Skip to main content

more options

Research Highlights 2017

Research Highlights 2016

Research Highlights 2015

Research Highlights 2014

Unusually deformable protein crystals

Application of high pressure to crystals of Snf7 has an unusual effect. Cryocooling protein crystals under pressure (HPC) can reduce damage due to the cooling process. Usually, HPC has negligible effect on the crystal structure. Occasionally, it causes a small change in packing of the molecules in the crystal. For crystals containing a lot of solvent, pressure may cause them to collapse, destroying their diffraction. HPC on crystals of Snf7, however, results in a large change in molecular packing without destroying the crystallinity. In fact, the pressure-cooled crystals diffract better than the normally cooled crystals, in spite of a 30% decrease in unit cell volume!

Snf7 is a component of ESCRT-III, which is one of the endosomal sorting complexes required for transport (ESCRTs). These complexes are membrane remodeling machineries that mediate diverse fundamental cellular processes, including the biogenesis of multivesicular bodies (MVBs) during receptor down-regulation, enveloped virus budding, cytokinesis, plasma membrane repair, nuclear pore complex assembly and nuclear envelope reformation. The job of ESCRTs is to package transmembrane proteins tagged with ubiquitin into vesicles that merge with the “late endosome”, creating an MVB that ultimately delivers cargos to the lysosome for destruction (Figure 1). Upstream ESCRT components (ESCRT-0, I & II) assemble into complexes which sort ubiquitinated cargo on the endosomal surface. In addition, ESCRT-II initiates the assembly of the ESCRT-III complex, which together with Vps4 is responsible for remodeling endosomal membranes.


Figure 1: The ESCRT pathway.

ESCRT-III is a dynamic hetero-polymer, composed of four core subunits (Vps20, Snf7/Vps32, Vps24 & Vps2) and four accessory subunits (Ist1, Did2/Vps46, Vps60 & Chm7). Although all ESCRT-III subunits share a common domain organization, each subunit appears to contribute a specific function. All ESCRT-III subunits are inactive monomers in the cytoplasm; when activated they assemble into spiraling polymers on endosomes to drive cargo sequestration, membrane invagination and constriction.

The group of scientists led by Prof. Scott Emr of Cornell University were able to crystallize and solve the structure of the core region (residues 12-150) of the most abundant ESCRT-III subunit, Snf7 (Snf7core), in its activated conformation. X-ray diffraction data were collected at the F1 beamline at CHESS. At ambient pressure Snf7core crystals diffracted to 2.4 Å at best; 2.7 Å was typical. HPC was used in an attempt to improve this resolution limit. A crystal cryocooled under the typical HPC pressure of 200 Mpa for 30 minutes diffracted out to 1.55Å. Surprisingly, this crystal showed a reduction of about 30% in unit cell volume, due to repacking of the layers of Snf7core molecules (Figure 2). Further HPC experiments on more Snf7core crystals confirmed the increase in diffraction resolution of pressure-cooled Snf7core crystals (most diffracted to better than 2Å), coupled with a large change in molecular packing. The unit cell dimensions varied as a function of pressure applied in the HPC process.

cellular_machinery cellular_machinery

Figure 2: Packing of Snf7 molecules (6 molecules are shown) in crystals cryocooled at ambient pressure (left) and at 200 MPa (right). In both forms, molecules are tightly packed within layers (shades of red or shades of green), but the stacking of the layers is quite different, as well as the orientation of the a4 helix (the short helix projecting away from the main axis of the molecule).

Unlike crystal structures of other ESCRT-III subunits (Vps24 & Ist1), which have been determined in their inactive conformations and adopt a canonical four a-helical core domain fold (closed conformation) (Figure 3A), the Snf7core contains only three a-helices packed into a highly elongated structure (open conformation) (Figure 3B). The a3 and a4 helices undergo large-scale structural rearrangements from the proposed autoinhibited ESCRT-III fold observed in the structures of Vps24 and Ist1. The open form contains an ~90Å long a-helix combining the a2 and a3 segments that were distinct a-helices in structures of Vps24 and Ist1. The orientation of the a4 helix varies substantially in the structures processed at different pressures, but the open conformation is maintained. The closed-to-open conformational change not only extends a cationic membrane-binding surface, but also exposes hydrophobic and electrostatic protein interacting surfaces for polymerization. This is the first crystal structure of an ESCRT-III subunit in its activated conformation. Using in vitro reconstitution and pulsed dipolar electron spin resonance spectroscopy techniques, the Emr group found that full-length Snf7 adopted the same active (open) conformation, and assembled into ~30Å periodic protofilaments, in a near-native lipid environment.


