A deep learning approach to identify gene targets of a therapeutic for human splicing disorders.

Pre-mRNA splicing is a key controller of human gene expression. Disturbances in splicing due to mutation lead to dysregulated protein expression and contribute to a substantial fraction of human disease. Several classes of splicing modulator compounds (SMCs) have been recently identified and establish that pre-mRNA splicing represents a target for therapy. We describe herein the identification of BPN-15477, a SMC that restores correct splicing of ELP1 exon 20. Using transcriptome sequencing from treated fibroblast cells and a machine learning approach, we identify BPN-15477 responsive sequence signatures. We then leverage this model to discover 155 human disease genes harboring ClinVar mutations predicted to alter pre-mRNA splicing as targets for BPN-15477. Splicing assays confirm successful correction of splicing defects caused by mutations in CFTR, LIPA, MLH1 and MAPT. Subsequent validations in two disease-relevant cellular models demonstrate that BPN-15477 increases functional protein, confirming the clinical potential of our predictions.

Conformational Fine-Tuning of Pore-Forming Peptide Potency and Selectivity.

To better understand the sequence-structure-function relationships that control the activity and selectivity of membrane-permeabilizing peptides, we screened a peptide library, based on the archetypal pore-former melittin, for loss-of-function variants. This was accomplished by assaying library members for failure to cause leakage of entrapped contents from synthetic lipid vesicles at a peptide-to-lipid ratio of 1:20, 10-fold higher than the concentration at which melittin efficiently permeabilizes the same vesicles. Surprisingly, about one-third of the library members are inactive under these conditions. In the negative peptides, two changes of hydrophobic residues to glycine were especially abundant. We show that loss-of-function activity can be completely recapitulated by a single-residue change of the leucine at position 16 to glycine. Unlike the potently cytolytic melittin, the loss-of-function peptides, including the single-site variant, are essentially inactive against phosphatidylcholine vesicles and multiple types of eukaryotic cells. Loss of function is shown to result from a shift in the binding-folding equilibrium away from the active, bound, α-helical state toward the inactive, unbound, random-coil state. Accordingly, the addition of anionic lipids to synthetic lipid vesicles restored binding, α-helical secondary structure, and potent activity of the "negative" peptides. While nontoxic to mammalian cells, the single-site variant has potent bactericidal activity, consistent with the anionic nature of bacterial membranes. The results show that conformational fine-tuning of helical pore-forming peptides is a powerful way to modulate their activity and selectivity.

Independent contribution of extracellular proton binding sites to ASIC1a activation.

Acid-sensing ion channels (ASICs) are a group of trimeric cation permeable channels gated by extracellular protons that are mainly expressed in the nervous system. Despite the structural information available for ASIC1, there is limited understanding of the molecular mechanism that allows these channels to sense and respond to drops in extracellular pH. In this report, we employed the substituted cysteine accessibility method and site-directed mutagenesis to examine the mechanism of activation of ASIC1a by extracellular protons. We found that the modification of E238C and D345C channels by MTSET reduced proton apparent affinity for activation. Furthermore, the introduction of positively charged residues at position 345 rendered shifted biphasic proton activation curves. Likewise, channels bearing mutations at positions 79 and 416 in the palm domain of the channel showed reduced proton apparent affinity and biphasic proton activation curves. Of significance, the effect of the mutations at positions 79 and 345 on channel activation was additive. E79K-D345K required a change to a pH lower than 2 for maximal activation. In summary, this study provides direct evidence for the presence of two distinct proton coordination sites in the extracellular region of ASIC1a, which jointly facilitate pore opening in response to extracellular acidification.

Synthetic molecular evolution of pore-forming peptides by iterative combinatorial library screening.

