Chemical and Antibodies
GalNAz, Ac4GalNAz, Ac4GlcNAz, Ac36AzGlcNAc, GlcNAz, 6AzGlcNAc, and thiamet-G were synthesized and characterized to be > 95% pure as previously reported76–79. Commonly used chemicals and solvents were purchased from MilliporeSigma or Thermo Fisher. Other chemicals and antibodies are listed in Supplementary Table 7.
Molecular Cloning
To generate pgPDLIM7 for protein purification, OGT and cMyc-tagged PDLIM7 were subcloned into pETDUET (MilliporeSigma) in which PDLIM7 was fused with an N-terminal His6-tag. The shRNAs targeting the 3’-UTR of OGA (TRCN0000285410) or PDLIM7 (TRCN0000350814) were selected from the Genetic Perturbation Platform (Broad Institute) and cloned into the EZ-Tet-pLKO-Puro (Addgene #85966) to generate EZ-Tet-pLKO-Puro-shOGA or EZ-Tet-pLKO-Puro-shPDLIM7 as previously described80. Human full-length OGA expression plasmid pcDNA5-FRT-TO-Flag-OGA-HA was generated by PCR amplification of OGA using primers encoding the N-terminal Flag3 (DYKDHDGDYKDHDIDYKDDDDK)-tag and C-terminal HA (YPYDVPDYA)-tag, and subcloned into the pcDNA5-FRT-TO (Addgene #40998). pcDNA5-FRT-TO-Flag-OGA was generated similarly using primers encoding the Flag3-tag. Human full-length PDLIM7 expression plasmid pcDNA3.1-PDLIM7 was generated by PCR amplification of PDLIM7 from PDLIM7_pLX304 (Harvard PlasmID #HsCD00411962) and subcloned into pcDNA3.1 fused with a C-terminal cMyc (EQKLISEEDL)-His6-tag (Addgene #64281). pcDNA5-FRT-TO-PDLIM7-cMyc, pLenti-PDLIM7-cMyc, and Tet-O-PDLIM7-cMyc-Hygro were similarly generated from pcDNA5-FRT-TO, pLenti (a kind gift from Dr. Meyer Jackson), and Tet-O-Hygro (Addgene #117271), respectively. Human HA-tagged p53 expression plasmid pcDNA3-p53 was purchased from Addgene (#69003), and subcloned into pLenti to generate pLenti-p53.
All plasmids encoding OGA and PDLIM7 mutants were generated by the QuikChange II XL Site-Directed Mutagenesis Kit (Agilent) according to manufacturer’s instructions, using the wild-type (WT) PDLIM7 or OGA plasmids as the DNA templates. All the constructs generated in this study were verified by DNA sequencing. All the primers and plasmids used in this study are listed in Supplementary Table 8 and 9, respectively.
Cell Lines, Cell Culture, and Doxycycline Induction
The Flp-In T-REx-293 (T-REx-293) cell line was purchased from Invitrogen (#R78007). HEK293, HEK293T, Hela, NCI-H1299, and NCI-H460 cell lines were obtained from American Type Culture Collection (ATCC). The stable cell lines used in this study were generated from T-REx-293, H1299, or H460 cells (Supplementary Table 10). All cells were maintained in 5% CO2 at 37 °C. H1299 and the derived stable cell lines were cultured in RPMI-1640 media (Corning) supplemented with 10% fetal bovine serum (FBS, MilliporeSigma #F0926) and 0.8% penicillin-streptomycin unless otherwise indicated. Other cell lines were cultured in Dulbecco's Modified Eagle Medium (DMEM, Corning) supplemented with 10% FBS and 0.8% penicillin-streptomycin. To induce expression or knockdown of protein target in the stable cell lines, cells were treated with doxycycline ranging from 2.5-60 ng/mL for the indicated time.
