Fig 2: Identification of co-translational NMT substrates in HeLa cells.(a) Log2-fold change for maximal response of YnMyr enrichment of MG-initiated proteins to NMT inhibition (compound 1, six concentrations, 0–5 µM, n=5) quantified by spike-in SILAC, plotted against enrichment relative to myristic acid fed control (by LFQ, see Supplementary Fig. 2). Circles plot substrates (green), substrates identified with lower confidence/fewer unique peptides (yellow) and non-substrates (grey circle); triangles indicate substrates for which a YnMyr-modified peptide was identified directly by MS/MS using AzRTB and/or AzKTB. Vertical scatter plot (at right) shows responses for all proteins identified, including those lacking quantifiable peptides in the myristate control. (b) Response of enrichment to NMT inhibition quantified by spike-in SILAC for the three known NMT substrates in Fig. 2d; error bars, s.d. (n=5). (c) Enrichment for three non-substrates shows no significant response up to the highest concentration of 1 tested; error bars, s.d. (n=5). (d) Dose-response curves for NMT substrates identified by quantitative chemical proteomics. Plots were generated by non-linear regression of IC50 values and slope factors to a standard dose-response equation (GraFit 7.0). Three known (PRKACA, c-Src and PSMC1) and three novel (FLOT2, MARC2 and HCCS) substrates are highlighted at the right. (e) Proportion of substrates identified here for the first time at their endogenous (native) expression level in live mammalian cells (green slices) vs those previously identified under the same conditions (white slice). (f) Biological functions for 70 NMT substrates identified in this study (Gene Ontology annotations).

Fig 3: Validation and high-throughput biomarker assays by Western blotting and ELISA. A, Pull-down and Western blot analysis of SrtA-labeled proteins. MDA-MB-231, Panc-1 and HeLa cells were treated with DMSO or increasing concentration of IMP-1088 NMT inhibitor, lysed and labeled with SrtA overnight; biotinylated proteins were enriched on NeutrAvidin beads, resolved by SDS-PAGE and blotted against YES1 (60 kDa), PRKACA (40 kDa), ARL1 (20 kDa) and GAPDH as loading control (36 kDa). SN: supernatant. NMT substrate proteins show concentration-dependent increase in enrichment with increasing concentration of NMT inhibitor, whereas the low molecular weight of ARL1 also allows for direct identification of the SrtA-labeled fraction in the input sample because of the gel shift induced by ligation to Biotin-ALPET. B, Streptavidin shift analysis of SrtA-labeled proteins. SrtA-labeled samples were briefly incubated with streptavidin before SDS-PAGE and blotting against ARL1 and PRKACA. SrtA biotinylated ARL1 (B-ARL1) shows a molecular weight shift induced by the Biotin-ALPET label that gets shifted by 30 kDa on Streptavidin binding. Biotinylated PRKACA (B-PRKACA) shows no apparent shift in the absence of streptavidin, but a clear shift of 30 kDa upon addition of streptavidin. These shifts match the apparent molecular weight of streptavidin alone, as shown by Ponceau staining. C, SrtA-ELISA analysis of labeled proteins. After protein precipitation to remove excess depsipeptide, 1 µg labeled protein lysate was applied to each well of a streptavidin-coated 96-well plate and incubated for 3 h at RT. After primary (anti-ARL1) and secondary (anti-rabbit HRP) antibody incubations, turnover of QuantaBlu fluorogenic HRP substrate was monitored for 30 min. Slopes were calculated from the first 10 min (linear range) after fitting to a straight line (Y = Slope · X + Yintercept) by nonlinear regression (Insert). Data were analyzed by one-way ANOVA followed by Dunnett's multiple comparison test.

Fig 4: Generation and characterization of AML12DNAJ-PKAc cell lines.(A) CRISPR-Cas9 gene editing of mouse chromosome eight in AML12 cells deleted a 400 kb region between intron 1 of the gene for Hsp40 (Dnajb1) and intron 1 of the gene for PKAc (Prkaca). (B) PCR detection of transcripts for the Gipc1, Ddx39 and Lphn1 genes encoded on the non-engineered strand of mouse chromosome 8. (C–E) Quantitative PCR detection of native mRNA transcripts in AML12 (black) and gene-edited (orange) cell lines. (C) Detection of native Dnajb1 mRNA transcripts, (D) Prkaca transcripts and (E) Dnajb1-Prkaca mRNA transcripts. Data (n = 3) is normalized to Gapdh (C–E) and relative to (C,D) wildtype AML12 or (E) clone 2. Error bars indicate mean ±s.d. (F) Amino acid sequence of the fusion protein DNAJ-PKAc is shown in orange and blue. Nucleotide sequence of the fusion gene from clone 14 AML12DNAJ-PKAc cells is shown below. (G) Immunoblot detection of both native and mutant PKAc in four clonal AML12DNAJ-PKAc cell lines. Top) DNAJ-PKAc fusion proteins (upper bands) and wildtype PKAc (lower bands) are indicated. The distribution of PKAc in wildtype AML12 cells, normal liver and FLC are included. Bottom) Actin loading control. (H) Immunoblot detection of PKA in Hsp70 immune complexes isolated from wildtype (AML12) and clone 14 AML12DNAJ-PKAc cells. Lysate loading controls indicate both forms of PKA (middle) and levels of Hsp70 (bottom). (I and J) Proximity Ligation (PLA) detection of proteins within 40–60 nm of each other in (I) AML12 and (J) AML12DNAJ-PKAc cells. Yellow puncta identify Hsp70-kinase sub-complexes. Actin stain (green) marks cytoskeleton and DAPI staining (blue) marks nuclei. (K) Box-whisker plots of Hsp70-kinase sub-complexes. Amalgamated data (PLA puncta/cell) from AML12 (black) and AML12DNAJ-PKAc (orange) cells. Number of cells analyzed over three independent experiments is indicated below each plot; data are shown as mean ±s.d., p<0.0001 by Student’s t-test (t = 14.16, df = 105).