Poster Session #1

Conference: 2022: 72nd ACA Annual Meeting
07/30/2022: 5:30 PM - 7:30 PM
PS1 
Poster Session 
Portland Marriott Downtown Waterfront 
Room: Exhibit Hall 

Presentations

A comprehensive strategy for quick determination of protein structures by MR-native SAD method

A special advantage of protein crystallography is the quick structure determination of target proteins. The high-throughput structure determination of protein-compound complexes is critical for pharmaceutical and life sciences. To achieve rapid structure determination, we need to overcome two bottlenecks in protein crystallography, crystallization and phasing. We have developed several methods for obtaining good crystals and reported them in ACA meetings (1, 2, 3). Here, we present a useful method for phasing: the MR-native SAD method using a diffraction data collection system with low-energy X-ray, which has been developed at PF in KEK.
It is relatively easy to obtain initial phases by the molecular replacement (MR) method using a starting model from PDB or AlphaFold2. However, since the starting model is not perfect, model bias arising from the imperfect model frequently hampers quick model building and crystallographic refinement. So, we have combined the MR method and an experimental phasing method using anomalous diffraction from sulfur atoms (native SAD). While a typical MR-SAD method needs derivative (typically Se-Met proteins) crystals for measuring anomalous signals, the MR-native SAD method does not need derivative crystals. Since we have developed a beamline for the native SAD method, it is possible to use anomalous diffractions from sulfur atoms for phasing. Diffraction data collection using low-energy X-ray of 1.9 Å or 2.7 Å wavelength is possible with BL-1A, which provides high-quality data for the SAD phasing (4). A crystal shaping machine can also be utilized to improve diffraction data quality (5). However, since the determination of the substructure is frequently difficult in the native SAD phasing, initial phases from the MR method are valuable for the substructure determination. Here, we present several examples of crystal structure determination using the MR-native SAD method (6, 7). MR-native SAD frequently gave a high-quality model without manual model building, even when the MR method gave a poor model that was too difficult to fix problems manually (Figure 1).

References
1. Senda et al., A comprehensive strategy for efficient generation of well-diffracting crystals. ACA2021
2. Senda et al. (2016) Advanced Method in Structural Biology, Springer, pp. 139.
3. Senda et al. (2016) Cryst. Growth Des. 16, 1565. doi: 10.1021/acs.cgd.5b01692
4. Liebschner et al. (2016) Acta Cryst. D 72, 728. doi: 10.1107/S2059798316005349
5. Kawano et al. (2022) Acta Cryst. F78, 88. doi: 10.1107/S2053230X2101339X
6. Kamimura et al. (2022) New Biotechnol. 68, 57. doi: 10.1016/j.nbt.2022.01.007
7. Kumano et al. (2021) PNAS 118, e2106580118 doi: 10.1073/pnas.2106580118 

View Abstract 1052

Poster Author

Miki Senda, High Energy Accelerator Research Organization (KEK) Tsukuba

Additional Author(s)

Atsushi Kawaguchi, Department of Infection Biology, Faculty of Medicine, University of Tsukuba Tsukuba, Ibaraki 
Toshiya Senda, Structural Bio Research Ctr Inst of Materials Structure , High Energy Accelerator Research Org Tsukuba

An autoregulation of DNA binding model of ZNF410 revealed by biophysical study of small-angle X-ray scattering

ZNF410 is a unique and remarkable transcription factor in that it recognizes a 15-base pair DNA element but has only one single target gene in the mammalian genome in erythroid cells. ZNF410 is composed of uncharacterized N- and C-terminal domains with a tandem array of ordered five zinc fingers (ZFs). Unexpectedly, full-length ZNF410 has reduced DNA binding affinity, compared to that of isolated DNA binding ZF array. AlphaFold predicts a partially folded N-terminal subdomain including a 30-residue long helix and its proceeding hairpin loop, which is rich in acidic (aspartate/glutamate) and serine/threonine residues. The hairpin loop is placed into the DNA binding interface of the ZF array. In solution, ZNF410 is a monomer and binds to DNA in 1:1 stoichiometry. Surprisingly, the single best-fit model from the experimental small-angle X-ray scattering profile, in the absence of DNA, is the original AlphaFold model with the N-terminal long-helix and the hairpin loop occupying the ZF DNA binding interface. Upon the DNA binding, the hairpin loop must be displaced. By using a combination of biophysical, biochemical, bioinformatic and artificial intelligence-based AlphaFold approaches, we suggest that the hairpin loop might mimic the structure and electrostatics of DNA, and provides an additional mechanism, in supplementary to the sequence specificity, to regulate the DNA binding of ZNF410. 

View Abstract 1269

Poster Author

Gundeep Kaur, MD ANDERSON CANCER CENTER HOUSTON, TX 

Additional Author(s)

Ren Ren, MD Anderson
Michal Hammel, LBNL Berkeley, CA 
John Horton, MD Anderson Cancer Center of The University of Texas Houston, TX 
Robert Bluementhal, The University of Toledo College of Medicine and Life Sciences Toledo, OH 
Xing Zhang, The University of Texas -M.D. Anderson Cancer Center Houston, TX 
Xiaodong Cheng, The University of Texas -M.D. Anderson Cancer Center Houston, TX 

Analysis of Primary Citations (References) of PDB Deposits

Researchers worldwide from almost every biomedical discipline perform basic searches of the PDB, so the essential information in a PDB deposit must be as informative as possible. On a larger scale, inaccurate or misleading metadata can skew data mining efforts. The title and keywords of PDB deposits may play an essential role in the data mining of the PDB. The primary citation (reference) title may help in such a search, yet many deposits have notable discrepancies between the structure title and the primary reference title. Moreover, we have observed that the fraction of deposits with the status "To be published" has grown in recent years. We also analyze the similarity of titles, the number of citations for various classes of structures, and the primary reference keywords. Finally, the information about crystallization conditions is compared between PDB and the methods section from the primary citation. Several noteworthy examples are presented. 

View Abstract 989

Poster Author

Joanna Lenkiewicz, University of Virginia Charlottesville, VA 

Additional Author(s)

Michal Gucwa, University of Virginia Charlottesville, VA 
David Cooper, University of Virginia Charlottesville, VA 
Wladek Minor, University of Virginia Charlottesville, VA 

Arrangement of Molecules for Photoreactions in Layered Peptide Crystals

"Solid-state" or "solvent-free" organic synthesis is a promising alternative to the traditional "wet" synthetic techniques [1-2]. The solid state / solvent free methods generally comply with the principles of Green Chemistry [3] and therefore can help make the chemical industry more sustainable and therefore better for the future. However the main problem to be solved in the solid state organic synthesis is the creation of suitable mutual arrangements of reactant molecules for a desired reaction to occur. The problem was emphasized and studied by Schmidt and coworkers who introduced so-called topochemical principles for the case of [2+2] photodimerization and used the term "crystal engineering" to foresee the design of crystals making a target solid state reaction possible [4].
Short hydrophobic peptides may form microporous frameworks with persistent H-bonded layers and interlayer cavity space [5-9]. These crystal frameworks may be suitable for accommodation and appropriate arrangement of reactant molecules. Our studies have focused on creating and utilizing the peptide crystalline materials for the solid state reactions. Both the peptide molecules themselves [10] and reactant molecules accommodated in the cocrystals [11] can react in the solid state showing a promise that the peptide frameworks may become useful in the design of new solid state synthetic methods.
A range of potential reactant molecules have been screened for the formation of cocrystals with several hydrophobic dipeptides. Crotonic acid and dimethylacrylic acid were identified as the most interesting as they displayed photoreactivity in both their pure crystalline state and in cocrystals. Remarkably, the reaction pathways, product yields, and the kinetics of the reactions were different in the cocrystals. In addition, the course of the reactions depended on the irradiation wavelength used and the size of the crystals in the sample. Overall, our results show that the solid state photoreactions in peptide crystals can be controlled, although further studies are needed to understand how various conditions of the experimental setup affect the photoreaction outcome.

