, WA).

For dual-FRAP analysis, cells were plated onto Delta T dishes (Bioptechs, Butler, PA) 24 h before transfection at a concentration of 2 x 10⁵ cells per dish. The expression of CFP- or YFP-tagged plasmids was performed using GeneJuice (Novagen) as described previously (Dong et al., 2004; Qiu et al., 2006). FRAP was performed on the heated stage of an LSM 510 confocal laser scanning microscope (Carl Zeiss) with a 63x, 1.4 numerical aperture Plan-Apochromat oil immersion objective. For both CFP and YFP fluorescent signals, the bleach was accomplished with the laser set at 458 and 514 nm and at 70% laser output for 30 iterations of a small circle of each lac operator array in A03_1 cells. Simultaneous images corresponding to the CFP and YFP fluorescence were obtained before and at different time intervals after the bleach by using the multitracking function of the microscope (typically 5 prebleach and 100 postbleach frames). Fluorescent intensities of regions of interest (ROI) were measured with LSM software, and exported to Excel for analysis. Fluorescent intensities of the ROI were corrected for background and normalized to the mean of the last three prebleach values by using the following equationfor each time point: (ROI_bleach,t – ROI_background)/(ROI_{bleach, prebleach} – ROI_background). The normalized recovery curve was adjusted so that the lowest point in the curve was assigned a value of 0, and the highest point was assigned a value of 1. The t_1/2 of observed recovery was defined as the time required for reaching half-maximum recovery, and it was calculated from the normalized recovery curve.

Statistical Analysis
LacO array volume and area data were transformed by taking logarithms to stabilize variances and reduce skewness. Summary statistics (geometric means and 95% confidence intervals) on the raw scale were computed by back-transforming the means and 95% confidence intervals of the transformed data. One-way analysis of variance (ANOVA) was used to determine whether there were any differences between groups, followed by specific linear contrasts to identify which groups differed. Analyses were conducted with SAS, version 9.1 (SAS Institute, Cary, NC).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Generation of Functional CFP-LacR–tagged Wild-Type RAR and X-RAR
To study the intracellular function of wild-type RAR, PML-RAR, and PLZF-RAR in the context of chromatin at the single-cell level, we created vectors expressing CFP-LacR (lac repressor) chimeras fused to the N-terminal ends of RAR, PML-RAR, and PLZF-RAR. Immunoblot analysis of whole cell extracts from transfected 293T cells showed that all CFP-LacR–tagged constructs encoded proteins of expected size (Figure 1A). To determine whether these CFP-LacR–tagged proteins could bind to lac operator DNA sequence contained within the lac operator array in A03_1 and RRE_B1 cell lines (Li et al., 1998), we performed gel-shift assays by using radiolabeled lac operator duplex oligonucleotide. As expected, all of the CFP-LacR–tagged proteins tested bound to lac operator duplexes and could be supershifted by antibody against RAR (Figure 1B). To be certain that the addition of CFP-LacR to RAR, PML-RAR, or PLZF-RAR did not significantly alter the functional characteristics of the RAR-containing portion of the construct, we examined the DNA binding activity of each using gel-shift assays with RARE (DR5G); each bound to RARE not only as a homo-oligomer, but also as a hetero-oligomer with RXR, similar to results reported by us and other investigators with each of the untagged counterparts (Figure 1, C and D) (Dong and Tweardy, 2002; Zeisig et al., 2007). In addition, these RAR-related hetero-oligomer and homo-oligomer complexes were further studied using the LacO array, A03_1 cell system. A03_1 cells were cotransfected with CFP-LacR–tagged RAR, PML-RAR, and PLZF-RAR plus their YFP-tagged counterpart or YFP-tagged RXR (Supplemental Figure 1). As expected, YFP-tagged RXR bound to all of CFP-LacR–tagged RAR constructs bound to the array (Supplemental Figure 1A). Also, YFP-tagged X-RAR (PML-RAR and PLZF-RAR) were able to form homo-oligomer complexes with their respective CFP-LacR–tagged counterpart bound to the array (Supplemental Figure 1, C and D) in contrast to YFP-tagged wild-type RAR, which did not bind to CFP-LacR-RAR bound to arrays (Figure 1B). These results confirm at the single-cell level previous in vitro studies by us and others (Dong et al., 1996; Dong and Tweardy, 2002; Kwok et al., 2006; Zeisig et al., 2007).

