In order to deduce the molecular mechanisms of biological function, it is necessary to monitor changes in the sub-cellular location, activation and interaction of proteins within living cells in real time. Förster resonance energy transfer (FRET)-based biosensors that incorporate genetically encoded, fluorescent proteins permit high spatial resolution imaging of protein–protein interactions or protein conformational dynamics. However, non-specific fluorescence background often obscures small FRET signal changes, and intensity-based biosensor measurements require careful interpretation and several control experiments. These problems can be overcome by using lanthanide (Tb(III) or Eu(III)) complexes as donors and green fluorescent protein (GFP) or other conventional fluorophores as acceptors. Essential features of this approach are the long-lifetime (~ms) luminescence of Tb(III) complexes and time-gated luminescence microscopy. This allows pulsed excitation followed by a brief delay that eliminates nonspecific fluorescence before detection of Tb(III)-to-GFP emission. The challenges of intracellular delivery, selective protein labeling, and time-gated imaging of lanthanide luminescence are presented, and recent efforts to investigate the cellular uptake of lanthanide probes are reviewed. Data is presented showing that conjugation to arginine-rich, cell penetrating peptides (CPPs) can be used as a general strategy for cellular delivery of membrane impermeable lanthanide complexes. A heterodimer of a luminescent Tb(III) complex, Lumi4, linked to trimethoprim (TMP) and conjugated to nonaarginine via a reducible disulfide linker rapidly (~10 min) translocates into the cytoplasm of Maden Darby canine kidney cells from culture medium. With this reagent, the intracellular interaction between GFP fused to FK506 binding protein 12 (GFP-FKBP12) and the rapamycin binding domain of mTOR fused to Escherichia coli dihydrofolate reductase (FRB-eDHFR) was imaged at high signal-to-noise ratio with fast (1–3 s) image acquisition using a time-gated luminescence microscope. The data reviewed and presented here show that lanthanide biosensors enable fast, sensitive and technically simple imaging of protein-protein interactions in live cells.

Generative artificial intelligence (AI) has the potential to greatly increase the speed, quality and controllability of antibody design. Traditional de novo antibody discovery requires time and resource intensive screening of large immune or synthetic libraries. These methods also offer little control over the output sequences, which can result in lead candidates with sub-optimal binding and poor developability attributes. Several groups have introduced models for generative antibody design with promising in silico evidence, however, no such method has demonstrated de novo antibody design with experimental validation. Here we use generative deep learning models to de novo design antibodies against three distinct targets, in a zero-shot fashion, where all designs are the result of a single round of model generations with no follow-up optimization. In particular, we screen over 400,000 antibody variants designed for binding to human epidermal growth factor receptor 2 (HER2) using our high-throughput wet lab capabilities. From these screens, we further characterize 421 binders using surface plasmon resonance (SPR), finding three that bind tighter than the therapeutic antibody trastuzumab. The binders are highly diverse, have low sequence identity to known antibodies, and adopt variable structural conformations. Additionally, these binders score highly on our previously introduced Naturalness metric, indicating they are likely to possess desirable developability profiles and low immunogenicity. We open source the HER2 binders and report the measured binding affinities. These results unlock a path to accelerated drug creation for novel therapeutic targets using generative AI combined with high-throughput experimentation.

Statins are commonly used to control hypercholesterolemia and to prevent cardiovascular diseases. Among the statins, Simvastatin is one of the most frequently prescribed statins because of its efficacy in reducing LDL lipoprotein cholesterol levels, its tolerability, and its reduction of cardiovascular risk and mortality. Conflicting results have been reported with regard to benefits (pleiotropic effects) as well as risks (adverse effects) of simvastatin on different soft and hard tissues. In the current study, Attenuated Total Reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy was used to obtain detailed information about protein conformational changes due to simvastatin therapy of soft tissues namely liver, testis, sciatic nerve and hard tissues such as femur and tibia. Protein secondary structural changes were predicted by intensity calculations from second derivative spectra and neural network (NN) analysis, using the amide I band (1700-1600 cm(-1)) of FTIR spectra. Moreover, based on protein secondary structural differences, hierarchical cluster analysis was carried out in the 1700-1600 cm(-1) region. The results of our study in liver, testis and sciatic nerve tissues revealed that simvastatin treatment significantly decreased alpha helix structure and beta sheet structure at 1638 cm(-1), while increased the anti-parallel and aggregated beta sheet and random coil structures implying a simvastatin-induced protein denaturation in treated groups. Different to soft tissues, the results of hard tissue studies on femur and tibia bones revealed increased alpha helix structure and decreased anti-parallel beta sheet, aggregated beta sheet and random coil structures implying more strengthened bone tissues in simvastatin-treated groups. Finally, the simvastatin-treated and control groups for all soft and bone tissues were successfully differentiated using cluster analysis. According to the heterogeneity values in the cluster analysis of these tissues, the sciatic nerve tissue was found to be the most affected tissue from simvastatin treatment among the studied soft tissues. In addition, the high heterogeneity value implied high secondary structural difference between control and simvastatin-treated groups in tibia bone tissues. These findings reveal that FTIR spectroscopy with bioinformatic analyses such as neural network and hierarchical clustering, allowed us to determine the simvastatin-induced protein conformational changes as adverse and pleiotropic effects of the drug on different soft and hard tissues.

Macrocycles and cyclic peptides are increasingly attractive therapeutic modalities as they often have improved affinity, are able to bind to extended protein interfaces and otherwise have favorable properties. Macrocyclization of a known binder molecule has the potential to stabilize its bioactive conformation, improve its metabolic stability, cell permeability and in certain cases oral bioavailability. Herein, we present an in silico approach that automatically generates, evaluates and proposes cyclizations utilizing a library of well-established chemical reactions and reagents. Using the three-dimensional (3D) conformation of the linear molecule in complex with a target protein as starting point, this approach identifies attachment points, generates linkers, evaluates the conformational landscape of suitable linkers and their geometric compatibility and ranks the resulting molecules with respect to their predicted conformational stability and interactions with the target protein. As we show here with several prospective and retrospective case studies, this procedure can be applied for the macrocyclization of small molecules and peptides and even PROTACs and proteins.The presented approach is an important step towards the enhanced utilization of macrocycles andcyclic peptides as attractive therapeutic modalities. File list (2) download file view on ChemRxiv DesignOfMacrocycles_chemrxiv.pdf (1.85 MiB) download file view on ChemRxiv SUPPORTING INFORMATION.pdf (458.87 KiB)

Protein–protein interactions (PPIs) are central to biological processes and represent an important class of therapeutic targets. Here we show that the interaction between FK506 binding protein 12 fused to green fluorescent protein (GFP-FKBP) and the rapamycin-binding domain of mTor fused to Escherichia coli dihydrofolate reductase (FRB-eDHFR) can be sensitively detected (signal-to-background (S:B) >100) and accurately quantified within an impure cell lysate matrix using a luminescence resonance energy transfer (LRET) assay. Ascomycin-mediated inhibition of GFP-FKBP/rapamycin/FRB-eDHFR complex formation was also detected at high S:B (>80) and Z’-factor (0.89). The method leverages the selective, stable binding of trimethoprim (TMP)-terbium complex conjugates to eDHFR, and time-resolved, background-free detection of the long-lifetime (~ms) terbium-to-GFP LRET signal that indicates target binding. TMP/eDHFR labeling can be adapted to develop high-throughput screening assays and complementary, quantitative counter-screens for a wide variety of PPI targets with a broad range of affinities that may not be amenable to purification.