Transfection, Lentiviral Particle Production, and Generation of Stable Cell Lines
For transient transfection of HEK293T, HEK293, T-REx-293, and T-REx-293 derived stable cell lines, cells were seeded in 6-well plate 22-24 h prior to transfection (around 70% confluency). The plasmids were delivered into the cells using calcium phosphate as previously reported81. Transient transfection of Hela, H460, H1299, and their derived stable cell lines were performed by Trans IT-X2 (Mirus) or Trans IT-2020 (Mirus) following manufacturer’s instructions. Cells were seeded in 6-well plate 20-24 h prior to transfection (around 80-90% confluency). All cells were cultured in antibiotic-free media during the transfection process. In general, 0.5-4 μg DNA of each plasmid was applied for transfection in each well. Doxycycline was added 24 h post-transfection if the induction was needed. Cells were washed by PBS, harvested by scrapping, and stored at -80 °C until use.
For producing lentiviral particles, HEK293T cells were co-transfected with the plasmid encoding target cDNA or shRNA, psPAX2 (Addgene #12260), and pCMV-VSV-G (Addgene #8454) (Supplementary Table 9). Lentivirus were harvested at 48 and 72 h post-transfection and stored at -80 °C until use.
To generate stable cell lines with Flp-In T-RΕx inducible expression, T-REx-293 cells were co-transfected with plasmids encoding the target genes and pOG44 (Invitrogen #V600520) at a ratio of around 1:10 using calcium phosphate as mentioned above (Supplementary Table 9). Other stable cell lines were generated by lentiviral transduction with 0.08 μg/mL hexadimethrine bromide for 48 h. Cells were then replated in 10-cm plate and screened for 2-3 weeks with antibiotics-containing media.
Western Blot
To prepare the whole cell lysates, cell pellets were lysed in RIPA buffer (50 mM Tris pH 8.0, 150 mM NaCl, 0.2 mM EDTA, 0.1% SDS, 0.5% sodium deoxycholate, and 1% NP-40). To prepare the nuclear extracts, cell pellets were first lysed in PBS containing 0.1% NP-40 followed by centrifugation at 10,000 g for 30 min at 4 °C. The remaining pellets were then lysed in nuclear extraction buffer (50 mM Tris pH 8.0, 400 mM NaCl, 20% glycerol, 0.5% sodium deoxycholate, 1% NP-40, and 0.5% SDS). All buffers contain protease inhibitor cocktail (MilliporeSigma). For O-GlcNAcylation detection, 10 μM thiamet-G was included in the buffer during cell lysis. Cell debris was removed by centrifugation at 16,000 g at 4 °C for 20 min. Protein concentration was measured by BCA protein assay kit (Pierce). Cell lysates were separated by SDS-PAGE and transferred onto a nitrocellulose membrane using iBlot (Life Technology). The membrane was then blocked with 0.9% bovine serum albumin followed by primary antibody hybridization overnight at 4 °C. Secondary antibody was applied to the blot with 1 h incubation at room temperature (r.t). The blot was developed by ECL substrate (Azure Biosystem or Bio-Rad). Signal was detected on C600 imaging system (Azure Biosystem) using chemiluminescent mode. Relative quantitation of the proteins was conducted by the ImageStudio Lite software (v 5.2, LI-COR). The overall intensity of the detected target was normalized to the corresponding loading control protein (actin for the whole cell lysates, lamin B1 or PCNA for nuclear extracts) in each sample.