Figure 3: Crystallographically determined structures of (A)Vps24 and (B)Snf7core.


S. Tang, W.M. Henne, P.P. Borbat, N.J. Buchkovich, J.H. Freed, Y. Mao, J.C. Fromme & S.D. Emr, “Structural basis for activation, assembly and membrane binding of ESCRT-III Snf7 filaments”, eLife 2015;10.7554/eLife.12548.

Submitted by: Marian Szebenyi and Qingqiu Huang, MacCHESS, Cornell University 01/13/2016

BioSAXS works to uncover cellular machinery involved in regulation of DNA transcription

The biology of human diseases and disorders is highly complex. In many cases, despite a great deal of detailed structural knowledge, understanding mechanisms is still a long way off. The MLL3 (mixed lineage leukemia 3) protein, for example, is a member of the SET1 family of histone-modifying enzymes, which plays a critical role in regulating transcription of genetic information in humans. Misregulation of histone modification is associated with different cancers and developmental disorders. MLL1 is well studied and has been used to predict how MLL3 and other SET1 proteins function, but the validity of these predictions is uncertain – do all family members really work the same way?

The Cosgrove laboratory at SUNY Upstate Medical University has developed an experimental reconstitution system that has allowed them to dissect the molecular mechanisms of methylation of histone 3 at lysine 4 (H3K4) by MLL family complexes, addressing questions about complex formation, mechanism of lysine methylation, and inhibitor design. In a recent publication Stephen Shinsky and Michael Cosgrove delved into the unique role that one of the subunits, WD-40 repeat protein-5 (WDR5), plays in the methyltransferase activity of the MLL3 core complex [1]. They deployed solution small-angle x-ray scattering (BioSAXS) at the CHESS G1 experimental station to characterize the MLL3 core complex and substructures. Information about this complex is limited, as no high resolution structures of MLL3 or its complexes are currently available. In addition, although some cryo-EM reconstructions of human MLL1 core complex have been reported, there is currently no information about solution structures and dynamics of any SET1 family complex.


On The Cover: The mixed lineage leukemia 3 (MLL3) core complex catalyzes monomethylation at histone H3 lysine 4 (H3K4). Unlike other MLL family core complexes, the WDR5 subunit of the MLL3 core complex inhibits H3K4 methylation activity. The cover shows the BioSAXS solution structure of the MLL3 core complex assembled with WDR5 (purple) superimposed with that of the core complex without WDR5 (blue). These structures give insight into the spatial location of WDR5 (orange) within the MLL3 core complex and its mechanism of inhibition (cover image by Victor Garcia).

The beautiful reconstructions of the MLL3 core complex and the complex with WDR5 omitted, aligned with each other and with the WDR5 crystal structure, appear on the cover of the Journal of Biological Chemistry this past month (see figure). The structures show that addition of WDR5 has little effect on the conformation of the rest of the complex. This is consistent with biochemical studies showing that MLL3 and other components of the core complex can assemble into a functional system in the absence of WDR5, and that adding WDR5 actually inhibits the activity of the complex. This situation is quite different from that in MLL1, where WDR5 is required for assembly of an active complex. Hence different members of the SET1 family use homologous components in distinct ways, and need to be analyzed individually.

In a bigger context, the Cosgrove group focuses on understanding the molecular mechanisms of “epigenetic gene regulation”, a field of biology that strives to understand how cells inherit information that is independent of changes in DNA sequence. The importance of this topic is shown by the fact that mutations in genes coding for MLL family enzymes are responsible for human developmental disorders such as Kabuki Syndrome, Weidman-Steiner Syndrome, and Kleefstra Syndrome. MLL family enzymes are also found to be mutated in cancers such as non-Hodgkin lymphomas, pediatric medulloblastomas, lung, renal, and prostate carcinomas. This work presents new information that will facilitate future studies to identify the cellular machinery involved in regulating MLL3 core complex assembly and the molecular mechanisms by which WDR5 reduces MLL3 methyltransferase activity. In addition, this work adds to our understanding of how additional regulatory mechanisms can influence the methyltransferase activity of important SET1 family members.


