Experimental Procedures | Northern vs. Microarray Results | BCL-6 Gel Mobility Shifts | Target Gene Summary
|
Experimental Procedures | Northern vs. Microarray Results | BCL-6 Gel Mobility Shifts | Target Gene Summary |
|
I. Experimental Procedures Cell Culture, Retroviral Constructs and Retrovirus Infection The cell lines K562 (erythroleukemia), WI-L2 (mature B cell, EBV+), Raji (mature B cell, EBV+), SUDHL5 (DLCL line, EBV+), BJAB (mature B cell, EBV-) cell lines were maintained in RPMI 1640 media (GIBCO-BRL) with 10% fetal calf serum (HyClone) and penicillin/streptomycin (GIBCO-BRL). In experiments involving estradiol, cells were maintained in media without phenol red, with charcoal/dextran-adsorbed serum (HyClone). Additionally, we used a stable transfectant of WI-L2 (WI-L2 FLERD) expressing a FLAG-tagged form of full length BCL-6 fused to a portion of the estrogen receptor, maintained in 1mg/ml G418 to select for the presence of the expression construct (pCXN2neo-BCL6ERDFlag). The estradiol-inducible form of BCL-6 (FLERD) was engineered as follows: The human estrogen receptor in pSG5neo was used as a template to PCR-amplify the ligand binding domain (ERD) using primers that introduce KpnI restriction sites at each end. This product was cloned into the TA-vector (Invitrogen) and sequence verified. A plasmid with a FLAG-tagged version of full length human BCL-6 (pCGN BCL-6FLAG, ) was digested with KpnI, which cuts between the 3' end of BCL-6 and the FLAG sequence to allow the insertion of the ERD. After verification of the fusion construct, it was shuttled by digestion with XhoI and SalI, into the XhoI site of stable transfection vector pCXN2neo. WI-L2 NS cells were calcium phosphate transfected with this vector, and clones were selected with 1mg/ml G418. A truncated form of BCL-6 containing only the DNA-binding zinc fingers (ZnF) was PCR amplified with the Flag-tag from the full length BCL-6 construct. We also engineered an estradiol-inducible form of ZnF (ZnFERD) by PCR amplifying the zinc finger through the ERD to the 5' end of the IRES from the full length BCL-6 (FLERD) construct and cloning that into pXYpuro after sequence verification. Expression and b -estradiol-induced (10 m M, SIGMA) nuclear localization of fusion proteins were confirmed by western blot (data not shown) and immunofluoresence (see below). The Phoenix retroviral system was used to express various proteins in target cells . We modified the bicistronic pBMN-IRES-Lyt2 by replacing the mouse CD8a sequence (Lyt2) with a puromycin resistance gene as follows: the pBMN-IRES-Lyt2 vector was linearized with SalI, and blunt-ended. It was then cut with HinDIII in the Internal Ribosmomal Entry Site (IRES) to accept the puromycin cassette with a complementary partial IRES sequence (XbaI, blunt-ended to HinDIII fragment of pIRESpuro, Clontech). After ligation, the pBMN/puro hybrid was recut with HinDIII to allow reconstruction of the full IRES sequence using the HinDIII to HinDIII IRES fragment from pIRESpuro. This produced a new bicistronic vector (pXYpuro) with a multiple cloning site (MCS) upstream of the IRES, followed by the puromycin resistance gene. Various forms of FLAG-tagged BCL-6 described above were cloned into the MCS of pXYpuro. The basic viral construct (pXYpuro) was used to generate control cell populations for subsequent analysis. Each construct was transfected by the calcium phosphate method (GIBCO-BRL) into the amphitrophic Phoenix packaging line to make a viral supernatant. Supernatants were harvested 48 hrs after transfection, and target cells were spin-infected . Briefly, viral supernatants were harvested and centrifuged to remove live cells and debris, then mixed with a 1:100 dilution of DOTAP (Roche) and placed on ice for 10 mins. Cells were pelleted by centrifugation, media was aspirated, and cells were mixed with viral supernatant (1ml supernatant/ 1 million cells). Cells were spun in 6-well plates at 2500 rpm, 30 o C for 1.5 hrs. Supernatant was aspirated and replaced with fresh media. After 48hrs, cells were selected with empirically determined concentrations of puromycin (0.5-4 m g/ml, SIGMA). Populations were analyzed by Western blot for expression of the transduced protein (data not shown) and then expanded for further analysis. Cell Stimulation Cells were stimulated with phorbol myristate acetate (PMA, 