Table 1.
Antibodies used in this work.
Table 2.
Conditions for the expression of recombinant proteins.
Table 3.
Data of TaqMan probes.
Fig 1.
PDRG1 interacts with methionine adenosyltransferase α1.
(A) Growth of yeast cotransfectants harboring pGBKT7-MATα1 (bait) and pACT2 plasmids (prey) including ORFs of MATα1, PDRG1, clone M2, clone M6 or laminin (negative control) in low (-LW) and high (-AHLW) stringency SC media. Additional controls including the empty pGBK plasmid are shown on the right. (B) Representative anti-FLAG immunoprecipitation results from four independent experiments using total lysates of CHO cells transiently cotransfected with pFLAG-MAT and pHA-PDRG1 or the empty plasmids (mock). The size of the standards is indicated on the left side of the panel. (C) Representative anti-HA immunoprecipitation data from three independent experiments utilizing total lysates of HEK 293T cells transiently cotransfected with pFLAG-MAT and pHA-PDRG1 or the empty plasmids (mock). Western blots of the input fractions were developed using anti-FLAG and anti-HA, whereas immunoprecipitates were analyzed using anti-HA or anti-FLAG with mouse TrueBlot ULTRA, as required. The arrow indicates an unspecific band recognized by anti-FLAG slightly over the FLAG-MATα1 signal in HEK 293T samples. The size of the standards is indicated on the left side of the panel. (D) Pull-down confirmation of the interaction using glutathione Sepharose beads loaded with GST or GST-PDRG1 and incubated with recombinant MATα1 plus excess GST. Results shown correspond to a typical experiments out of the five carried out; input fractions of the recombinant proteins used (left) and pull-down results (right) are shown. The size of the standards is indicated on the left side of the panel.
Fig 2.
Structural model of rat PDRG1 and interaction of PDRG1 truncated forms with MATα1.
(A) PDRG1 structural model comprising residues K27-Q106 obtained with PHYRE. (B) Schematic representation of PDRG1 and the truncated forms prepared; the modeled area (white box) and deleted sequences (crossed box) are indicated. (C) Representative western blots of pull-down experiments carried out with recombinant truncated PDRG1 forms and MATα1 using anti-GST and anti-MATα1. Incubations with MATα1 were carried out in the presence of excess GST to avoid unspecific binding. The size of the standards is indicated on the left side of the panels. (D) Quantification of the MATα1/GST-PDRG1 signal ratio (mean ± SEM) from seven independent pull-down experiments (*p≤0.05 vs GST-PDRG1).
Fig 3.
Subcellular localization of PDRG1 in mammalian cell lines.
(A) Representative confocal immunofluorescence images of HA-PDRG1 localization using mouse anti-HA and anti-mouse Alexa Fluor 488; a minimum of three independent experiments were carried out in cuadruplicate. CHO (Chinese hamster ovary), COS-7 (monkey kidney), H35 (rat hepatoma) and N2a (mouse neuroblastoma) cells were transiently transfected with pHA, pHA-PDRG1, pEGFP or pPDRG1-EGFP. (B) Representative results of direct fluorescence localization using confocal microscopy of EGFP and PDRG1-EGFP; three independent experiments were performed in duplicate. Both panels show colocalization with Hoechst nuclear staining in white. (C) Histograms (mean ± SEM) show quantification results of nuclear (N) and cytoplasmic (C) fluorescence signals of a minimum of 200 cells per condition, using the Leica confocal software. Results of the C/N signal ratio calculated from immunofluorescence experiments are depicted. (D) Data of the C/N signal ratio from direct fluorescence observations. Cells were classified as: C>N with a ratio above 1.2; C = N when the ratio was 1 ± 0.2; and C<N with ratios below 0.8. Statistical evaluation was done by means of one-way ANOVA with Bonferroni post-hoc (*p<0.05 vs C = N). Scale bar = 10 μm.
Fig 4.
Subcellular distribution of HA-PDRG1 and the HA-PDRG1/ MATα1 interaction.