In-Gel Fluorescence
For metabolic labeling, cells were seeded in 6-well plates to reach 70% confluency. After around 20 h of seeding, the growth media was replaced with low-glucose, antibiotic-free DMEM (0.5 g/L glucose, 10% FBS) containing different sugar analogues (200 μM for Ac4GalNAz, Ac4GlcNAz, and Ac36AzGlcNAc; 2 mM for GalNAz, GlcNAz, and 6AzGlcNAc). After 24 h incubation, cells were washed by PBS, harvested by scrapping, and stored at -80 °C until use. Cell pellets were lysed in NP-40 buffer (1% NP-40, 150 mM NaCl, and 50 mM HEPES pH 7.5). Cell debris was removed by centrifugation at 16,000 g at 4 °C for 20 min. Protein concentration was measured as described above. 35 μg lysates were diluted to 1 μg/μL by 20 mM HEPES pH 7.5. The freshly prepared click chemistry reagent mix (1 mM CuSO4, 5 mM Tris(3-hydroxypropyltriazolylmethyl)amine (THPTA), 50 μM fluor 488-alkyne (MilliporeSigma), and 7.5 mM sodium ascorbate) was then added to each sample and incubated for 1 h in the dark. Samples were precipitated by MeOH overnight at -80 °C and re-dissolved in 4% SDS. For in vitro non-enzymatic labeling, lysates similarly prepared from the HEK293 cells without sugar treatment were incubated with each sugar analogue (200 μM) at 37 °C for 2 h followed by click chemistry and protein precipitation. The re-solubilized proteins were separated by SDS-PAGE and detected by fluorescence scanning. Coomassie blue staining was then applied to obtain the total protein loading. All the imaging was performed on C600 imaging system.
Co-Immunoprecipitation (co-IP)
For OGA interactome analysis, T-REx-293 cells with inducible OGA-D175N variant were cultured in 15-cm plate at 80% confluency for 17-18 h followed by 8 h of doxycycline induction. Cells were then washed by PBS, harvested by scrapping, snap-frozen in liquid N2, and stored at -80 °C until use. The cell pellet was lysed in IP buffer (1% NP-40, 50 mM HEPES pH 7.5, 150 mM NaCl, 0.1% sodium deoxycholate). Protein concentration was determined as mentioned above. Lysates were then diluted to 1 μg/μL by PBS. All buffers contained protease inhibitor cocktail. The affinity purification of Flag-, HA-, or cMyc-tagged protein complexes were performed as previously reported using anti-Flag (M2, MilliporeSigma), anti-HA (Pierce) or anti-cMyc agarose (Pierce), respectively, at 4 °C for 16-19 h with gentle rotation82. The resins were gently washed three times with 20 mM Tris-HCl pH 7.5, 150 mM NaCl, and 0.05% NP-40 followed by washing once with 20 mM Tris-HCl pH 7.5 and 150 mM NaCl. The resin bound proteins were eluted by 0.1 M glycine pH 2.6 containing 0.5% NP-40, and immediately neutralized by 1 M NH4HCO3 followed by MeOH precipitation overnight at -80 °C. Protein pellets were washed with MeOH and solubilized in 8 M urea/0.1 M triethylammonium bicarbonate (TEAB) with rotation and sonication. Protein solution was then subjected to the sample preparation for proteomic analysis (please see “Sample Preparation for Proteomic Analysis” section). Sample for PDLIM7 phosphopeptide analysis was similarly prepared using T-REx-293 cells with inducible knockdown of endogenous PDLIM7 and co-transfection with HA-tagged p53, ubiquitin, and cMyc-tagged WT PDLIM7 or mutant. IP samples for western blot analysis were prepared similarly as above with the following modifications: 10 μM thiamet-G was included in the buffer during cell lysis and affinity purification to retain the protein O-GlcNAcylation if needed; 10 mM iodoacetamide was included in the buffer during cell lysis for ubiquitination detection; 1/20 of the total lysates for affinity purification was used as the input control; three-time wash was applied to affinity purification for O-GlcNAcylation detection; the resin bound proteins were eluted twice by incubating with two-fold SDS-loading buffer at 35 °C for 15 min. For the second elution, resins were boiled for 5 min prior to centrifugation. All protein samples were subjected to SDS-PAGE for western blot analysis. Relative quantitation of the protein was performed using ImageStudio Lite software. The intensity of detected target was normalized to the intensity of corresponding actin loading control or the total affinity purified tagged proteins.