View larger version (78K): [in this window] [in a new window]   Figure 1. Characterization of CFP-LacR-containing constructs. (A) Western blot analysis of CFP-LacR–tagged RAR and X-RAR (PML-RAR and PLZF-RAR) proteins expressed in 293T cells by using antibody against RAR. (B–D) Gel-shift assay analysis of whole cell extracts of 293T cells transiently transfected with CFP-LacR-RAR or CFP-LacR-X-RAR (PML-RAR and PLZF-RAR) alone or plus RXR, as indicated, were performed using lac operator (B) or RARE sequence (C and D). The filled arrowheads indicate the positions of the CFP-LacR-X-RAR homodimer/RARE complex. The empty arrowheads indicate the positions of the CFP-LacR-RAR/RXR heterodimer/RARE complex or CFP-LacR-X-RAR/RXR/RARE complex. Extracts of 293T cells transiently transfected with CFP-PML-RAR and CFP-PLZF-RAR were included as size controls (C). The RAR antibody was used for supershift experiments as indicated (B and D). (E) Luciferase activity in lysates of HeLa cells transiently transfected with CFP-LacR–tagged RAR or X-RAR proteins incubated in the absence or presence of ATRA at 10–6 M. The amounts of plasmid DNA used per well were 250 ng of 8xLacO luciferase reporter vector, 500 ng of CFP-LacR–tagged expression vector and 200 ng of -galactosidase expression vector as transfection control. Luciferase activity within cell lysates was measured in a luminometer, expressed in arbitrary units and normalized to the transfection control. Data presented are the mean ± SD of at least three independent experiments. (F) Fluorescence microscopy of representative cells demonstrating nuclear localization and distribution of CFP-tagged or CFP-LacR–tagged RAR and X-RAR protein within transfected HeLa cells or A03_1 cells transiently transfected with the indicated vector construct.

Effect of CFP-LacR-RAR and CFP-LacR-X-RAR Binding on the lac Operator Array
To assess the effect of wild-type RAR versus PML-RAR or PLZF-RAR on large-scale mammalian chromatin structure at the single-cell level, we made use of a CHO cell line, A03_1, in which multiple copies of the lac operator were engineered to produce an integrated 90-Mb heterochromatic array within the genome (Li et al., 1998). The molecular organization of this array consists of 400-kb repeats of the 14-kb vector insert that contains the lac operator repeat and the dihydrofolate reductase selectable marker (Robinett et al., 1996; Li et al., 1998). This lac operator array system has been applied by Dr. Belmont's group for studying the transcriptional responses of chromatin structure during transcriptional activation with enhanced green fluorescent protein (EGFP)-lac repressor (LacR)-VP16 (Tumbar et al., 1999), which is a potent viral transcription activator protein, and EGFP-lac repressor-estrogen receptor (ER) (Nye et al., 2002). Targeting this VP16 fusion protein to the lac operator array in A03_1 cells resulted in visualization of chromatin decondensation within 15 min, accompanied by the recruitment of histone acetyltransferases and by histone hyperacetylation (Tumbar et al., 1999). In our studies, CFP-LacR transiently transfected into A03_1 cells distributed almost exclusively on the array consistent with high-affinity binding of LacR to the LacO and the previous results of others (Li et al., 1998; Stenoien et al., 2001). The array volume was 17.25 µm³ (95% confidence interval [CI] = 12.26–24.31) in cells untreated with ATRA, and it did not change in cells treated with ATRA at 10^–7 or 10^–6 M (Figure 2). Similar to CFP-LacR, CFP-LacR-RAR, and CFP-LacR-X-RAR (CFP-LacR-PML-RAR, and CFP-LacR-PLZF-RAR) proteins transiently expressed in A03_1 cells each were distributed almost exclusively on the array. However, in contrast to CFP-LacR, the array volumes in cells transfected with CFP-LacR-RAR (7.19 µm³; 95% CI = 5.91–8.76), CFP-LacR-PML-RAR (7.38 µm³; 95% CI = 6.35–8.58), and CFP-LacR-PLZF-RAR (7.42 µm³; CI = 6.22–8.85) were reduced in each instance by 60% within cells untreated with ligand (p < 0.001 for each). To confirm these results in a cell line containing a lac operator array that is more euchromatin-like, we repeated the experiment using the RRE_B1 cell line. The RRE_B1 cell line, like A03_1, is derived from CHO cells, and it contains amplified genomic domains consisting of lac operator binding sites, in an extended, often fibrillar conformation (Robinett et al., 1996; Verschure et al., 2005). Previous studies that targeted an EGFP-LacR-HP1 construct onto this array caused local chromatin condensation (Verschure et al., 2005). Identical to our results with A03_1 cells, we observed array condensation in RRE_B1 cells transfected with each CFP-LacR-RAR–containing construct compared with CFP-LacR (Supplemental Figure 2). Thus, results in both array-containing cell lines are consistent with the hypothesis that wild-type RAR, PML-RAR, and PLZF-RAR each recruit corepressors with histone deacetylase activity (Figure 3B).