[1] Shinsky SA, Cosgrove MS, “Unique Role of the WD-40 Repeat Protein 5 (WDR5) Subunit within the Mixed Lineage Leukemia 3 (MLL3) Histone Methyltransferase Complex”. J Biol Chem. 2015 Oct 23; 290 (43):25819-33. doi: 10.1074/jbc.M115.684142. Epub 2015 Aug 31.

Submitted by: Ernest Fontes, CHESS and Marian Szebenyi, MacCHESS, Cornell University 11/13/2015

MacCHESS runs ACA course on small angle scattering


Following the successful BioSAS training workshop held at the 2013 American Crystallographic Association annual meeting, the ACA approved a new proposal from MacCHESS for a small-angle scattering workshop bringing together experts in both biology and soft matter. While the foundational physics of scattering is common to all application areas, individual fields have diverged over the years to develop many specialized tools appropriate to the type of matter under investigation. As science advances, however, areas like structural biology, materials science, and engineering have greater overlap. We organized a dual-track workshop this year aimed at getting soft-matter scientists and biologists in one room to promote exchange.

The workshop was held at the 2015 Philadelphia ACA meeting and drew 46 participants. All students were provided with lunch, a set of printed course notes, and a memory stick containing course notes, tutorial data, and other useful material. The two groups shared a morning session on the common basic physics of scattering followed by divided afternoon sessions devoted to practical specifics of the two fields.

Jan Ilavsky (APS) led the morning with a lecture entitled “Biology Meets Soft Matter.” Angela Criswell (Rigaku) gave the first tutorial of the day on basic data processing. Sergio Rodrigues (Xenocs) and Peter Worsch (Anton Paar) spoke about laboratory sources.

The group divided after lunch for field-specific lectures and tutorials. We were fortunate once again to be joined by Jill Trewhella (U. Sydney) who gave two lectures: one on biological specifics of SAS and the other on essential publishing guidelines and good research practice (“Publishing Data: what you should know”). Richard Gillilan (MacCHESS, Cornell) covered P(r) functions and envelope calculations. Kushol Gupta (UPENN) discussed sample preparation, with some introduction to the popular new method SAXS-SEC-MALS.

To provide some instruction on advanced data processing, Susan Krueger and Joseph Curtis (NIST) conducted a SASSIE software tutorial using the new online CCP-SAS platform. This easy-to-use online system bypasses the need for students to install software on their laptops.

Jan Ilavsky (APS) and Kevin Yager (BNL) led the afternoon soft-matter session. Topics covered included physics basics, an introduction to grazing incidence methods, models and instrumentation. The track concluded with an in-depth hands-on tutorial of data processing using the Irena package.


A website containing software installation instructions, the course schedule, speaker information, and links to tutorial data and online tools will remain active after the course as a general resource for SAS students (http://meetings.chess.cornell.`edu/ACABioSAS).

Thanks go to Barbara Herrman and Nika Ablao for website design. Thanks also to Kathy Dedrick and Irina Kriksunov for administrative and logistical support. In addition to our speakers, the non-lecturing members of the organizing committee were Shuo Qian and Volker Urban (ORNL), Thomas Weiss (SSRL), Zhang Jiang (APS), and Andreas Keilbach (Anton Paar).

The workshop would not have been possible without the generous support and participation of our corporate sponsors: Anton Paar, Dectris, Rigaku, and Xenocs.

MacCHESS’s regularly-scheduled BioSAXS Essentials training workshops will continue in late Spring 2016. Be sure to check the CHESS webpage for the official announcement.

Submitted by: Richard Gillilan, MacCHESS, Cornell University 10/13/2015

MacCHESS post-doc applies pressure

TK Chua





MacCHESS post-doc TK Chua seeks to understand the operation of enzymes that operate on gases, such as carbonic anhydrase and nitric oxide synthase, by applying the pressure-cryocooling technique to trap gas molecules in their active sites.

Pressure cryocooling, originally developed by Chae Un Kim and Sol Gruner (1) and now supported as a resource by MacCHESS, has seen most use as a means of reducing the damage caused to macromolecular crystals when they are cooled to 100 K. By pressurizing crystals with helium gas at about 2 kbar before and during cooling, it is often possible to maintain the original, pre-cooling, crystal quality. Moreover, it is usually not necessary to add cryoprotectants (such as glycerol, sucrose, or polyethylene glycol) to the crystal to prevent ice formation; the high pressure takes care of this. MacCHESS offers a pressure-cryocooling service to users whose crystals suffer damage during cryocooling, and a number of successes have been recorded.