References
[1] Tanaka, K & Toda, F. (2000). Chem. Rev. 100, 1025-1074.
[2] Kaupp, G. (2005). Top. Curr. Chem. 254, 95–183.
[3] Tundo, P., Anastas, P., Black, D.S., Breen, J., Collins, T., Memoli, S., Miyamoto, J., Polyakoff, M. & Tumas, W. (2000). Pure Appl. Chem. 72, 1207-1228.
[4] Schmidt, G.N.J. (1971). Pure Appl. Chem. 27, 647–678.
[5] Soldatov, D.V. (2008). Nanoporous Materials, Vol. edited by A. Sayari & M. Jaroniec, pp. 213-224. N.J.: World Scientific.
[6] Burchell, T.J., Soldatov, D.V., Enright, G.D. & Ripmeester, J.A. (2007). CrystEngComm 9, 922-929.
[7] Burchell, T.J., Soldatov, D.V. & Ripmeester, J.A. (2008). J. Struct. Chem. 49, 188-191.
[8] Soldatov, D.V. & Smith, A.J. (2018). Acta Cryst. A74, a367.
[9] Soldatov, D.V., Smith, A.J. & Ali, F.I. (2019). Acta Cryst. A75, a216.
[10] Smith, A.S., Ali, F.I. & Soldatov, D.V (2014). CrystEngComm 16, 7196-7208.
[11] Esmaeili, M., Paget, B. & Soldatov, D. (2021). Acta Cryst. A77, a290. 

View Abstract 1300

Poster Author

Mehdi Esmaeili, University of Guelph Guelph, ON 

Additional Author

Dmitriy V Soldatov, Department of Chemistry, University of Guelph Guelph, Ontario 

Conformational dynamics in modular enzyme systems studied by SAXS and cryo-EM

Although proteins are often depicted by a single conformation or set of discrete conformations, they are constantly in motion. In solution, proteins adopt ensembles of conformations as described by a continuous thermodynamic landscape. Some, including those involved in the biosynthesis of small molecules, require large-scale domain rearrangements during turnover, and thus a high degree of flexibility is intrinsic to these systems. These types of systems are especially challenging to study, and a multi-faceted approach is required to describe how their dynamics relate to their function. In this talk, I will discuss how information from high-resolution techniques and computational modeling is most powerful when combined with solution techniques, such as small-angle X-ray scattering (SAXS), that measures the full solution conformational ensemble. I will show how structural data from AlphaFold2 predictions, fragment crystal structures, and single-particle cryo-electron microscopy (cryo-EM) can be effectively used concurrently with direct measurements of protein conformational ensembles by SAXS to capture the full extent of motions in a flexible, multi-domain enzyme. 

View Abstract 1154

Poster Author

Maxwell Watkins Princeton, NJ 

Additional Author

Nozomi Ando, Cornell University Ithaca, NY 

Cryo-EM Sample Preparation of native Myosin Filament from Striated Muscle

The detailed structure of filaments formed from myosin II is poorly understood. In so far as it is known, vertebrate thick filaments follow a single structural model; invertebrate thick filaments show high variability between species and between muscles of the same species. To date, high resolution structures have been reported only for myosin filaments isolated from invertebrates and no structures have been reported for myosin filaments in which one or more thick filament proteins have been mutated. A structure of thick filaments from vertebrates would be the most useful for understanding the effects of mutations on muscle function in humans. Despite numerous cryo-EM advances, structures of vertebrate thick filaments are limited to low resolution negatively stained specimens. The main problem seems to arise from difficulties in preserving the apparently more fragile vertebrate thick filaments using techniques that work well for invertebrates. Here, we report progress in the cryo-EM sample preparation of thick filaments which preserves the structure of both backbone and myosin heads. We are using rabbit psoas myofibrils as the model system to utilize the sensitivity of the ordered head arrangement to biochemical conditions. The approach uses an overnight incubation of myofibrils in a calcium-free relaxing buffer plus 1% Triton X-100 and calcium-insensitive gelsolin to remove as much of the sarcomeric actin as possible. All subsequent steps incorporate calcium-insensitive gelsolin in the buffers. Calpain-digested myofibrils were sheared by pulling the suspension through a 26-gauge needle with a syringe only 5 times to release filaments. Thick filaments were applied to a 1.2/1.3 Quantifoil grid and washed with relaxing buffer plus 0.5% glutaraldehyde to preserve ordered heads which otherwise disorder rapidly in vertebrate thick filament cryo-EM samples. This same method without the glutaraldehyde works well with invertebrate thick filaments. Supported by NIH. 

View Abstract 962

Poster Author

Hosna Rastegarpouyani, Florida State University Tallahassee, FL 

Additional Author(s)

Dianne W. Taylor, Institute of Molecular Biophysics, Florida State University Tallahassee, FL 
Fatemeh Abbasi Yeganeh, Institute of Molecular Biophysics, Florida State University Tallahassee, FL 
Alimohammad Hojjatian, Institute of Molecular Biophysics, Florida State University Tallahassee, FL 
Kenneth A. Taylor, Institute of Molecular Biophysics, Florida State University Tallahassee, FL 

Cryo-EM Structure of Pre-liganded NAIP5 reveals activation mechanism of NAIP/NLRC4 Inflammasome

Inflammasome is a cytosolic multiprotein complex formed in response to abnormal or pathogenic stimuli, and it initiates immune response to establish the innate immunity in mammals. Among the known inflammasomes, NAIP/NLRC4 inflammasome is specifically responsible for conferring immunity against various pathogenic bacteria. NAIP (nucleotide-binding domain, leucine rich repeat domain containing protein family (NLR family) apoptosis inhibitory protein) acts as a pathogen recognition receptor whereas NLRC4 (NLR family CARD containing protein 4) serves as a downstream molecule to undergo oligomerization to propagate and amplify the signal initiated by the activated NAIP. NAIP recognizes various bacterial ligands present in the cytosol of phagocytes, and the ligand bound NAIP further binds to NLRC4 and activates it by relieving its auto-inhibition. The activation of one NLRC4 molecule leads to exposure of its buried nucleation surface, which then serves as a binding site for another inactive NLRC4, to get activated. As a result of this process, a wheel-like NAIP/NLRC4 inflammasome composed of only one ligand bound NAIP and numerous NLRC4 molecules is formed. The CARD domain of NLRC4 in inflammasome interacts with CARD domain of pro-caspase-1 through CARD-CARD interactions and mediates the activation of pro-caspase-1. Caspase-1 activates the pro-inflammatory cytokines such as pro-IL-1β and pro-IL-18, which eventually induce the immune response that leads to pyroptosis (a form of programmed cell death) of the infected cell. Although, the information on overall inflammasome formation and its downstream signaling is known, no knowledge on how NAIP exits in inactive state in the absence of pathogenic ligand is available. The information on whether NAIP adopts the same auto-inhibited state as similar to NLRC4 or not, and how it is activated upon binding to ligand would provide valuable insights into the understanding of the activation of inflammasome formation and that further help design and develop the therapeutics for various inflammasome associated auto-immune diseases. Therefore, in this study, we determined the structure of pre-liganded mouse NAIP5 at the resolution of 3.3 Å by cryo-electron microscopy and found that it adopts an unprecedented wide-open conformation, with the nucleating surface fully exposed and accessible to recruit inactive NLRC4. Upon comparing it with the available liganded NAIP5 structures, we further found that the ligand binding could induces ~20° rotation of the winged helix domain (WHD) of NAIP5 and that triggers the conformational change of NLRC4 to propagate the inflammasome signal. Moreover, in our biochemical assays, we observed that the WHD loop of NAIP5 plays key roles in the inflammasome activation by relieving NLRC4 auto-inhibition, and stabilizing the formation of initial-encounter complex between liganded NAIP and active NLRC4. Overall, these data provide key insights into the understanding of the structural mechanisms of pre-liganded NAIP5, the process of NAIP5 activation, and the NAIP-dependent NLRC4 activation. 