View larger version (44K): [in this window] [in a new window]   Figure 2. Effect of CFP-LacR-RAR or CFP-LacR-X-RAR binding on the lac operator array volume within A03_1 cells. (A) A03_1 cells were transfected with CFP-LacR–tagged RAR and X-RAR, followed with or without ATRA treatment. The volume of the arrays was determined as described under Materials and Methods and presented as the geometric mean ± 95% CI. (B) Representative images are shown for arrays bound by CFP-LacR (first row panels), CFP-LacR-RAR (second row panels), CFP-LacR-PML-RAR (third row panels), and CFP-LacR-PLZF-RAR (bottom row panels) within A03_1 cells incubated without ATRA (left column panels), with ATRA at 10–7 M for 3 h (middle column panels) and with ATRA at 10–6 M for 3 h (right column panels). The number of cells examined, the geometric mean (GM), and a histogram showing the distribution of array volumes is presented below each representative image. The nuclei were stained with DAPI (blue).

View larger version (181K): [in this window] [in a new window]   Figure 3. Effect of CFP-LacR-RAR and CFP-LacR-X-RAR expression and ATRA on the localization and distribution of nuclear receptor corepressor SMRT, HDAC3, and nuclear receptor coactivator SRC-1 within A03_1 cells. Cells cotransfected with YFP-SMRT (A), YFP-HDAC3 (B), or YFP-SRC-1 (C), and CFP-LacR-RAR or CFP-LacR-X-RAR was subjected to deconvolution microscopic analysis after treatment with ATRA for 3 h at the indicated concentrations. Nuclei were visualized by DNA staining with DAPI (blue). DAPI and CFP, YFP, or merged CFP and YFP images from representative nuclei are shown in the top left, top right, and bottom left panels. The percentage of cells (y-axis) that showed complete binding (yellow bar), partial binding with partial diffusion (purple bar), and diffusion (blue bar) are shown in the bottom right panel. The number at the top of bar graphs represent the number of nuclei examined for each group; the data presented are the combined results of three independent experiments.

We and others have reported that the protein products of X-RAR fusion genes were degraded within cells upon retinoic acid treatment (Koken et al., 1999; Melnick and Licht, 1999; Dong and Tweardy, 2002). To assess whether differences in protein stability participated in the variable ATRA-induced increase in array volumes, we performed immunoblotting analysis of whole cell extracts from cells transiently transfected with CFP-LacR–tagged RAR and X-RAR before and after ATRA treatment (Supplemental Figure 3). The level of each of the RAR-containing proteins in A03_1 cells incubated with ATRA was virtually unchanged, indicating that ATRA-mediated changes in protein stability did not contribute to changes in array volumes.

Effect of CFP-LacR-RAR and CFP-LacR-X-RAR Expression and ATRA on the Distribution of Nuclear Receptor Corepressor SMRT and Coactivator SRC-1 within A03_1 Cells
The lac operator array A03_1 cell system has been used previously to study nuclear receptor biology (Stenoien et al., 2001), and it revealed interactions between the ER and both SRC-1 and CREB-binding protein (CBP) on the array within A03_1 cells. Transient triple-transfection of A03_1 cells with CFP-LacR–tagged RAR or X-RAR plus YFP-SRC-1 and ChFP-SMRT construct established that each nuclear receptor coregulator demonstrated ATRA-dependent interactions with each CFP-LacR–tagged RAR-containing construct on the lac operator array in the expected manner, confirming earlier studies at the single-cell level (Supplemental Figure 4).

To determine whether the tagged RAR-containing constructs recruited nuclear corepressors to the array and whether this contributed to ATRA-induced changes in the lac operator array volume, we imaged A03_1 cells coexpressing YFP-SMRT and each of the CFP-LacR–containing constructs. Coexpression of YFP-SMRT and CFP-LacR in cells untreated with ATRA revealed CFP-LacR localized to the array, whereas YFP-SMRT distributed diffusely throughout the nucleus, and it did not colocalize with CFP-LacR on the array (data not shown). In contrast, coexpression of YFP-SMRT and CFP-LacR-RAR in A03_1 cells untreated with ATRA revealed virtually all the YFP-SMRT colocalized with CFP-LacR-RAR on the array (Figure 3A) consistent with findings summarized above of smaller array volumes in cells expressing CFP-LacR–tagged RAR versus CFP-LacR alone. Treatment of these cotransfected cells with ATRA at either 10^–7 or 10^–6 M resulted in partial or complete dissociation of YFP-SMRT from the array, which correlated with the increase in array volume observed in cells transfected with CFP-LacR-RAR and treated with ATRA at these concentrations (Figure 3A). Coexpression of YFP-SMRT with either CFP-LacR-PML-RAR or CFP-LacR-PLZF-RAR in A03_1 cells untreated with ATRA resulted in virtually all the YFP-SMRT colocalized with CFP-LacR-X-RAR on the array (Figure 3A). Unexpectedly, treatment of both sets of cotransfected cells with ATRA at either 10^–7 or 10^–6 M revealed partial or complete dissociation of YFP-SMRT from the array, similar to the results obtained with ATRA-treated cells cotransfected with YFP-SMRT and CFP-LacR-RAR. Thus, YFP-SMRT associated with RAR-containing constructs on the arrays and this association correlated to array volumes in the absence of ATRA. Furthermore, ATRA-induced dissociation of YFP-SMRT from the array correlated well with the ATRA-induced increase in array volume in cells expressing CFP-LacR–tagged wild-type RAR; however, this was not the case for cells expressing either CFP-LacR-PML-RAR or CFP-LacR-PLZF-RAR.

