Targeted protein degradation (TPD) has rapidly emerged as a powerful modality for drugging previously “undruggable” proteins. TPD employs small molecules like PROTACs and molecular glue degraders (MGD) to induce target protein degradation via the formation of a ternary complex with an E3 ligase. However, the rational design of these degraders is severely hindered by the difficulty of obtaining these ternary structures. Here we introduce DeepTernary, a novel end-to-end deep learning approach using an SE(3)-equivariant encoder and a query-based decoder to accurately and rapidly predict these critical structures. Trained on carefully curated TernaryDB, DeepTernary achieves state-of-the-art performance on PROTAC benchmarks without prior exposure to known PROTACs and shows notable prediction capability on the more challenging MGD benchmark with a blind docking protocol. Remarkably, the buried surface areas calculated from predicted structures correlate with experimental degradation potency metrics. Overall, DeepTernary offers a powerful tool for the development of targeted protein degraders.
MAPK inhibitor (MAPKi) therapy in melanoma leads to the accumulation of tumor-surface PD-L1/L2, which may evade antitumor immunity and accelerate acquired resistance. Here, we discover that the E3 ligase ITCH binds, ubiquitinates, and downregulates tumor-surface PD-L1/L2 in MAPKi-treated human melanoma cells, thereby promoting T-cell activation. During MAPKi therapy in vivo, melanoma cell–intrinsic ITCH knockdown induced tumor-surface PD-L1, reduced intratumoral cytolytic CD8+ T cells, and accelerated acquired resistance only in immune-competent mice. Conversely, tumor cell–intrinsic ITCH overexpression reduced MAPKi-elicited PD-L1 accumulation, augmented intratumoral cytolytic CD8+ T cells, and suppressed acquired resistance in BrafV600MUT, NrasMUT, or Nf1MUT melanoma and KrasMUT-driven cancers. CD8+ T-cell depletion and tumor cell–intrinsic PD-L1 overexpression nullified the phenotype of ITCH overexpression, thereby supporting an in vivo ITCH–PD-L1–T-cell regulatory axis. Moreover, we identify a small-molecular ITCH activator that suppresses acquired MAPKi resistance in vivo. Thus, MAPKi-induced PD-L1 accelerates resistance, and a PD-L1–degrading ITCH activator prolongs antitumor response. Significance: MAPKi induces tumor cell–surface PD-L1 accumulation, which promotes immune evasion and therapy resistance. ITCH degrades PD-L1, optimizing antitumor T-cell immunity. We propose degrading tumor cell–surface PD-L1 and/or activating tumor-intrinsic ITCH as strategies to overcome MAPKi resistance. This article is highlighted in the In This Issue feature, p. 1825
Deep proteome coverage in bottom-up proteomics requires peptide-level fractionation to simplify the complex peptide mixture before analysis by tandem mass spectrometry. By decreasing the number of coeluting precursor peptide ions, fractionation effectively reduces the complexity of the sample leading to higher sample coverage and reduced bias toward high-abundance precursors that are preferentially identified in data-dependent acquisition strategies. To achieve this goal, we report a bead-based off-line peptide fractionation method termed CIF or carboxylate-modified magnetic bead–based isopropanol gradient peptide fractionation. CIF is an extension of the SP3 (single-pot solid phase–enhanced sample preparation) strategy and provides an effective but complementary approach to other commonly used fractionation methods including strong cation exchange and reversed phase–based chromatography. We demonstrate that CIF is an effective offline separation strategy capable of increasing the depth of peptide analyte coverage both when used alone or as a second dimension of peptide fractionation in conjunction with high pH reversed phase. These features make it ideally suited for a wide range of proteomic applications including the affinity purification of low-abundance bait proteins.
High-field asymmetric waveform ion mobility spectrometry (FAIMS) enables gas-phase separations on a chromatographic time scale and has become a useful tool for proteomic applications. Despite its emerging utility, however, the molecular determinants underlying peptide separation by FAIMS have not been systematically investigated. Here, we characterize peptide transmission in a FAIMS device across a broad range of compensation voltages (CVs) and used machine learning to identify charge state and three-dimensional (3D) electrostatic peptide potential as major contributors to peptide intensity at a given CV. We also demonstrate that the machine learning model can be used to predict optimized CV values for peptides, which significantly improves parallel reaction monitoring workflows. Together, these data provide insight into peptide separation by FAIMS and highlight its utility in targeted proteomic applications.
The formation and plasticity of neuronal circuits relies on dynamic activity-dependent gene expression. Although recent work has revealed the identity of important transcriptional regulators and of genes that are transcribed and translated in response to activity, relatively little is known about the cell biological mechanisms by which activity alters the nuclear proteome of neurons to link neuronal stimulation to transcription. Using nucleus-specific proteomic mapping in silenced and stimulated neurons, we uncovered an understudied mechanism of nuclear proteome regulation: activity-dependent proteasome-mediated degradation. We found that the tumor suppressor protein PDCD4 undergoes rapid stimulus-induced degradation in the nucleus of neurons. We demonstrate that degradation of PDCD4 is required for normal activity-dependent transcription and that PDCD4 target genes include those encoding proteins critical for synapse formation, remodeling, and transmission. Our findings highlight the importance of the nuclear proteasome in regulating the activity-dependent nuclear proteome and point to a specific role for PDCD4 as a regulator of activity-dependent transcription in neurons.