10ng/ml, SIGMA) and the calcium ionophore ionomycin (1 m g/ml, SIGMA) for 4 hrs. Cells were also treated with goat anti-human IgM (Fab'2, Jackson Immunoresearch) and interferon g (10ng/ml, R&D systems) for 4hrs. In some cases, the stable transfectant of WI-L2 with FLERD and cells transduced with ZnFERD virus were treated with cycloheximide (1 m g/ml, SIGMA) for 30 min to stop de novo protein synthesis, then treated for 1 hr with estradiol (0.01 m M) to induced nuclear redistribution of ERD-fusion proteins. RNA Preparation and cDNA Microarray Analysis Ten million cells were pelleted by centrifugation and total RNA was prepared by the Trizol method (GIBCO-BRL) with yields of 2-6 mg/ml. Lymphochip cDNA microarrays were prepared from PCR-amplified material as described . cDNA probes from the two cell populations to be compared on a given microarray were prepared as indicated in . Briefly, 30-70 m g of each total RNA sample was converted to cDNA using Superscript (GIBCO-BRL) in the presence of either CY-3 (green)- or CY-5 (red) -labeled dUTP. Unincorporated nucleotides were removed by spin column purification (AMICON). Probes were combined with SSC, along with non-specific nucleic acid carriers: polyA RNA (Pharmacia), yeast tRNA (Pharmingen), and COT-1 DNA (GIBCO-BRL). The probe was heated to 100 o C and applied to the microarray. A coverslip was placed on the array which was sealed in a hybridization chamber and placed at 65 o C for 16-20 hrs. The microarray removed from the chamber and washed in a series of SSC solutions (2xSSC down to 0.2xSSC, 2-fold dilutions) and spun dry prior to scanning. Arrays were analyzed by scanning on a GenePix (AXON) scanner at 635nm (CY-5, red) and 523nm (CY-3, green) wavelengths. Results were quantitatively analyzed using in-house software tools developed at the NCI and at Stanford University . Below is a list of the various cell types (with constructs) and conditions analyzed in this study. V = virally transduced, L = stable transfectant.
For primary B Cell arrays:
Data for Figure 7A. were generated by analyzing RNA from Raji control virus- and ZnFERD virus-transduced cells before (time 0) and after estradiol induction at the times specified. Northern Blots and Semiquantitative RT-PCR Total RNA (10 m g) was size-separated on a formaldehyde-agarose gel, blotted to nitrocellulose, and hydridized with radioactively-labeled, PCR-generated probes (see below). Free probe was washed away and blots were exposed to both Xray film (BIO-MAX MR, Kodak) and Phosphoimager screens (Molecular Dynamics) for analysis. Radioactive, nick-translated probes (Hi Prime Kit, Roche) were generated using PCR products amplified from the coding regions of the human GAPDH, cyclin D2, CD69, and blimp-1 genes with the RT-PCR primers listed below; the MIP-1 a probe was generated using T3 and T7 primers flanking the cDNA of human MIP-1 a in pUC19. For RT-PCR cDNA was generated from 5 m g of total RNA using the cDNA Choice System (GIBCO-BRL). cDNAs were diluted 1:1 with sterile water and, 1 m l of each was added to a 50 m l PCR reaction containing 1xPCR buffer, 2mM MgCl 2 and dNTPS (Sigma) as well as appropriate primers (25ng/reaction, see below) that amplify the coding regions of selected target genes. Reaction tubes were placed in a PCR machine pre-warmed to 95 o C and 5 m l of a mix of 1x PCR buffer and 0.2 units of Taq polymerase (Sigma) were added. PCR cycle conditions were: 95 o C 1 min., 56 o C 30 sec., 72 o C 1.5 min. for 25 cycles (to keep the amplification in the linear range). PCR products were separated on a 1% agarose gel, stained with ethidium bromide and visualized by UV transillumination. RT-PCR Primers
(5' to 3'): 5'-GAPDH: GGGCGCCTGGTCACCAGGGCTG Promoter Analysis
Portions of the human cyclin D2 and MIP-1
a
promoters were PCR amplified using Klentaq (Clontech) and sequence confirmed
prior to cloning into the promoter-less pGL3 luciferase reporter vector
(Promega). Deletion constructs of the MIP-1
a
promoter were made using the following primers:
Common 3' primer:
AGCATAGTCTCGAGAAGCATGGACTGACCACTCTCTGCTG The cyclin D2 promoter was assembled PCR amplification from two halves of the promoter (-1560 to -540 with -1490 to -5) united by an internal BamHI site. The PCR primers used were: 5'-CCCCTTCCTCCTGGAGTGAAATAC 3'-GACAGGCAACCCACAAAAAACTAC 5' GAAACGCCACCAGATCGTATCTC 3'CAGCCCGGCGG CCCCTCTCCC.