HDAC3 is known to form SMRT-HDAC3 functional complexes (Li et al., 2002). Examination of A03_1 cells cotransfected with the CFP-LacR construct revealed HDAC3 present at very low levels in the amplified chromosomal region (data not shown). A03_1 cells cotransfected with YFP-HDAC3 and either CFP-LacR-RAR or CFP-LacR-X-RAR demonstrated virtually all of the YFP-HDAC3 partially colocalized to the lac operator array in the absence of ATRA (Figure 3B). After addition of ATRA, YFP-HDAC3 completely dissociated from the array. Thus, YFP-HDAC3 followed the identical pattern of array colocalization as YFP-SMRT.

To determine whether the tagged RAR-containing constructs recruited nuclear coactivators to the array and whether this contributed to changes in the lac operator array volume, we imaged A03_1 cells coexpressing YFP-SRC1 and CFP-LacR, CFP-LacR-RAR, CFP-LacR-PML-RAR or CFP-LacR-PLZF-RAR. Coexpression of YFP-SRC-1 and CFP-LacR in A03_1 cells revealed CFP-LacR localized to the array, whereas YFP-SRC-1 distributed diffusely throughout the nucleus; no colocalization of YFP-SRC-1 with CFP-LacR on the array was observed (data not shown). Coexpression of YFP-SRC-1 and CFP-LacR-RAR in A03_1 cells untreated with ATRA revealed CFP-LacR-RAR localized to the array, and most of the YFP-SRC-1 distributed diffusely throughout the nucleus (Figure 3C). Examination of cotransfected cells treated with ATRA at either 10^–7 or 10^–6 M revealed nearly complete localization of YFP-SRC-1 on the array, which correlated well with the increase in array volume observed in cells transfected with CFP-LacR-RAR summarized above (Figure 3C). Coexpression of YFP-SRC-1 with either CFP-LacR-PML-RAR or CFP-LacR-PLZF-RAR in A03_1 cells untreated with ATRA resulted in YFP-SRC-1 distributed diffusely throughout the nucleus. However, the increasing complete binding of coactivator SRC-1 into X-RAR-bound chromatin area (23% PML-RAR– and 44% PLZF-RAR–expressing cells, in contrast to only 7% wild-type RAR-expressing cells) was observed, indicating a ligand-independent association between SRC-1 and X-RAR (Figure 3C). Treatment of both sets of cotransfected cells with ATRA at either 10^–7 or 10^–6 M revealed nearly complete localization of YFP-SRC-1 on the array, again similar to the results with cells cotransfected with YFP-SRC-1 and CFP-LacR-RAR. Thus, ATRA treatment of cotransfected A03_1 cells induced association of SRC-1 with all tagged RAR constructs on the array. Furthermore, although ATRA-induced colocalization of YFP-SRC-1 with CFP-LacR–tagged constructs on the arrays correlated with an increase in array volume in cells expressing tagged wild-type RAR, this was not the case for either tagged PML-RAR or PLZF-RAR. Thus, when assessed at the single-cell level, neither association nor dissociation of either SMRT/HDAC3 or SRC-1 was sufficient to explain differences in ATRA-induced array volumes between wild-type RAR and X-RAR.

