Fig 2: Interplay of dynamic interactions between PTPN2 and CRBN is important for modulating the complex formation.a Comparison of optimized structures of PTPN2/Cmpd-1/CRBN-DDB1 derived from three replicates of MD simulations (2.2 μs in total) with the cryo-EM structure as template. The left panel shows a snapshot at t = 700 ns from replicate 1, the middle panel shows a snapshot at t = 541 ns from replicate 2, and the right panel shows a snapshot at t = 323 ns from replicate 3. The isolated ternary complex PTPN2/Cmpd-1/CRBN-DDB1 in all represented snapshots underwent energetic minimization using the MOE QuickPrep module49. This involved adjusting the protonation states of all atoms and performing structure optimization using the Amber10:EHT force field. The top panels display the side view, while the bottom panel shows the top view of the complex. PTPN2 position in cryo-EM is shown as green cartoon within a transparent green surface. PTPN2 positions in MD replicates are shown as blue cartoons within a transparent blue surface. CRBN is shown as a purple cartoon. DDB1 is shown as a yellow cartoon. b (top and middle panels) analysis of Root mean squared deviation (RMSD) of the ternary complex and its components over time. All RMSD values were calculated relative to the initial configuration (cryo-EM positions) using the Calpha atoms of proteins and all heavy atoms (non-H atoms) of Cmpd-1. For RMSD values with respect to DDB1, the DDB1 proteins from the MD simulation trajectory were superimposed onto the DDB1 structure solved by cryo-EM, and then the RMSD values were calculated (bottom panel). The center of mass (COM) to COM distances were measured between the PTPN2 structures from the MD simulation trajectory and the cryo-EM structure. COM calculations were performed using the Calpha atoms of PTPN2. Data for complex is shown in blue, PTPN2 is shown in green, CRBN is shown in purple and Cmpd-1 is shown in yellow. c Distinct interfaces and detailed interactions between PTPN2 and CRBN resulting from MD simulations. This panel showcases three distinct interfaces between PTPN2 and CRBN, each derived from snapshots presented in Fig. 4a. To obtain optimized structures, energetic minimization steps were performed using the MOE QuickPrep module. The detailed interactions within each interface highlight the specific residue-level contacts and molecular interactions between PTPN2 and CRBN. d Binding affinity of PTPN2R35E/E41K/R43E with CRBN-DDB1 in the presence of Cmpd-1 and Cmpd-2 as measured by BLI assay. Biotinylated-PTPN2 (8 µg/mL) was immobilized on SAX sensors, dipped in a solution containing 1 µM of Cmpd-1 or Cmpd-2 and 20 µg/mL of Biocytin and further dipped in serially diluted solutions of CRBN-DDB1 (1000 nM to 62.5 nM). KD was calculated using global fit analysis using a 1:1 binding model. Data points are representative of n = 2 independent replicates. Source data is provided as a source data file. Data curve colors from highest to lowest concentration in order: red, blue, green, yellow and purple. Fits are shown as black curves.

Fig 3: Characterization of nanoparticles. (a) and (b) Size distribution, zeta potential (inset graph), and morphology of HD@RPR and M1HD@RPR (inset image). (c) Protein profiles in the M1, HD@RPR, and M1HD@RPR determined via SDS-PAGE assay. (d) DNA fragment migration by agarose gel electrophoresis. Lane 1, DNA ladder; lane 2, shRNA-Ptpn2 plasmid; lane 3-6: RR, RPR, HD@RPR, M1HD@RPR. (e) DOX-release profiles in the presence different pH values.

Fig 4: (a) Histological images of TUNEL, Ki67, Ptpn2, and CRT expression in tumor sections of different treatment groups (scale bar: 200 μm). (b) to (d) mRNA expression of Ptpn2, CRT, and HMGB1 were detected by qPCR after treatment.

Fig 5: Relationship between PTPN2 and tumor heterogeneity and its prediction for immunotherapy. Correlation analysis of PTPN2 expression with (A) TMB, (B) MSI, (C) tumor purity, (D) neoantigens, (E) ploidy, and (F) HRD. (G) PTPN2 expression predicts immunotherapy response in cancer patients based on the TIDE algorithm. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.