Biotin Conjugation of O-GlcNAcylated Proteins for O-GlcNAcome Analysis
T-REx-293 cells with inducible Flag-tagged WT OGA or mutants were seeded at 6×106 cells/plate in 15-cm dish for 27-28 h. The growth medium was then changed to low-glucose (0.5 g/L), antibiotic-free DMEM supplemented with 10% FBS and 2.5 mM GalNAz. Doxycycline was also added to the cells for Flag-tagged OGA induction. After 24 h incubation, cells were washed by PBS, harvested by scrapping, snap-frozen in liquid N2, and stored at -80 °C until use. Cell pellets were first lysed in NP-40 buffer for 1 h followed by centrifugation (16,000 g, 20 min, 4 °C). The remaining pellets were further lysed in SDC buffer (0.5% sodium deoxycholate, 400 mM NaCl, 20% glycerol, and 50 mM HEPES pH 7.5) followed by centrifugation to remove cell debris. All buffers contained protease inhibitor cocktail and 10 μM thiamet-G. Protein concentration was determined as mentioned above. 10 and 1 mg cell lysates from the first and the second lysis step were diluted to 2.5 and 1 μg/μL, respectively, by 50 mM HEPES pH 7.5. Freshly prepared click chemistry reagent mix (5 mM THPTA, 1 mM CuSO4, 100 uM Biotin-PEG4-alkyne, and 7.5 mM ascorbate) was then added to each lysate sample followed by 2 h incubation at r.t with gentle rotation. Lysates from the same cell sample were then combined for MeOH precipitation and re-solubilization as mentioned above. The re-solubilized proteins were subjected to O-GlcNAcome and whole proteome analyses as described in “Sample Preparation for Proteomic Analysis”.
Sample Preparation for Proteomic Analysis
Protein Digestion
Proteins were first reduced by 10 mM dithiothreitol for 30 min at r.t. followed by alkylation with 55 mM iodoacetamide in the dark at r.t. for 30 min. The protein solution was then seven-fold diluted using 25 mM TEAB and digested by trypsin/Lys-C mix (100:1, protein/protease, w/w, Promega) at r.t. for 20 h. Protease activity was quenched by 0.5% formic acid (FA) followed by SDB-XC StageTip desalting as previously reported83. The final peptide samples were dried by vacuum centrifugation and stored at -80 °C until use.
O-GlcNAcylated Peptide Enrichment
The dried peptides were resuspended in 0.1 M TEAB with incubation and sonication. The peptide concentration was estimated by BCA assay. For whole proteome analysis, 10 μg peptides from each condition were directly subjected to dimethyl labeling (please see “Dimethyl Labeling” section). For O-GlcNAcome analysis, the same amount of peptides from each condition was diluted to approximately 2 mg/mL with PBS. The enrichment of biotin-conjugated peptides was performed using the reported DiDBiT method with slight modifications84. Briefly, peptide solution was mixed with NeutrAvidin agarose resin (Life Technology) for 3 h at r.t. with rotation. The resins were pelleted by centrifugation (1,000 g, 5 min) to remove unbound peptides and washed once with 1 mL PBS, four times with 5% ACN/PBS, and once with 1 mL HPLC water. The biotinylated peptides were then eluted three times by 80% ACN/0.3% FA. For the last elution, resins were heated at 80 °C for 5 min prior to centrifugation. The combined eluents were dried by vacuum centrifugation and stored at -80 °C until dimethyl labeling.
Dimethyl Labeling
Tryptic peptide samples were redissolved in 100 μL of 0.1 M TEAB solution. The labeling was conducted as previously reported85. Briefly, every 100 μL peptide sample from WT OGA, R586A, and S652F mutants were added with 6 μL of 4% formaldehyde-H2 (L), 4% formaldehyde-D2 (M), and 4% formaldehyde-13CD2O (H), respectively. Samples were then mixed with 6 μL of freshly prepared sodium cyanoborohydride (for “L” and “M” labeling) or sodium cyanoborodeuteride (for “H” labeling). Each sample was vigorously mixed and incubated at r.t. for 60 min with rotation. The reaction was quenched by adding 1% ammonium hydroxide and 10% FA. The L-, M- and H-labeled samples were then combined and subjected to SDB-XC StageTip desalting. The final peptide samples were dried by vacuum centrifugation and stored at -80 °C until peptide fractionation.