Figure 1

In addition to improved cryocooling, application of pressure can change equilibria in a crystal, leading to the trapping of states that could not otherwise be observed. TK Chua, a post-doc from Singapore who joined MacCHESS in October 2014, is focusing on the trapping of gases in enzymes. In this process, helium is replaced by the gas to be trapped, and the pressure applied only needs to be high enough to ensure that the active site of the enzyme is fully occupied by the gas. The first system TK worked on, in collaboration with Rob McKenna from U. Florida and Mayank Aggarwal of Oak Ridge, was carbonic anhydrase (CA). CA is an enzyme that converts carbon dioxide to bicarbonate; in humans this acts to regulate the pH of the blood, and in industry a potential CO2 sequestrator. In a previous collaboration between McKenna and Kim, the structure of human CA II (hCA) loaded with CO2 was determined (2); see Figure 1. Recently, Aggarwal provided TK with a ß-CA from the bacterium Pseudomonas aeruginosa (bCA). TK grew crystals, pressurized them with CO2 at ~10 atmospheres, and cooled them to liquid nitrogen temperature under pressure. Then he collected data and solved the crystal structure. The placement of CO2 relative to the catalytic zinc ion and critical water molecules is quite similar in hCA and bCA, in spite of completely different active site structure in the human and bacterial enzymes (Figure 2). A paper comparing the structures in detail has been submitted.


Figure 2: CO2 in the active site of (Left) hCA II, and (Right) bCA. Residues are as labeled. Similar CO2 binding orientation relative to the zinc-bound solvent.

For his next project, TK is collaborating with Brian Crane of Cornell, attempting to trap NO in nitric oxide synthase (NOS). This system has not previously been studied using pressure cryocooling, and there is a complex set of factors to consider, including selection of an appropriate species for study, crystallization conditions, whether to add co-factors, etc. It's a work in progress. TK is also helping to upgrade the MacCHESS pressure-cryocooling apparatus, following the lead of the Carpentier group at ESRF (3); these upgrades should both extend the capabilities of the apparatus and make it easier to use.


[1] Kim, C. U.; Kapfer, R.; Gruner, S. M. Acta Crystallogr. D61, 881–890 (2005).

[2] Domsic, J. F.; Avvaru, B. S.; Kim, C. U.; Gruner, S. M.; Agbandje-McKenna, M.; Silverman, D. N.; McKenna, R. J. Biol. Chem. 283, 30766-71 (2008).

[3] Van der Linden, P.; Dobias, F.; Vitoux, H.; Kapp, U.; Jacobs, J.; McSweeney, S.; Mueller-Dieckmann, C.; Carpentier, P. J. Appl. Crystallogr. 47, 584-592 (2014).

Submitted by: Marian Szebenyi, MacCHESS, Cornell University 07/31/2015

Small Angle Scattering: Structural Biology and Soft Matter (WK.04)

2015 Annual Meeting of the American Crystallographic Association, July 25-29. Philadelphia, PA.


Organizer: Richard Gillian Cornell University.

Workshop sponsored, in part, by the following: Anton Paar, Dectris, Rigaku

Small angle x-ray solution scattering (SAXS) continues to experience dramatic growth within both the structural biology and soft matter communities. While there tends to be relatively little interaction between these communities historically, the two share essentially the same basic theoretical foundation as well as a number of tools and techniques. This workshop will bring together leading SAS experts in both areas to prepare students for successful experiments.

The morning portion of the dual-track/two-room workshop will be a joint session covering theory and practice common to both fields. After the joint session, the rooms will be divided and the two parallel sessions will cover specifics of the individual fields of soft matter and structural biology. In addition to synchrotron sources, this workshop is expected to have some content devoted to laboratory x-rays sources, but also particularly neutron sources and techniques. The workshop format will include lectures, and a selection of hands-on practical exercises. Students will be expected to bring laptops with appropriate pre-installed software as necessary. Prior to the workshop, a website (prepared by Cornell) will be configured containing installation instructions and software for each tutorial. Throughout the workshop, the emphasis will be on knowing how to judge data quality, what to do about problematic samples, and basic requirements for acceptable publication of first-time data. Students will also learn tips and tricks for home laboratory data collection and be introduced to the various national synchrotron BioSAXS beamlines and neutron sources. This workshop will also aim to educate students about the particular advantages of neutron scattering and the extra steps necessary to carry out a first SANS experiment. Selected advanced modelling techniques will also be covered.

