Mass Spectrometry and Machine Learning Reveal Determinants of Client Recognition by Antiamyloid Chaperones.

The assembly of proteins and peptides into amyloid fibrils is causally linked to serious disorders such as Alzheimer's disease. Multiple proteins have been shown to prevent amyloid formation in vitro and in vivo, ranging from highly specific chaperone-client pairs to completely nonspecific binding of aggregation-prone peptides. The underlying interactions remain elusive. Here, we turn to the machine learning-based structure prediction algorithm AlphaFold2 to obtain models for the nonspecific interactions of ß-lactoglobulin, transthyretin, or thioredoxin 80 with the model amyloid peptide amyloid ß and the highly specific complex between the BRICHOS chaperone domain of C-terminal region of lung surfactant protein C and its polyvaline target. Using a combination of native mass spectrometry (MS) and ion mobility MS, we show that nonspecific chaperoning is driven predominantly by hydrophobic interactions of amyloid ß with hydrophobic surfaces in ß-lactoglobulin, transthyretin, and thioredoxin 80, and in part regulated by oligomer stability. For C-terminal region of lung surfactant protein C, native MS and hydrogen-deuterium exchange MS reveal that a disordered region recognizes the polyvaline target by forming a complementary ß-strand. Hence, we show that AlphaFold2 and MS can yield atomistic models of hard-to-capture protein interactions that reveal different chaperoning mechanisms based on separate ligand properties and may provide possible clues for specific therapeutic intervention.

A druggable conformational switch in the c-MYC transactivation domain.

The c-MYC oncogene is activated in over 70% of all human cancers. The intrinsic disorder of the c-MYC transcription factor facilitates molecular interactions that regulate numerous biological pathways, but severely limits efforts to target its function for cancer therapy. Here, we use a reductionist strategy to characterize the dynamic and structural heterogeneity of the c-MYC protein. Using probe-based Molecular Dynamics (MD) simulations and machine learning, we identify a conformational switch in the c-MYC amino-terminal transactivation domain (termed coreMYC) that cycles between a closed, inactive, and an open, active conformation. Using the polyphenol epigallocatechin gallate (EGCG) to modulate the conformational landscape of coreMYC, we show through biophysical and cellular assays that the induction of a closed conformation impedes its interactions with the transformation/transcription domain-associated protein (TRRAP) and the TATA-box binding protein (TBP) which are essential for the transcriptional and oncogenic activities of c-MYC. Together, these findings provide insights into structure-activity relationships of c-MYC, which open avenues towards the development of shape-shifting compounds to target c-MYC as well as other disordered transcription factors for cancer treatment.

A "spindle and thread" mechanism unblocks p53 translation by modulating N-terminal disorder.

Disordered proteins pose a major challenge to structural biology. A prominent example is the tumor suppressor p53, whose low expression levels and poor conformational stability hamper the development of cancer therapeutics. All these characteristics make it a prime example of "life on the edge of solubility." Here, we investigate whether these features can be modulated by fusing the protein to a highly soluble spider silk domain (NT∗). The chimeric protein displays highly efficient translation and is fully active in human cancer cells. Biophysical characterization reveals a compact conformation, with the disordered transactivation domain of p53 wrapped around the NT∗ domain. We conclude that interactions with NT∗ help to unblock translation of the proline-rich disordered region of p53. Expression of partially disordered cancer targets is similarly enhanced by NT∗. In summary, we demonstrate that inducing co-translational folding via a molecular "spindle and thread" mechanism unblocks protein translation in vitro.

Radiosafe micro-computed tomography for longitudinal evaluation of murine disease models.

Implementation of in vivo high-resolution micro-computed tomography (µCT), a powerful tool for longitudinal analysis of murine lung disease models, is hampered by the lack of data on cumulative low-dose radiation effects on the investigated disease models. We aimed to measure radiation doses and effects of repeated µCT scans, to establish cumulative radiation levels and scan protocols without relevant toxicity. Lung metastasis, inflammation and fibrosis models and healthy mice were weekly scanned over one-month with µCT using high-resolution respiratory-gated 4D and expiration-weighted 3D protocols, comparing 5-times weekly scanned animals with controls. Radiation dose was measured by ionization chamber, optical fiberradioluminescence probe and thermoluminescent detectors in a mouse phantom. Dose effects were evaluated by in vivo µCT and bioluminescence imaging read-outs, gold standard endpoint evaluation and blood cell counts. Weekly exposure to 4D µCT, dose of 540-699 mGy/scan, did not alter lung metastatic load nor affected healthy mice. We found a disease-independent decrease in circulating blood platelets and lymphocytes after repeated 4D µCT. This effect was eliminated by optimizing a 3D protocol, reducing dose to 180-233 mGy/scan while maintaining equally high-quality images. We established µCT safety limits and protocols for weekly repeated whole-body acquisitions with proven safety for the overall health status, lung, disease process and host responses under investigation, including the radiosensitive blood cell compartment.