Peptide Fractionation
Peptide fractionation was conducted by SDB-RPS StageTip as previously reported86,87. Peptides were eluted with the following buffers: 60 mM ammonium acetate/20% ACN/0.5% FA, 100 mM ammonium acetate/30% ACN/0.5% FA, 150 mM ammonium acetate/80% ACN/0.5% FA, and 5% ammonium hydroxide/80% ACN. For O-GlcNAcome analysis, two additional fractions were collected using 30 mM ammonium acetate/10% ACN/0.5% FA and 80 mM ammonium acetate/60% ACN/0.5% FA. For whole proteome analysis, an additional fraction was collected using 80 mM ammonium acetate/60% ACN/0.5% FA.
NanoLC-MS/MS Analysis
Peptides were resuspended in 0.1% FA and analyzed on either a Thermo Scientific Orbitrap Fusion Lumos Tribrid mass spectrometer (for O-GlcNAcome analysis) or Q Exactive mass spectrometer (for analyses of whole proteome and OGA interactome) equipped with a Dionex Ultimate 3000 UPLC system (Thermo Scientific). Peptides were separated on a 75 μm × 15 cm homemade column packed with BEH C18 (1.7 μm, 150 Å, Waters). Mobile phase A consisted of 0.1% FA, and B consisted of ACN/0.1% FA. A linear gradient of 3–30% B in 18 min, 30–75% B in 80 min, and 75-95% B in 10 min was employed throughout the study. For glycosylated peptide analysis on Fusion Lumos, mass spectra from survey full scans were acquired on the Orbitrap (m/z 300–1800) at a resolution of 60,000 with automated gain control (AGC) value of 2×105 and maximum injection time of 50 ms. The data dependent HCD was acquired on the most intense precursor ions at a resolution of 30,000 and a normalized collision energy of 30%. If the biotin fragment ions60 derived from the oxonium ion at m/z 270.127 or m/z 420.217 (± m/z 0.01) were detected within the top 10 most abundant peaks, the EThcD collision-induced dissociation MS/MS scan would be triggered with a resolution of 60,000 and a normalized collision energy of 25% at AGC value of 5×104. For other peptide analyses by Q Exactive, mass spectra from survey full scans were acquired on the Orbitrap (m/z 300-1500) at a resolution of 70,000 with AGC value of 106 and maximum injection time of 100 ms. The top 15 precursor ions were selected for the subsequent data dependent HCD MS/MS fragmentation at a resolution of 17,500 and the AGC value of 105. All proteomic analyses were performed with three biological and two technical repeats.
Proteomic Data Analysis
Raw MS spectra were processed for peak detection and quantitation using MaxQuant88 (v. 1.6.1.0) and Perseus89 (v. 1.6.7.0) software with default settings. Peptides were searched against the Uniprot database90 (release 2017_09). Basic search criteria including trypsin specificity, dimethyl labeling, variable modifications of oxidation (Met) and carbamidomethyl (Cys), and up to two missed cleavages were used for all data processing. Specific variable modifications were applied in O-GlcNAcome dataset: glycosylation of Cys/Ser/Thr (H47C29N7O11S, +701.305 Da). Using a decoy database strategy, peptide identification was accepted based on the posterior error probability with a false discovery rate of 1%. Precursor intensities of identified peptides were further searched and recalculated using the “match between runs” option in MaxQuant. For peptides with post-translational modifications (PTMs), the localization probability of all putative modification sites was calculated by the MaxQuant PTM score algorithm based on the peptide spectral match and the potential modification sites in each peptide91. Sites with a PTM probability > 0.75 were considered unambiguously identified. Data imputation algorithm incorporated in Perseus was applied to enable statistical evaluation. For the quantitative analysis of O-GlcNAcome, the log2 ratio (mutant/WT) of O-GlcNAc site was normalized to the log2 ratio of proteins from the whole proteome (proteins not detected in the whole proteome were given log2 ratio = 0). Regarding the average standard deviation and the size of dataset from each proteomic analysis, O-GlcNAc sites that displayed a minimum of average 1.5-fold change (p ≤ 0.05) between WT OGA and mutant cells were considered dysregulated. For quantitative analysis of the whole proteome and OGA interactome, protein groups with a minimum of average 1.75-fold change (p ≤ 0.05) between WT OGA and mutant cells were considered dysregulated. The p-value was calculated by one-tailed paired student t-test using Excel (Microsoft).
