Cell line

The P19 cell line (mouse embryonal carcinoma cell line) was purchased from ATCC.

Realtime RTPCR

P19 cells were grown in MEM alpha medium containing 7.5% bovine calf serum, 2.5% fetal bovine serum, with penicillin and streptomycin. On the day prior to treatment, 200,000 cells were seeded into 6-well plates pre-coated for 1 h at room temperature with a sterile 2% gelatin solution. Cells were allowed to attach overnight before 6-h serum starvation to synchronize the cells prior to treatment in medium containing 2% serum, Wnt3a, ATRA, or a combination of Wnt3a plus ATRA. Cells were treated for 24 h and harvested with Trizol Reagent (Life Technologies). Total RNA was extracted per manufacturer’s protocol. 1 microgram of total RNA was used to carry out first strand cDNA using the cDNA Synthesis Kit (Quanta). Realtime RTPCR analysis was performed on the Bio-Rad MyiQ Single Color Real-Time PCR Detection System using PerfeCTa SYBR Green SuperMix (Quanta) and the following forward and reverse primers: Stra6–5`-AGCCAAGTCAGACTCCAA GAG-3’ and 5’-CAGAGAGCACACTAACTTCT TTCA-3’; Stra8–5’-GAGGCCCAGCATATGTC TAAC-3’ and 5’-GCTCTGGTTCCTGGTTTAAT G-3’; Ephrin B1–5’-GTAGGCCAGGGCTATTT CTG-3’ and 5’-TGTCTCATGAGGGTCCAAAA-3’; Gapdh– 5’-GGTGCTGAGTATGTCGTGGA -3’ and 5’-ACAGTCTTCTGGGTGGCAGT-3’; Survivin– 5’-GACAACCCGATAGAGGAGCAT -3’ and 5’-TCACTGACGGTTAGTTCTTCCA-3’.

Immunohistochemistry/differentiation of P19 cells

P19 cells were grown in MEM alpha medium containing 7.5% bovine calf serum, 2.5% fetal bovine serum, with penicillin and streptomycin on 10cm tissue culture plates. To induce differentiation, cells were subjected to a previously described [23] protocol with minor modification. Briefly, cells were cultured for 48 h in the presence of 1μM ATRA and then dispersed with 0.25% trypsin/EDTA. The P19 cells were then seeded into fresh ATRA-containing medium and additional 24 h. Cells formed small aggregates from which individual aggregates were seeded into 6-well plates containing 20mm² gelatin-coated coverslips in ATRA-containing medium. Incubation was at 37°C, 5% CO₂ and medium was changed every two days. On day 8 (day 11 of ATRA treatment), cells were fixed and permeabilized with 4% paraformaldehyde, then stained for EphrinB1 (Santa Cruz) with secondary CF488A donkey anti-mouse IgG (Biotium), and counter-stained for DAPI (Vector Laboratories, Inc.). Immunostained cells were visualized using the Leica AF6000 Modular System fitted with a Leica CTR 6000 light/laser source and Leica DMI 6000B microscope running the Leica Application Suite Advanced Fluorescence Software. Cells were photographed under 20X magnification.

Transient transfection

Prior to transfection, 48-well plates were coated with a 2% gelatin solution for 1 h at room temperature. P19 cells were seeded into gelatin-coated 48-well plates at 30,000 cells per well in a total volume of 200 microliters. The cells were transfected using the CalPhos Mammalian Transfection Kit (Clontech) per manufacturer’s protocol. pCDNA3.1 (empty/control vector) and pCDNA3.1-p300 wild type expression vector were provided by D. Livingston (Dana-Farber Cancer Institute, Harvard Medical School) and transfection with these plasmids was performed as previously described [24].