Nuclear Mobility of YFP-SRC-1 and YFP-SMRT in A03_1 Cells Expressing CFP-LacR-RAR or CFP-LacR-X-RAR
We previously demonstrated that X-RAR protein expression resulted in aberrant localization and reduced mobility of key nuclear receptor coregulators (Dong et al., 2004). To assess whether reduced mobility of nuclear receptor corepressors contributed to aberrant ATRA-mediated chromatin remodeling in cells expressing X-RAR, we performed dual-FRAP of A03_1 cells cotransfected with YFP-SMRT and CFP-LacR–tagged wild-type RAR or X-RAR. In the absence of ATRA, the mobility for YFP-SMRT is very slow (t_1/2 > 10 min; data not shown). However, as shown in Figure 3A, treatment with ATRA at 10^–7 M for 3 h resulted in nearly complete dissociation of YFP-SMRT from CFP-LacR–tagged RAR and X-RAR on the array. To obtain measurable recovery times, cells were examined after treatment with ATRA at this concentration for 30 min; nuclei selected for analysis were those in which the YFP signal was both on and off the array. After the 2-s bleach, there was little or no recovery of the CFP signal over 99 s (Figure 4A) or even over 20 min (data not shown), indicating that the LacR-containing proteins were essentially immobilized on the array due to the high-affinity binding of the lac repressor domain to the lac operator sites within the array. The mobility of YFP-SMRT localized on the CFP-LacR-RAR–bound array assessed by its recovery t_1/2 was 13.57 ± 4.53 s, whereas the mobility of nonarray bound YFP-SMRT was very fast with recovery t_1/2 = 0.69 ± 0.14 s (Table 1). The mobility of YFP-SMRT localized on the CFP-LacR-PML-RAR–bound array was essentially identical to that for CFP-LacR-RAR (t_1/2 = 13.50 ± 4.21 s). In contrast, the mobility of YFP-SMRT on the CFF-LacR-PLZF-RAR-bound array was reduced by 68% (t_1/2 = 22.76 ± 7.87 s; p < 0.001; Figure 4A and Table 1) compared with YFP-SMRT on the CFR-LacR-RAR–bound array, indicating that YFP-SMRT bound more tightly to PLZF-RAR than to wild-type RAR. We demonstrated reduced intranuclear mobility of SMRT bound to PLZF-RAR, which most likely reflects increased binding affinity of SMRT to PLZF-RAR compared with either wild-type RAR or PML-RAR. The biochemical and molecular basis for a higher affinity interaction between SMRT and PLZF-RAR in vitro has been attributed to the PLZF POZ domain (Dong et al., 1996; Lin et al., 1998). Although consistent with these in vitro findings, our in vivo results do not explain the resistance to the ATRA-induced increase in array volumes in cells expressing both X-RAR, in particular PML-RAR (Figure 2).

View larger version (36K): [in this window] [in a new window]   Figure 4. Nuclear mobility of YFP-SRC-1 and YFP-SMRT in A03_1 cells expressing CFP-LacR-RAR or CFP-LacR-X-RAR. Dual-FRAP analysis was performed in A03_1 cells cotransfected with YFP-SMRT and CFP-LacR-RAR (top), YFP-SMRT and CFP-LacR-PML-RAR (middle), and YFP-SMRT and CFP-LacR-PLZF-RAR (bottom) (A) after incubation with ATRA at 10–7 M for 30 min and in A03_1 cells cotransfected with YFP-SRC-1 and CFP-LacR-RAR (top), YFP-SRC-1 and CFP-LacR-PML-RAR (middle), and YFP-SRC-1 and CFP-LacR-PLZF-RAR (bottom) (B) after incubation with ATRA at 10–7 M for 30 min. Images show single Z-sections, and they were obtained before bleaching and at the indicated time points after bleaching.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
We used two integrated lac operator array cell systems (Belmont et al., 1999b) to examine the hypothesis at the single-cell level that RAR-containing APL fusion proteins remodel chromatin aberrantly compared with wild-type RAR. Fluorescent microscopic examination of arrays after expression of CFP-LacR–tagged, RAR-containing proteins demonstrated that each of the tagged proteins bound to arrays resulting in condensation of the array volume to less than half that of arrays bound by lac repressor alone. All-trans retinoic acid treatment resulted in array decondensation in cells expressing tagged wild-type RAR. However, cells expressing tagged X-RAR were insensitive to the array-decondensing effect of ATRA at different level (Figure 2).

