Biology subjects

Hoogerheide, D. P.

Publications and source records attributed to Hoogerheide, D. P..

6 recordsLinked to original sources

bioRxiv2026

XAMPGen: An Interpretable SI Pipeline For Novel Antimicrobial Peptide Discovery

Antimicrobial peptides (AMPs) are short peptides with a broad spectrum of activity against bacterial pathogens, representing a promising avenue for combating the growing threat of antibiotic resistance. This study focuses on two main components: 1) generating novel AMPs with desired properties using generative SI (GenSI) and 2) utilizing explainable SI (XSI) methods to validate the GenSI model and derive biological insights. In this work, we fine-tuned ProGen, a decoder-only protein language model, on known AMPs to generate novel peptide sequences with strong antimicrobial properties. To advance the interpretability and reliability of the generative model, we applied Explainable SI (XSI) techniques. Specifically, we used a probing classifier and a regressor to identify cornerstone layers that encode functionally critical information. Self-attention analysis within these layers revealed residue level interdependencies and patterns underlying antimicrobial functions, which we validated against well-characterized AMPs from the literature. Overall, these methods provide mechanistic insight into the model's decision-making process, offering a principled basis for trusting the sequences it generates rather than treating the model as a black box. Candidate sequences were filtered based on predicted Minimum Inhibitory Concentration (MIC) along with human insights into the structural activities of the AMPs in bacterial membranes, and a final set of peptides was selected for experimental testing. Biological assays for determining MIC confirmed antimicrobial activity in all generated and tested peptides, with fluorescence and circular dichroism measurements providing structural insight. With our XSI approach, we are able to reveal correlations between amino acid sequence patterns and binding properties of these sequences to bacterial membrane mimics, that directly correlate with biological activities in cell-based assays. This builds confidence in that the learned amino acid interdependencies as applied to AMPs carries physico-chemical meaning and is relevant for discoveries of novel AMPs with therapeutic potential. This work demonstrates that interpretable generative language models can serve as effective and transparent tools for rational AMP discovery.

bioinformatics↗

bioRxiv2026

Towards VDAC pharmacology: Developing small-peptide channel inhibitors inspired by nature

The Voltage-Dependent Anion Channel (VDAC) forms a unique pathway across the mitochondrial outer membrane for small ions and water-soluble metabolites, such as ATP and ADP. Despite VDAC's involvement in both normal regulation of respiration and in a wide variety of mitochondria-associated pathologies, there are, surprisingly, no known FDA-approved pharmacological agents targeting VDAC. Earlier, we have shown that VDAC can be regulated through its interactions with well-known cytosolic proteins--dimeric tubulin and -synuclein (Syn)--that are able to reversibly block this channel at nanomolar concentrations. We hypothesized that the architecture of a prospective VDAC inhibitor should fulfill two main requirements: possess a negatively charged disordered polypeptide C- or N-terminus as a pore blocking domain and feature a mitochondrial membrane-binding domain. Using Syn as a template, we engineered three peptides and demonstrated that they all characteristically block VDAC reconstituted into planar lipid membranes in a controllable manner, thereby mimicking the actions of tubulin and Syn. The VDAC blockage times induced by the synthetic peptides are comparable to Syn-induced blockages, though the half-effect solution concentration for synthetic blockers is a hundred times higher than that of Syn. Our work can be considered a proof-of-principle study that informs future development of mitochondria-targeting pharmacological tools.

biophysics↗

bioRxiv2026

Species- and Topic-aware Representation Learning for Antimicrobial Peptide Discovery

Antimicrobial resistance poses a major global health challenge, necessitating efficient strategies to discover potent antimicrobial peptides (AMPs). While recent generative models can produce many candidate sequences, experimentally validating all generated peptides in wet labs is impractical due to the high costs and time involved in such measurements. As a result, there is a strong demand for accurate predictions of peptide efficacy, typically measured as the minimum inhibitory concentration (MIC). We introduce STAMP, a framework for Species- and Topic-aware Representation Learning in AMP Discovery. This unified machine learning framework allows for cross-species predictions of AMP activity. STAMP integrates protein language model embeddings with species conditioning and topic-aware representations that capture sequence-level patterns, enabling generalizable predictions across multiple bacterial species within a single model. We evaluated STAMP on three benchmark datasets, which include two previously published datasets and a newly curated dataset derived from DBAASP, addressing duplicates and inconsistencies systematically. STAMP achieved strong predictive performance across these datasets, demonstrating a Pearson correlation coefficient (PCC) of 0.837 and an R2 of 0.70, outperforming several baseline models. Importantly, we further validated our prediction model using peptides that were experimentally tested for their antimicrobial activity against E.coli. and S.epidermidis bacteria, demonstrating its real-world applicability. Furthermore, residue-level importance analyses provide insights into the sequence determinants governing antimicrobial activity. Together, these results establish STAMP as a scalable framework for MIC prediction and an effective computational tool for accelerating AMP discovery and optimization.

