Oral Presentation Abstracts

Day 1 | May 11th

Scientific Session 1: An Eclipse of Proteins

Dr. James Geiger; Michigan State University, Dept. of Chemistry

The retinal binding protein family: a powerful template for design and discovery

The field of protein design has incredible potential for the development of myriad new and improved chemical and biochemical function, from the development of powerful new catalysts to new biological tools for the detection and treatment of disease. It can also be a valuable tool for understanding the structure, function and biophysics of proteins. We have focused on using members of the intracellular lipid binding protein (iLBP) family as templates for protein design, due to their small size, relative stability to mutation, ease of structural study and comparatively large internal binding cavity, suitable to binding a wide range of large hydrophobic ligands. We have focused on the study of chromophore-bound proteins that mimic the rhodopsin family in that binding of an aldehyde results in conjugated protonated Schiff base (PSB) formation. The conjugated PSB’s absorbance is exquisitely sensitive to the protein electrostatic environment, allowing us to control absorbance, sometimes over a 200 nm range. It also offers the potential of photoswitching via protonation/deprotonation of the PSB. We have used these properties to create new fluorescent proteins with a wide variety of properties, including near IR emission, photoswitching over an adjustable range of wavelengths, large Stokes shift FP’s to name a few. We have also used the system to study the detailed mechanisms of wavelength tuning and photoisomerization. This talk will focus on the unravelling of the mechanisms of these systems structurally, where we are able to see photoswitching happen in the crystal at atomic resolution. In the course of this work we have discovered how small structural changes can lead to enormous, dynamic changes in pKa within a protein binding pocket.

Hui Ma; Purdue University, Weldon School of Biomedical Engineering

Protein-protein interaction measurement using particle diffusometry in a low-volume microfluidic chip

Microfluidic techniques are widely applied in biomolecular interaction analysis and disease diagnostics. Although the volume of sample that is directly used in such assays is often only femto- to microliters of precious reagents, the "dead volume" solution supplied in syringes and tubing can be much larger, even up to milliliters, making analysis significantly more expensive. To reduce the difficulty and cost, we designed a new chip using a low volume solution for analysis; and applied it to obtain real-time data for protein-protein interaction measurements. The chip takes advantage of air/water two-phase droplet flow, on-chip fast mixing within milliseconds, and droplet capture, and only requires 2mL reagent solution. The interaction is analyzed by particle diffusometry (PD), a non-intrusive and precise optical detection method to compute properties of micro-particle in solution. Herein, we demonstrate the characterization of nanoparticle-bound protein-protein interactions on chip to analyze human immunodeficiency virus (HIV) p24 antibody-antigen binding kinetics via fluorescence microscopy. The measured kon and koff are 5.17 ´ 105M-1s-1 and 6.87 ´ 10-4s-1, respectively, and are in line as previously calculated p24-antibody binding kinetics. This new microfluidic chip and the protein-protein interaction analysis method can also be applied in other fields which require low volume solutions to perform an accurate measurement, analysis, and detection.

Owen Tyoe; University of Cincinnati, Dept. of Physics

α-Synuclein - lipid membraneinteraction and the binding mechanismof the N-terminal region

α-Synuclein (αSyn) is an intrinsically disordered protein whose aggre-gation is associated with Parkinson’s disease, dementia, and other neu-rodegenerative diseases known as synucleinopathies. However, the functional role of αSyn is still unclear, although it has been shown to be involved in the regulation of neurotransmitter release via the interaction with synaptic vesicles (SVs), vesicle clustering, and vesicle fusion. We report here a detailed characterization of the physical properties of αSyn bound to phospholipid bilayer membranes, using molecular dynamics simulation. Our results suggest that αSyn preferentially binds to anionic lipid membranes, with a dependence on lipid composition of membrane as well as the ion concentration in solution. On the basis of these studies we characterize a mechanism by which the N-terminus of αSyn binds in a concentration-dependent manner, to membranes with compositions similar to synaptic vesicles and presynaptic plasma membranes. The study then provides further evidence that changes in the lipid composition of such membranes (typically associated with neurodegenerative diseases), affect the binding properties of αSyn, specifically in the N-terminal region. We also use single molecule imaging methods including total internal reflection fluorescence (TIRF) microscopy to quantify the dependence of lipid composition and charge on vesicle clustering mediated by αSyn which is based on the methods developed previously in OT1. Together, these results reveal how lipid composition modulates the interaction of αSyn with lipid membranes and is a factor underlying its functional and pathological behaviours in vitro.