View Abstract 1286

Poster Author

Bhaskar Paidimuddala, Oregon Health and Science University Portland, OR 

Additional Author(s)

Jianhao Cao, Oregon Health and Science University Portland, OR 
Qing Xie, Oregon Health and Science University Portland, OR 
Hao Wu, Harvard Medical School and Boston Children's Hospital
Liman Zhang, Oregon Health and Science University Portland, OR 

Crystallographic analysis of hydrazine coordination modes with dichlorotris(triphenylphosphine)ruthenium(II)

Hydrazine and its derivatives are important chemicals in the aerospace industry due to their application in rocket propulsion. The unique chemical properties of hydrazine, such as hypergolicity and its ability to rapidly decompose while passing over a catalytic surface (i.e. a transition metal), make it a versatile choice for many propulsion applications. Although the decomposition mechanisms of propellants have been a topic of interest since the field's inception, little work has focused on examining hydrazine interactions with catalytic surfaces in the monopropellant regime. Serving as a model substrate, the prominent catalyst, RuCl2(PPh3)3, was explored with various hydrazines. The products were subsequently studied using single crystal X-ray diffraction in order to gain insight into the coordination behavior and reactivity. This poster illustrates the bonding modes and complex chemistry of the materials from this investigation. 

View Abstract 1327

Poster Author

Kamran Ghiassi, Air Force Research Laboratory Edwards AFB, CA 

Additional Author

Nicolas Cena, HX5, LLC Edwards AFB, CA 

Dare to spin – well diffracting protein nanocrystals through on-vortex crystallisation

Protein crystallisation has been extensively studied for X-ray and neutron diffraction.
The hunt for optimised conditions yielding large single crystals has become the
staple of protein crystallography labs around the world. Recently new advances in
synchrotron beam lines and free electron X-ray lasers have opened the world of
smaller (<40 μm a side) crystals being used for structure determination.
Electron diffraction (3D ED / microED) brings altogether new challenges to the field.
Not only can it utilise even smaller crystals, it even requires one dimension to be sub
1 μm to allow the electron beam to pass through. Additionally, it benefits from the
other dimensions being significantly larger, since that allows for more protein to
participate in the diffraction measurement - which in turn improves signal strength
while reducing beam damage effects. Such ideal plate shaped crystals are often
created by focused ion beam milling of larger crystals, but this approach is time
consuming and requires specialised machinery.
We have found that leaving behind the careful and slow techniques of X-ray protein
crystallography and instead employing rapid crystallisation on a vortex mixer can
offer surprisingly good control over crystal size in the range required for 3D ED. The
optional addition of metal or PTFE beads to create shear forces capable of
fragmenting larger crystals provides a feedback loop that seeds new crystal growth
should the seed count be insufficient at first.
With an optimized protocol taking only minutes, we have succeeded in creating
solutions containing countless well diffracting crystal plates of a urate oxidase
(UOX), a ribonucleotide reductase R2 and two arginine kinases amongst others.
Diffraction data sets to 2 Å are commonly obtainable from these plates with both the
diffraction quality and resolution improving with decreasing plate thickness. In the
case of UOX the achievable resolution changed from 3 Å to 1.4 Å when switching to
the new method. 

View Abstract 1301

Poster Author

Gerhard Hofer, Stockholm University Graz

Additional Author(s)

Laura Calmanovici Pacoste, Stockholm University Stockholm
Lei Wang, Stockholm University Stockholm
Hongyi Xu, Department of Materials and Environmental Chemistry Stockholm
Xiaodong Zou, Stockholm University Stockholm, Sweden 

Defining the structural basis of ASCC2’s specificity for K63-linked polyubiquitin chains using 3D NMR analysis

DNA damage requires precise repair mechanisms that function in a timely manner to maintain genomic integrity. The ALKBH3-ASCC complex is a DNA alkylation damage repair complex that binds to K63-linked polyubiquitin chains which are assembled at damage sites. ASCC2, a subunit in the complex, selectively binds K63-linked polyubiquitin chains by interacting with two ubiquitins simultaneously. Although the residues on ASCC2 that interact with polyubiquitin chains are known, the specific interactions between ASCC2 and ubiquitin that impart specificity remains unclear. We are using 3D NMR to determine intermolecular distances which will guide our modeling of the ASCC2:K63Ub2 complex to elucidate how ASCC2's specificity for K63-linked polyubiquitin chains is achieved. 

View Abstract 1281

Poster Author

Rita Anoh, Mount St. Mary's University Emmitsburg, MD 

Additional Author

Patrick Lombardi, Mount St. Mary's University Emmitsburg, MD 

Elucidation of the Geometric Properties of the Pdu Microcompartment by Cryo-Electron Tomography

Approximately 20% of all bacteria contain giant supramolecular structures called bacterial microcompartments, or MCPs. MCPs are organelle-like structures that carry out various metabolic functions and are composed of a large proteinaceous shell that encapsulates enzymes and other proteins. These shells are roughly polyhedral in shape and consist of tessellating hexameric, pentameric and trimeric protein oligomers. Besides organizing an interior space inside the cell, the protein shell of an MCP plays a vital role in controlling diffusive transport of metabolic substrates and products from and to the cytosol. While some MCPs, such as carboxysomes, have been observed to have a nearly icosahedral shell, others have ambiguous geometries that have not been characterized in detail. Compared to carboxysomes, the 1,2-propanediol utilization (Pdu) microcompartment is a type of MCP that is more irregular in shape and size, with polyhedral properties that are not yet well-understood. Additionally, the hexagonally shaped protein oligomers that make up the shell of the Pdu MCP have distinct properties, with one side that is flat and one side that is dimpled. The native orientation of these shell proteins has not been confirmed in endogenous MCPs. We present ongoing efforts to use cryo-electron tomography (cryo-ET), combined with X-ray crystallographic data, to investigate these unsolved mysteries of the geometric shape of the Pdu MCP. 

View Abstract 1270

Poster Author

Kevin Cannon, UCLA Los Angeles, CA 

Additional Author(s)

Todd Yeates, University of California Los Angeles Los Angeles, CA 
Jessica Ochoa, California Institute of Technology Pasadena, CA 
Justin Miller, UCLA Los Angeles, CA 
James Evans, Pacific Northwest National Lab Richland, WA 
Trevor Moser, Pacific Northwest National Lab Richland, WA 

Enabling structure-based drug discovery for NUAK kinases

Novel (nua) kinase (NUAK1 and NUAK2) are serine/threonine kinases belonging to the AMPK family. Despite the rich history of structural biology around other AMPK kinases (AMPK, MARK, BRSK, MELK, etc.) there are no publicly available structures for NUAKs. NUAKs are implicated in the progression of neurodegenerative tauopathies and multiple cancers making them attractive targets for drug discovery. Development of robust structure-based drug discovery systems for NUAK kinases necessitated extensive protein characterization efforts. These included limited proteolysis, HDX-mass spectrometry, phosphorylation mapping, tool compound screening, and purification development to obtain homogenous and thermostabilized NUAK1 and NUAK2. Here we report the kinase domain crystal structures of phosphorylated human NUAK1 and dephosphorylated human NUAK2, both in complex with an orthosteric small molecule inhibitor. Structural comparisons between the NUAKs and the other AMPK family kinases are explored. 

View Abstract 1330

Poster Author

Robert Hayes, Merck Boston, MA 

Fixed-Targets for Serial Protein Crystallography at SwissFEL

X-ray free electron lasers (XFELs) have enabled the overcoming of limitations of classical crystallography by outrunning radiation damage with their highly coherent and brilliant femtosecond pulses. As a consequence of crystal destruction, large numbers of new crystals must be introduced sequentially to the beam, establishing methods such as serial femtosecond crystallography (SFX) and time-resolved serial femtosecond crystallography (TR-SFX) [1]. Both SFX and TR-SFX have provided new modes of structural biology. However, the constant need to refresh crystals in the beam path is a constant challenge and improvements to sample delivery can greatly improve experimental outcomes.

Fixed-target sample delivery methods allow for a reduction of sample consumption and optimization of sample density without issues such as clogging. Fixed-targets also lend themselves to high throughput technologies and an increased ability to locate and position crystals. Silicon wafers are the most common fixed-target and offer an inert support for the immobilized crystals and a precise aperture array for rapid alignment strategies [2]. However, the silicon wafers are brittle, expensive and can give strong Si(111) reflections in misaligned [3, 4].

Here we present preliminary data on the fabrication of polymer-based fixed-targets being developed for TR-SFX at SwissMX, the new end station dedicated to fixed-target SFX at SwissFEL, Switzerland's X-ray free-electron laser. The polymer-based film provides low x-ray absorption and scattering background, high design flexibility and the potential mass-fabrication at low cost. Using silicon microfabrication and polymer replication technologies, we have designed inverted pyramidal shaped wells in membranes ranging from 25-50 µm in thickness. This design enables single crystals to funnel into predefined positions, optimizing the hit-rate of the probing X-ray beam.