bioinformatics↗

bioRxiv2022

Restricting α-Synuclein Transport into Mitochondria by Inhibition of α-Synuclein-VDAC Complexation as a Potential Therapeutic Target for Parkinson's Disease Treatment

Involvement of alpha-synuclein (Syn) in Parkinsons disease (PD) is complicated and difficult to trace on cellular and molecular levels. Recently we established that Syn can regulate mitochondrial function by voltage-activated complexation with the Voltage-Dependent Anion Channel (VDAC) of the outer mitochondrial membrane. When complexed with Syn, the VDAC pore is partially blocked, reducing the transport of ATP/ADP and other metabolites. Further, Syn can translocate into the mitochondria through VDAC, where it interferes with mitochondrial respiration. Recruitment of Syn to the VDAC-containing lipid membrane appears to be a crucial prerequisite for both the blockage and translocation processes. Here we report an inhibitory effect of HK2p, a small membrane-binding peptide from the mitochondria-targeting N-terminus of hexokinase 2, on the Syn membrane binding, and hence on Syn complex formation with VDAC and translocation through it. In electrophysiology experiments, addition of HK2p at micromolar concentrations to the same side of the membrane as Syn results in dramatic reduction of the frequency of blockage events in a concentration-dependent manner, reporting on complexation inhibition. Using two complementary methods of measuring protein-membrane binding, bilayer overtone analysis and fluorescence correlation spectroscopy, we found that HK2p induces detachment of Syn from lipid membranes. Experiments with live HeLa cells using proximity ligation assay confirmed that HK2p impedes Syn entry into mitochondria. Our results demonstrate that it is possible to regulate Syn-VDAC complexation by a rationally designed peptide, thus suggesting new avenues in the search for peptide therapeutics to alleviate Syn mitochondrial toxicity in PD and other synucleinopathies.

biophysics↗

bioRxiv2021

MspA Porin as a Local Nanopore Probe for Membrane-bound Proteins

Nanopore sensing is based on detection and analysis of nanopore transient conductance changes induced by analyte capture. We have recently shown that -Synuclein (Syn), an intrinsically disordered, membrane-active, neuronal protein implicated in Parkinson disease, can be reversibly captured by the VDAC nanopore. The capture process is a highly voltage dependent complexation of the two proteins where transmembrane potential drives the polyanionic C-terminal domain of Syn into VDAC--exactly the mechanism by which generic nanopore-based interrogation of proteins and polynucleotides proceeds. The complex formation, and the motion of Syn in the nanopore, thus may be expected to be only indirectly dependent on the pore identity. Here, we confirm this prediction by demonstrating that when VDAC is replaced with a different transmembrane pore, the engineered mycobacterial porin M2MspA, all the qualitative features of the Syn/nanopore interaction are preserved. The rate of Syn capture by M2MspA rises exponentially with the applied field, while the residence time displays a crossover behavior, indicating that at voltages >50 mV M2MspA-bound Syn largely undergoes translocation to the other side of the membrane. The translocation is directly confirmed using the selectivity tag method, in which the polyanionic C-terminal and neutral N-terminal regions of Syn alter the selectivity of the M2MspA channel differently, allowing direct discrimination of translocation vs retraction for single Syn molecules. We thus prove that the physical model of the motion of disordered protein chains in the nanopore confinement and the selectivity tag technique are not limited to VDAC but are broadly applicable to nanopore-based protein detection, analysis, and separation technologies.

biophysics↗

bioRxiv2021

Highly basic clusters in the HSV-1 nuclear egress complex drive membrane budding by inducing lipid ordering

During replication of herpesviruses, capsids escape from the nucleus into the cytoplasm by budding at the inner nuclear membrane. This unusual process is mediated by the viral nuclear egress complex (NEC) that deforms the membrane around the capsid by oligomerizing into a hexagonal, membrane-bound scaffold. Here, we found that highly basic membrane-proximal regions (MPRs) of the NEC alter lipid order by inserting into the lipid headgroups and also promote negative Gaussian curvature. We also find that the electrostatic interactions between the MPRs and the membranes are essential for membrane deformation. One of the MPRs is phosphorylated by a viral kinase during infection, and the corresponding phosphomimicking mutations block capsid nuclear egress. We show that the same phosphomimicking mutations disrupt the NEC/membrane interactions and inhibit NEC-mediated budding in vitro, providing a biophysical explanation for the in-vivo phenomenon. Our data suggest that the NEC generates negative membrane curvature by both lipid ordering and protein scaffolding and that phosphorylation acts as an "off" switch that inhibits the membrane-budding activity of the NEC to prevent capsid-less budding.

microbiology↗