Fig 1: PTP4A1 recruitment to MET is essential for LECT2-mediated RIG-I dependent innate immune responses.a HepG2 cells were treated with r-LECT2 (0, 5, 10, 50 ng/mL). At 6 h after treatment, lysates were immunoprecipitated with an anti-MET antibody. Immunoblot analysis of MET, p-MET (Tyr1349), p-SHP2, SHP2, PTP4A1, c-Cbl and β-actin in immunoprecipitated samples and input cell lysate. b HepG2-MET-KO cells were transfected with Control plasmid (Cont), MET WT plasmid, MET Y1349F plasmid, or PTP4A1 plasmid. At 48 h after transfection, lysates were immunoprecipitated with an anti-MET antibody. Immunoblot analysis of MET, p-MET (Tyr1349), p-SHP2, SHP2, PTP4A1, c-Cbl and β-actin in immunoprecipitated samples and input cell lysate. c HepG2 cells were transfected with 50 nM c-Cbl siRNA. The concentration of siRNA was adjusted to 50 nM with control siRNA. At 24 h after siRNA transfection, siRNA-transfected HepG2 cells were treated with 50 ng/mL r-LECT2 for 1 h and then transfected with HCV-RNA. c (upper) Immunoblot analysis of c-Cbl, RIG-I and β-actin at 12 h after HCV-RNA transfection. c (lower) qRT-PCR analysis of IFNB1 at 12 h after HCV-RNA transfection. Results were normalised to those of ACTB. Data shown are means of 3 technical replicates ± SEM. ***p < 0.001, not significant (ns) by two-way ANOVA with Tukey’s multiple comparison test. d HepG2-MET-KO cells were transfected with Control plasmid or MET plasmid; at 48 h after transfection, the cells were treated with 50 ng/mL r-LECT2 and/or 10 ng/mL HGF. At 1 h after r-LECT2 and/or HGF treatment, lysates were immunoprecipitated with an anti-MET antibody. Immunoblot analysis of MET, p-MET (Tyr1234/1235), p-MET (Tyr1349), p-MET (Tyr1356), PTP4A1, p-SHP2, SHP2 and β-actin in immunoprecipitated samples and input cell lysate. Source data are provided as a Source data file.
Fig 2: PTP4A1 is essential for the enhancement of innate immune response by LECT2.a Immunoblot analysis of MET, PTP4A1, p-SHP2, RIG-I and β-actin in immunoprecipitated samples and input cell lysate. b HepG2-MET-KO cells were transfected with Control, MET and PTP4A1 plasmid or co-transfected with MET and PTP4A1 plasmid. These cells were treated with 50 ng/mL r-LECT2 for 1 h and then transfected with HCV-RNA. qRT-PCR analysis of IFNB1 in these cells. Results were normalised to those of ACTB. Data shown are means of 3 technical replicates ± SEM. ***p < 0.001, not significant (ns) by two-way ANOVA with Tukey’s multiple comparison test. c–e KH-MET-KO cells were transfected with Control, MET and PTP4A1 plasmid or co-transfected with MET and PTP4A1 plasmid. These cells were treated with 50 ng/mL r-LECT2 for 1 h and then transfected with H77S.3/GLuc-RNA. c Immunoblot analysis of MET, PTP4A1 and β-actin in these cells. d qRT-PCR analysis of IFNB1, DDX58 and IFIH1 in these cells. Data shown are means of 3 technical replicates ± SEM. ***p < 0.001, **p = 0.002, not significant (ns) by two-way ANOVA with Tukey’s multiple comparison test. e GLuc activity in these cells at 48 h after H77S.3/GLuc-RNA transfection. Data shown are means of 3 technical replicates ± SEM. ***p < 0.001, **p = 0.002, not significant (ns) by two-way ANOVA with Tukey’s multiple comparison test. f (left) Co-immunofluorescence staining of MET and PTP4A1 in HepG2 cells. DAPI was used to stain nuclei. f (right) Co-localization was estimated using Pearson’s correlation coefficient and Manders’ co-localization coefficients M1 (red per green) and M2 (green per red). Boxplots: centre line, median; box limits, 25 to 75th percentiles; whiskers, min to max. g (left) Immunoblot analysis of PTP4A1, RIG-I and β-actin at 12 h after HCV-RNA transfection. g (right) qRT-PCR analysis of IFNB1 in HepG2-WT and HepG2-PTP4A1-KO cells at 12 h after HCV-RNA transfection. Data shown are means of 3 technical replicates ± SEM. ***p < 0.001, not significant (ns) by two-way ANOVA with Tukey’s multiple comparison test. Source data are provided as a Source data file.