[1] Chapman, Henry N., et al. "Femtosecond X-ray protein nanocrystallography." Nature 470.7332 (2011): 73-77.
[2] Sherrell, Darren A., et al. "A modular and compact portable mini-endstation for high-precision, high-speed fixed target serial crystallography at FEL and synchrotron sources." Journal of synchrotron radiation 22.6 (2015): 1372-1378.
[3] Cheng, Robert KY. "Towards an optimal sample delivery method for serial crystallography at XFEL." Crystals 10.3 (2020): 215.
[4] Martiel, Isabelle, Henrike M. Müller-Werkmeister, and Aina E. Cohen. "Strategies for sample delivery for femtosecond crystallography." Acta Crystallographica Section D: Structural Biology 75.2 (2019): 160-177. 

View Abstract 1293

Poster Author

Melissa Carrillo Chicago, IL 

Additional Author(s)

John Beale, Paul Scherrer institut
Celestino Padeste, Paul Scherrer Institute Villigen Aargau, AG 

In situ architecture of the human kinetochore visualized by cryo-electron tomography

Chromosome segregation during mitosis relies on a carefully coordinated interplay between the centromere, kinetochore, and spindle microtubules. Despite its importance, the architecture of this interface remains elusive. Here we combined in situ cryo-electron tomography and cryo-light microscopy to visualize the native architecture of the kinetochore-microtubule interface in human U2OS cells at different stages of mitosis. Our data reveal that upon microtubule binding, the centromere forms a pocket-like structure around kinetochore microtubules. This centromeric pocket contains sparsely distributed nucleosome chains and two morphologically-distinct fibrillar densities from the kinetochore that form both lateral and end-on attachments to the plus-ends of microtubules within the pocket. Our data shows that the curling of protofilaments is impeded by the thick end-on fibrils, suggesting a direct relationship between these fibrils and microtubule depolymerization. Our data thus suggest that the pocket configuration of centromere scaffolds a dynamic kinetochore-microtubule interface in which multiple interactions facilitate stable attachment to microtubule plus-ends that are continually switching between growing and shrinking states. 

View Abstract 1309

Poster Author

WEI ZHAO, California Institute of Technology Pasadena, CA 

Integrative Modeling of the ASCC2:K63Ub2 Complex to Better Understand DNA Alkylation Damage Repair

The ALKBH3-ASCC complex plays a vital role in repairing alkylation damage in DNA. ASCC2, a subunit of the ASCC complex, localizes the ALKBH3-ASCC complex by binding to K63-linked polyubiquitin chains assembled at alkylation damage sites. ASCC2 has been shown to bind K63-linked polyubiquitin chains with enhanced affinity compared to monoubiquitin or other types of polyubiquitin chains. The purpose of this study is to determine the structural basis for ASCC2's enhanced affinity for K63-linked polyubiquitin chains. To study this, models of the ASCC2:K63Ub2 complex will be generated using NMR, ITC, and SAXS experimental data to visualize the interaction interfaces and elucidate the basis for this specificity. A better understanding of the alkylation damage repair pathway will prove useful to studies working toward treating alkylation damage diseases, as well as utilizing alkylation damage as a type of chemotherapy. 

View Abstract 1262

Poster Author

Zachary Beck, Mount St. Mary's University Cumberland, MD 

Additional Author

Patrick Lombardi, Mount St. Mary's University Emmitsburg, MD 

KHNYN is a Zinc-Finger Antiviral Protein (ZAP) co-factor that degrades ZAP-bound RNA

The human genome is CpG-suppressed due to the methylation and subsequent deamination of cytosines to uracil in open chromatin, leading to the conversion of the complementary nucleotide to an adenine by DNA repair pathways. This leads to CpG-poor mRNA, which is used as a molecular marker by host cells to distinguish between self mRNA and CpG-enriched viral RNA during infection. A protein called Zinc finger Antiviral Protein (ZAP) binds CpG dinucleotides, leading to degradation of viral RNA1–3. However, ZAP has no intrinsic RNase activity. It is thought that this activity is provided by KHNYN, a ZAP co-factor4. KHNYN has three domains – a KH (K-Homology) domain with predicted RNA binding activity; an NYN (N4BP1, YacP-like Nuclease) domain with predicted RNase activity; and a C-terminal domain (CTD). However, how KHNYN and ZAP act together to promote RNA degradation is not understood and is the focus of these studies.
We are using biochemical and structural approaches to address the function of three KHNYN domains and to test whether KHNYN acts as an RNase for ZAP. Our biochemical and structural results show that the KH domain does not bind RNA. We also found that while the KHNYN NYN domain is a poor RNase on its own, when complexed with ZAP it rapidly hydrolyzes RNA irrespective of the CpG content. Finally, the KHNYN CTD is essential for binding to ZAP. Together, these results support a model where ZAP and KYNYN form a complex that facilitates the binding and degradation of single-stranded RNA.

Supported by NIH grant U54-AI150470. ZCY is supported by training grant NIH T32-GM132046 and JAB by fellowship NIH F32 AI160904.

References:
Gao et al. Science 297, 1703-1706 (2002).
Takata et al. Nature 550, 124-127 (2017).
Meagher et al. PNAS 116, 24303-24309 (2019).
Ficarelli et al., eLife, 8, e46767 (2019). 

View Abstract 1000

Poster Author

Zoe Yeoh, University of Michigan Ann Arbor, MI 

Additional Author(s)

Jennifer Meagher, University of Michigan
Chia-Yu Kang, Biophysics at Michigan, University of Michigan, Ann Arbor, MI, USA Ann Arbor, MI 
Jennifer Bohn, Laboratory of Retrovirology, The Rockefeller University, New York, NY, USA New York, NY 
Paul Bieniasz, The Rockefeller University - HHMi New York, NY 
Melanie Ohi, University of Michigan Ann Arbor, MI 
Janet Smith, Life Sciences Institute, University Of Michigan

Microgravity Crystallization and Neutron Diffraction of Perdeuterated Tryptophan Synthase

The ubiquitous cofactor, pyridoxal 5'-phosphate (PLP), is the metabolically active and phosphorylated form of vitamin B6 present in all forms of life. PLP-dependent enzymes are functionally diverse, catalyzing transamination, racemization, β- and γ-elimination, α-decarboxylation, replacement reactions, and phosphorylation. Due to their significance in metabolic pathways and amino acid synthesis, PLP-dependent enzymes are attractive targets for specific inhibitor design. Such developments require atomic-level structural studies to understand how PLP is modulated to perform specific chemistry. Because neutron diffraction provides the ability to directly visualize the position of hydrogens and assign protonation states, it is a favorable technique for studying PLP-dependent enzymes. The bottleneck of neutron diffraction, however, is the growth of large crystals (≥ 0.5 mm...3) to overcome the flow flux of neutron sources. The inclusions and high mosaicity of crystals of tryptophan synthase (TS), a Fold Type II PLP-dependent enzyme, are only amplified when increasing the size, resulting in poor diffraction quality. Microgravity crystallization provides the opportunity to grow large, well-ordered crystals by reducing gravity-driven convection currents that permit variable crystal feeding and impede crystal growth. We developed the Toledo Crystallization Box (TCB), a membrane-barrier capillary-dialysis device, to grow neutron diffraction quality crystals of perdeuterated TS in microgravity. Here, we present the design of the TCB and the results from Center for Advancement of Science in Space (CASIS) supported International Space Station (ISS) Missions Protein Crystal Growth (PCG)-8 and PCG-15. From perdeuterated TS crystals grown on the ISS, we were able to collect a 2.1 Å neutron diffraction data set and solve the joint X-ray/neutron structure, revealing key protonation states important for understanding the mechanism of TS. 

