Fig 2: PDPK1 interacts with WDR5 in nuclear lysates(A) Endogenous PDPK1 was recovered from lysates of the indicated cell lines and probed for co-precipitating WDR5 by IB. Inputs for PDPK1 are 10%–20%. Inputs for WDR5 are 1%–5%. n = 3 biological replicates.(B) Proximity ligation assay with FLAG and WDR5 antibodies in U2OS cells stably expressing FLAG-tagged PDPK1. Cells were treated overnight (30 µM C6/C6nc) before analysis; scale bar, 20 µm. n = 3 biological replicates.(C) HEK293 cells were treated overnight with 30 µM C6 or 5 µM GSK470, lysates prepared, and a PDPK1 IP performed. IB was then performed for the indicated proteins. Inputs are 5% for PDPK1 and 1% for all others. n = 3 biological replicates.(D) HEK293 cells were fractionated into cytosolic (S2), soluble nuclear (S3), and chromatin-associated (P3) fractions. Equal amounts of each fraction were analyzed by IB with the antibodies against the indicated proteins. H3 (nuclear) and a-tubulin (cytosolic) are controls for fractionation. n = 3 biological replicates.(E) Cytosolic and nuclear lysates from HEK293 cells were subject to IP with PDPK1 antibody or an IgG control and immunoblotted with antibodies against the indicated proteins. A short and long exposure of the WDR5 IB are shown. n = 3 biological replicates.(F) PDPK1 possesses two WIN-like motifs centered on R3 and R238.(G) FLAG-tagged PDPK1 (WT and the R3A and R238A mutants) were transiently expressed in HEK293 cells; lysates were prepared and subject to IP with anti-FLAG beads. Immune complexes were probed for PDPK1 or endogenous WDR5 by IB. n = 3 biological replicates.(H) FLAG-tagged PDPK1 (WT and the R3A) was transiently expressed in HEK293 cells, recovered by FLAG-IP, resolved by SDS-PAGE, and transferred to polyvinylidene fluoride (PVDF) membrane. Membranes were then incubated with recombinant WDR5 followed by anti-WDR5 antibody. n = 3 biological replicates.(I) In vitro-transcribed and -translated PDPK1-FLAG variants were incubated with recombinant 6xHis-SUMO-WDR5 proteins, recovered with Ni-NTA agarose, and analyzed by IB. n = 2 biological replicates.PH, pleckstrin homology domain. See also Figure S4.

Fig 3: PDPK1 is a high-affinity WIN site binding protein(A) Tandem mass spectrum of N-terminally acetylated PDPK1 peptide, residues 2–9. The doubly protonated precursor, [M+2H]+2, with m/z 491.2556 was fragmented with higher-energy collisional dissociation. The identified amino acid sequence is provided above the annotated spectrum; brackets indicate sites of dissociation at the peptide backbone. Observed product ions are assigned to their corresponding m/z peaks in the mass spectrum.(B) Binding constants of PDPK1 peptides were determined using a TR-FRET-based KMT2A peptide competition assay. All peptides are amidated at the C terminus. Two or more repeats were obtained; average Ki values and standard deviations are reported.(C) Structure of WDR5 in complex with the acetylated-PDPK1 WIN peptide. The PDPK1 WIN peptide is shown in stick representation (magenta, colored by atom type); WDR5 is shown as cartoon with semitransparent surface representation (gray). 2.7Å resolution.(D) As in (C) but rotated along a 90°axis.(E) The Fo-Fc omit map of PDPK1 peptide bound with WDR5 domain contoured at 2.0 s level. PDPK1 peptide is shown in magenta sticks.(F) Close-up of the first three residues of PDPK1 (ART) in the WIN site of WDR5. The PDPK1 peptide is green sticks; WDR5 is gray ribbons. Key WDR5 residues F133, S175, C261, and F263 are indicated in pink stick representation. Yellow dotted lines indicate intermolecular hydrogen bonds.(G) WIN motif of PDPK1 aligned with established WIN motifs. The related histone methyltransferase enzymes are grouped with brackets.(H) Table summarizing consensus sequences for internal (WINI) and N-terminal (WINN) WIN motifs.See also Figure S5 and Tables S3 and S4.

Fig 4: PDPK1 and WDR5 oppositely influence the expression of cell-cycle genes(A) U2OS cells expressing PDPK1-FKBP(F36V)-2xHA were treated for 24 h with 500 nM dTAG47 or DMSO, lysates prepared, and PDPK1, WDR5, and GAPDH levels determined by IB.(B) Number of transcripts significantly (FDR < 0.05) altered by 24-h treatment of cells in (A) with 500 nM dTAG47, compared with DMSO control. n = 3 biological replicates.(C) As in (A) but for the cells expressing WDR5-FKBP(F36V)-2xHA.(D) Number of transcripts significantly (FDR < 0.05) altered by 24-h treatment of cells in (C) with 500 nM dTAG47, compared with DMSO control. n = 4 biological replicates.(E) GO analysis of decreased transcripts identified by RNA-seq of U2OS cells depleted of PDPK1 for 24 h. Biological Process GO terms were ranked by adjusted p value, and the 15 most significant enriched terms are presented; the color indicates the Bonferroni-corrected Fisher exact p value; the size indicates the number of genes in that category; the x axis is the ratio of genes in the category over total analyzed genes.(F) GO term analysis of increased transcripts identified by RNA-seq of U2OS cells depleted of WDR5 for 24 h. Ranking and presentation are as in (E).(G) Enriched Hallmark gene sets (Liberzon et al., 2015), determined by GSEA of RNA-seq from 24-h PDPK1 or WDR5 depletion. Eight of the top Hallmarks are shown. Color indicates the normalized enrichment score (NES); size indicates the FDR value.(H and I) Examples of GSEA enrichment plots summarized in (G). FDR = 0.000 for all plots shown.(J) Overlap of transcripts that are decreased with PDPK1 depletion and increased with WDR5 depletion.(K) Scatterplot of RNA-seq data from PDPK1 and WDR5 depletions. The 246 genes from (J) are highlighted in purple.(L) GO term analysis of the 246 genes represented in (J). Biological Process GO terms were sorted hierarchically, and the most specific subclasses were ranked by adjusted p value. The 10 most significantly enriched subclasses are presented. Presented as in (E).See also Figure S6.