Bioinformatics Analysis
Volcano plot was generated using VolcaNoseR92. Information of protein-protein interactions and the functional term enrichment were retrieved from STRING database93,94 (STRING consortium, v.1.5.1) with a confidence threshold of 0.5. Visualization of protein association network was conducted by Cytoscape95 software (Cytoscape Consortium, v. 3.7.1). Protein clustering was performed using MCL (Markov Clustering) algorithm96. Six extra protein nodes were randomly added for network generation to achieve better clustering and connection between proteins. The sequence logo of peptides flanking the O-GlcNAc site was generated by pLogo97. O-GlcNAc sites near C- or N-terminus of proteins with the flanking sequence shorter than seven residues were excluded from pLogo analysis. Venn diagram was generated by Free Venn Diagram Marker software (Media Freeware, v. 1.0.0).
Protein Purification
To purify O-GlcNAcylated PDLIM7 (gPDLIM7), the pgPDLIM7 plasmid was transformed into E. coli Rosetta2(DE3) competent cells, and the transformants were grown at 37 °C in Luria-Bertani (LB) media. After the optical density reached 0.6 at 600 nm, the culture was induced by 0.3 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) at 16 °C for 16 h. The cells were pelleted, resuspended in buffer containing 50 mM Na3PO4 pH 8.0, 300 mM NaCl, and 1 mM phenylmethylsulfonyl fluoride (PMSF). Cells were lysed with ultra-high-pressure cell disruptor EmulsiFlex-C5 (Avestin). After centrifugation, the supernatant was subjected to an Ni-NTA column (Qiagen) for affinity purification. The recombinant protein was subsequently eluted by buffer containing 50 mM Na3PO4 pH 8.0, 300 mM NaCl, 250 mM imidazole. Further purification was performed by size-exclusion chromatography (Superdex 200 increase 10/300; GE Healthcare) in the buffer containing 50 mM Na3PO4 pH 8.0, 300 mM NaCl, and 0.5 mM Tris(3-hydroxypropyl)phosphine (THP). The O-GlcNAcylation of purified gPDLIM7 protein was validated by western blot analysis using O-GlcNAc antibody (CTD110.6) as mentioned above.
To purify OGA, the plasmid pET21b-OGA was transformed into E. coli BL21(DE3) competent cells, and the transformants were grown at 37 °C in LB media. After optical density reached 0.6 at 600 nm, the culture was induced by 0.3 mM IPTG at 16 °C for 16 h. The cells were pelleted, resuspended in buffer containing 20 mM sodium phosphate pH 7.0, 150 mM NaCl, and 1 mM PMSF. Cell lysis and affinity purification with Ni-NTA column were performed similarly as above. The recombinant protein was eluted with buffer containing 20 mM sodium phosphate pH 7.0, 150 mM NaCl, and 250 mM imidazole. The eluted OGA was further purified by size-exclusion chromatography in buffer containing 20 mM sodium phosphate pH 7.0, 150 mM NaCl, and 0.5 mM THP. OGA mutant proteins were expressed and purified similarly. All purified proteins were stored at -80 °C until use.
OGA Enzymatic Assay
The steady-state kinetics of WT OGA and mutants were measured with the fluorogenic substrate 4-methylumbelliferyl-N-acetyl-β-D-glucosaminide (4MU-GlcNAc) as previously reported57. Briefly, 2 nM enzyme was incubated with each of 10, 50, 150, 300, 600, and 1200 µM 4MU-GlcNAc as the substrate in 25 µL reaction (50 mM NaH2PO4 pH 6.5, 100 mM NaCl, 0.1 mg/mL BSA) for 10 min at 37 °C. The reaction was quenched by 150 µL of 200 mM glycine pH 10.75. The fluorescence of liberated 4-methylumbelliferone was measured using BioTek Synergy H1 Hybrid microplate reader (Agilent) with excitation and emission wavelengths of 360 and 450 nm, respectively. Enzymatic activity was converted to µM/min using a standard curve of free 4-methylumbelliferone. All analyses were carried out in triplicate. Michaelis-Menten kinetic parameters were calculated by Prism 5 software (GraphPad).