(A) Representative western blots of nuclear (N) and cytoplasmic (C) fractions (50–70 μg) obtained from HEK 293T cells transiently transfected with pHA or pHA-PDRG1 in three independent experiments. Membranes were analyzed using anti-HA, anti-TBP (nuclear marker) and anti-tubulin (cytoplasmic marker). (B) Representative confocal immunofluorescence images (N = 50 per cell line) of CHO, COS-7, HEK 293T and H35 cells transiently cotransfected with pHA-PDRG1 and pFLAG-MAT obtained using mouse anti-HA, rabbit anti-MATα1 and the corresponding secondary antibodies coupled to Alexa Fluor 488 (green) or 546 (red). Colocalization of both proteins is shown in yellow (scale bar = 12 μm) (C) Representative results of anti-FLAG immunoprecipitations carried out in N and C fractions obtained from HEK 293T cells transiently cotransfected with pFLAG-MAT and pHA-PDRG1 or the empty plasmids (mock). Input fractions were analyzed using anti-FLAG, anti-HA, and antibodies against nuclear and cytoplasmic markers, whereas immunoprecipitates were examined using anti-HA. Results correspond to a minimum of three independent experiments. (D) Representative confocal immunofluorescence images of nuclear matrix preparations of HEK 293T and CHO cells overexpressing HA-PDRG1 using rat anti-HA (green), mouse anti-SC-35 (red) and appropriate secondary antibodies conjugated to Alexa Fluor dyes (N = 50 per cell line); colocalization appears in orange. Scale bar = 12 μm.
Fig 5.
Evaluation of the PDRG1/ MATα1 association in nuclear extracts by analytical gel filtration chromatography.
(A) Elution profile of nuclear extracts overexpressing HA-PDRG1 obtained on a Superose 12 10/300 GL column and analyzed by dot-blot using anti-HA. (B) Elution profile of nuclear FLAG-MATα1 detected using anti-MATα1. (C) Elution profile of nuclear extracts overexpressing HA-PDRG1 and FLAG-MATα1; the arrow indicates the new peak recognized by both antibodies (anti-HA (▲) and anti-MAT (●)). Elution of the protein standards was as follows: blue dextran (7.4 ml); ferritin (9.82 ml); β-amylase (a; 10.62 ml); aldolase (11.1 ml); alcohol dehydrogenase (b; 11.34 ml); conalbumin (c; 12.78 ml); ovalbumin (13.3 ml); carbonic anhydrase (14 ml); lysozyme (17.31 ml); and ATP (17.65 ml). The figure shows representative profiles obtained in five independent experiments.
Fig 6.
Pdrg1 expression evaluated by real-time RT-PCR in rat tissues and models of hepatic disease.
(A) Pdrg1 expression levels in several rat tissues (N = 3), using kidney levels as reference for graphical purposes. (B) Comparison of Mat1a (white) and Pdrg1 (black) expression levels using kidney levels as reference for graphical purposes. (C) Changes in Pdrg1 hepatic expression at early stages of Wilson’s disease using Long Evans Cinnamon rats 9-weeks old (LEC, N = 6) and matched control Long Evans rats (LE, N = 5). (D) Changes in Pdrg1 hepatic expression upon D-galactosamine intoxication for 48 h (control group N = 13, galactosamine group N = 11). (E) Pdrg1 expression differences between rat hepatoma H35 cells (N = 12) and normal livers of Wistar (W; N = 13) and LE (N = 5) rats. Histograms show the mean ± SEM of the fold change calculated against the control group using 18s data as reference. Statistical evaluation of the change in the animal models was performed by Students t-test against the appropriate control group (*p≤0.05).
Fig 7.
Pull-down analysis of PDRG1 interaction with MATα2 and MAT II.