Elisabeth Garland-Kuntz; Purdue University, Dept. of Chemistry

Structural Insights into Phospholipase Cε

PLC (phospholipase C) enzymes hydrolyze phosphatidylinositol lipids to generate inositol phosphates (IPx) and diacylglycerol (DAG). These second messengers stimulate intracellular calcium release and protein kinase C activation, respectively. PLCε is one of 13 isoforms of the PLC family and is activated downstream of G protein-coupled receptors (GPCRs) and receptor tyrosine kinases (RTKs) through direct interactions with G proteins. In the cardiovascular system, PLCε is necessary for normal cardiac contractility, and its dysregulation contributes to cardiac hypertrophy and heart failure. PLCε shares a highly conserved core that is observed in other PLC subfamilies and contains a unique N-terminal CDC25 domain and two C-terminal Ras association (RA) domains. However, structural insights into the full-length protein, or catalytically active fragments that contain two or more regulatory domains, have been lacking. Using a combination of structural and functional studies, we previously showed that the PH domain and first two EF hands are dynamic in solution, providing an explanation as to why high-resolution structures have been elusive. We are now pursuing high-resolution structural studies of larger PLCε variants using single-particle cryo-electron microscopy (cryo-EM). To facilitate this process, we are utilizing conformation-specific Fab antibodies as fiducial markers that may also stabilize this conformationally heterogeneous protein. Understanding the structure of PLCε will be critical in identifying unique sites that can be targeted by small molecule chemical probes, ultimately validating this enzyme as a therapeutic target in cardiovascular disease. Here, we present cryo-EM reconstructions of PLCε PH-COOH, the largest fragment of PLCε characterized to date, in complex with a Fab.

Keynote Lecture | Dr. Janet Smith; University of Michigan, Dept. of Biological Chemistry & Biophysics

Human Hosts can detect viral pathogens with a variety of molecular surveillance systems that trigger the release of interferon proteins, resulting in the expression of interferon-stimulated genes. Some of these gene products are restriction factors that director or indirectly destroy viral RNA. The talk will describe structure-function studies of two restriction factors for RNA viruses: APOBEC3H and ZAP. The APOBEC proteins (apolipoprotein B editing complex) are cytidine deaminases (C ⇒ U). APOBEC3F, G, and H restrict retroviruses such as HIV-1 by hypermutation of retroviral DNA. The structure of APOBEC3H - RNA complex showed how the protein associates with viral RNA and is positioned to act on single-stranded viral DNA upon sythnesis by the viral reverse transcriptase. ZAP (zinc-finger antiviral protein) acts in a different manner by recognizing CpG dinucleotides in viral RNAs and targeting the RNAs for degradation. The structure of the ZAP RNA-binding domain with an RNA oligomer revealed how ZAP is specific for CpG, and subsequent studies demonstrate how ZAP recruits nucleases to the bound RNA.

Scientific Session 2: Macromolecular Milky Way

Dr. Betsy Parkinson; Purdue University, Dept. of Chemistry

Synthetic Natural Product Inspired Cyclic Peptidesfor Discovery of Bioactive Natural Products and Biocatalysts

Cyclic peptide natural products are a fantastic source of medicines, including antibiotics and immunosuppressants. Unfortunately, discovering novel cyclic peptides can be an arduous task due to the low expression of many natural product biosynthetic gene clusters and challenging purifications. For this reason, we developed SNaPP (Synthetic Natural Product Inspired Cyclic Peptides). SNaPP expedites bioactive molecule discovery by combining bioinformatics predictions of non-ribosomal peptide synthetases with chemical synthesis of the predicted natural products (pNPs). Head-to-tail cyclic peptides were targeted by using a recently discovered cyclase, the penicillin binding protein-like cyclase, as the initial search input. Analysis of 500 biosynthetic gene clusters allowed for identification of 131 novel pNPs. 51 diverse pNPs were synthesized using solid phase peptide synthesis and solution-phase cyclization. Antibacterial testing revealed 14 pNPs with antibiotic activity, including activity against multidrug-resistant Gram-negative bacteria. Overall, SNaPP demonstrates the power of combining bioinformatics predictions with chemical synthesis to accelerate the discovery of bioactive molecules.Additionally, SNaPP allowed for identification of PBP-like cyclases potentially capable of performing challenging cyclizations, such as for tetrapeptides. We have experimentally confirmed these activities and found one enzyme with a greatly expanded substrate scope that could have great utility as a biocatalyst.