Eukaryotic cells use multiple steps at several structural levels to effectively package their genome within the nucleus. It is now well established that unpacking or remodeling of chromatin structure locally is a critical step in the initiation of multiple chromosome functions notably transcription (Belmont et al., 1999a). Whereas intense research in the past decade has provided a wealth of information regarding the biochemical basis for ligand-dependent transcriptional properties of RAR (Chambon, 1995), much less is known about reorganization of local chromatin structure by RAR protein and how this compares with X-RAR oncogenic proteins. Our studies are the first to report the effects of RAR-containing fusion proteins on chromatin structure and nuclear receptor coregulator recruitment in vivo at the single-cell level. The finding that expression of LacR-tagged RAR within lac repressor-tethering cell systems in the absence of ligand results in lac operator array condensation is consistent with in vitro evidence that, in the absence of ligand, RAR can recruit corepressor with histone deacetylase activity (Melnick and Licht, 1999). Cotransfection of lac operator array-containing cells with CFP-LacR-RAR and YFP-SMRT or YFP-HDAC3 confirmed this at the single-cell level. We also demonstrated that lac operator arrays bound by wild-type RAR decondensed after incubation of cells in ATRA, consistent with in vitro evidence that ligand addition results in release of corepressors and recruitment of coactivators with histone acetylase activity (Licht, 2001, 2006; Zelent et al., 2001; Melnick et al., 2002). Examination of lac operator array-containing cells cotransfected with CFP-LacR-RAR and YFP-SMRT or YFP-SRC-1 as well as cells transfected with all three constructs revealed that the addition of ATRA resulted in both the release of SMRT from the RAR-bound lac operator array and the recruitment of coactivator SRC-1 at the single-cell level.

These studies provide intriguing insight at the single-cell level regarding how APL fusion proteins may function as oncogenes. Similar to wild-type RAR, expression of both PML-RAR and PLZF-RAR proteins in lac repressor-tethering systems resulted in lac operator array condensation. In the absence of ligand, each also was shown to recruit nuclear receptor corepressor SMRT to the array, thereby confirming in vivo at the single-cell level previous in vitro studies that demonstrated the ability of corepressor to bind to APL fusion proteins in the absence of ligands (Lin et al., 1998; Dong and Tweardy, 2002). However, somewhat unexpectedly, based on previously reported responses in APL patients to ATRA (Melnick and Licht, 1999), cells containing PML-RAR were more, not less, insensitive to ATRA-induced chromatin decondensation compared with cells containing PLZF-RAR. In addition, the complete dissociation of corepressor SMRT from PLZF-RAR-bound chromatin area in retinoic acid (RA)-treated cells was observed, which was in line with the RA-dependent transactivation effect by this tagged PLZF-RAR, although we found the decreased SMRT nuclear mobility with chromatin-bound PLZF-RAR in comparison with that with wild-type RAR or PML-RAR. The integration of our data with earlier findings, in which RA treatment led to PLZF-RAR degradation (Koken et al., 1999; Dong and Tweardy, 2002), strongly suggests that PLZF-RAR chimeric protein responds well to the treatment of retinoic acid. However, these effects were not accompanied by terminal differentiation or apoptosis of APL cells (Koken et al., 1999). These findings suggest that the molecular basis for differences in clinical response to ATRA for these two forms of APL may lie beyond differences in ATRA-mediated changes in large-scale chromatin remodeling. Combined with the transgenic studies (He et al., 2000), in which, Pandolfi's group found that double transgenic mice with RAR-PLZF and PLZF-RAR but PLZF-RAR alone developed leukemia with typical APL features, our single-cell studies highlighted the importance of reciprocal RAR-PLZF fusion protein, whose mRNA was existed in all t(11;17) APL patients (Melnick and Licht, 1999), in leukemogenesis as well as clinical RA resistance.

There is growing evidence that the dynamic exchange of proteins on chromatin is essential for transcriptional activators to gain access to chromatin and that controlling the exchange rate of a protein on chromatin might contribute to regulation of gene expression (Misteli, 2001). Although we found evidence of ATRA dependence of SMRT and SRC-1 interactions with APL fusion proteins in vivo using lac repressor-tethering system as reported previously in vitro (Lin et al., 1998; Dong and Tweardy, 2002), no differences were observed in the ATRA concentration dependence of the interactions of either SMRT or SRC-1 with X-RAR compared with wild-type RAR that would explain differences in ATRA-induced array decondensation. However, we found that chromatin-bound X-RAR (PML-RAR and PLZF-RAR) fusion proteins associated with nuclear receptor coactivator SRC-1 in a retinoic acid-independent way. Furthermore, the impaired ATRA-dependent decondensation in cells expressing X-RAR correlated with reduced intranuclear mobility of SRC-1. Reduced intranuclear mobility presumably is due to increased binding affinity of SRC-1 to PML-RAR and PLZF-RAR. In line with our studies, very recently, Dr. Kao's group demonstrated that PML-RAR oncogenic protein aberrantly interacts with nuclear receptor coactivator SRC-3 and CBP in a hormone-independent manner, which correlate with the ability of PML-RAR to decrease the transcriptional activation by the glucocorticoid receptor and Notch signaling pathway (Reineke et al., 2007). By binding nuclear receptor coactivators such as SRC-1, SRC-3, and CBP with higher affinity than wild-type RAR, APL fusion proteins may contribute to transcriptional alterations resulting in diseases by reducing the availability of these coactivators to bind to sites on chromatin important for maintaining the normal transcriptome of the cell.

