Article Figures & Data
Figures
- Figure 1. Global absolute quantification of protein expression in Ramos B cells.
(A) Ramos B cell protein copy number plot. Shown are the mean log10-transformed copy number values of 8,086 proteins. The dataset was divided into five equal-sized quantiles (Q1–Q5) according to copy number rank. (B) Gene ontology term enrichment analysis of proteins in Q1–Q5, as shown in (A), against the human proteome. For each quantile, significantly enriched processes of the Gene Ontology term “biological process” and the Reactome pathway are shown with Bonferroni step-down corrected P-values (n = 3) and the number of associated genes. (C) BCR downstream signaling network. The mean copy numbers of individual proteins per Ramos B cell (n = 3) are indicated by their color. The pathway was adapted from Satpathy et al (2015). See also Fig S1 and Table S1.
- Figure S1. Absolute quantitative analysis of the Ramos B cell proteome.
Related to Fig 1. (A) Workflow of the global absolute quantitative proteome analysis. Ramos B cells were lysed in urea buffer and proteins digested in-solution using Lys-C and trypsin. Proteolytic digests were fractionated using high-pH reversed-phase liquid chromatography and analyzed by LC-MS. Copy numbers of proteins were calculated based on MS1 data from three independent biological replicates using the proteomic ruler method (Wiśniewski et al, 2014). (B) Multi-scatterplot depicting the correlation of the determined log10 copy number of proteins per Ramos B cell between individual replicates. Pearson’s correlation coefficient, r. (C) Piechart depicting the share of the 50 most abundant proteins on the total copy number per Ramos B cell. The bar chart shows the classification of these 50 proteins in different functional categories (see Table S1).
- Figure 2. Evaluation of protein tyrosine kinases (PTKs) and classical protein tyrosine phosphatases (PTPs) in Ramos B cells at quantitative scale.
(A) Copy number profile of 28 different PTKs identified and quantified in Ramos B cells. Src family kinases and other PTKs known to be involved in BCR signaling are shown in light blue, all others in dark blue. (B) Copy number profile of the 13 classical PTPs identified and quantified in Ramos B cells. CD45 and SHP-1, known to be involved in BCR signaling are highlighted in yellow, the other classical PTPs in red. (C) Density plot comparing mean log2-transformed protein copy numbers of the Ramos B cell proteome analysis against transcript per million values of the RNA sequencing dataset. A Pearson correlation coefficient of 0.52 was calculated. TPM, transcripts per million. (D) Comparison of the abundance of classical PTPs shown in (B) according to protein (mean log10-transformed copy numbers, blue) and mRNA levels (mean log2 transcript per million values, red) of 7,944 proteins and transcripts, respectively. Mass spectrometric and RNA-sequencing datasets were divided into 12 equal-sized abundance rank intervals, with the least and most abundant 662 proteins/transcripts in interval 1 and 12, respectively.
- Figure S2. Evaluation of absolute quantitative proteome data of human Ramos B cells.
Related to Fig 2. (A) Protein copy number profile of protein tyrosine phosphatases (PTPs), without the “classical” PTPs shown in Fig 2B, identified in Ramos B cells. (B, C) Scatterplots comparing mean log10 protein copy numbers of naïve B cells (Rieckmann et al, 2017) and Ramos B cells reported in this study. Protein tyrosine kinases present in the dataset are depicted in blue (B) and “classical” PTPs in red (C). (D) Mean log10 copy numbers of the three most abundant “classical” PTPs in naïve B cells compared to Ramos B cells and the difference in expression given as fold-change. (E) Kernel-smoothed distribution curves of RNA sequencing (seq) data of human Ramos B cells calculated from raw data reported in Qian et al (2014). Grey, distribution of transcripts of protein-coding genes; light blue, distribution of transcripts of protein-coding genes with a mean log2 transcript per million value > 0; dark blue, protein-coding transcripts for which proteins were identified by MS analysis in this work (see Figs 1 and S1 and Table S1). (F) Comparison of log2-transformed mean transcript per million values and protein copy numbers of ribosomal proteins (purple), proteasome components (red), and enzymes of canonical glycolysis (blue).