Trevor Boram; Purdue University, Dept. of Biochemistry

Probing the catalytic interactions of ketosynthases using near-natural substrate analogs

Synthesis of fatty acids and specialized metabolites, such as polyketides, is dependent on several enzymatic activities, most interesting is the ketosynthase (KS). KS perform a complex carbon-carbon bond forming reaction via a ping-pong mechanism; the catalytic interactions of which are highly debated. Their reaction involves the Claisen condensation of an acylated cysteine with a malonyl-thioester, driven by the energy of the malonyl-thioester decarboxylation. A challenge in deducing the catalytic details of enzymes acting on malonyl-thioesters in general is the “hyper-reactivity” of their β-ketoacid and thioester substrates, which are prone to hydrolysis and decarboxylation. Many structures of malonyl-CoA bound enzymes feature hydrolysis of the thioester, preventing deduction of enzyme:substrate interactions in structure-function studies. In the case of KSs, capturing malonyl-thioester interactions is a fool’s errand as the activity of the enzyme is to perform these very decarboxylation and transthiolation reactions that plague structure-function experiments. To work around this reactivity problem, groups have synthesized a variety of acyl-thioester analogs for probing the details of KS catalysis with mixed success. The success of these KS:analog mechanistic studies appears to hinge upon the similarity of the analog to the natural substrate. Here, we demonstrate the synthesis of near-natural, acyl-thioester analogs, featuring single atom substitutions. Using a novel UV-vis assay, we have determined Ki values of our analogs with paradigmatic KSs E. coli FabH and G. hybrida 2-pyrone synthase. These Ki values approximate the substrate Km values, suggesting the KSs bind the analogs as they would natural substrates. Using this information, we have conducted preliminary X-ray crystallography experiments to determine the KS:analog catalytic interactions, which will allow for new insight into debated KS catalytic details.

Samadhi C. Kulathunga; Purdue University, Dept. of Chemistry

In-vitro characterization and inhibitionstudies of Sulfotransferase 1A1 as a potential therapeutic target for structure-based design of drugs toovercome drug resistance in breast cancer.

Despite the advancement of early diagnosis and various treatment options, breast cancer is the most commonly diagnosed and the second leading cause of cancer death among women in the US as of 2022.The majority of diagnosed cases correspond to estrogen receptor-positive (ER+)breast cancers, which are typically treated with anti-estrogens such as tamoxifen, the gold standard anti-estrogen for ER+ breast cancers, and fulvestran. However, studies have shown the acquisition of anti-estrogen resistance in patients after several years of anti-estrogen treatments. Thus, it is imperative to identify novel drug targets to broaden the breast cancer treatment efficacy aiming the reduction of prevailing breast cancer death rate.Human sulfotransferase 1A1 (SULT1A1) is one of the majordrug metabolizing enzymes in our body that facilitates the excretion of phenolic drugs by sulfo-conjugatingthe phenolic groups. This phenomena involves in reducingthe bio-availabilityof phenolic drugs. Significantly, SULT1A1 has been shown overexpressed in ER+ breast cancer cell lines and tamoxifen-resistant tumor tissues. Therefore, we hypothesized that SULT1A1 may be involved in anti-estrogen resistance in ER+ breast cancer patients by increasing the efflux of active drugs from the target cell.Inthis study, we focused on the in-vitrosulfonation of fulvestrant, and the two most reactive metabolites of tamoxifen using Desorption Electrospray Ionization Mass Spectrometry (DESI-MS) by measuring various kinetic parameters including substrate affinity(Km) and turn over number (kcat). Further, two SULT1A1 inhibitors, were identified and the reduction of SULT1A1 mediated sulfonation of tamoxifen-metabolites and fulvestrant in the presence of these inhibitors was assessed using DESI-MS. Finally,usingX-ray crystallographic studies SULT1A1-inhibitor co-crystal structures were determined to 1.27 Å and 1.51 Å resolutions. These high-resolution co-crystal structureswill provide a molecular platform for the structure-based design ofmore potent inhibitors against SULT1A1.