Co-immunoprecipitation

Methods are as previously reported [13]. Cells were washed twice in ice-cold PBS, and then transferred into 1.5 mL eppendorf tubes chilled on ice. Cell pellets were then subjected to cytoplasmic and nuclear fractionation using the NE-PER (Pierce). Protein concentration was determined using the Bio-Rad Protein Assay. 200μg of nuclear protein from each treated sample were diluted with 1mL of co-immunoprecipitation buffer (25mM Tris, HCl, pH7.5, 150mM NaCl, 5%glycerol, 0.5%NP40, 1mM EDTA, 2mM DTT, 1X protease inhibitor cocktail). 2μg of CBP, p300, or normal rabbit IgG antibody were added to the diluted protein sample and incubated at 4°C on a rotating mixer overnight. 50μl of a 50% slurry of protein A-agarose beads, equilibrated in co-ip buffer, were added to each diluted sample containing antibody. The protein-antibody complexes were then allowed to bind to protein A-agarose on a rotating mixer for 1 h at 4°C. Protein A-agarose beads containing bound antibody/protein complex were washed three times using 500μl of co-ip buffer each time. After the final wash, the agarose beads were resuspended in 30μl of 2X Laemmli Buffer and boiled for 10 minutes for SDS-PAGE. For proteomic analysis, the agarose beads were suspended in 5% Sodium Deoxycholate (SDC) (Sigma-Aldrich). The supernatant, containing bound protein complexes were separated from the agarose beads using Illustra microspin columns. The boiled supernatants were loaded onto 8% polyacrylamide gels and subjected to SDS-PAGE. Electrophoresed proteins were then transferred onto PVDF membrane overnight under low voltage (25V) in a cold room. PVDF membranes were then probed using a mouse monoclonal antibody for β-catenin (1:5000 dilution). Detection of β-catenin was performed on the Bio-Rad ChemiDoc MP Imaging System.

Stra6, CBP and p300 Western blotting

For detection of Stra6, P19 cells under the treated conditions were harvested and lysed using RIPA buffer (25mM Tris, pH8, 150mM NaCl, 0.1% SDS, 0.5% sodium deoxycholate and 1% TritonX-100). 50μg of whole cell lysate was subjected to SDS-PAGE and Western blotting was performed using Stra6 antibody (Santa Cruz). For CBP and p300 Western blotting, 10μg of nuclear lysate were used with antibodies to CBP and p300 (Santa Cruz).

CRISPR/Cas9 editing system

Two guide RNAs were selected for targeting the murine Ep300 N-terminal sequence. The sequences used were: forward guide RNA 112A AGTCA GCTTCAGACAAGTCT (Chr15: 81600997–816 01016); reverse guide RNA 112B CTGACTGATATCGCCACCAT (Chr15:8160 1002–81600983). The guide RNAs were inserted into the backbone vector pCas-Guide-EF1a-GFP (Origene Technologies, Inc.). The guide RNA sequences are located within or across the boundary of the motif to be edited in murine Ep300. The vector pCas-Guide-EF1a-GFP-GE200112A (F) carrying the forward sequence AGTCAGCTTCAGACAAGTCT was chosen for the editing experiment. All sequence annotations are based on the UCSC Genome Assembly version GRCm38/mm10. The donor DNA fragment was generated by PCR using mouse genomic DNA as template and cloned into the plasmid pBluescript II sk(+) (Stratagene) to form the donor vector pBL-Donor1. The vector carries 1.8 kb of genomic DNA of murine Ep300 spanning from (Chr15:81600121) to (Chr15:81601983) (Chr15:81600991–81601023) within which the Switch motif of 33 nucleotides (Chr15:81600991–81601023) was deleted and replaced with restriction endonuclease HincII site (GTTAAC) to generate the edited sequence. For transfection of pCas-Guide-EF1a-GFP-GE200112A (F) and pBL-Donor1: P19 cells were grown in MEM alpha medium containing 7.5% bovine calf serum, 2.5% fetal bovine serum, with penicillin and streptomycin. Cells were grown on 60mm dishes. At approximately 30% confluency, the cells were transfected with pCas-Guide-EF1a-GFP-GE200112A (F) and pBL-Donor1 at 1:1 molar ratio. Transfection was performed with 8μg of DNA and 20μl of lipofectamine 2000 (Thermo Fisher Scientific).