In vitro OGA Deglycosylation Assay
Recombinantly purified gPDLIM7 protein (5 μM) was mixed with 1 μM recombinantly purified WT OGA or mutant in a 10 μL reaction containing 50 mM sodium phosphate pH 8.0, 250 mM NaCl, and 0.5 mM THP. The reaction was conducted at 37 °C for 30 min with gentle rotation. One sample incubated with 10 μM thiamet-G was used as a negative control. The reaction was quenched by SDS-loading buffer, boiled, and subjected to SDS-PAGE for western blot analysis as mentioned above.
Cell Growth and Anchorage-Independent Soft Agar Assays
To measure the cell growth with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), T-REx-293 cells with inducible WT OGA or mutant were cultured in 96-well plates at 30% confluency (100 μL/well) in DMEM supplemented with 10% FBS (tetracycline-free, Takara Bio) 24 h prior to doxycycline induction. Cell viability was measured at 0 h and the indicated time point post-induction by adding 20 μL of MTT to each well. After 3 h of incubation at 37 °C, the formazan dye formed in intact cells was solubilized by isopropanol containing 40 mM HCl and measured at 560 nm with background subtraction at 670 nm on BioTek Synergy H1 Hybrid microplate reader. Cell density measured at the indicated post-induction time was normalized to the density at 0 h. Results from nine wells of each condition were used for quantitation. For resazurin assay, cells with doxycycline induction were similarly prepared as above. Cells were then incubated with 50 μg/mL of resazurin for 4 h at 37 °C. Cell density was then measured on the plate reader with an excitation and emission wavelength of 544 and 590 nm, respectively. Data from four wells were used for quantitation.
For soft agar assay, cells were cultured in DMEM supplemented with 10% FBS (tetracycline-free) in 6-well plates at 60% confluency. After 48 h of doxycycline induction, cells were trypsinized and mixed with 0.48% low-melt point agarose in culture media, and then plated on top of the base layer of media containing 0.8% agarose in 6-well plates. The agarose layer was covered with 1 mL of growth media (+/- doxycycline) and replaced every 3 d. After 3-4 weeks, the colonies were dyed by crystal violet (0.005%) with 4% formaldehyde and imaged by C600 imaging system. The total colony density of each well was quantified by Image J (v 1.48, Public Domain) or AzureSpot (v 2.2.167, Azure Biosystem) software.
Wound Healing Assay
Cells were seeded in 6-well plates after 48 h of doxycycline induction and reached 100% confluency before wound making. The scratch wounds were made using a sterile 200 μL pipette tip. Doxycycline was added in the media during healing if needed. At 0 and 24 h post-scratching, cells were imaged using an inverted microscope AE2000 (Motic) at 100x magnification with the ocular lens attached to a digital camera. The wound area at 0 and 24 h post-scratching was quantified by Image J.
RNA Extraction and cDNA Synthesis
For RNA extraction, cells were cultured in 6-well plates at 30-40% confluency. After 48 h of doxycycline induction, RNA purification was performed using TRIzol reagent (Invitrogen). 1.5 μg RNA with Oligo(dT)18 and random hexamer primers were used for cDNA synthesis. cDNA synthesis was conducted by Maxima H Minus First Strand cDNA Synthesis Kit with DNase (Life Technology) or GoScript™ Reverse Transcriptase kit (Promega) according to manufacturer’s instructions. cDNA samples were stored at -80 °C and used for real-time qPCR analysis within one week.