	ACKNOWLEDGMENTS

 
We thank Dr. Andrew S. Belmont (University of Illinois, Champagne-Urbana, IL) for kindly providing the A03_1 and RRE_B1 cell lines and sharing reagents; Dr. David L. Spector (Cold Spring Harbor Laboratory) for vector p3216PECMS2 with which we can facilitate to make 8xLacO luciferase reporter vector; and Dr. Roger Y. Tsien (University of California at San Diego) for mCherry expression vector. We also appreciate the technical assistance and helpful discussions of Dr. Adam Szafran, Maureen Mancini, and Dr. Michael A. Mancini (Department of Molecular and Cellular Biology, Baylor College of Medicine). This work was supported in part by R21 grant CA-119080 from the National Cancer Institute (NCI) and National Natural Science Foundation of China 30470360 (to S.D.); grant IRG 93-034-06 from the American Cancer Society (to S.D.); Chao Award from the Department of Medicine of Baylor College of Medicine (to S.D.); and R01 grant CA-72261 (to D.J.T.) and CA-86430 (to D.J.T.) from NCI.

	Footnotes

 
This was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E07-03-0245) on August 1, 2007.

The online version of this article contains supplemental material at MBC Online (http://www.molbiolcell.org).

Address correspondence to: Shuo Dong (sdong{at}bcm.tmc.edu) or David J. Tweardy (dtweardy{at}bcm.tmc.edu)

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Belmont, A. S. (2001). Visualizing chromosome dynamics with GFP. Trends Cell Biol 11, 250–257.[CrossRef][Medline]

Belmont, A. S., Dietzel, S., Nye, A. C., Strukov, Y. G., and Tumbar, T. (1999a). Large-scale chromatin structure and function. Curr. Opin. Cell Biol 11, 307–311.[CrossRef][Medline]

Belmont, A. S., Li, G., Sudlow, G., and Robinett, C. (1999b). Visualization of large-scale chromatin structure and dynamics using the lac operator/lac repressor reporter system. Methods Cell Biol 58, 203–222.[Medline]

Chambon, P. (1995). The molecular and genetic dissection of the retinoid signaling pathway. Recent Prog. Horm. Res 50, 317–332.[Medline]

Chen, J. D., and Evans, R. M. (1995). A transcriptional co-repressor that interacts with nuclear hormone receptors. Nature 377, 454–457.[CrossRef][Medline]

Dong, S., Chen, S. J., and Tweardy, D. J. (2003a). Cross-talk between retinoic acid and STAT3 signaling pathways in acute promyelocytic leukemia. Leuk. Lymphoma 44, 2023–2029.[CrossRef][Medline]

Dong, S., Qiu, J., Stenoien, D. L., Brinkley, W. R., Mancini, M. A., and Tweardy, D. J. (2003b). Essential role for the dimerization domain of NuMA-RARalpha in its oncogenic activities and localization to NuMA sites within the nucleus. Oncogene 22, 858–868.[CrossRef][Medline]

Dong, S., Stenoien, D. L., Qiu, J., Mancini, M. A., and Tweardy, D. J. (2004). Reduced intranuclear mobility of APL fusion proteins accompanies their mislocalization and results in sequestration and decreased mobility of retinoid X receptor alpha. Mol. Cell. Biol 24, 4465–4475.[Abstract/Free Full Text]

Dong, S., and Tweardy, D. J. (2002). Interactions of STAT5b-RARalpha, a novel acute promyelocytic leukemia fusion protein, with retinoic acid receptor and STAT3 signaling pathways. Blood 99, 2637–2646.[Abstract/Free Full Text]

Dong, S. et al. (1996). Amino-terminal protein-protein interaction motif (POZ-domain) is responsible for activities of the promyelocytic leukemia zinc finger-retinoic acid receptor-alpha fusion protein. Proc. Natl. Acad. Sci. USA 93, 3624–3629.[Abstract/Free Full Text]

He, L., Bhaumik, M., Tribioli, C., Rego, E. M., Ivins, S., Zelent, A., and Pandolfi, P. P. (2000). Two critical hits for promyelocytic leukemia. Mol. Cell 6, 1131–1141.[CrossRef][Medline]

Koken, M. H., Daniel, M. T., Gianni, M., Zelent, A., Licht, J., Buzyn, A., Minard, P., Degos, L., Varet, B., and de The, H. (1999). Retinoic acid, but not arsenic trioxide, degrades the PLZF/RARalpha fusion protein, without inducing terminal differentiation or apoptosis, in a RA-therapy resistant t(11;17)(q23;q21) APL patient. Oncogene 18, 1113–1118.[CrossRef][Medline]