Colony screening

P19 cells were subcultured after 48 h following the transfection. The detached cells were suspended in growth medium and seeded into 96-well plates at a density of 30 cells per 96 wells. Noticeable single colonies formed in about 7–10 days. The colonies were subcultured in duplicate into 24-well plates at day 14 for further propagation. After nearly reaching confluency, one culture was frozen down for stock and the other prepared for PCR screening. For this purpose, cells were trypsinized, neutralized with serum-containing medium, and collected by centrifugation. The cell pellet was dissolved in 30μl lysis solution (25mM NaOH, 0.2mM EDTANa2, pH 12). The lysate was heated at 95°C for 30 min, and then neutralized with an equal volume of neutralization solution (40mM Tris-HCl, pH 5).

PCR was performed in a volume of 25μl. In each reaction, 1–2μl of cell lysate as template, 12.5μl of EmeraldAmp GT PCR Master Mix (Clontech), 0.2μM each of the forward and reverse primers were used. PCR conditions: 98°C 1 min; 98°C 10sec, 58°C 30sec, 72°C 1.5 min, 35 cycles; 72°C 5min. The PCR products were subjected to agarose gel electrophoresis. The resolved DNA bands were visualized by ethidium bromide staining and documented by photography.

The PCR primers used for screening clones and distinguishing the edited insertion region from the wild-type were specifically the forward primer detecting the edited EP300: p300-Switch-Donor3-F2 CGGCGGAGTTAACGTA CAAG (editing-generated sequence only, not present in the wild-type Ep300, located within exon 2); and the reverse primer P300-Donor-RA-R3 AGATGTTCCCAAGGCTGCTA (within intron 2–3, Chr15:81602195–81602176). The PCR product of the edited fragment was 1185bp. In the wild type primer pair, the forward primer used for detecting wild-type Ep300 was P300-Donor-LA-F3 TAAGCCTGTTCTGTCCTC AGC (within intron 1–2, Chr15: 81599715–81599735); the reverse primer specifically detecting the wild-type Ep300 was p300-Switch-R TGCCAAG ACTTGTCTGAAGCT (present in wild-type Ep300 only within exon 2, Chr15:81601021–81601001). The PCR product size for the wild-type fragment was 1307 bp.

Sequencing genomic DNA from positive clones to confirm the results of the PCR screening

Genomic DNA samples were prepared using the Qiagen Flexigene DNA kit. The PCR enzyme PrimeStar (Clontech) was utilized. The forward primer sequence used was GGTTCACTGTTTGA CCTGGAA; the reverse primer was AGATGTT CCCAAGGCTGCTA. PCR conditions: 98°C 1min; 98°C 10sec, 58°C 15sec, 72°C 35sec, 35 cycles; 72°C 2min. The 1.3 kb PCR product generated was used as the sequencing template. The PCR products were purified from agarose gel using a Gel Extraction Kit (QIAGEN). 40ng each of DNA was sequenced by GENEWIZ. The forward sequencing primer used was GGTTCACTGTTTGACCTGGAA; the reverse sequencing primer used was GGCTGGTTC ACTGGACTGTT.

Sequencing the cDNA edited sub-clones to define the expression ratio between different alleles

Total RNA was prepared using Trizol (Thermo Fisher Scientific) from edited clone 10. Reverse transcription of the RNA was performed with a qScript cDNA Synthesis Kit (Quanta) according to the manufacturer's protocol. PCR was performed using ClonAmp HiFi PCR Premix (Clontech) to generate a 676 bp product including the locus of edited motif. The primer sequences were forward GTTGAGTCCGCATCCCTCTC and reverse GGCTGGTTCACTGGACTGTT. PCR conditions: 98°C 10sec, 58°C 15sec, 72°C 15sec, 35 cycles; 72°C 2min. The gel-purified PCR product was inserted into the plasmid pBluescript II sk(+) and the resulting sub-clones were sequenced with T7 primer by GENEWIZ. The ratio of three modified regions in Ep300 is 1:1:1.

Clone 10: All three Switch motifs have been modified. One was precisely edited as designed; the other two were mutated by sequence specific targeting, random deletion and ligation. The size of Switch in the edited allele is 33 bp shorter than the wild type with an addition of HincII site (6 bp) as designed. The size of truncation in the two mutant alleles is 6 bp and 33 bp respectively. All three modified sequences are overlapping and the open reading frame of three Ep300 alleles remains unchanged.