View Abstract 1294

Poster Author

Victoria Drago, Oak Ridge National Laboratory Oak Ridge, TN 

Additional Author(s)

Timothy Mueser, Dept of Chemistry and Biochemistry, Univ. Of Toledo
Andrey Kovalevsky, Oak Ridge National Lab Oak Ridge, TN 
Constance Schall, University of Toledo Toledo, OH 
Juliette Devos, Life Sciences Group, Partnership for Structural Biology (PSB), Institut Laue-Langevin Grenoble
Matthew Blakeley, Large-Scale Structures Group, Institut Laue-Langevin Grenoble
Trevor Forsyth, Faculty of Medicine, Lund University, and LINXS Institute for Advanced Neutron and X-ray Science Scheelevägen

New Opportunities for Structural Biology Research at LCLS and SSRL

Serial femtosecond crystallography (SFX) is an emerging method that expands the structural information accessible from very small or very radiation sensitive macromolecular crystals. Utilizing extremely bright, short-time-scale X-ray pulses produced by an X-ray free electron laser (XFEL), this method exploits a 'diffraction before destruction' phenomenon where a still diffraction image is produced by a single X-ray pulse before significant radiation induced electronic and atomic rearrangements occur within the crystal. Furthermore, methods originally developed for serial diffraction experiments at XFELs, are proving valuable at synchrotron sources to study protein dynamics.
Similarities in instrumentation, existing and new sample delivery systems, and software environments form the foundation of a synergistic relationship between micro-focus beam line 12-1 at the SSRL synchrotron and the Macromolecular Femtosecond Crystallography (MFX) instrument at the LCLS XFEL. General user facilities for SFX at MFX include equipment for liquid-crystal injector-based sample delivery and a goniometer-based setup which supports fully automated sample exchange and data collection at room temperature and controlled humidity or at cryogenic conditions. The goniometer setup provides a suite of efficient automated experimental strategies tailored to handle a variety of sample requirements, crystal sizes and experimental goals. These developments coupled with improvements in data processing algorithms make it possible to derive high resolution crystal structures using only 100 to 1000 still diffraction images. Advanced capabilities to support serial and time-resolved crystallography are available at the SLAC laboratories and SSRL beam lines including crystal injectors, equipment for single crystal UV-Visible Absorption Spectroscopy (UV-Vis AS) and anaerobic setups for crystal growth, characterization and mounting. 

View Abstract 1290

Poster Author

Darya Marchany-Rivera, SLAC/SSRL-SMB Menlo Park, CA 

Oxidation State Change of Cerium-based UiO-66 Architecture

Over the past decade, the UiO-66 series, constructed from M6O4(OH)4 nodes (M = Zr, Hf, Th, Ti, Ce) and ditopic carboxylate linkers have aroused attention because of their great stability and diverse applications in the field of metal-organic frameworks (MOFs). Researchers discovered that not all the M-UiO-66s possess similar stimulus-responsive behaviors. The process of photo-excitation via Ligand-to-metal charge transfer (LMCT) is favorable only in the Ce-based UiO-66 crystal. Computational scientists proposed that LMCT is promoted from the empty 4f orbitals of Ce4+.
However, single-crystal X-ray diffraction and Thermogravimetric Analysis (TGA) data revealed that a significant number of missing ligands is present in our Ce-UiO-66 crystals, suggesting that the Ce6 node is likely composed of Ce4+ and Ce3+, instead of pure Ce4+. This result indicates that the performance of photo catalysis using Ce-UiO-66 will be unreliable if the oxidation state of Ce-UiO-66 is incomplete. In this work, we present a method to confirm the oxidation state using single-crystal X-ray diffraction. As a consequence, single-crystal data demonstrated that the Ce-Ce bond length of the Ce6 cluster is shrinking with air exposure time. This finding provides a facile way to rapidly judge the oxidation state of Ce-UiO-66 prior to catalysis use. 

View Abstract 1288

Poster Author

Ying-Pin Chen, ChemMatCARS Darien, IL 

Additional Author(s)

Tieyan Chang, The University of Chicago Lemont, IL 
Yu-Sheng Chen, University of Chicago Limont, IL 

Radiolytic Damage in Small-Molecule 3D Electron Crystallography

Every electron diffraction experiment is fundamentally limited by radiation damage. Immediately as the crystal of interest is illuminated by the incident beam, a complex set of inelastic scattering events initiates a cascade of radiolytic reactions within the sample, breaking chemical bonds and ultimately destroying the structural integrity of the crystal lattice. In 3D electron crystallography, an irradiated specimen is unidirectionally rotated within a transmission electron microscope while reciprocal space is periodically sampled in regular intervals, generating a tomographic series of diffraction patterns. Here we analyze a series of diffraction datasets acquired from repeated, consecutive sampling of single nanocrystals formed by organic and organometallic compounds. These species represent groups of small-molecule structures featuring site-specific modifications to an otherwise conserved scaffold. Our results indicate that chemically inspired substitutions can exert a significant effect on either accelerating or arresting the onset and progression of radiolytic damage, thus diminishing or enhancing the dose tolerance of specific crystalline specimens. Motifs explored include loss of aromaticity and removal of heavier atoms with relatively favorable elastic-to-inelastic cross-section ratios. 

View Abstract 1307

Poster Author

Ambarneil Saha, University of California, Los Angeles Los Angeles, CA 

Additional Author(s)

Niko Vlahakis, University of California, Los Angeles Los Angeles
Jose Rodriguez, UCLA Los angeles, CA 

Structural and Enzymatic Comparison of Faecalibacterium prausnitzii GH31 α-glycosidases

The gut microbiome is home to thousands of species of bacteria, that are essential for human digestion, immunity, and physiology. Faecalibacterium prausnitzii makes up about 5% of a healthy human gut microbiome and a lower abundance of this bacterium has been found in patients with IBD and Crohn's disease. Among an extensive repertoire of carbohydrate active enzymes, F. prausnitzii has 2 GH31 α-glycosidases, which are from the same family as Sucrase-Isomaltase and Maltase-Glucoamylase, human digestive enzymes with overlapping and distinguishing substrate specificities. This project aims to characterize the substrate specificity and preference of F. prausnitzii GH31 α-glycosidases to better understand the structural features of GH31 enzymes and the biological capabilities of these bacteria. AlphaFoldV2.1.0 was used to create computational models of F. prausnitzii α-glycosidases, and the substrate specificity and kinetics parameters are reported. Structurally, these α-glycosidases have the same identified conserved N-terminal and (β/α)8 barrel domains, but FpAG1 has an additional conserved domain of unknown function at the C-terminus which is not found in the FpAG2 structure. Both FpAG1 and FpAG2 have α-glucosidase and oligo-1,6-glucosidase activity. The comparative kinetic studies show that FpAG1 has a greater preference for α-1,6 glycosidic linkages, and FpAG2 has a greater preference for α-1,4 glycosidic linkages. Gaining insight on the GH31 α-glycosidases as a component of F. prausnitzii metabolism can further our understanding of this community in the human gut microbiome. 

View Abstract 1194

Poster Author

Anna Jewczynko, University of Waterloo Waterloo, ON 

Additional Author

David Rose, Dept of Biology, Univ of Waterloo Waterloo

Structural Dynamics of Non-Ribosomal Peptide Synthetases

Non-ribosomal peptide synthetases (NRPS) are multi-domain modular enzymes that catalyze the synthesis of small molecules using amino acids or non-proteinogenic amino acid substrates. Each module consists of 3 core domains which are condensation, adenylation, and the peptidyl carrier protein. With the exception of the terminating module containing an additional thioesterase domain which is responsible for releasing the small molecule from the assembly line. The adenylation domain serves to activate the loaded substrate which is then transported by the peptidyl carrier protein to either the condensation or other catalytic domains which serve to modify the substrate. This occurs in an assembly line process until a functional molecule is formed and completely constructed. The structures of each domain have been isolated and solved which characterized their catalytic activity. Multidomain structures have also been solved in hopes of elucidating the functional interaction between the multiple domains. Although structures of each domain and multidomain complexes exist, there is no tangible evidence for how these modular enzymes are dynamically functioning. Using small angle x-ray scattering (SAXS) on these highly dynamic modules will allow us to see differences in the scattering profile of the varying conformational states that they can adopt. Then using DENSS, a standalone software which calculates electron density from scattering profiles, we will be able to visualize these changes. However, there is a need for a version of this software that will be able to construct an ensemble of states that will be present in SAXS data which I will develop. Coupling this with the crystal structures of the complex we are performing SAXS with will allow us to clearly visualize conformational changes in theoretically high resolution and see exactly how these modular enzymes assemble these important peptide products. 