Yu-Chen Yen; Purdue University, Dept. of Biological Sciences

Cryo-EM structures of ACKR3 reveal molecular mechanisms forbiasedsignaling and promiscuity

Atypical chemokine receptor 3 (ACKR3) is a seven-transmembrane chemokine receptor that signals exclusively through β-arrestin when activated by its endogenous ligand, CXCL12. CXCR4, a canonical receptor that also responds to CXCL12, however activates both G-protein and β-arrestin signaling pathways. ACKR3 is also known to display constitutive internalization and recycling, and acts as a decoy receptor that transports CXCL12 to lysosomes for degradation. The scavenging process is important for regulating extracellular CXCL12 concentration and CXCR4 signaling, and hence to cell migration and metastasis. In addition to CXCL12, ACKR3 scavenges a large variety of other ligands, most of which activate ACKR3.To understand the molecular mechanism underlying its biased signaling and this promiscuous activation, we determined seven structures of ACKR3 bound to three different ligands using cryo-electron microscopy (cryo-EM). Our structural models represent the first of atypical chemokine receptor and reveal a unique chemokine binding mode. In the orthosteric binding pocket, several side chains of ACKR3 are rearranged to engage different ligands ,underscoring its plasticity for ligand recognition. The cytoplasmic binding pocket of ACKR3 overall adopts an active-like configurationbut has a more compact cytoplasmic cleft and a unique conformation of intracellular loop 2 (ICL2), both of which could prevent ACKR3 from binding to heterotrimeric G-proteins and instead promote biased signaling.These features may be more broadly representative of how other receptors can be induced to preclude heterotrimeric G proteins by biased agonists.

Day 2 | May 12th

Scientific Session 3: Black Hole of Biochemistry

Dr. Jonathan P. Schlebach; Indiana University, Dept. of Chemistry

Pharmacological Profiling of Cystic FibrosisVariantsby Deep Mutational Scanning

Over 400 mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene are known to cause cystic fibrosis (CF). Though all of these mutations cause a loss of CFTR function, they vary considerably with respect to the nature of their mechanistic effects on the structure and function of the CFTR chloride channel. Moreover, these variants exhibit considerable differences in their response to current therapeutics. Here we utilize deep mutational scanning to compare the plasma membrane expression of 129 CFTR variants in HEK293T cells. We identify 82 variants, including 10 variants of unknown significance (VUS), that exhibit diminished plasma membrane expression relative to WT .These findings confirm that most CF variants enhance CFTR misfolding and degradation. Of these apparent class II variants, we identify 77 that measurably respond to VX-661 and/ or VX-445, two different FDA-approved “corrector” molecules designed to rescue the expression of misfolded CFTR variants. Nevertheless, the magnitude of their effects varies widely across the spectrum of clinical variants. Our preliminary analyses suggest the most responsive mutants are structurally coupled to the corrector binding pocket. Together, our findings identify previously uncharacterized CF variants that are potentially amenable to therapeutic intervention and provide insights into the molecular basis for the mutation-specific efficacy of CFTR modulators.

Chun-Liang Chen; Purdue University, Dept. of Biological Sciences

Structural and mechanistic basis for Gβγ-mediated activation of phosphoinositide 3-kinase γ