Sample preparation for proteomics

Methanol solutions of whole cell extracts were dried. To this was added 100μl of 5% SDC. Both the whole cell extracts and 100 μl of immunoprecipitates (IPs) were processed as follows. The solutions were diluted to 200μl of 1% SDC, 2M urea, 50mM NH4HCO3, 1% with 1μg trypsin and incubated at 37°C for 20 h with gentle agitation. Detailed methods are as previously reported [25] and summarized as follows. The solutions of digested peptides were treated by phase-transfer method to remove the SDC [26]. The obtained peptides were resolved with 0.1% formic acid (FA) and then quantified using a Qubit® Protein Assay Kit (Life Technologies). 3μg of peptide solution was desalted according to the manufacturer’s protocol using C18 StageTip (Thermo Scientific) [27], dried, and redissolved in 10μl of 0.1% FA. The obtained peptide solution (2μl) was subjected to nanoLC-Ultra 2D (AB SCIEX) coupled with to a TripleTOF®5600 (AB SCIEX) mass spectrometer. The subjected peptides were directly injected onto a C18 reverse-phase column (75μm×150mm; ChromXP C18-CL 3μm120 A°) (Eksigent) equipped with a nanoLC-Ultra 2D and then separated by a binary gradient consisting of mobile phase (A) 0.1% FA and 1% Acetonitrile (ACN) and mobile phase (B) 0.1% FA and 99% ACN. The gradient condition used was: 2% of B initially, increased to 25% of B at 130 min, to 90% of B at 131–135 min, and to 2% of A at 135.1 min (total run time was 155 min). The masses of the eluted peptides were determined using the TripleTOF5600.

Proteomics analysis by 2DICAL

MS peaks were detected and quantified using 2DICAL [28, 29]. Detailed methods are as previously reported [30] and summarized as follows. 2DICAL was developed as a shotgun proteomics analysis system. It analyzes the data of mass to charge ratio (m/z), retention time (RT) and peak intensity generated by liquid chromatography and mass spectrometer (LC/MS), and each sample as elemental data; it deploys various 2-dimensional images with different combinations of axes using these four elements. From the m/z–RT image, peaks derived from the same peptide in the direction of the acquiring time are integrated. By adding algorithms to ensure reproducibility of m/z and RT, the same peak can be compared precisely across different samples, and a statistical comparison of identical peaks in different samples leads to the discovery of specific differentially expressed peptide peaks. Specific peaks are designated by their m/z and RT coordinates, and further analysis is based on these identifiers. Isotopic labeling is not necessary, and a large numbers of samples can be analyzed in this way. The peptide search engine used in 2DICAL is MASCOT® software (version 2.5.1; Matrix Science) using the Uniprot/Swiss-Prot human database (H. sapiens, 471472 sequences in Sprot_57.5 fasta file). Statistical Analyses was performed with the open-source statistical language R (version 2.7.0) [31]. The 2DICAL intensity data was converted to protein value averaging the intensity data of peptides derived from the protein.

Three-dimensional structural modeling of the N-terminus of CBP/p300

There are no structures available for the N-terminus of CBP and p300. To understand the structural aspects of the N terminus region in CBP and p300a molecular modeling study was performed. Briefly, three-dimensional structures of the N-terminus of CBP and p300 were determined using Robetta [32]. Although the confidence levels for overall structure prediction by Robetta was low (14–16%), the secondary structure prediction was determined with higher confidence for the LXXLL motif (over 50%).

To assess whether CBP/p300 can form a complex with both β-catenin and nuclear receptor, best models, generated by the Robetta structure prediction algorithm, were used to dock with the full-length humanized β-catenin structure using β-catenin from zebrafish (PDB:2Z6G) as a template. The docking of CBP/p300 was performed using ZDOCK [33]. The top model for CBP/p300 from the ten predicted models from the ZDOCK calculations was selected for further analysis. To estimate potential binding of nuclear receptor to CBP/p300/ β-catenin complex, RXR alpha ligand binding domain (LBD) (PDB code: 1RDT) was docked to either CBP-β-catenin or p300-β-catenin complexes. In the final analysis, the docking ensembles were not optimized and only relative orientations of CBP/p300, β-catenin and RXR alpha LBD were used.