We previously reported the de novo design of a combinatorial peptide library that was subjected to high-throughput screening to identify membrane-permeabilizing antimicrobial peptides that have ß-sheet-like secondary structure. Those peptides do not form discrete pores in membranes but instead partition into membrane interfaces and cause transient permeabilization by membrane disruption, but only when present at high concentration. In this work, we used a consensus sequence from that initial screen as a template to design an iterative, second generation library. In the 24-26-residue, 16,200-member second generation library we varied six residues. Two diad repeat motifs of alternating polar and nonpolar amino acids were preserved to maintain a propensity for non-helical secondary structure. We used a new high-throughput assay to identify members that self-assemble into equilibrium pores in synthetic lipid bilayers. This screen was done at a very stringent peptide to lipid ratio of 1:1000 where most known membrane-permeabilizing peptides, including the template peptide, are not active. In a screen of 10,000 library members we identified 16 (~0.2%) that are equilibrium pore-formers at this high stringency. These rare and highly active peptides, which share a common sequence motif, are as potent as the most active pore-forming peptides known. Furthermore, they are not α-helical, which makes them unusual, as most of the highly potent pore-forming peptides are amphipathic α-helices. Here we demonstrate that this synthetic molecular evolution-based approach, taken together with the new high-throughput tools we have developed, enables the identification, refinement, and optimization of unique membrane active peptides.

Interactions of membrane active peptides with planar supported bilayers: an impedance spectroscopy study.

Membrane active peptides exert their biological effects by interacting directly with a cell's lipid bilayer membrane. These therapeutically promising peptides have demonstrated a variety of activities including antimicrobial, cytolytic, membrane translocating, and cell penetrating activities. Here, we use electrochemical impedance spectroscopy (EIS) on polymer-cushioned supported lipid bilayers constructed on single crystal silicon to study two pairs of closely related membrane active peptides selected from rationally designed, combinatorial libraries to have different activities in lipid bilayers: translocation, permeabilization, or no activity. Using EIS, we observed that binding of a membrane translocating peptide to the lipid bilayer resulted in a small decrease in membrane resistance followed by a recovery back to the original value. The recovery may be directly attributable to peptide translocation. A nontranslocating peptide did not decrease the resistance. The other pair, two membrane permeabilizing peptides, caused an exponential decrease of membrane resistance in a concentration-dependent manner. This permeabilization of the supported bilayer occurs at peptide to lipid ratios as much as 1000-fold lower than that needed to observe effects in vesicle leakage assays and gives new insights into the fundamental peptide-bilayer interactions involved in membrane permeabilization.

Determining the mechanism of membrane permeabilizing peptides: identification of potent, equilibrium pore-formers.

To enable selection and characterization of highly potent pore-forming peptides, we developed a set of novel assays to probe 1) the potency of peptide pores at very low peptide concentration; 2) the presence or absence of pores in membranes after equilibration; 3) the interbilayer exchangeability of pore-forming peptides; and 4) the degree to which pore-forming peptides disrupt the bilayer organization at equilibrium. Here, we use these assays to characterize, in parallel, six membrane-permeabilizing peptides belonging to multiple classes. We tested the antimicrobial peptides LL37 and dermaseptin S1, the well-known natural lytic peptides melittin and alamethicin, and the very potent lentivirus lytic peptides LLP1 and LLP2 from the cytoplasmic domain of HIV GP41. The assays verified that that the antimicrobial peptides are not potent pore formers, and form only transient permeabilization pathways in bilayers which are not detectable at equilibrium. The other peptides are far more potent and form pores that are still detectable in vesicles after many hours. Among the peptides studies, alamethicin is unique in that it is very potent, readily exchanges between vesicles, and disturbs the local bilayer structure even at very low concentration. The equally potent LLP peptides do not exchange readily and do not perturb the bilayer at equilibrium. Comparison of these classes of pore forming peptides in parallel using the set of assays we developed demonstrates our ability to detect differences in their mechanism of action. Importantly, these assays will be very useful in high-throughput screening where highly potent pore-forming peptides can be selected based on their mechanism of action.