Kwok, C., Zeisig, B. B., Dong, S., and So, C. W. (2006). Forced homo-oligomerization of RARalpha leads to transformation of primary hematopoietic cells. Cancer Cell 9, 95–108.[CrossRef][Medline]

Lallemand-Breitenbach, V., Zhu, J., Kogan, S., Chen, Z., and de The, H. (2005). Opinion: how patients have benefited from mouse models of acute promyelocytic leukaemia. Nat. Rev. Cancer 5, 821–827.[CrossRef][Medline]

Li, G., Sudlow, G., and Belmont, A. S. (1998). Interphase cell cycle dynamics of a late-replicating, heterochromatic homogeneously staining region: precise choreography of condensation/decondensation and nuclear positioning. J. Cell Biol 140, 975–989.[Abstract/Free Full Text]

Li, J., Lin, Q., Wang, W., Wade, P., and Wong, J. (2002). Specific targeting and constitutive association of histone deacetylase complexes during transcriptional repression. Genes Dev 16, 687–692.[Abstract/Free Full Text]

Licht, J. D. (2001). Targeting aberrant transcriptional repression in leukemia: a therapeutic reality? J. Clin. Invest 108, 1277–1278.[CrossRef][Medline]

Licht, J. D. (2006). Reconstructing a disease: what essential features of the retinoic acid receptor fusion oncoproteins generate acute promyelocytic leukemia? Cancer Cell 9, 73–74.[CrossRef][Medline]

Lin, R. J., Nagy, L., Inoue, S., Shao, W., Miller, W. H., Jr, and Evans, R. M. (1998). Role of the histone deacetylase complex in acute promyelocytic leukaemia. Nature 391, 811–814.[CrossRef][Medline]

Lin, R. J., Sternsdorf, T., Tini, M., and Evans, R. M. (2001). Transcriptional regulation in acute promyelocytic leukemia. Oncogene 20, 7204–7215.[CrossRef][Medline]

Look, A. T. (1997). Oncogenic transcription factors in the human acute leukemias. Science 278, 1059–1064.[Abstract/Free Full Text]

Melnick, A., Carlile, G., Ahmad, K. F., Kiang, C. L., Corcoran, C., Bardwell, V., Prive, G. G., and Licht, J. D. (2002). Critical residues within the BTB domain of PLZF and Bcl-6 modulate interaction with corepressors. Mol. Cell. Biol 22, 1804–1818.[Abstract/Free Full Text]

Melnick, A., and Licht, J. D. (1999). Deconstructing a disease: RARalpha, its fusion partners, and their roles in the pathogenesis of acute promyelocytic leukemia. Blood 93, 3167–3215.[Free Full Text]

Misteli, T. (2001). Protein dynamics: implications for nuclear architecture and gene expression. Science 291, 843–847.[Abstract/Free Full Text]

Nye, A. C., Rajendran, R. R., Stenoien, D. L., Mancini, M. A., Katzenellenbogen, B. S., and Belmont, A. S. (2002). Alteration of large-scale chromatin structure by estrogen receptor. Mol. Cell. Biol 22, 3437–3449.[Abstract/Free Full Text]

Prasanth, K. V., Prasanth, S. G., Xuan, Z., Hearn, S., Freier, S. M., Bennett, C. F., Zhang, M. Q., and Spector, D. L. (2005). Regulating gene expression through RNA nuclear retention. Cell 123, 249–263.[CrossRef][Medline]

Qiu, J., Wong, J., Tweardy, D. J., and Dong, S. (2006). Decreased intranuclear mobility of acute myeloid leukemia 1-containing fusion proteins is accompanied by reduced mobility and compartmentalization of core binding factor beta. Oncogene 25, 3982–3993.[CrossRef][Medline]

Reineke, E. L., Liu, H., Lam, M., Liu, Y., and Kao, H. Y. (2007). Aberrant Association of promyelocytic leukemia protein-retinoic acid receptor-{alpha} with coactivators contributes to its ability to regulate gene expression. J. Biol. Chem 282, 18584–18596.[Abstract/Free Full Text]

Robinett, C. C., Straight, A., Li, G., Willhelm, C., Sudlow, G., Murray, A., and Belmont, A. S. (1996). In vivo localization of DNA sequences and visualization of large-scale chromatin organization using lac operator/repressor recognition. J. Cell Biol 135, 1685–1700.[Abstract/Free Full Text]

Sasmor, H. M., and Betz, J. L. (1990). Specific binding of lac repressor to linear versus circular polyoperator molecules. Biochemistry 29, 9023–9028.[CrossRef][Medline]

Shaner, N. C., Steinbach, P. A., and Tsien, R. Y. (2005). A guide to choosing fluorescent p