Data analysis

Numerical data were expressed as the means ±S.E.M. unless otherwise noted. Student’s t-test or One-way ANOVA with post-hoc Tukey test was performed as appropriate. p-values < 0.05 were considered significant.

Nuclear receptor ligand binding domain (RXR alpha) binds to p300 without any steric hindrance

De novo structure prediction by Robetta showed that the N-terminus of both CBP and p300 is highly disordered and that the nuclear receptor binding motif (LXXLL) adopted a helix. In CBP, the LXXLL motif was predicted to be part of a longer helix containing the putative β-catenin binding motif. In contrast, in p300 the LXXLL motif kept its helical structure and was solvent exposed (i.e. extended further due to the insertion of 9aa) poised for protein interaction. Further docking analyses against either β-catenin/p300 or β-catenin/CBP show that the nuclear receptor ligand binding domain (RXR alpha) binds to p300 without any steric hindrance (Fig 2C, right). On the other hand, in CBP since the LXXLL motif was part of the longer helix, the nuclear receptor did not bind (Fig 2C, left); the C-terminal of β-catenin sterically prevented both β-catenin/CBP and nuclear receptor binding.

p300 editing affects Wnt signaling

We initiated our studies utilizing CRISPR/Cas9 editing in P19 mouse embryonal carcinoma cells to delete the 27bp insertion in the amino terminus of p300 (Fig 3A). After careful screening, we selected clone 10 in which all alleles of p300 had been edited to effectively delete the insertion (see experimental details in the Materials and Methods section). We first confirmed that in clone 10 (subsequently termed edited P19 cells) that the expression levels of both Kat3 coactivators CBP and p300 were not affected by the CRISPR/Cas9 editing. Immunoblotting (Fig 3B) demonstrated that the protein expression levels of both CBP and p300 in the edited P19 cells were essentially equivalent to those in the wild type P19 cells. We next tested the effects of the p300 editing on two Wnt responsive luciferase reporter constructs i.e., Topflash and Survivin/ luciferase, which we have previously utilized for first pass assessment of differential CBP/p300 Wnt/catenin coactivator usage [13]. The edited P19 cells exhibited significantly higher Topflash/Fopflash activity compared with the wild type cells under basal conditions (Fig 4A). In both the wild type and edited P19 cells, Wnt3a also significantly activated the Survivin/luciferase reporter construct (Fig 4B and 4C). As previously observed [34], this activation was effectively inhibited by ICG-001 in the wild type cells (Fig 4B). However, in the P19 edited cells, although Wnt3a increased Survivin/luciferase activity, ICG-001 did not reduce the activity (Fig 4C). Furthermore, the expression of survivin/BIRC5 message, a gene that we have previously demonstrated to be highly CBP dependent for transcriptional activation, displayed Wnt3a enhanced expression in the wild type and to a lesser extent in the p300 edited P19 cells (Fig 4D). Whereas treatment with ICG-001 at 5uM in the wild type cells significantly decreased survivin/BIRC5 message, consistent with the reporter gene assay data, there was essentially no effect of ICG-001on survivin/BIRC5 expression in the edited cells (Fig 4D).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. CRISPR/Cas9 editing/deletion of p300 insertion sequence in P19 mouse embryonal carcinoma cells.

A. Schematic outlining the design of CRISPR/Cas9 editing in P19 mouse embryonal carcinoma cells to delete the 27bp insertion in the amino terminus of p300 (as detailed in the Experimental Procedures). Guide RNA (gRNA), CRISPR RNA (crRNA), protospacer adjacent motif (PAM). B. Immunoblotting demonstrated that protein expression levels of both CBP and p300 in the “edited P19 cells” (Edited) are equivalent to those in the wild type P19 cells (WT), confirming that the expression levels of both Kat3 coactivators were not affected by the CRISPR/Cas9 editing. n = 3.

https://doi.org/10.1371/journal.pone.0200714.g003

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Edited P19 cells exhibited higher β-catenin/TCF transcription, as well as increased survivin/BIRC5 expression.