- Figure S3. Verification of PTP1B knockout in human Ramos B cells.
Related to Fig 3. (A) Western blot analysis of different PTP1B KO clones (clone numbers are indicated). The loss of PTP1B protein was controlled with an antibody directed against the C-terminal part of PTP1B. As control, non-gene–edited Ramos B cells (Ramos) were used. Equal loading of samples was monitored by the detection of GAPDH. (B) Extracted ion chromatograms (XICs) of selected doubly charged unique peptides of PTP1B. Indicated in orange and blue are the XICs of the light- and heavy-labelled peptide versions originating from non-gene–edited and gene-edited PTP1B KO Ramos B cells. The data confirm the absence of PTP1B at the protein level in the KO cells. (C) Quantitative MS analysis to confirm the loss of PTP1B at the protein level in PTP1B KO Ramos B cell lines. Ramos B cells and two PTP1B KO cell lines made from two different clones (clone 1 and 5) were differentially labelled with stable isotope-coded amino acids (SILAC technology) and proteins digested in solution using trypsin and LysC followed by high pH-reversed phase fractionation and quantitative LC-MS analysis. Mean log2 ratios of Ramos B cells and PTP1B KO cells were calculated over three out of four biological replicates and a two-sided t test was applied to determine P-values. PTP1B is indicated in orange and proteins with a minimum fold change of two and a P-value ˂ 0.05 were labelled. For PTP1B, a fold change of 3.42 was calculated by MaxQuant (Cox & Mann, 2008).
- Figure 3. Loss of PTP1B leads to an increase in phosphorylation of SYK-Y525/526, ERK-T202/Y204, PLCγ2-Y759, and BTK-Y223 in Ramos B cells.
(A) Immunoblot analysis of SYK-Y525/526 and ERK-T202/Y204 phosphorylation in WT and PTP1B KO Ramos cells stimulated with anti-λ for the indicated time points. Representative immunoblots of three independent biological replicates are shown. (B, C) Quantification of immunoblot data of SYK-pY525/526 (B) and ERK-pT202/Y204 (C) shown in (A). Signal intensities were normalized to the respective total protein signals and a t test was performed (n = 3). Error bars represent the SEM. (D) Immunoblot analysis of PTP1B KO and FLAG-PTP1B induced (PTP1B rescue) Ramos cells stimulated with anti-λ for the indicated time points. Representative immunoblots of three independent biological replicates are shown. (E, F, G, H) Quantification of immunoblot data of SYK-pY525/526 (E), ERK-pT202/Y204 (F), BTK-pY223 (G), and PLCγ2-pY759 (H) shown in (D). Signal intensities were normalized to the respective total protein signals and a t test was performed (n = 4). Error bars represent the SEM.
Source data are available online for this figure.
Source Data for Figure 3[LSA-2021-01084_SdataF3.pdf]
- Figure S4. PTP1B re-introduction delays the return of the intracellular calcium level to the baseline.
Related to Fig 3. (A, B) Intracellular calcium measurements were performed using PTP1B KO cells with inducible expression of either empty vector control (GFP only) (A) or FLAG-PTP1B WT (B). Basal calcium levels were recorded for 30 s before the addition of 1 μg/ml anti-λ to stimulate BCR-induced calcium mobilization. Δ indicates the baseline corrected difference of the indo-1 (bound/unbound) ratio between the samples at 300 s. (C) Quantification of the calcium response in B cells at 300 s. Statistical analysis was performed using a t test (two-sided, n = 6, error bars = SEM).
- Figure S5. Analysis of FLAG-tagged PTP1B and PTP1B substrate mutant versions in human Ramos B cells by immunofluorescence microscopy.
