Fig 2: Swapping PD-1-ITIM with BTLA-ITIM induced PD-1:SHP1 interaction in T cells.(A) Cartoons depicting the domains and motifs of PD-1WT-mGFP and PD-1BTLA-ITIM-mGFP. (B) Representative immunoblots (IBs) showing the levels of SHP1 bound to mGFP-tagged PD-1 variants pulled down from the indicated coculture lysates via GFP IP. SHP1 IB and SHP2 IB of the whole cell lysates (WCL) indicate their inputs. GAPDH IB of the WCL served as a loading control. Bar graphs summarize SHP1 optical density (OD) normalized to the fluorescence intensity (FI) of each PD-1 variant, based on flow cytometry data in Figure 7—figure supplement 1. Error bars are s.d. from three independent coculture experiments performed on three different days. (C) Left: representative TIRF images of both PD-1 (GFP) and endogenous SHP1 (stained with anti-SHP1) in a supported lipid bilayer (SLB)-associated SHP2 KO Jurkat expressing indicated PD-1 variants. Right: dot plots summarizing anti-SHP1 FI normalized to GFP FI of 40 Jurkat cells under each condition recorded on the same day using the same microscope setting; Error bars: s.d. Scale bars: 5 µm. ***p<0.001; ****p<0.0001; Student’s t-test. Figure 7—source data 1.Uncropped IBs for Figure 7B and Figure 7—figure supplement 2B. Figure 7—source data 2.Raw data for quantification graphs in Figure 7 and Figure 7—figure supplement 2.

Fig 3: Glycine to alanine substitution at the pY+1 position of PD-1 immunoreceptor tyrosine-based inhibitory motif (ITIM) promoted SHP1 recruitment.(A) Cartoons depicting the amino acid (AA) alignment of PD-1-ITIM and BTLA-ITIM. (B) Representative immunoblots (IBs) showing the levels of SHP1 bound to mGFP-tagged PD-1 variants pulled down from the indicated coculture lysates via GFP IP. GAPDH IB of whole cell lysates (WCL) served as a loading control. Bar graphs summarize SHP1 optical density (OD) normalized to the fluorescence intensity (FI) of each PD-1 variant, based on flow cytometry data in Figure 8—figure supplement 1A. Error bars are s.d. from three independent coculture experiments conducted on three different days. (C) Left: representative TIRF images of both PD-1 (GFP) and endogenous SHP1 (stained with anti-SHP1) in a supported lipid bilayer (SLB)-associated SHP2 KO Jurkat cell expressing indicated PD-1 variants. Right: dot plots summarizing anti-SHP1 FI normalized to GFP FI of 35 Jurkat cells under each condition recorded on the same day using the same microscope setting (see Materials and methods). (D) Cartoons showing the AA sequences of ITIM of indicated receptors. (E) Representative IBs showing the levels of SHP1 co-precipitated with mGFP-tagged PD-1 variants, with the original ITIM replaced by the indicated ITIM, from the indicated coculture lysates via GFP IP. GAPDH IB of WCL served as a loading control. Bar graphs summarize SHP1 OD normalized to the FI of each PD-1 variant, based on flow cytometry data in Figure 8—figure supplement 1C. Error bars are s.d. from three independent coculture experiments conducted on three different days. (F) Left: representative TIRF images of both PD-1 (GFP) and endogenous SHP1 (stained with anti-SHP1) in an SLB-associated SHP2 KO Jurkat cell expressing PD-1 variants with indicated ITIM. Right: dot plots summarizing anti-SHP1 FI normalized to GFP FI of 35 Jurkat cells under each condition recorded on the same day using the same microscope setting (see Materials and methods); Error bars: s.d. Scale bars: 5 µm. **p<0.01; ***p<0.001; ****p<0.0001; ns, not significant. Student’s t-test. Figure 8—source data 1.Uncropped IBs for Figure 8B and E. Figure 8—source data 2.Raw data for quantification graphs in Figure 8.

Fig 4: PD-1 recruits and signals through SHP2, but not SHP1, whereas BTLA prefers to recruit SHP1.(A, B) Left are representative immunoblots (IBs) showing the levels of bound SHP1 and SHP2 in PD-1-mGFP (A) or BTLA-mGFP (B) pulled down by GFP IP from indicated cell lysates, with the duration of stimulation prior to lysis indicated (see Materials and methods). IBs of GFP and phosphotyrosines (pY) of the same samples were shown to indicate PD-1 or BTLA input and their degrees of phosphorylation. Right are quantification graphs. (C) Relative IL-2 levels produced by PD-1-mGFP-expressing WT, SHP1 KO, SHP2 KO, or SHP1/2 DKO Jurkat cells stimulated with PD-L1-mCherry-expressing Raji cells in the presence of increasing concentrations of pembrolizumab (0, 0.4, 1.3, 4.4, 13.3, 44, 133, or 267 nM). For each type of Jurkat cells, IL-2 data were normalized to the condition with the highest IL-2 value in each replicate. Error bars are s.d. from three independent coculture assays run in three different days, with each assay run in technical triplicates. ****p<0.0001; ns, not significant; two-way ANOVA test. Figure 1—source data 1.Uncropped IBs for data shown in Figure 1A and B. Figure 1—source data 2.Raw data for Figure 1 graphs.

Fig 5: Both immunoreceptor tyrosine-based inhibitory motif (ITIM) and immunoreceptor tyrosine-based switch motif (ITSM) are required for BTLA to recruit SHP1/SHP2.(A) Flow cytometry histograms showing BTLA surface expressions in the indicated Jurkat cells. (B) Left: representative immunoblots (IBs) showing the levels of SHP1 and SHP2 bound to the mGFP-tagged BTLA variants captured by GFP IP. GAPDH IB of the whole cell lysates (WCL) served as a loading control. Right: bar graphs summarizing SHP1 optical density (OD) and SHP2 OD normalized to the fluorescence intensity (FI) of the corresponding BTLA variants, based on flow cytometry data in (A). Error bars are s.d. from three independent coculture experiments performed on different days. **p<0.01; ***p<0.001; ns, not significant; Student’s t-test (n = 3). Figure 5—source data 1.Uncropped IBs for Figure 5B and Figure 5—figure supplement 1B. Figure 5—source data 2.Raw data for bar graphs in Figure 5 and Figure 5—figure supplement 1.