A. Edited P19 cells exhibited significantly higher β-catenin/TCF transcription compared with wild type (WT) cells under basal conditions, as assessed by Topflash/Fopflash (TOP/FOP) activity. n = 3, *p < 0.05. B and C. In both wild type and edited P19 cells, Wnt3a significantly activated the Survivin/luciferase reporter construct (compared with empty vehicle). This activation was effectively inhibited by specific, direct small molecule CBP/catenin antagonist ICG-001 in wild type cells (B). However, in the P19 edited cells, although Wnt3a increased Survivin/luciferase activity, ICG-001 did not reduce the activity (C). Representative of n = 3, *p < 0.05, **p < 0.01. D. Expression of survivin/BIRC5 mRNA, which is highly dependent on CBP, was enhanced by Wnt3a in wild type (WT) and to a lesser extent in p300 edited P19 cells. Whereas treatment with ICG-001 in WT cells significantly decreased survivin/BIRC5 message, there was essentially no effect of ICG-001on survivin/BIRC5 expression in edited cells. n = 3, **p < 0.01. Vehicle/control data (set to 1) as indicated with a red, dashed horizontal.

https://doi.org/10.1371/journal.pone.0200714.g004

We attribute this lack of efficacy of ICG-001 in the edited cells to the fact that although ICG-001 still binds to CBP and inhibits CBP/β-catenin transcription, it does not bind to the edited p300. (We have previously demonstrated that ICG-001 binds specifically to N-terminus of CBP but not p300 [12].) However, the edited p300, which is now more “CBP-like” due to the 9aa deletion, as opposed to wild type p300, potently drives TCF/β-catenin driven activation of both Survivin/luciferase and survivin/BIRC5 message. To further explore differential Kat3 coactivator usage in the P19 edited cells, we used a β-catenin co-immuno-precipitation assay. β-catenin was associated with both CBP and p300 in wild type cells, but appears to be associated with only p300 in edited P19 cells (compared to IgG control). However, substantially more β-catenin appears to be associated with p300 in the edited P19 cells than in the wild type cells (Fig 5). This could potentially explain both the enhanced Topflash/Fopflash reporter activity in the edited cells, as the Topflash reporter appears to be able to utilize either Kat3 coactivator effectively and the lack of efficacy of ICG-001 in decreasing either Survivin/luciferase or survivin/BIRC5 message [13].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. More β-catenin is associated with p300 in edited P19 cells than in wild type (WT) cells, as demonstrated by immunoblot for β-catenin using samples of nuclear protein subjected to immunoprecipitation with CBP, p300, or (control) rabbit IgG antibody.

n = 3. (5% input/control immunoblotted for β-catenin is depicted on the far left).

https://doi.org/10.1371/journal.pone.0200714.g005

p300 editing affects Wnt and retinoic acid signaling interactions

As discussed in the introduction, ATRA via RAR/RXR has been shown to antagonize aberrant Wnt/β-catenin signaling in cancer cells [15, 21]. However, ATRA and Wnt’s synergistic induction of expression of the gene Stra6 has also been reported [16]. We first examined the expression of Stra6 in both wild type and p300 edited P19 cells. Although Stra6 was expressed in both cell lines, basal expression in the p300 edited cells was 16 fold lower than in the control cells (Ct value of 28 in control versus 32 in the edited cells), consistent with the inability of the edited p300 to concomitantly bind both β-catenin and RAR/RXR to drive Stra6 expression. Furthermore, whereas individually Wnt3a and ATRA modestly induced Stra6 in both cell lines, there was a strong additive effect on the expression of Stra6 with the addition of Wnt3a and ATRA in the wild type cells. However, no additivity could be demonstrated in the p300 edited cells (Fig 6A). Moreover, in edited P19 cells (S1 Fig, right) which have been transfected with empty/control vector (pcDNA), Wnt3a+ATRA treatment does not show any additive effect on Stra6 mRNA expression. However, wild type p300 expression vector (p300) tends to restore the additive effect (observed in WT P19 cells) of Wnt3a+ATRA on Stra6 mRNA expression in edited P19 cells. Indeed, these results further support that the 9 amino acids of p300 are important for mediating the additive effect on Stra6 mRNA expression in P19 cells. The additivity of Wnt3a and ATRA was also reflected at the protein level for Stra6 in the wild type but not the edited P19 cells (Fig 6B). Similarly, in the case of Stra8 (stimulated by retinoic acid 8), a protein important for gametogenesis in mammals [35], although there is a clear additive effect of Wnt3a and ATRA in the wild type cells, there is no additive effect in the edited cells (Fig 6C).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. p300 editing affects Wnt and retinoic acid signaling interactions.

