MYC targeted de novo binders impair tumor growth
Nora K. McNamara-BordewickiD, Kejia WuiD, Yulong Su, Assmaa ElSheikhiD, Xiaoyan WangiD, Melissa CunninghamiD, Colin Daniel, Rosalie C. SearsiD, Robert N. EisenmaniD, David A. BakeriD
Abstract
MYC is one of the most frequently dysregulated oncogenic drivers across many cancers, yet direct therapeutic targeting has remained difficult because its function is mediated by disordered regions and transient protein-protein interactions. Here we design proteins that bind epitopes within regions of MYC whose sequence is conserved between human and mouse, including intrinsically disordered motifs required for oncogenic activity. In vitro, the designs bind MYC with nanomolar affinity and in cells they engage MYC, disrupt interactions with MAX, and remodel MYC-associated protein complexes. Expression of the binders decreases cell proliferation across diverse MYC-dependent hematopoietic, liver, pancreatic, and patient-derived tumor models, and in an orthotopic pancreatic cancer model, binder-expressing tumor cells were selectively depleted. This work establishes de novo protein design as a generalizable strategy to therapeutically target intrinsically disordered transcription factors.
02 /Specification & Classification
Click any tag to browse matching papers- Citations
- 1
- Organism
- Assay
- Research Focus
- Approach
In-depth summary
Claude Opus 5.5 · xhigh effort · 7 Nov 2026 · from the full text of version 2 (22 Oct 2026) · CC BY 4.0
Designed protein binders to conserved MYC regions reach low-nanomolar affinity, reduce MYC-MAX proximity in cells, and are depleted from MYC-dependent cancer cells in culture and mouse pancreatic tumors.
The paper says direct MYC targeting has centred on the dimerization interface, leaving conserved regions that govern MYC stability, transactivation and chromatin targeting unexplored. It uses de novo design to target disordered loops as well as a helical epitope, and compares the binders with Omomyc.
Key results
ShownOf 192 designed binders tested, 15 bound full-length human MYC below 500 nM. They covered three of the four epitopes: loop 1, loop 2 and the helical epitope. None came from loop 3.
SPR on 192 candidates gave 15 binders under 500 nM (5 loop 1, 1 loop 2, 9 helical). Initial designs had KD of 10-200 nM, and off-target peptide binding was minimal and fast-dissociating. · De novo protein design yields high-affinity…; Surface plasmon resonance and binding analysis
ShownA second design round (motif diffusion for loops, partial diffusion for the helix) improved affinity substantially while keeping selectivity for the intended peptide.
174 further designs were ordered. The best loop-epitope binder improved from 119 nM to 5.46 nM (MYCb-L1.e). The best helix-epitope binder improved from 15 nM to 1.23 nM (MYCb-H.f). · De novo protein design yields high-affinity…; Structure-guided affinity optimization; Surface plasmon resonance and binding analysis
ShownBinders engage endogenous MYC in cells. Four of five tested binders lowered MYC-MAX proximity. IP-MS pulldowns of two binders enriched known MYC interactors, with different partners depending on the epitope.
PLA in HEK293T cells (Mann-Whitney U vs no-binder control). IP-MS in HEK293 cells, n = 3 per condition: BioGRID MYC-interactor overlap P = 5.35 x 10^-43 for MYCb-L2.a and P = 1.17 x 10^-2 for MYCb-H.a. · Designed binders engage MYC in cells and disrupt…; In situ proximity ligation assays; Anti-FLAG immunoprecipitation and mass…; Proteomics statistical analysis and…
ShownIn pooled barcoded competitive growth assays, nine of ten binders were depleted relative to GFP in all four MYC-dependent lines, including a patient-derived PDAC line. The same five were most depleted everywhere, spanning all three active epitopes, and all five exceeded Omomyc.
HepG2, K562, KMCz682 and ST-00015919, mean of 3 independent infections, log2 fold change normalized to GFP. The strongest design exceeded Omomyc by 2.4- to 62-fold. · MYC binders suppress proliferation across…; Pooled lentiviral competitive growth screens; Statistical analysis and reproducibility
SuggestedActive binders were also selectively depleted from orthotopic KMCz682 pancreatic tumors in immunocompetent mice, in line with the culture results.