To generate site-directed mutations in the BCL-6 site (underlined): Flow Cytometry One million cells were pelleted, resuspended in 10 m l of titrated, fluor-conjugated antibody, and placed on ice for 15 min. Cells were washed with 1ml of FACS buffer (1xPBS, 5% FCS), resuspended in 300 m l of FACS buffer and analyzed using a FACSorter (Becton Dickinson) and CellQuest software (Becton Dickinson). With the exception of the anti-9-27/LEU13 antibody (a gracious gift of Dr. Sharon Evans), antibodies were purchased from Pharmingen and were controlled by staining with species- and isotype- matched irrelevant antibodies. For primary cell analysis, spleens were dissected into PBS and single cell suspensions were made by manual disruption. Splenocytes were separated on a Ficoll density gradient. The buffy coat was extracted and washed in FACS buffer (see above). Splenocytes were analyzed as described above. Immunofluoresence For detection of FLAG-tagged BCL-6 proteins (FLERD and ZnFERD), cells were treated with EtOH (non-inducer) or estradiol (inducer) for 1hr., then cytospun onto slides and processed for immunofluoresence by first fixing in ice cold acetone for 10 min. at 4 o C, then transfer to 10% formalin at room temperature for another 10 min. Slides were rinsed in 1xPBS and placed in cold methanol for 10 min. at -20 o C. Cells were rinsed in Tris-buffered saline (20mM Tris, 137mM NaCl, pH 7.0) with 0.01% Tween-20 (TBST). Blocking was performed for 1 hr. at room temperature in a humidified chamber using 20 m l of a mixture of 5%FCS, and 10% serum (from the appropriate species) with a 1:50 dilution of FcBlock (Pharmingen) in TBST. Staining was performed by diluting antibodies (isotype control or anti-FLAG, M2, Kodak) in the blocking solution and adding 20 m l of this mix to the sections for 45 mins. at 4 o C. Slides were washed 3x in TBST and covered with FluorSave (CalBiochem) and a coverslip. Cells were visualized by confocal microscopy. ELISA Quantitation of Chemokine Secretion MIP-1 a levels were measured in WI-L2 control cells and WI-L2 FLERD cells. Cells were washed with PBS and placed in fresh media (to remove existing MIP-1 a ) along with EtOH (control) or estradiol, to induce BCL-6 relocalization in WI-L2 FLERD. After 2hr, cells were washed again with PBS and placed in fresh media, at one million cells/ml, for an additional 8hr. Supernatants were harvested, transferred to a fresh tube with 2mM PMSF (SIGMA), and stored at –20 o C. MIP-1 a levels in diluted of the supernatant (1:3) were determined with a Quantikine Kit (R&D Systems) for human MIP-1 a analyzed on a SpectraMax 250 (Molecular Devices) plate reader at 450nm. Chemokine secretion (MIP-1 a and IP-10) were also measured in Raji control or ZnFERD transduced cells. One million cells were plated in 1ml of media/well in a 24 well plate. For MIP-1 a measurement, cells were stimulated, after 30 mins. of EtOH or estradiol induction, with PMA and ionomycin for an additional 5hr. In the case of IP-10, cells were treated with EtOH (non-inducer) or estradiol (inducer) for a total of 6.5hr. MIP-1 a levels were measured using a 1:50 dilution of Raji supernatants as above. To measure human IP-10 production, we coated 96 well plates with anti-IP-10 antibody (5 m g/ml, R&D Systems) overnight at room temperature. The plates were washed with wash buffer (1xPBS, 0.05% Tween-20) and blocked with a solution of 1xPBS, 1% BSA, 5% sucrose, 0.05% sodium azide for 1 hr. Plates were washed again, blotted dry, and 100 m l of undiluted cell supernatant was added per well for 2hr at room temp. Plates were washed and 100 m l of a second biotinylated anti-IP-10 antibody (50ng/ml, R&D Systems) were added for 2hr at room temp. Plates were washed and 100 m l of a streptavidin-horseradish peroxidase conjugate (1:20,000 dilution of 1.25 mg/ml, Zymed) was applied for 20 min at room temp. Plates were washed a final time before the addition of 100 m l of colorimetric substrate solution (R&D Systems) for 20 min, followed by addition of 50 m l of STOP solution (2N sulfuric acid). Plates were analyzed as above. Amounts of IP-10 produced were calculated based on a standard curve generated by 4 fold serial dilutions (4ng/ml and below) of recombinant human IP-10 (R&D