View Abstract 969

Poster Author

Jitendra Singh, University at Buffalo Buffalo, NY 

Additional Author

Thomas D. Grant, PhD Supervisor Buffalo, NY 

Structure determination of novel quantum spin liquid crystal at NSF’s ChemMatCARS

NSF's ChemMatCARS is a third-generation synchrotron user facility. It is located at Advanced Photon Source, Argonne National Laboratory, IL. NSF's ChemMatCARS is founded by National Science Foundation and operated by the University of Chicago. Advanced Crystallography Program at NSF's ChemMatCARS is dedicated to small-molecule crystallography. It provides high brilliance X-ray resources to explore novel structures of small molecular materials. Here we studied the structure of a new compound Cs14Cu4V16Cl8O48 crystallized from the quantum spin liquid system CsCl-V2O5-CuO. This system has potential applications such as quantum computing, and it's a candidate of high-temperature superconductive materials. Since the crystal is easy to deform and has large units, it will be benefited from the high-energy and high-resolution single crystal diffraction experiment performed at NSF's ChemMatCARS. A green single crystal Cs14Cu4V16Cl8O48 with size of 50×40×10 μm3 was mounted on the tip of fiber. Single crystal diffraction experiment was performed with Pilatus 1M (CdTe) detector at 30 keV. We succeeded in solving the crystal structure, which belongs to the monoclinic P2/n space group with lattice parameters of a=7.7669(3) Å, b=7.7610(3) Å, c=33.5028(12) Å, and β=96.608(1)°. Each V atom is coordinated with four oxygen atoms to form VO4 polyhedra, and each Cu is surrounded by five oxygen atoms. This quasi-two-dimensional structure shows interesting V4O12 rings, which are formed by VO4 polyhedra, and these rings are linked by CuO5

View Abstract 1297

Poster Author

Tieyan Chang, The University of Chicago Lemont, IL 

Additional Author(s)

Ying-Pin Chen, ChemMatCARS Darien, IL 
Yu-Sheng Chen, University of Chicago Limont, IL 

Sub-100 Microsecond Time Resolved SAXS at BioCAT

Time resolved SAXS (TR-SAXS) allows the measurement of kinetic intermediates after an initiating event. The BioCAT beamline (Sector 18) at the Advanced Photon Source uses chaotic and laminar flow microfluidic mixers to measure time ranges from ~80 us to 1.5 s. Recent advances include: new mixer designs to optimize accessible time ranges and sample consumption; improved microbeam focusing and mixer fabrication techniques to reduce parasitic scattering; improved positioning and exposure triggering for optimal reliability; and a new, easy to use GUI for controlling all aspects of the experiments. These advances have significantly improved data quality and ease of use. Time resolved experiments can now be done with as little as ~200 uL of sample at modest concentrations for slow (>1 ms) reactions, whereas ultra-fast time resolved measurements can be done with as little as ~1 mL of sample. The TR-SAXS program at BioCAT is open to general users. 

View Abstract 1116

Poster Author

Jesse Hopkins, BioCAT (Sector 18, APS), Illinois Institute of Technology Downers Grove, IL 

Additional Author(s)

Srinivas Chakravarthy, BioCAT (Sector 18, APS), Illinois Institute of Technology Argonne, IL 
Osman Bilsel, Microgradient Fluidics LLC Worcester, MA 

The enzyme EryM gives a helping hand in natural product biosynthesis

Polyketide synthases (PKS) and non-ribosomal peptide synthetases (NRPS) are essential enzymes responsible for the assembly of natural products, such as the common antibiotic erythromycin and the siderophore erythrochelin. Most bacterial PKS and NRPS biosynthetic pathways are encoded by gene located in a cluster. Interestingly, the enzyme EryM, encoded by a gene apart from any PKS or NRPS gene cluster, was identified to be crucial to producing erythrochelin and erythromycin.1, 2 EryM was first identified as a malonyl-CoA decarboxylase.3 Later studies led to the proposal that EryM is an acetyltransferase where the decarboxylation product, acetyl-CoA, is the acetyl group donor.4 The NRPS gene cluster for the erythrochelin pathway lacks an acetyltransferase to generate acetyl-hydroxyl-ornithine as the starter unit, thus EryM provides the missing link. Yet another study demonstrated that eryM is essential for erythromycin biosynthesis. The erythromycin PKS gene cluster lacks any gene for an enzyme to generate the starter unit, propionyl-CoA. Thus, EryM is proposed to form this starter unit by decarboxylation of methylmalonyl-CoA. However, it is still unknown how or whether EryM performs both malonyl- and methylmalonyl-CoA decarboxylation as well as acetyl transfer.

We used structural and biochemical techniques to probe the biological function of EryM. We found that EryM is composed of an N-terminal uncharacterized domain and a C-terminal domain containing the acetyltransferase active site. Purified EryM catalyzed decarboxylation of malonyl-CoA and subsequently used the intermediate acetyl-CoA as a substrate for acetyl transfer. These results suggest that EryM is a bifunctional enzyme that synthesizes the substrate for downstream natural product biosynthesis. 

View Abstract 1277

Poster Author

Yihua Li, University of Michigan Ann Arbor, MI 

Additional Author(s)

Xunkun Liu, Univerisity of Michigan Ann Arbor, MI 
Jacquelyn Roberts, Univerisity of Michigan Ann Arbor, MI 
Natalia Harris, Univerisity of Michigan Ann Arbor, MI 
Estefania Valdivia, Univerisity of Michigan Ann Arbor, MI 
Janet Smith, Life Sciences Institute, University Of Michigan

Time-resolved and Multi-temperature Crystallography of PEPCK Allows Observation of Previously Unobserved Dynamics and Structural States

Phosphoenolpyruvate carboxykinase (PEPCK), an essential enzyme that converts oxaloacetate to phosphoenolpyruvate and gates the gluconeogenesis pathway has recently been found to be upregulated in certain cancers.(1,2) Utilizing MMQX (Millisecond Mix-and-Quench Crystallography) we collected time-resolved crystallography data at timepoints of 40ms, 120ms, and 200ms. These datasets were able to capture PEPCK motions associated with substrate binding and catalysis as well as binding positions of the phosphoenolpyruvate and carbon dioxide products.(3) In addition to time-resolved crystallography, we also performed multi-temperature crystallography of PEPCK to better understand the energy landscape in steady-state conditions. These experiments captured the opening of the omega active site gating loop in PEPCK. Taken together, these experiments greatly improve our understanding of PEPCK's structural fluctuations.

1. Johnson, T. A. & Holyoak, T. The ω-loop lid domain of phosphoenolpyruvate carboxykinase is essential for catalytic function. Biochemistry 51, 9547–9559 (2012).
2. Leithner, K., Hrzenjak, A., Trötzmüller, M., Moustafa, T., Köfeler, H. C., Wohlkoenig, C., Stacher, E., Lindenmann, J., Harris, A. L., Olschewski, A. & Olschewski, H. PCK2 activation mediates an adaptive response to glucose depletion in lung cancer. Oncogene 2015 348 34, 1044–1050 (2014).
3. Clinger, J. A., Moreau, D. W., McLeod, M. J., Holyoak, T. & Thorne, R. E. Millisecond mix-and-quench crystallography (MMQX) enables time-resolved studies of PEPCK with remote data collection. IUCrJ 8, 784–792 (2021). 

View Abstract 1245

Poster Author

Jonathan Clinger, Cornell University Ithaca, NY 

Additional Author(s)

Matthew McLeod, Cornell University Ithaca, NY 
Robert Thorne, MiTeGen, LLC Ithaca, NY 

Towards solving the hydrogenase maturation mystery

We generate most of the energy from fossil fuels. Almost all fossil fuel energy sources generate harmful CO2. Hydrogen can be one of the main sources of fuel energy in the future. The hydrogenase enzyme is involved in the formation and degradation of hydrogen in various biochemical reactions and it can be used as an efficient source to generate hydrogen. The hydrogenase uses a special kind of iron-sulfur cluster (called the H-cluster) to generate hydrogen and this H-cluster is made by three maturase enzymes: HydF (~46 KDa), HydE (40KDa), and HydG (~55KDa). The missing link of the mechanism is the formation of the H-cluster (maturation process). Structural data on the maturation process is lacking which may be required to differentiate between the different mechanism hypotheses. First, we did chemical cross-linking mass spectrometry (CXMS) to determine interactions between different combinations of maturase enzymes. CXMS data suggests the three-way interaction between maturation enzymes at the same time. CXMS data is backed up with the SEC data which suggests the ~175KDa complex containing dimeric HydF, monomeric HydG, and HydE. We have an initial negative stain low-resolution structure of the maturation complex and a future step will be to find a high-resolution cryo-EM structure of the maturase complex. 