Related to Fig 4. (A, B, C) Localization analysis of FLAG-PTP1B (A), FLAG-PTP1B-D181A-Y46F (B), and FLAG-PTP1B-D181A-Y46F Δ406-435 (C) in Ramos PTP1B KO cells by immunofluorescence microscopy. The expression of the constructs was induced by doxycycline 24 h before the experiment. Cells were seeded on cover slips, fixed, and stained with DAPI (nucleus, blue), anti-calnexin antibody (ER marker, green), and anti-PTP1B antibody (PTP1B, red). Shown are representative excerpts of acquired images. For better visualization, single cells were zoomed in. FLAG-PTP1B WT localized mainly to the ER, as shown by overlay with the ER marker protein calnexin. The FLAG-PTP1B trapping mutant only marginally overlapped with the ER marker with a punctated ring-like structure. The ER domain-lacking PTP1B trapping mutant did not overlap with the ER marker and showed a diffuse staining, pointing to its expected cytosolic location in Ramos B cells. Scale bar, 10 μm.
- Figure 4. Identification of putative PTP1B substrates in Ramos B cells.
(A) Experimental design. Ramos PTP1B KO cells stably expressing the empty vector control, the FLAG-tagged PTP1B-D181A-Y46F trapping mutant or the FLAG-tagged truncated trapping mutant without the ER-targeting domain were subjected to stable isotope labelling by amino acids in cell culture (SILAC) using “light” (L), “medium-heavy” (M), and “heavy” (H) versions of arginine and lysine. Following SILAC, co-immunoprecipitations using anti-FLAG beads were performed. Bound proteins were eluted by competition with FLAG peptide and eluates were mixed in equal ratio. SILAC samples were separated by SDS–PAGE followed by in-gel digestion of proteins using trypsin and quantitative LC-MS analysis. (B, C) Scatterplot of proteins eluted with the trapping mutant FLAG-PTP1B-D181A-Y46F (B) and the truncated form of FLAG-PTP1B-D181A-Y46F lacking amino acid residues Δ406-435 (C). Mean log10-transformed SILAC ratios were plotted against −log10 P-values. Proteins with a minimum fold change of two and a P-value < 0.05 (n = 3; right-sided t test) are highlighted in yellow (B) or in blue (C). The bait PTP1B is shown in orange. (D) Scatterplot of proteins eluted with both trapping mutants. Mean log10-transformed SILAC ratios of proteins in (B, C) were plotted against each other. Proteins considered significantly enriched (minimum fold change of two and a P-value < 0.05) with both trapping mutants are marked in green, with FLAG-PTP1B-D181A-Y46F or FLAG-PTP1B-D181A-Y46F Δ406-435 only in yellow and blue, respectively. (E) STRING database analysis of proteins enriched with the PTP1B substrate trapping mutants. Candidates are grouped into three clusters according to the indicated Gene Ontology terms. src-homology 3-domain containing proteins are highlighted by bold margins. Known PTP1B (gene name PTPN1) interactors/substrates are underlined.
- Figure S6. PTP1B trapping mutants capture proteins involved in calcium signaling.
Related to Fig 5. (A, B, C) Intracellular calcium mobilization after anti-λ stimulation (1 μg/ml) in PTP1B knockout cells with inducible expression of empty vector control (GFP only) (A), FLAG-PTP1B-D181A-Y46F (B) or FLAG-PTP1B-D181A-Y46F Δ406-435 (C) substrate trapping mutants. (D) Quantification of the calcium response by the determination of the baseline-corrected area under the curve of the calcium kinetics. Statistical analysis was performed using a t test (two-sided, n = 3, error bars = SEM). AUC, area under the curve.
- Figure 5. Validation of PTP1B substrates in Ramos B cells.
