The clinical efficacy and safety of a drug is determined by its activity profile across many proteins in the proteome. However, designing drugs with a specific multi-target profile is both complex and difficult. Therefore methods to design drugs rationally a priori against profiles of several proteins would have immense value in drug discovery. Here we describe a new approach for the automated design of ligands against profiles of multiple drug targets. The method is demonstrated by the evolution of an approved acetylcholinesterase inhibitor drug into brain-penetrable ligands with either specific polypharmacology or exquisite selectivity profiles for G-protein-coupled receptors. Overall, 800 ligand–target predictions of prospectively designed ligands were tested experimentally, of which 75% were confirmed to be correct. We also demonstrate target engagement in vivo. The approach can be a useful source of drug leads when multi-target profiles are required to achieve either selectivity over other drug targets or a desired polypharmacology.

Anteroposterior (AP) polarity in the mammalian embryo is specified during gastrulation when naive progenitor cells in the primitive ectoderm are recruited into the primitive streak to form mesoderm and endoderm. At the opposite pole, this process is inhibited by signals previously induced in distal visceral endoderm (DVE). Both DVE and primitive streak formation, and hence positioning of the AP axis, rely on the TGFβ family member Nodal and its proprotein convertases Furin and Pace4. Here, we show that Nodal and Furin are initially co-expressed in the primitive endoderm together with a subset of DVE markers such as Lefty1 and Hex. However, with the appearance of extra-embryonic ectoderm (ExE), DVE formation is transiently inhibited. During this stage, Nodal activity is essential to specify embryonic VE and restrict the expression of Furin to the extra-embryonic region. Activation of Nodal is also necessary to maintain determinants of pluripotency such as Oct4, Nanog and Foxd3 during implantation, and to stimulate elongation of the egg cylinder, before inducing DVE and germ layer formation. We conclude that Nodal is already activated in primitive endoderm, but induces a functional DVE only after promoting the expansion of embryonic VE and pluripotent progenitor cells in the epiblast.

Bone morphogenetic proteins (BMPs) are derived from inactive precursor proteins by endoproteolytic cleavage. Here we show that processing of Nodal and Myc-tagged BMP4 is significantly enhanced by SPC1/Furin or SPC4/PACE4, providing direct evidence that regulation of BMP signaling is likely to be controlled by subtilisin-like proprotein convertase (SPC) activities. Nodal processing is dramatically enhanced if two residues adjacent to the precursor cleavage site are substituted with amino acids found at the equivalent positions of Activin, demonstrating that structural constraints at the precursor cleavage site limit the processing efficiency. However, in transfection assays, mature Nodal is undetectable either in culture supernatants or in cell lysates, despite efficient cleavage of the precursor protein, suggesting that mature Nodal is highly unstable. Domain swap experiments support this conclusion since mature BMP4 or Dorsalin are also destabilized when expressed in conjunction with the Nodal pro domain. By contrast, mature Nodal is stabilized by the Dorsalin pro domain, which mediates the formation of stable complexes. Collectively, these data show that the half-life of mature BMPs is greatly influenced by the identity of their pro regions.