View Abstract 1127

Poster Author

Parag Gajjar, Brigham Young University Provo, UT 

Additional Author(s)

Miles Callahan, Brigham Young University Provo, UT 
Maria Pedroza, Brigham Young University Provo, UT 
Nathan Redd, Brigham Young University Provo, UT 
Celeste Litchfield, Brigham Young University Salt Lake City, UT 
Supeshala Sarath Nawarathnage
Derick Bunn, Brigham Young University Provo, UT 
Sara Soleimani, Brigham Young University Provo, UT 
Tobin Smith, Brigham Young University Provo, UT 
James Moody

Two riboswitches that share a common ligand-binding fold show dramatic differences in the ability to accommodate mutations

Riboswitches are naturally occurring, structured RNAs that directly "sense" the levels of cellular metabolites to regulate downstream genes. PreQ1-III (class III) riboswitches sense the metabolite preQ1 (7-deaza-7-aminomethyl-guanine) to control translation of a downstream transporter required for preQ1 salvaging. Most preQ1-III riboswitches belong to the Ruminococaceae family of commensal gut bacteria, which produces the hypermodified nucleobase queuosine that is necessary for genetic decoding by tRNAs with GUN anticodons. We showed previously that despite a distinctive overall HLout pseudoknot fold, the binding pocket of the preQ1-III riboswitch from Faecalibacterium prausnitzii (Fpr) shares ten identical nucleotides with the phylogenetically distinct preQ1-II (class II) riboswitches.1,2 Notably, preQ1-II riboswitches are present in several human pathogens, such as Streptococcus pneumoniae (Spn) - a leading cause of bacterial meningitis in adults and children. Given the goal of targeting RNA motifs with therapeutics, we asked whether commensal and pathogenic riboswitches respond similarly to mutations in or around their binding pockets that could cause loss of gene regulation or antibacterial resistance. Specifically, we prepared A52G, A84G, Δ84, U8C/A8G mutations in the Fpr class III binding pocket based on homologous mutations in the preQ1-II riboswitch that proved detrimental to preQ1-binding and gene regulation.3 X-ray crystallographic analysis of each class III mutant revealed compensatory chemical networks in the binding pocket, including large conformational changes, that maintain ligand binding. Indeed, preQ1 binding analysis revealed that each mutation is tolerated better by preQ1-III riboswitches than preQ1-II riboswitches. Chemical probing of riboswitch flexibility in solution suggests base-pairing of mutants is consistent with the crystal structures, although flexibility is altered compared to the wildtype sequence. A take home message is that the context of the global RNA fold impacts mutation tolerance. Accordingly, larger class III riboswitches from commensal bacteria appear more resilient to mutations compared to their smaller class II counterparts found in human pathogens.

References
1Liberman et al. Wedekind (2013) Structure of a class II preQ1 riboswitch reveals ligand recognition by a new fold. Nat. Chem. Biol. 9, 353-5.
2Liberman et al. Wedekind (2015) Structural analysis of a class III preQ1 riboswitch reveals an aptamer distant from a ribosome-binding site regulated by fast dynamics. Proc. Nat'l Acad. Sci. U.S.A 112, E3485.
3Dutta & Wedekind (2020) Nucleobase mutants of a bacterial preQ1-II riboswitch that uncouple metabolite sensing from gene regulation. J. Biol. Chem. 295, 2555-2567. 

View Abstract 938

Poster Author

Kumari Yoshita Srivastava, Department of Biophysics and Biochemistry, University of Rochester Medical Center Rochester, NY 

Additional Author(s)

Joseph Wedekind, Dept of Biochemistry & Biophysics, University of Rochester Rochester, NY 
Jermaine Jenkins, University of Rochester

Tying the knot in the tetrahydrofolate (THF) riboswitch: A molecular basis for gene regulation

Effective gene regulation by the THF riboswitch has shown dependencies on both ligand affinity and the kinetics of ligand association. Moreover, crystal structures of the aptamer in the ligand-bound conformation have revealed two ligand-binding sites. Knowledge of ligand-free aptamer conformations that are distinct from the ligand-bound form, therefore, can aid in determining the mechanism by which the absence or presence of ligand elicits conformational switching. We have determined a 1.9 Å-resolution crystal structure of the THF riboswitch aptamer domain in the absence of ligand that shows significant differences from previously reported apo and holo structures, particularly in the conformation of the P1 and P3 helices. The pseudoknot binding site 'unwinds' in the absence of ligand, causing rotation and misalignment of the gene-regulatory P1 helix with respect to P3. The second binding site, however, located at the three-way junction, is structurally conserved between apo and holo forms. This suggests cooperativity for the two binding sites, one which is preformed to elicit kinetic control, and the other which is directly involved in conformational switching through winding and unwinding of the pseudoknot.

Wilt, H.M. et al. Journal of Structural Biology 213 (2021) 107703. https://doi.org/10.1016/j.jsb.2021.107703 

View Abstract 1126

Poster Author

Jason Stagno, Center for Structural Biology, Center for Cancer Research, National Cancer Institute Frederick, MD 

Additional Author(s)

Haley Wilt, MCL Frederick, MD 
Ping Yu, National Cancer Institute Frederick, MD 
Kemin Tan, Argonne National Laboratory Argonne, IL 
Yun-Xing Wang, Structural Biophysics Laboratory, National Cancer Institute

Understanding substrate binding and delivery through the bi-chaperone Hsp104_Hsp70 supercomplex

Yeast Hsp104 is a member of the AAA+ family of Hsp100 protein disaggregases and plays critical roles in maintaining proteostatsis by collaborating with the Hsp70 molecular chaperone machinery to bind and solubilize protein amyloids and aggregates. Like other related AAA+ unfoldases, Hsp104 forms a hexameric ring complex comprised of two rings of nucleotide-binding AAA+ domains (NBD1 and NBD2) that bind the substrate polypeptides and power translocation through it's central channel. Previous cryo-EM structural work from our lab revealed that Hsp104 adopts distinct helical spiral conformations bound to substrate. From this work we proposed a model for stepwise translocation in which the protomers sequentially release and re-bind substrate through ATP hydrolysis at the spiral seam, alternating between hexameric states defined by 5 and 6 substrate-bound protomers. Here we sought to expand these structural studies and characterize the Hsp70 collaboration mechanism by determining structures of an Hsp104-Hsp70 supercomplex. Through extensive cryo-EM classification methods and multi-variability analysis, we determined three substrate-bound conformational states that provide additional support of the stepwise translocation mechanism. We identify low-resolution views of the Hsp70 NBD contacting the middle domains (MD) and stabilizing a remarkable bent conformation of the coiled-coil that positions Hsp70 adjacent to the Hsp104 N-terminal domains (NTDs) and channel entrance. Finally, though 3D classification and focus refinement we achieved structures of the NTD ring and channel entrance that reveal a potential rearrangement for substrate transfer to the AAA+ core. Together these structures reveal a network of allosteric interactions and conformational changes that enable substrate hand-off by Hsp70 in coordination with the Hsp104 translocation motor. 