A. Wnt3a and all-trans retinoic acid (ATRA), individually, induced Stra6 (stimulated by retinoic acid 6) mRNA expression in both wild type (WT) and p300 edited P19 cells, as assessed by realtime RTPCR. Whereas there was a strong additive effect on the expression of Stra6 with combined Wnt3a and ATRA treatment in WT cells, no additivity was demonstrated in p300 edited cells. n = 3, *p < 0.05. Vehicle/control data (set to 1) as indicated with a red, dashed horizontal. B. The additivity of Wnt3a and ATRA was also reflected at the protein level for Stra6 in wild type but not edited P19 cells, as assessed by densitometry of immunoblot for Stra6. n = 3, *p < 0.05, **p < 0.01. C. Similarly, with Stra8 (stimulated by retinoic acid 8), there is a clear additive effect of Wnt3a and ATRA in wild type (WT) cells, but there is no additive effect in edited P19 cells, as assessed by realtime RTPCR. n = 3, *p < 0.05. Vehicle/control data (set to 1) as indicated with a red, dashed horizontal.

https://doi.org/10.1371/journal.pone.0200714.g006

p300 editing may affect recruitment of SWI/SNF

To more broadly investigate the biochemical consequences of editing p300 in P19 cells, we undertook a proteomic analysis of the wild type versus the edited cells and of proteins associated with CBP and/or p300 in wild type versus p300 edited cells. One interesting feature observed from proteomic analysis of the CBP and p300 immunoprecipitation experiments was that binding to the Kat3 coactivators by SWI/SNF, a group of proteins which alter chromatin structure to facilitate transcription, was affected. SMARCD1 (BAF60A) was associated with both CBP and p300 in the wild type cells, whereas its binding to both Kat3 coactivators in the edited P19 cells was decreased (S1 Table). SMARCD1 has previously been shown to contribute to maintaining the ES cell pluripotent state [36]. The related SWI/SNF protein SMARCE1 (BAF57) in the parental P19 cells was principally associated with p300. However, in the edited P19 cells, the association of SMARCE1 to p300 was significantly reduced. This is intriguing in that the recruitment of SMARCE1 has been shown to be important for the neuronal differentiation of ES cells [37].

Moreover, many members of BAF (mammalian SWI/SNF complex) complexes, which play a role in mammalian neural development [38], including ARID1A/BAF250a, ARID2/BAF200, SMARCA4/Brg1, BCL7/BAF40, SMARCB1/hSNF5, Brd7 and SMARCC1/ BAF155 were all found associated with the Kat3a coactivators, with a number being significantly affected by the editing of p300. The heterogeneous nuclear ribonucleoprotein A2/B1 (ROA2), was found associated with both CBP and p300 in the edited cells, however, it was associated only with p300 in the parental cell line. Knockdown of ROA2 (hnrnpa2b1) has previously been shown to result in loss of ES cell self-renewal [39].

Global proteomic analysis of p300 edited cells

We next performed global proteomic analysis which revealed that many proteins were also significantly differentially expressed in the P19 edited versus the parental cells. The expression of fatty acid-binding protein 5 (FABP5), which activates PPARδ, was significantly decreased (greater than 80%) in the edited cells (S2 Table). There was also a significant reduction of ALDH1A2, an enzyme that catalyzes the synthesis of ATRA from retinaldehyde in the edited cells. The mammalian ISWI chromatin remodeling protein SMARCA1 (Snf2l) was expressed only in the edited cells. Mice mutant for Snf2l display increased cellularity in the forebrain due to enhanced expansion of progenitor cells and postponed differentiation [40]. The expression level of several other interesting proteins was also dramatically down-regulated in the edited cells. For example, the expression of wdr43, a protein that is critical in ribosome biogenesis, was decreased greater than 90% in the edited cells compared with the parental cells [41], as was the SRSF8 (SFRS2B), a candidate splicing factor for the regulation of patterning and growth of the inner ear [42]. Interestingly, the expression of the histidine triad protein Hint1 was found to be significantly reduced in the p300 edited cells. Hint1 negatively regulates β-catenin/TCF-mediated transcriptional activity [43], and was increased by both Wnt3a and ATRA treatment only in the wild type P19 cells (S2 Table).