Course Topics Basic Theory and Methods · Theory Essentials: understanding the scattering profile · What are all those plots? (Guinier, Kratky, & Porod plots) · Do Try This at Home: tips and tricks for lab-source data collection · SAXS and SANS: unique strengths and differences · National sources: how to get time, what to expect. Tutorials (both parallel sessions) · Basic SAX/SANS analysis methods (molecular weights, plots, envelopes) · Advanced modelling Biological Topics · Sample preparation and characterization · Assessing data quality, finding basic parameters · distance distributions and molecular envelopes · Docking, EOM and flexibility · Analysis of mixtures and separation · SANS contrast variation · Publishing data: what you should know Soft Matter Topics · Sample preparation and characterization · SAXS, SANS and USAXS basics (Guinier, Porod approximations) · Introduction to Grazing Incidence SAXS (GISAXS) · Beyond linearization plots; hierarchical structures · Overview of models for specific systems · Advanced modeling

All participants must pre-register for the workshop and submit the following fee with their meeting registration form:

  • Student or Post-doc - before May 31, $130, on or after June 1 $180
  • Academic others - before May 31, $130, on or after June 1 $180
  • Corporate - before May 31, $250, on or after June 1 $300

Workshop on fast framing detectors and time-resolved x-ray biology

Tiit Lukk

Organizer: Tiit Lukk Cornell University.

The second workshop at the 2015 CHESS Users’ Meeting was organized by MacCHESS scientist Tiit Lukk on the topic of “Fast framing detectors and their applications in time-resolved crystallography/BioSAXS.” The advent of third generation light sources that produce bright X-rays allow scientists to carry out experiments at speeds orders of magnitude faster than was previously possible. This workshop focused on how advances in X-ray detector development have chartered new unexplored avenues into biological sciences, making it possible, for example, to follow enzymatic reactions in real-time as they occur at near atomic resolution.

In the first session Kate Shanks (Gruner detector group, Cornell) spoke of the state of the art of X-ray detector development, highlighting how the field has evolved through time and where the field is headed. She gave an introduction to how fast-framing detectors have benefited the fields of BioSAXS and protein X-ray crystallography in recent years. Robert Thorne (Cornell) gave a presentation on the advances of room temperature structural biology. Thorne reported that 98% of all protein structures are now determined at T=100K, but that crystallographers are starting to realize the value of structural dynamics data that can go missing during sample vitrification (including side chain conformers or subdomain movements). He went on to present evidence for ways to mitigate radiation damage of protein crystals during X-ray diffraction experiments at temperatures between ambient and 100K. Some, but not all, of the damage may be outrun by use of very high dose rates and fast detectors.

Lois Pollack (Cornell) opened the second session, which was dedicated to research and advances in time-resolved BioSAXS (TR-SAXS), a rapidly evolving field enabling scientists to study dynamic biological systems in solution. She presented two contrasting stories utilizing continuous- and stopped-flow set-ups to achieve a time-resolution suitable for determining intermediates in the assembly/disassembly of a nucleosome core particle and the dimerization event of a light-activated LOV domain protein. Tsutomu Matsui (SSRL) added detail to the timeline of TR-SAXS development, pointing out that this powerful technique is still not very well known and only 6% of beamtime at SSRL’s BL4-2 is utilized for TR-SAXS.

Philip Anfinrud (NIH) was the first speaker of the final session on time-resolved crystallography. He discussed recent advances in time-resolved picosecond Laue crystallography and TR-SAXS/WAXS at the BioCARS beamline at the APS. He showed that careful timing of the storage ring bunch spacings, and laser pulses for the activation of the photoactive yellow protein (PYP), enabled the tracking of PYP's reversible photocycle at a 150 picosecond resolution. The experiment enabled Anfinrud and colleagues to assign precise lifetimes to the individual intermediates within the photocycle. Uwe Weierstall (Arizona State University) spoke of recent developments in millisecond serial microcrystallography. Weierstall presented work on a gas shielded gel extrusion device that provides a steady stream of protein microcrystals in a gel-like matrix through a bright X-ray beam. Even though the device was initially designed for an XFEL source, Weierstall demonstrated that the technology can successfully be utilized at a synchrotron source for cases where protein crystals of a few µm in size are available.

The workshop was very well attended and all learned a great deal from the fields’ early adopters and practitioners.