Systems). Splenic B cells from heterozygous control and BCL-6 mutant littermates were prepared as above by positive selection on the basis of CD19 expression using magnetic beads (Miltenyi Biotech, ). B cells (5x10 5 /ml) were placed in media with LPS (25ug/ml SIGMA) for 3 days. Supernatants were harvested and MIP-1 a levels were measured by Quantikine ELISA (R&D Systems). Cell Viability and Cell Cycle Analysis Equal numbers of Raji control virus- and ZnFERD virus-transduced cells were added to fresh media. To determine viability, aliquots were mixed with trypan blue and numbers of live versus dead cells were counted. For cell cycle analysis, cells were harvested at the indicated times, pelleted in a FACS tube and resuspended in 0.5ml of PBS. Cells were fixed by gentle vortexing while adding 5mls of 100% EtOH. Cells were placed on ice for 20 minutes, pelleted and resuspended in 1ml of 1xPBS/1% FBS. Cells were pelleted again, resuspended in a solution of 1xPBS/2.5mM EDTA/10 m g/ml propidium iodide (Sigma), and 250 m g/ml RNAse A (preboiled for 15 mins.) for 40 minutes at 37 o C, then analyzed by flow cytometry to determine the proportion of cells in each phase of the cell cycle. Western Blotting One million cells were pelleted in microfuge tubes and supernatants were aspirated. 40ul of 1xPBS was added along with 50ul of 2xSDS buffer. Samples were boiled for 5 minutes, sonicated, and stored at –20 o C. Before loading, samples were boiled for 3 minutes. Ten m l of each sample were loaded onto a 7.5% SDS PAGE (BioRAD) and run at 70V for 2 hrs. Proteins were transferred to Immobilon-P nylon membranes (Millipore) by electrophoresis at 135V for 30 minutes. Membranes were blocked in a solution of TBST+5% dry milk for 60 minutes. Primary antibodies (rabbit-anti-c-myc and rabbit-anti-SP1, Santa Cruz) were added in TBST+5% milk for 45 minutes. Membranes were washed in TBST and secondary HRP-conjugated antibody (anti-rabbit-HRP, Amersham) was added for 30 minutes. Membranes were washed again, developed with ECL reagents (Amersham), and exposed to Xray film. Animal Generation and Handling Mice used in this study were kept under sterile conditions. BCL-6 mutant mice have been described . To generate litters with heterozygous control and homozygous knockout animals, mice heterozygous for the BCL-6 knockout allele (backcrossed to the C57B6 background for >5 generations) were mated. Pups were typed by PCR of genomic DNA as described . Healthy mice were sacrificed between the ages of 4 and 6 weeks for analysis. II. Northern vs. Microarray Results
Northern Blot Quantitation of Target Genes Fold repression of RNA levels shown by Northern Blot (in SUDHL5, WI-L2, and K562) was calculated by determining the ratio of expression without BCL-6 to expression in the presence of BCL-6 normalized by the GAPDH signals. Fold induction in Raji cells was calculated was calculated by comparing ZnFERD transduced cells to control cells, normalized similarly. Fold repression or induction from the microarrays is shown by calculating the Cy5/Cy3 (red:green) ratio from the appropriate arrays. III. BCL-6 Gel Mobility Shifts A. Potential BCL-6 Binding Sites
Using the Entrez search engine the longest 5' regulatory region for each human target gene was identified. When possible, the same region from other species was also identified. By inspection, potential BCL-6 binding sites were identified on the basis of matching to a consensus BCL-6 site (Seyfert, 1996 #277 and ALS unpublished observations). Shown are the core BCL-binding site and matches in the promoters of each target gene. B. Gel Mobility Shift Assays of Target Gene BCL-6 Sites:
 BCL-6 sites for the human target genes, along with the consensus BCL-6 binding site, used as gel shift probes. Probes were mixed with BCL-6-containing nuclear extract (lanes 6, 11, 16, 21). Reactions were also prepared with excess unlabeled specific (5,10,15,20) or mutant (4,9,14,19) competitor oligo or with BCL-6-specific (3,8,13,18) or irrelevant (SP1-2,7,12,17) antibodies for supershifting. The BCL-6 complex is indicated by an arrow on the left. Lane 1 is free probe.
IV. Target Genes:
|
|
National Institutes of Health | National Cancer Institute Center for Cancer Research | Lymphoid Malignancies Branch
|