Phosphoinositide 3-kinase γ (PI3Kγ) converts PIP2 to PIP3 as a key step in chemotaxis but is also highly expressed during metastasis, particularly in prostate, breast, and pancreatic cancer. Activation of PI3Kγ is coordinately regulated by G protein-coupled receptors (GPCRs) and receptor tyrosine kinases (RTKs). Upon GPCR activation, Gbγ subunits dissociate from heterotrimeric G proteins to promote membrane translocation and activation of PI3Kγ, thereby increasing the production of PIP3 and triggering downstream signaling via the AKT pathway. Growth factors binding to RTKs activate Ras, which also activates PI3Kγ. PI3Kγ is composed of a catalytic subunit (p110γ) and a regulatory subunit (p101), which form a constitutive heterodimer. Although a cryo-EM structure of PI3Kg bound to a nanobody has been recently reported, molecular mechanisms underlying the Gβγ- and Ras-mediated activation remain unclear. In addition, the nanobody used in the prior structure determination may cause conformational changes and binds to a domain believed to bind Gβγ. To try to resolve these mechanisms, we have purified the PI3Kγ heterodimer to homogeneity, demonstrated its Gβγ-dependent activation, and reconstructed a 3.1 Å cryo-electron microscopy (cryo-EM) map and built a de novo model for heterodimeric PI3Kγ in the absence of nanobody. We went on to obtain a cryo-EM reconstruction of the PI3Kγ–Gβγ complex, revealing distinct Gβγ binding sites on the p110γ and the p101 subunits and conformational changes that we hypothesize are linked to allosteric activation. These findings allow for functional investigation of Gβγ-mediated activation mechanisms and will aid in future drug development targeting PI3Kγ for cancer therapeutics.

Kaushik Muralidharan; Purdue University, Dept. of Biological Sciences

Understanding the molecular mechanisms of PLCε regulation

Haley R. Harrington; Indiana. University, Dept. of Chemistry

Determining the Constraints of -1 Programmed Ribosomal Frameshifting in Alphaviruses

Alphaviruses are positive-sense RNA viruses that are transmitted by arthropods to vertebrates. In vertebrates, infection can result in long-term viral polyarthritis and/ or encephalitis that can be fatal. Alphaviruses utilize -1 programmed ribosomal frameshifting (-1PRF) to generate a virulence factor known as the transframe (TF) protein. Beyond the canonical RNA slip-site and RNA structure element that are known to stimulate -1PRF, we have found that the efficiency of -1PRF is sensitive to the topology of the nascent polypeptide and the pulling forces on the nascent chain that occur during polyprotein synthesis. Our results reveal that modifications to the hydrophobicity and/ or the position of the second transmembrane domain of the E2 protein (TM2) alter the timing and magnitude of forces generated by its insertion into the ER membrane impact ribosomal frameshifting. To investigate the role of nascent chain pulling forces in alphavirus frameshifting, we characterized a series of chimeric -1PRF reporters bearing combinations of slip sites, RNA secondary structures, and upstream TM domains from different alphaviruses. Consistent with expectations, our results suggest that the slip-site and a correctly positioned RNA secondary structure are essential for -1PRF. However, our results also reveal that nascent chain pulling forces alone cannot overcome the absence of secondary structure. Interestingly, our data also suggest that not all alphaviruses have efficient -1PRF motifs, which implies that many are not reliant on the production of TF. This suggest the role of TF in infection may vary among alphaviruses. These observations merit follow up studies in the context of the viral life cycle. Together our work provides new insights into the -1PRF mechanism within alphaviruses.

Scientific Session 4: A Constellation of Methods

Dr. Lauren Ann Metskas; Purdue University, Depts. of Biological Sciences & Chemistry

Structures in situ: solving a Rubisco structure and interactions inside intact carboxysomes

Roughly 20% of bacteria employ bacterial microcompartments (BMCs) to sequester dangerous or inefficient enzymatic processes. In these structures, a cargo of enzymes and accessory proteins is encased within a semi-permeable protein shell that permits passage of substrates and products but restricts movement of intermediates. In addition to their importance as a component of many bacterial species’ metabolisms, BMCs have become a target of protein engineering due to their potential for enclosing cargos of choice for alternative pharmacological and biotechnological applications.Despite the importance of BMCs in prokaryotic biology and bioengineering, structural heterogeneity has prevented a complete understanding of the architecture, ultrastructure, and spatial organization of both the shell proteins and the cargo. Here, we employ cryo-electron tomography to image intact alpha-carboxysomes, a model BMC responsible for carbon fixationin cyanobacteria and chemoautotrophs. We developed an ultrastructure analysis toolkit to identify a previously unknown arrangement of the enzymatic cargo, Rubisco, and solved a highresolution subtomogram average of the enzyme in situ. Efforts are underway to identify and resolve additional cargo components and the lattice-like arrangement of the shell proteins to build a complete model of the entire assembly and the flow of small molecule substrates and products inside it. Taken together, these data offer new insights into conserved BMC organization and heterogeneity, and establish a new style of in situ tomographic analysis that can be applied to other complex protein architectures.