Real-Time qPCR for p53 Transcription Analysis
The PCR sample was prepared using 2 ng cDNA and PowerUp SYBR Green Master Mix (Life Technology) following manufacturer’s instruction. No-template reaction was included as negative control. The forward and reverse primers used for p53 and reference gene GAPDH were listed in Supplementary Table 8. Real-time qPCR was performed using StepOne Plus Real-Time PCR system (Applied Biosystems) with the following cycling parameters: 50 °C for 2 min, 95 °C for 2 min, 40 cycles of amplification at 95 °C for 15 sec, 55 °C for 30 sec, and 72 °C for 1 min. The calculation of cycle threshold (Cт) values was performed using StepOne software (v. 2.1, Applied Biosystems). Each sample was assessed with four to five technical repeats. The relative abundance of the transcripts was analyzed by 2-ΔΔCт method as previously reported98.
In-Cell Ubiquitination Assay
For cell lines with stable or inducible expression of cMyc-tagged WT PDLIM7 or mutant, cells were co-transfected with HA-tagged p53 and ubiquitin. For other cell lines, cells were co-transfected with plasmids containing HA-tagged p53, cMyc-tagged WT PDLIM7 or mutant, and ubiquitin as mentioned above. The culture medium was changed to fresh medium 24 h post-transfection. Doxycycline was also added to the media if the induction of protein expression or endogenous protein knockdown was needed. After culturing for another 24 h, medium was changed again to the one containing 35 μM MG132 (1% DMSO, +/- doxycycline). Cells were then incubated for another 3 h prior to harvesting. For cells with MDM2-p53 inhibition, 20 μM Nutlin-3a (0.5% DMSO, +/- doxycycline) was added to the cells 12 h prior to MG132 treatment. Cells were then washed by PBS, harvested by scrapping, snap-frozen in liquid N2, and stored at -80 °C until use.
Analysis of Global and PDLIM7 Phosphorylation
To analyze the global phosphoproteins, HEK293 cells were prepared following the procedure of in-cell ubiquitination assay. The whole cell lysates were then prepared and separated by SDS-PAGE as mentioned above. Phosphoproteins were stained using Pro-Q Diamond Phosphoprotein Gel Stain following manufacturer’s instructions (Life Technology) and detected by fluorescence scanning. The Coomassie blue staining was then applied to obtain the total protein loading. All imaging was performed on C600 imaging system.
For analysis of PDLIM7 phosphorylation, the immunoprecipitated cMyc-tagged WT PDLIM7 or mutant was lysed, digested, and dimethyl labeled as mentioned above. Phosphopeptides were enriched by immobilized metal affinity chromatography (IMAC) as previously reported99. Briefly, to freshly prepared IMAC slurry, Ni-NTA Superflow agarose (Qiagen) was stripped with 100 mM EDTA for 30 min, reloaded with 10 mM FeCl3 for 30 min, washed three times with HPLC water, and resuspended in IMAC solution (1:1:1:1, beads/ACN/MeOH/0.01% acetic acid, v/v/v/v). The dimethyl labeled peptides were resuspended in final 80% ACN/0.1% FA and incubated with IMAC beads for 30 min at r.t. with rotation. The beads were then washed with 80% ACN/0.1% FA and 0.1% FA. The phosphopeptides were eluted by 500 mM potassium phosphate buffer followed by C18 StageTip desalting. The final peptide samples were dried by vacuum centrifugation and stored at -80 °C. The peptide samples without phosphopeptide enrichment were used to analyze the total protein level. The nanoLC-MS/MS analysis of phosphopeptides was performed on Q Exactive mass spectrometer similarly as mentioned above. Data analysis and phosphopeptide quantitation were conducted by the aforementioned MaxQuant with phosphorylation of Ser/Thr/Tyr as a specific variable modification.
Statistical Analysis
The statistical analysis of data from cell growth, anchorage-independent growth, wound healing, and western blot were performed by Excel or Prism 5 software. Data comparison was analyzed by one-tailed paired or unpaired student t-test; n ³ 3. All error bars denote the standard deviation. The statistical significance cutoff was set at p < 0.05.