Submitted by: Tiit Lukk, MacCHESS and Ernest Fontes, CHESS Cornell University 07/13/2015

Poster Prize winners honored at CHESS Users’ Meeting


CHESS Director Joel Brock announces Qingqiu Huang winning a poster award from prize sponsor Alex Deyhim, president of ADC Inc.

An important part of the annual CHESS Users’ Meeting is the display and discussion surrounding users who present their work via posters. This year 42 posters covered the gamut of topics from new technical advances in hardware, software and beamline capabilities to the newest scientific results and achievements of students, post-docs and user groups. To recognize both types of achievements, CHESS has a tradition of awarding two poster prizes, one for best technical achievement and one for best science results.

This year’s award for best science display went to Mark C. Weidman and William A. Tisdale from the Massachusetts Institute of Technology collaborating with CHESS scientist Detlef-M. Smilgies for their poster "Transient Symmetry Breaking Observed During Real-Time Self-Assembly of a Nanocrystal Superlattice." Motivating the work, they discuss how during solvent evaporation, monodisperse colloids undergo a phase change in which the isolated, disordered, non-interacting colloidal particles self-assemble into a highly-ordered, close-packed superlattice. Although initial and final states can be readily characterized, little is known about the pathway and dynamics between these two diverse states. This study utilized real-time in-situ grazing-incidence X-ray scattering measurements (at the D1 beamline) to track the self-assembly of faceted lead sulfide nanocrystals as they progress from colloid to superlattice. After explaining nanocrystal characterization, they show strategies for in-situ experimental setup, real-time imaging, measuring the kinetics of self-assembly, and conclude with an overall picture of superlattice formation. They conclude that the pathway includes intermediate states, symmetry-breaking, and different timescales for orientation and densification.


The nanocrystals, as a colloidal suspension in toluene, show the typical GISAXS pattern for a suspension of nearly-spherical, non-interacting, disordered particles. The GIWAXS shows no preferential alignment of any of the atomic planes. When these same nanocrystals are spin-coated into a thin film, the GISAXS pattern is a highly-ordered BCC superlattice and the GIWAXS pattern indicates that the atomic planes of the nanocrystals are aligned as well. The (110)SL plane is parallel to the substrate, which is depicted in the illustration.

In the best technology category, Qingqiu Huang and Doletha M. Szebenyi of the MacCHESS group were chosen for their poster “High Pressure Cryocooling Improves Protein Crystal Diffraction Quality.” They provide first some historical perspective discussing how high pressure cryocooling – during which crystals are pressurized to 200MPa before being cooled to liquid nitrogen temperatures - was developed as an alternative method for cryopreservation of macromolecular crystals. The poster showed striking images of experimental results from the past year, documenting how high pressure cryocooling improves X-ray diffraction quality of protein crystals in two respects. First they showed how high pressure cryocooling dramatically improved the diffraction resolution of three protein crystals. They outlined procedures and specimen handling recipes under which diffraction resolution improves by almost a factor of two, for example from 2.4 to 1.5 Angstroms. Second, they show four specimens where high pressure cryocooling at pressures over 300MPa transformed non-merohedral twinned-crystals into single-crystals.

In both cases, CHESS users are invited to contact staff to learn more about in-house capabilities and how these advances can aid their research programs.

Submitted by: Ernest Fontes and Margaret Koker, CHESS, Cornell University 07/13/2015

Mutations in the canine parvovirus capsid make it more infectious

Canine parvovirus (CPV) is closely related to the feline panleukopenia virus (FPV), also a parvovirus, that infects domestic cats and some non-domestic carnivores. In the 1970s, canine parvovirus type 2 (CPV-2) came to the scene, and had spread around the world by 1978 to be almost completely replaced by a mutant CPV-2a variant by the end of the eighties. The CPV-2a variant has a broad host range infecting both domestic and wild carnivores (incl. dogs and cats). It’s been hypothesized that CPV-2a may be displacing FPV-like viruses in many wild carnivore hosts. Understanding the structural basis of virus-host recognition is therefore of utmost importance to be able to design strategies for intercepting infections with this high-fatality rate virus.