Tunde Aderinwale; Purdue University, Dept. of Computer Science

Real-Time Structure Search and Structure Classification for AlphaFold Protein Models

Last year saw a breakthrough in protein structure prediction, where the AlphaFold2 method showed a substantial improvement in the modeling accuracy. Following the software release of AlphaFold2, predicted structures by AlphaFold2 for proteins in 21 species were made publicly available via the AlphaFold Database. Here, to facilitate structural analysis and application of AlphaFold2 models, we provide the infrastructure, 3D-AF-Surfer, which allows real-time structure-based search for the AlphaFold2 models. In 3D-AF-Surfer, structures are represented with 3D Zernike descriptors (3DZD), which is a rotationally invariant, mathematical representation of 3D shapes. We developed a neural network that takes 3DZDs of proteins as input and retrieves proteins of the same fold more accurately than direct comparison of 3DZDs. Using 3D-AF-Surfer, we report structure classifications of AlphaFold2 models and discuss the correlation between confidence levels of AlphaFold2 models and intrinsic disordered regions.

Matthew Clark; Purdue University, Dept. of Chemistry

Imaging and controlling chemical processes in live cells with real-time precision opto-control (RPOC)

Achieving precision control of molecular activities and chemical reactions in live cells is a long-sought capability by life scientists. Currently, no existing technology can probe molecular targets in cells and simultaneously control the activities of only these targets with high spatial precision and in real-time. We develop a technology dubbed real-time precision opto-control (RPOC) technology that uses a chemical-specific optical response from target molecular species during laser scanning imaging to trigger an ultrafast optical shutter, which allows a separate laser beam to only interact with the molecules of interest without interacting with other parts of the sample. Combining multiple optical signal outputs creates digital logical functions for additional molecular and region selectivity. RPOC allows the analyst to automatically probe and control biomolecular activities and chemical processes in dynamic living samples with submicron spatial accuracy, nanosecond response time, and high chemical specificity.

Sajad Shiekh; Kent State University, Dept. of Physics

FRET-PAINT for Studying Long Telomeric DNA Overhangs and their Interactions with Shelterin Proteins

I will be presenting thesingle molecule FRET-PAINT measurements and computational modeling studies where we investigated the accessibility of range ofdifferent human telomeric overhangs. These overhangs can form 1-7 tandem G-quadruplex (GQ) structures, which, to our knowledge, is the most comprehensive range studied to date and covers a significant portion of the physiologically relevant telomeric overhang length scale. Our measurements demonstrate novel accessibility maps where certain regions of telomeric overhang are significantly more accessible than others. We also observe folding frustration patterns with a well-defined periodicity and constructs with a certain number of telomeric repeats demonstrate elevated levels of frustration compared to others. These patterns have significant implications for telomere organization, protection of free 3’-end against exonuclease activity and telomerase-catalyzed extension, and folding cooperativity between neighboring GQ structures.We further investigated theimpact of POT1 and a four-protein Shelterin complex on the accessibility of human telomeric DNA constructs.with physiologically relevant overhang lengths(28-150 nt). To quantify telomere accessibility, we monitored transient binding events of a Cy5-labeled, short peptide nucleic acid strand to available sites on the telomere using FRET-PAINT methodology. When averaged over 11 constructs investigated in this study, we observed approximately 2.5-fold reduced accessibility in the presence of POT1 (compared to DNA-only case) and about 5-fold reduced accessibility in the presence of Shelterin. These results suggest the protection of exposed telomeric segments by POT1 was effective enough to dominate over its mild GQ unfolding activity. The more effective protection in the presence of Shelterin compared to POT1-only case suggests the restructuring of the junction region between single and double stranded telomere, which is otherwise the most accessible part of the overhang, serves an important function in telomere maintenance.