(A) Representative western blots of pull-down experiments using glutathione Sepharose beads loaded with GST or GST-PDRG1 and recombinant MATα2, MATβ or the hetero-oligomer MAT II; anti-GST, anti-MATα2 and MATβ were used for detection. The size of the standards is indicated on the left side of the panels. (B) Quantification of the MATα2/GST-PDRG1 signal ratio (mean ± SEM) from five independent pull-down experiments. (C) Representative western blots of pull-down experiments carried out with the truncated PDRG1 forms and recombinant MATα2 using anti-GST and anti-MATα2. The size of the standards is indicated on the left side of the panels. (D) Quantification of the MATα2/GST-PDRG1 signal ratio (mean ± SEM) from five independent pull-down experiments. All the incubations with MAT subunits or MAT II were carried out in the presence of excess GST to avoid unspecific binding. (*p≤0.05 vs GST-PDRG1).
Fig 8.
Effects of PDRG1 in DNA methylation and MAT activity.
(A) Global DNA methylation levels of CHO cells transiently transfected with pHA-PDRG1, pFLAG-MAT or both plasmids evaluated with the inverse radioactive assay and compared to mock transfected cells. Incorporation of methyl groups into DNA (mean ± SEM) of five independent experiments carried out in triplicate is shown. For graphical purposes, the data are expressed as percentage of the pFLAG control taken as 100% (23392.65 ± 1790.07 cpm). Statistical analysis was performed using GraphPad Prism and changes were considered significant when p≤0.05 (*vs. pFLAG; ** vs. pHA; ***vs.FLAG-MAT). (B) Purified recombinant MATα1 (0.7 μM) was incubated with 0–5.6 μM PDRG1 (black) and S-adenosylmethionine synthesis determined; the panel shows results (mean ± SEM) of a typical experiment out of five carried out in triplicate. Controls including MATα1 and histone IIA (red) were also performed (C) Results (mean ± SEM) of a typical activity assay out of three performed in triplicate using MATα2 (0.7 μM). (D) Effects of PDRG1 (mean ± SEM) on MAT II activity (0.7 μM) from a typical experiment out of three carried out in triplicate. (E) Typical profile of a Biogel A purification of the MATα1/GST-PDRG1 complex followed by MAT activity. Elution of the standards is indicated with sticks that correspond to: Blue dextran (40 ml); ferritin (48 ml); aldolase (69 ml); conalbumin (81 ml); ovalbumin (84 ml); and ATP (105 ml). The upper part of the panel shows a stained SDS-PAGE gel of the relevant fractions as indicated on the top; the molecular size of the markers shown in the last lane (right) is indicated next to the corresponding stained band. (F) Comparison of the MAT activity shown by the MATα1/GST-PDRG1 (MATα1-HE; top) and MATα2/GST-PDRG1 complexes (MATα2-HE; bottom) vs. MATα1 or MATα2 homo-oligomers as correspond. The results shown are mean ± SEM of three independent experiments; *p<0.05.
Fig 9.
Differential expression analysis of Pdrg1 silenced clones prepared in H35 cells.
(A) Real-time RT-PCR analysis of Pdrg1 expression using the 18s gene as reference in the stable silenced clones (3–44 and 4–18) and the negative control clone (CN-10) prepared in H35 cells. The results shown are the mean ± SEM of four independent experiments carried out in triplicate. (B) Growth curves for H35 wild type cells (♦), the CN-10 (■) and 4–18 (●) clones; the figure shows the mean ± SEM of eight replicates of a representative independent experiment from the four carried out. (C) Pathway analysis of genes exhibiting expression changes ≥2-fold using Gene Ontology; only pathways with p<0.05 are indicated. (D) Real-time RT-PCR verification of expression changes (mean ± SEM; N = 4) in selected genes using the Rn18s gene as reference.
Fig 10.
Schematic representation of PDRG1 interactions.
PDRG1 interactions from the literature and the present work are presented. Arrows indicate interactions identified by a single method, bait (tail) and prey (head). Double-headed arrows indicate interactions confirmed by several methods (solid lines) or by the same method using either protein as bait (dashed lines). Proteins indicated in red have been identified in several studies by different authors. Arrow’s color codes indicate: yeast two-hybrid (YTH; red); affinity-purification coupled to mass spectrometry (AP-MS; violet); immunoprecipitation (IP) and YTH (black); YTH, IP, pull-down (PD) and activity (green); AP-MS and IP (blue); PD and activity (brown).