The group of scientists led by Prof. Susan Hafenstein of Pennsylvania State University were able to crystallize and solve the structure of the capsid of the CPV-2a variant (Figure 1, A). The 3.5Å crystal structure revealed many interesting details about the host interaction sites, giving new insights into why the CPV-2a variant has a range of hosts broader than its predecessor (CPV-2). The most significant structural difference between CPV-2 and CPV-2a is caused by an Ala300Gly mutation, which results in a 3Å movement of the GH loop and the loss of a stabilizing hydrogen bond interaction (Figure 1, B). This loop movement likely influences the binding between capsid and its major host cell receptor. Additionally, Gly300 is also within one of the major antigenic sites on the capsid.


Figure 1: A) Capsid reconstruction of the CPV-2a variant; B) The 3Å GH loop movement is caused by a single point mutation from Ala to Gly.

In addition to others, one more noteworthy mutation for CPV-2a is the Asn to Aps change at positon 426. The Asp residue introduces a negative charge that prevents the interaction of several of the antigenic site monoclonal antibodies. The seemingly minor rearrangements and subtle conformational changes that enable enhanced flexibility of the capsid have had a profound impact on the success of the virus in nature and its ability to globally replace CPV-2. The findings were published in the Journal of Virology.

X-ray diffraction data was collected at the F1 beamline at CHESS


[1] Organtini LJ, Allison AB, Lukk T, Parrish CR, Hafenstein S, "Global displacement of canine parvovirus by a host-adapted variant: structural comparison between pandemic viruses with distinct host ranges", J Virol. 2015 Feb; 89(3):1909-12.

Submitted by: Tiit Lukk, MacCHESS, Cornell University 05/29/20

'Shield' gives tricky proteins a new identity



Cornell Chronicle


Bacterial proteins co-opt host cell skeletal elements to spread infection

Rickettsia bacteria are transmitted by the bites of infected ticks and other arthropods, and cause diseases such as typhus and spotted fever. Like many other bacteria, Rickettsia uses the actin cytoskeleton of the host cells to move within a cell and spread from one cell to another. It does this through the agency of a “comet tail” assembled from actin filaments, and inhibition of comet tail formation reduces the virulence of Rickettsia. The bacterial transporter protein Sca2 is required for assembly of comet tails; it functions similarly to the eukaryotic formin proteins in promoting actin filament formation. The structure of Sca2 is important both as a simpler model for the formin-based system and as a possible point of attack for drugs to treat Rickettsia infection.

The Dominguez group (U. Penn.) determined the structure of a fragment, “SCA400” (residues 34-400 of the N-terminal “NRD” domain), of Sca2 from R. conorii. Although comprising less than 1/3 of the full protein, SCA400 was (weakly) active in nucleating actin filaments. Diffraction data were collected at CHESS A1 station, on native and Se-Met derivatized crystals, and the structure was solved using the SAD (Single-wavelength Anomalous Diffraction) method. The final refinement of the native data, to a limiting resolution of 2.18 Å, gave an R/Rfree of 17.9/22.7%.

SCA400 has a crescent shape composed mainly of a repeated helix-loop-helix motif; there are seven copies of the motif, of which three are incomplete. The N-terminal portion of the first helix of each motif is variable in conformation, but the remainder is very consistent, in spite of a lack of sequence conservation across repeats. This fold has not been previously observed, and was particularly unexpected among bacterial transporters, which are generally based on curved ß-sheets rather than a-helices.


Figure 1. Structure of SCA400. The unique N-terminal region is yellow and the repeating helix-loop-helix units are magenta.

A second domain, “CRD”, of Sca2 is predicted to have a similar conformation to NRD, and the authors suggest a model whereby crescent-shaped NRD and CRD domains fit together to create a ring which promotes assembly and elongation of an actin filament passing through it. A somewhat similar ring is formed by a dimer of the FH2 domains of eukaryotic formins, but the fold of FH2 is entirely different from that of NRD and CRD. The other domains of Sca2, as indicated by mutation experiments, may function similarly to other domains of formin in recruiting actin monomers and interacting with profilin-actin complexes. Bacterial Sca2 and eukaryotic formin represent a striking instance of equivalent functionality achieved through distinct molecular mechanisms.


Figure 2. Top, domains in Sca2. Bottom, model for Sca2 surrounding a growing actin filament.


[1] Y. Madasu, C. Suarez, D.J. Kast, D.R. Kovar, R. Dominguez, "Rickettsia Sca2 has evolved formin-like activity through a different molecular mechanism," Proc. Nat'l Acad. Sci. USA 110, E2677-E2686 (2013).

Submitted by: Marian Szebenyi, MacCHESS, Cornell University 2014