Fig 3: Proteomics analysis implicates PTP4A1 in regulation of tonic IFN signaling(A) Immunofluorescence microscopy evaluating PTP4A1 localization in NCI-H1944 using two different antibodies.(B) Western blot and densitometry analysis of PTP4A1 in cytosolic and non-cytosolic fractions with and without tipifarnib treatment (1 μM, 24 h) in NCI-H1944 cells. Densitometric values were normalized to GAPDH and then the control mean and are represented as mean ± SD (n = 3 independent cell cultures). p values (p-val) were calculated using Welch’s two-sided t test.(C) Western blot analysis of PTP4A1 upon PTP4A1 silencing using four different siRNAs and negative control siRNA (siNegCtrl), n = 4 independent cell cultures were performed, and one randomly selected replicate was used for the western blot analysis. GAPDH is shown as the loading control.(D) Experimental design for MS-based evaluation of PTP4A1 silencing effects on the proteome. HiRIEF-LC-MS, high-resolution isoelectric focusing-liquid chromatography-mass spectrometry.(E) Density plots showing fold changes (FC) in protein levels upon PTP4A1 silencing using different siRNAs (n = 10,431 proteins, including n = 158 proteins in IFN-γ and IFN-α response hallmark gene sets, n = 4 independent cell cultures per group).(F) Top 20 proteins commonly regulated by tipifarnib treatment (see Figure 3) and PTP4A1 silencing in NCI-H1944 cells. Fold change (FC) and adjusted p values (DEqMS p-val) were calculated using the DEqMS method.(G) Western blot analysis of STAT1, pSTAT1, and PTP4A1 upon PTP4A1 silencing and/or IFN-γ treatment (10 ng/mL), n = 3 independent cell cultures were performed, and one randomly selected replicate was used for the western blot analysis. β-actin is shown as the loading control.(H) Experimental design for MS-based evaluation of PTP4A1 silencing and IFN-γ treatment effects on the proteome.(I) Density plots showing fold changes (FC) in protein levels upon PTP4A1 silencing and/or IFN-γ treatment (10 ng/mL) (n = 8,644 proteins, including n = 150 proteins in IFN-γ and IFN-α response hallmark gene sets, n = 3 independent cell cultures per group).
Fig 4: Publicly available data highlight relevance of multiple PFPs in cancer(A) Tipifarnib AUC scores (GDSC1 dataset) in solid tumor cell lines (n = 720).(B) Tipifarnib AUC scores by NRAS, KRAS, and HRAS mutation status (WT, wildtype; mut, mutant) per Sanger Institute’s Cell Model Passport annotations in solid tumor cell lines (n = 708).(C) Tipifarnib AUC by cell line primary disease (Broad Institute’s DepMap annotation). NSCLC, non-small cell lung cancer; NET, neuroendocrine tumor. p value was calculated using Welch’s t test; the size of the groups is indicated in the figure.(D) Cell line growth rate (Sanger Institute’s Cell Model Passport data) by tipifarnib AUC score in solid tumor cell lines (n = 711).(E) Essentiality of prenylation candidates (n = 114, see Figure 1D) and farnesyltransferase (FTase) components FNTA and FNTB in Broad Institute’s and Sanger Institute’s datasets (n = 578 and n = 305 solid tumor cell lines [CLs], respectively). Essentiality was defined as a scaled loss of fitness score <0 upon CRISPR-based gene knockout.(F) Heatmap displaying PFPs (see Figure 1D) that were found to be differentially abundant in the non-small cell lung cancer (NSCLC) proteome subtypes published by Lehtiö et al. (2021) (n = 65 proteins, n = 141 tumors). Differential protein abundance was defined as DEqMS adjusted p value <0.01 and absolute log2 ratio >0.5 in at least one pairwise comparison between the subtypes. Cell-line-specifically essential proteins as per panel (E) and proteins relocalizing upon farnesyltransferase inhibition (FTI) as per Figure 2E are indicated.(G) PTP4A1 protein levels in the Lehtiö et al. (2021) NSCLC cohort. p value was calculated using Kruskal-Wallis test, and the number of samples per subtype is indicated.(H) PTP4A1 mRNA levels (CCLE 21Q4 release) versus loss of fitness score upon PTP4A1 knockout (KO) in lung cancer cell lines (n = 83), colored by STK11 mutation status (WT, wildtype; mut, mutant, Sanger Institute’s Cell Model Passport annotation). The corresponding Pearson correlation coefficient (Pearson Cor.) and a linear regression are displayed.For panels (B–D), p values (p-val) were calculated using Welch’s two-sided t test. In panels (C and D), the NCI- prefix has been omitted in the cell line names.
Supplier Page from Abcam for Anti-PRL-1 antibody