p300 edited P19 cells do not undergo neuronal differentiation

P19 embryonal carcinoma cells have been thoroughly validated as a system for in vitro neuronal differentiation by ATRA [44]. Stra6 expression, amongst a number of genes encoding proteins regulating retinoid uptake and signaling is increased upon the induction of neuronal differentiation [45]. Based upon the critical role of p300/β-catenin transcription in differentiative events [1, 10, 11, 13, 46], the lack of synergistic expression of Stra6 by Wnt3a and ATRA and the differential core transcriptional component interactions with p300 in the edited cells, we anticipated that the p300 edited cells would exhibit neuronal differentiation defects. From the proteomic analysis, the expression of a number of proteins that are known to be important for neuronal maturation and survival, including Bcl2-associated athanogene 2 (BAG2) [47] and nardilysin (NRD1) [48] was significantly reduced in the edited cells (S2 Table). Mutations in nardilysin have previously been linked to neurodegeneration [49]. It was recently shown that NRD1 localizes to mitochondria, where it participates in folding of the enzyme α-ketoglutarate dehydrogenase, which catalyzes a rate-limiting step in the TCA cycle. Loss of Nrd1 leads to increased α-ketoglutarate, mTORC1 activation and a subsequent reduction in autophagy associated with neurodegeneration [50]. Interestingly, in this manner, p300 coactivator usage couples neuronal protection via increased autophagy to a metabolic change and increase OXPHOS [1]. Metabolic remodeling and increased OXPHOS is also a hallmark of the loss of pluripotency [51].

Similar to Stra6, ephrin B1 message expression was induced by Wnt and ATRA in an additive manner in the wild type P19 cells. Although expression of ephrin B1 was also induced by Wnt3a and ATRA, individually, in the edited cells, there was antagonism between the two treatments. (Fig 7A). Ephrin B1 has been demonstrated to be critical for neuronal migration and the maintenance of neuronal progenitors [52, 53] and mutations in EFNB1 are associated with the X-linked disorder Craniofrontonasal syndrome (CFNS) (OMIM number 30411). Finally, we used a previously described neuronal differentiation protocol for P19 cells in both the wild type and edited cells. The wild type cells, as anticipated, stop proliferating and neurite outgrowth ensued, whereas the edited cells continue to proliferate and neurite outgrowth was not observed (Fig 7B) [44]. By both immuno-cytochemistry and proteomic analysis, ephrin B1 is clearly expressed after differentiation in the wild type but not the edited P19 cells (Fig 7C) (S2 Table).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. p300 edited P19 cells do not undergo neuronal differentiation.

A. Expression of ephrin B1, a gene critical for neuronal migration and the maintenance of neuronal progenitors, was induced by Wnt and all-trans retinoic acid (ATRA) in an additive manner in wild type P19 cells, as assessed by realtime RTPCR. Although expression of ephrin B1 was induced by Wnt3a and ATRA, individually, there was antagonism between the two treatments in the edited cells. n = 3, *p < 0.05, **p < 0.01. Vehicle/control data (set to 1) as indicated with a red, dashed horizontal. B and C. Wild type (WT) and edited P19 cells were subjected to an established neuronal differentiation protocol (as detailed in the Experimental Procedures). As demonstrated with light microscopy, WT cells stopped proliferating and neurite outgrowth ensued, whereas edited cells continued to proliferate and neurite outgrowth was not observed (20X magnification) (B). By immuno-fluorescence microscopy, ephrin B1 was clearly expressed after subjected to the differentiation protocol in WT but not edited P19 cells (20X magnification) (C).

https://doi.org/10.1371/journal.pone.0200714.g007