(A, B) Co-immunoprecipitation experiments were performed using PTP1B KO cells with inducible expression of empty vector control, FLAG-PTP1B WT, FLAG-PTP1B-D181A-Y46F, or FLAG-PTP1B-D181A-Y46F Δ406-435 substrate trapping mutants. Eluates were subjected to SDS-PAGE and selected PTP1B substrate candidates (see Fig 4) were analyzed by immunoblotting (A). To further confirm substrates of PTP1B, cells were lysed in the presence of 1 mM orthovanadate competing for binding and, thus, promoting the release of bound substrate (B). IP, immunoprecipitation; ortho, orthovanadate. (C) PTP1B KO and WT Ramos B cells were analyzed for the interaction of PTP1B with endogenous Igα, CD22 or APLP2 by proximity ligation assay. Data analysis was performed with the CellProfiler software. Each dot represents one cell. Statistical analysis was performed using a Mann-Whitney test (****P < 0.0001, n = 3). Scale bar 10 μm.
Source data are available online for this figure.
Source Data for Figure 5[LSA-2021-01084_SdataF5.pdf]
- Figure 6. PTP1B dephosphorylates CD22 at pY807.
(A) Schematic representation of the human/mouse CD22 molecule with domain structure and intracellular tyrosine (Y) residues. ITIM, immunoreceptor tyrosine-based inhibitory motif; TM, transmembrane domain; IG, immunoglobulin domain; GRB2, growth factor receptor-bound protein 2; SYK, spleen tyrosine kinase. (B) Peptide-based phosphatase assay showing the dephosphorylation of four synthetic CD22 peptides by GST-tagged PTP1B. Error bars represent the standard deviation of three independent experiments. (C) PTP1B KO Ramos cells with inducible FLAG-PTP1B-D181A-Y46F trapping mutant were stably transduced with Myc-tagged CD22 WT or Myc-tagged CD22 site mutants in which single tyrosine residues were replaced by phenylalanine (Y-to-F mutants) as indicated. Following inducible expression of FLAG-PTP1B-D181A-Y46F, co-immunoprecipitations using anti-Myc antibody were performed and bound proteins were eluted (n = 3). CD22 WT and phosphosite mutants were detected in cell lysates and eluates by SDS–PAGE immunoblotting using anti-Myc antibody. The FLAG-tagged PTP1B trapping mutant was detected using anti-FLAG antibody. Co-immunoprecipitation of GRB2 was detected using a GRB2 antibody (n = 1). (D) Quantification of immunoblot data from (C). Anti-FLAG signals (FLAG-PTP1B-D181A-Y46F) were normalized to anti-myc signals (myc-tagged CD22 and site mutants thereof). A t test was performed testing each mutant against the control and P-values were calculated (n = 3). Error bars represent the SEM. (E) Phosphatase assay in FLAG-PTP1B-inducible Ramos cells. Myc-tagged CD22 was immunoprecipitated from untreated or anti-λ stimulated cells, with or without the expression of FLAG-PTP1B. Phosphorylation levels of CD22-Y807 were monitored by immunoblot. (F) Quantification of immunoblots from (E). CD22-pY807 signals were normalized to the signals for immunoprecipitated CD22-myc. A t test was performed and P-values were calculated (n = 3). Error bars represent the SEM.
Source data are available online for this figure.
Source Data for Figure 6[LSA-2021-01084_SdataF6_FS7.pdf]
- Figure S7. CD22 phosphatase assay.
Related to Fig 6. Lysates used as an input for the CD22 phosphatase assay. As indicated cells were stimulated with 10 μg/ml anti-λ antibody for 5 min after 30 min of serum starvation. The expression of FLAG-PTP1B was induced by the addition of doxycline to the cell culture media.
Source data are available for this figure.
Source Data for Figure S7[LSA-2021-01084_SdataF6_FS7.pdf]
Supplementary Materials
Table S3 Comparison of the copy numbers in Ramos B cells to naïve B cells from Rieckmann et al (2017).
Table S4 RNA sequencing data.
Table S9 Results of the PTP1B Co-IPs.
Table S10 Plasmids and peptides.
In this Issue
Subjects
Related Articles
Cited By...
- No citing articles found.