50,000 cells implanted per mouse, harvest at day 14. Four mice were pooled into 2 sequencing samples, compared with 3 day-zero input samples, normalized to GFP. · De novo MYC binders suppress tumor growth in…; Orthotopic pancreatic tumor competitive fitness…; Statistical analysis and reproducibility
Methods, data and reproducibility
Binders to four conserved MYC epitopes were designed with Logos, RFdiffusion and ProteinMPNN, and filtered with AlphaFold2/3. They were then screened by SPR, refined by diffusion-based redesign, and tested in cells by PLA, IP-MS and pooled barcoded growth assays, plus an orthotopic mouse tumor assay.
- Target epitopes on human MYC (residues 55-77, 187-208, 286-324, 356-379), selected from human and mouse MYC conservation · human · n = 4 epitopes · UniProt P01106; PDB 1NKP; AF-P01106
- SPR screening of first-round designed binders against full-length recombinant human MYC · human (recombinant MYC) · n = 192 designs
- SPR testing of second-round optimized designs (96 helical, 78 loop) · human (recombinant MYC) · n = 174 designs
- Anti-FLAG IP-MS of MYCb-L2.a and MYCb-H.a vs FLAG-only control in HEK293 cells · human · n = 3 biological replicates per condition
- Pooled competitive growth assays in HepG2, K562, KMCz682 (mouse PDAC) and patient-derived PDAC line ST-00015919 · human and mouse · n = 10 binders plus Omomyc and GFP; 3 independent infections per model
- Orthotopic KMCz682 tumor competitive fitness assay in syngeneic C57BL/6 mice · mouse · n = 4 mice, 2 tumor-pool sequencing samples, 3 day-zero input samples
- Methods & tools
- Logos, RFdiffusion, ProteinMPNN, AlphaFold2, AlphaFold3, Biacore 8K, Biacore Insight Evaluation Software, BioGRID, proximity ligation assay, data-independent acquisition mass spectrometry, Omomyc
- Reproducibility gaps
- Binder sequences, design models, scripts, proteomics data and barcode sequencing data are stated as to be deposited before journal publication. Plasmids are available on request under material transfer agreements.
Limitations (4)
- Delivery into cells is described as critical for therapeutic use. Binders were expressed intracellularly from lentiviral or plasmid constructs, and LNP-mRNA delivery is only proposed. · stated by the authors
- How loop-targeting binders reduce MYC-MAX proximity is unresolved. PLA reflects both proximity and antibody accessibility, and extra IP-MS hits need further mechanistic validation. · stated by the authors
- The in vivo result rests on 4 mice pooled into 2 sequencing samples, and the readout is barcode depletion rather than tumor size or survival. · noted in this summary
- The text reports no MYC-independent control cell line, so depletion cannot be confirmed as MYC-specific from the data shown. · noted in this summary
Who should read:Read this if you work on de novo protein design against disordered targets or transcription factors and want affinity, specificity and cell-based functional data for MYC binders.
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- Development of DuoMYC: a synthetic cell penetrant miniprotein that efficiently inhibits the oncogenic transcription factor MYC
Brecht D. Ellenbroek, Jan Pascal Kahler, … Sebastian J. Pomplun·Angewandte Chemie International EditionPeer-reviewed·Sep 24, 2024
- Undruggable oncoproteins cMyc and NMyc bind to mediator of transcription with superior high affinity
Andrea Knight, Josef Houser, … Martin Piskáček·Molecular MedicinePeer-reviewed·Nov 13, 2024
- Sulfinyl Aziridines as Stereoselective Covalent Destabilizing Degraders of the Oncogenic Transcription Factor MYC
Hannah T. Rosen, Kelvin Li, … Daniel K. Nomura·Angewandte Chemie International EditionPeer-reviewed·Sep 25, 2025
- Intrinsically disordered regions in the transcription factor MYC:MAX modulate DNA binding via intramolecular interactions
S. Schuetz, C. Bergsdorf, … W. Jahnke·bioRxiv·Jun 21, 2023
- De novo protein design enables targeting of intractable oncogenic interfaces
Varshika Ram Prakash, Yusuf Najy, … Benjamin L. Kidder·bioRxiv·Oct 22, 2025
- De novo designed pMHC binders facilitate T cell induced killing of cancer cells
Kristoffer Haurum Johansen, Darian Stephan Wolff, … Timothy P. Jenkins·SciencePeer-reviewed·Jul 24, 2025