View Abstract 1009

Poster Author

Alexandrea Rizo Washington, MI 

Additional Author

Daniel Southworth, UCSF San Francisco, CA 

Using Structural Biology to Elucidate Differences in Kinetic Inhibition Data Between Different Isozymes of PEPCK

Our lab studies the phosphoenolpyruvate carboxykinase (PEPCK) family of enzymes. PEPCK enzymes catalyze the conversion of oxaloacetic acid to phosphoenolpyruvate as one of the key reactions of gluconeogenesis. These enzymes can be categorized into different subclasses according to the phosphoryl donor. Two of these subclasses, the ATP- and GTP-dependent isozymes, while having low sequence similarity, share a global structural homology with the same three mobile active site loops and key active site binding residues.
The majority of our kinetic and structural research has been done on the rat cytosolic enzyme (rcPEPCK), as a proxy for the GTP-dependent class of PEPCKs. A complete characterization of kinetic parameters and inhibition constants has been determined for the rat isozyme, as well as extensive collection of both wildtype and mutant structures with inhibitors and other ligands bound. While the E. coli ATP-dependent enzyme was the first PEPCK to have its structure determined, a similar body of comparable structural and functional data is lacking. Recently we have undertaken studies to rectify this knowledge gap for the E. coli/ATP-dependent enzyme class which has yielded several interesting and contradicting functional differences when comparing kinetic parameters between the two classes of nucleotide-dependent enzymes.
Surprisingly, despite the residues comprising the OAA/PEP and M1 metal binding sites being completely conserved between the ATP- and GTP-dependent enzymes, the ATP-dependent E. coli enzyme exhibits some significant differences in its functional properties as well as its inhibition by known substrate/intermediate analogues. One of the initial differences observed for the E. coli enzyme was the increase in catalytic activity, appearing to be significantly more dynamic than its GTP-dependent counterpart. In addition, the ability of the E. coli enzyme to carry out two alternative chemistries not catalyzed by the GTP-dependent enzyme is also observed. Based upon these functional differences, we present structural data to that are consistent with the idea that the functional differences between the two enzyme classes (as seen in the kinetic data) are due to differences in the energetic barriers for conformational changes that are required for chemical function rather than differences in enzyme ligand/substrate interactions. Using structural data in addition to enzyme kinetics can help characterize and compare minute differences between isozymes belonging to the same enzyme family. 

View Abstract 949

Poster Author

Sarah Barwell, University of Waterloo Waterloo, ON 

Additional Author

Todd Holyoak, Biology Dept, Univ of Waterloo

UV related cataract formation: insights from serial synchrotron crystallography

Cataracts are a leading cause of blindness, characterised by opacification of the eye lens, due to the aggregation of crystallin proteins. Human γ-D-Crystallin (HGD) is the most abundant monomeric crystallin, a long-lived, small compactly folded protein of 173 amino acids. Localised to the lens fibre cells, which lack the cellular machinery for translation, HGD must therefore remain correctly folded and soluble for the entire human lifespan (Slingsby et al. 2013). Although highly stable repeated exposure to UV-radiation has been implicated in the aggregation of HGD and the subsequent formation of cataracts. UV radiation has been reported to destabilise HGD via two mechanisms; primary photodamage caused by cleavage of the indole ring leading to the conversion of tryptophan to kynurenine, partial unfolding and inevitably aggregation and secondary photodamage caused by the formation of reactive oxygen species (ROS) is thought to accelerate cataract formation through oxidation of surface thiol residues, leading to intermolecular disulphide bond formation (Craghill et al. 2004).
Glutathione (GSH) is present in high concentrations within the lens and acts as a ROS scavenger. GSH may prevent aggregation by reversing the oxidation of surface thiols, preventing intermolecular disulphide bond formation. Aging is associated with a reduction in GSH levels as the oxidised form GSSG increases alongside an increase in oxidised surface thiols. GSSG may be regenerated to GSH via formation of S-glutahionlyated cysteine, as is reported in γ-S and γ-C crystallins. UV radiation may result in reduction of disulphide contributing to the replenishment of GSH (Zetterberg et al. 2006).
Using serial crystallography (Schulz et al. 2022) and the readily crystallisable R36S mutant of HGD, and DTT as a proxy for GSH, we show that aged crystals of HGD accumulate covalent modifications on surface cysteines. Studies of oxidised crystals, both with and without UV irradiation revealed that UV irradiation disrupted the covalent modification of surface cysteines.

Schulz, E.C., Yorke, B.A., Pearson, A.R. and Mehrabi, P. (2022) Best practices for time-resolved serial synchrotron crystallography. Acta Crystallographica. Section D, Structural Biology 78 (Pt 1), 14–29.
Slingsby, C., Wistow, G.J. and Clark, A.R. (2013) Evolution of crystallins for a role in the vertebrate eye lens. Protein Science 22 (4), 367–380.
Zetterberg, M., Zhang, X., Taylor, A., Liu, B., Liang, J.J. and Shang, F. (2006) Glutathiolation enhances the degradation of gammaC-crystallin in lens and reticulocyte lysates, partially via the ubiquitin-proteasome pathway. Investigative Ophthalmology & Visual Science 47 (8), 3467–3473. 

View Abstract 937

Poster Author

Jake Hill, University of Bradford Bradford

Additional Author(s)

Yvonne Nyathi, University of Bradford Bradford
Briony Yorke, University of Bradford Bradford

X-Ray Crystallography as a Tool to Understand the Structure-Property Relationship in Metal-Organic Frameworks for the Synthesis of Desired Sensor Materials

Metal–organic framework (MOF)-based sensors for the detection of various analyte molecules has been a subject of absolute importance. However, most of these sensors rely on the turn-off (quenching) transduction response, while those reporting turn-on response are very rare. Here, we present two new MOF-based sensors, {[Zn2(oxdz)2(tpbn)]·14H2O}n (1) and {[Zn2(oxdz)2(tpxn)]·10H2O·2C2H5OH}n (2), via the self-assembly of Zn(II) metal ions, a fluorogenic oxdz2– linker, and bis(tridentate) ligands (tpbn and tpxn) under ambient conditions. Their formation from such a self-assembly process has been evaluated on the basis of the geometry around the five-coordinated Zn(II), preferential meridional binding of the bis(tridentate) ligands, and diverse binding of the carboxylate groups in oxdz2– studied using single-crystal X-ray diffraction. The bulk phase purity and the similarity of the bulk materials of 1 and 2 is studied through powder X-ray diffraction analysis. Although 1 and 2 are isostructural, a difference in the transduction mechanism for the sensing of acetylacetone in organic solvents (turn-on for 1 and turn-off for 2) is observed and can be attributed to the spacer in the bis(tridentate) ligands. We have demonstrated the competing effect of the nonradiative interactions and photoinduced electron transfer toward the sensing mechanism. The results are well-supported by the Fourier transform infrared spectroscopy study, intensity versus concentration plots, spectral overlap measurements, time-resolved fluorescence studies, and MM2 and density functional theory calculations. Furthermore, we have showcased the utilization of 1 for the sensing of trace amounts of water in organic solvents. 

View Abstract 1271

Poster Author

Alisha Gogia, New Mexico Highlands University Las Vegas, NM 

A Room temperature Polar and Weak-ferromagnetic Oxide with Low Dielectric Loss

Single-phase materials that are simultaneously ferroelectric and ferromagnetic at room temperature are promising for non-volatile random access memory devices. Perovskite BiFeO3 which crystallizes in the polar rhombohedral structure (R3c), is ferroelectric and antiferromagnetic at room temperature. Here, we report a family of perovskite oxides in the BiFeO3 – Bi2/3TiO3 – ATiO3 (where A2+ = Ca2+, Sr2+, Ba2+) ternary phase diagram that is polar as well as weak ferromagnetic. We achieved nearly pure A-site bismuth-based perovskite phase Bi0.9167R0.075Fe0.9Ti0.1O3 that crystallizes in the space group R3c similar to BiFeO3 as corroborated by powder X-ray and neutron diffraction analysis. Their polarity was confirmed by second harmonic generation (SHG) experiments. Room-temperature powder neutron diffraction confirms G-type antiferromagnetic ordering consistent with weak ferromagnetism that onsets at TN = 557 K. These perovskites show a low dielectric loss, and the electrical response is dominated by grain contributions below 723 K. 

View Abstract 973

Poster Author

Alicia Manjon Sanz, Oak Ridge National Laboratory Oak Ridge, TN 

Additional Author(s)

Nagamalleswari Katragadda, SRM University AP Amaravati, Andhra Pradesh
Pranab Mandal, SRM University AP Amaravati, Andhra Pradesh
Premakumar Yanda, School of Advanced Materials, Chemistry, and Physics of Materials Unit Bangalore
A Sundaresan, School of Advanced Materials, Chemistry, and Physics of Materials Unit Bangalore
S. D. Kaushik, UGC-DAE Consortium for Scientific Research Mumbai Centre Mumbai
Weiguo Zhang, University of Houston Houston, TX 
P. Shiv Halasyamani, University of Houston Houston, TX