Theme 5 abstracts: Foundational Technologies

DNA assembly design software and automation devices
Dr. Nathan Hilson

The production of clean renewable biofuels from cellulosic starting material requires concerted feedstock engineering, deconstruction of plant matter into simple sugars, and microbial fermentation of the sugars into biofuel. These three efforts share significant molecular biological challenges, including the construction of large enzymatic libraries (e.g. vast collections of glycosyl transferases, cellulases, and efflux pumps), the generation of combinatorial libraries (e.g. multi-functional enzyme domain fusions; variations in copy number, promoter and ribosomal binding site strength), and the concurrent
assembly of multiple biological parts (e.g. the incorporation of an entire metabolic pathway into a single target vector). With these challenges in mind, we have developed two on-line software tools, j5 and DeviceEditor (http://j5.jbei.org), that automate the design of
sequence agnostic, scar-less, multi-part assembly methodologies and translates them to robotics-driven protocols. Given a target library to construct, the software provides automated oligo, direct synthesis, and cost-optimal assembly process design, and integrates with liquid-handling robotic platforms to set up the PCR and multi-par  assembly reactions.  This work reduces the time, effort and cost of large scale cloning and assembly tasks, as well as enables research scales otherwise unfeasible without the assistance of computer-aided
design tools and robotics. We are also pursuing the development of microfluidic devices that couple DNA library construction with functional assessment, a compelling process integration that is anticipated for the next round of scale-enabling foundational technologies.

Determining structures of membrane protein through combined cryo-electron microscopy and molecular dynamics
James Gumbart
Argonne National Laboratory, 9700 S. Cass Avenue, Argonne, IL 60439
e-mail: gumbart@ks.uiuc.edu

Many nascent proteins, including nearly all membrane and secreted proteins, must traverse a membrane-bound protein-conducting channel prior to their full maturation.  This channel, the Sec translocon, is found in all domains of life and possesses the novel ability to direct nascent proteins to the membrane or to the extracellular space, often concomitant with their synthesis by the ribosome.  By combining atomic structures with cryo-electron microscopy data using the molecular dynamics flexible fitting method (MDFF), we have developed some of the first views of inactive and active translocons in complex with the ribosome.  Most recently, by placing the translocon inside a so-called nanodisc, a self-contained, soluble lipid-protein particle, the role of the translocon in the insertion of a nascent membrane protein was visualized. 
The structure of the ribosome-channel complex reveals a translocon laterally open to the bilayer with the nascent protein’s signal anchor helix inserting into the membrane from the channel; key interactions observed between membrane and ribosome are predicted to mediate this insertion process.  Because the nanodisc technology permitted the use of a real membrane bilayer, as opposed to the detergents typically used to solublize membrane proteins, insights into membrane protein function in a native context were made possible. Similarly, as data on increasingly large biomolecular complexes in a myriad of functional states will come from sources other than X-ray crystallography alone, the need for hybrid computational methods such as MDFF to obtain atomic-scale detail will increase significantly.

Using Small-angle scattering and nanodiscs as an enabling platform to determine the low-resolution structure of membrane proteins.
Lise Arleth
Nanobio Science, Dept. Basic Sciences and Environment, faculty of Life Sciences, Thorvadsensvej 40, 1871 Frederiksberg, Denmark
e-mail: lia@life.ku.dk

About half of all proteins may be categorized as membrane proteins. This obviously makes membrane proteins crucial for biological function and central to pharmaceutical applications: Presently about 50% of all marketed small-molecule drugs have a membrane protein as their main target. In this context, it is in particular worth mentioning the family of so-called G-protein coupled receptors, which remains a leading target for the development of pharmaceutical molecules. However, little is known about the general structure-function relationship of membrane proteins: While there are presently about 60.000 atomic structures of water-soluble proteins in the
Protein Data Bank (PDB), less than 200 of these may be categorized as unique membrane protein structures. This discrepancy is generally explained by the notorious difficulties in crystallizing membrane proteins. Small-angle scattering comes in as a very relevant alternative structural method here. It does obviously not provide the same structural resolution as protein crystallography, but it does, however, allow for investigating the structure of a general non-crystallized protein with a resolution down to about 10 Å. In the context of the UNIK synthetic Biology project at University of Copenhagen, Denmark, we are presently in the process of developing the so-called nanodisc system into a platform for small-angle scattering based structural studies of membrane proteins. We combine Small-Angle X-ray scattering (SAXS) and Small-angle Neutron Scattering (SANS) and obtain structural information at different contrast situations. Furthermore we develop new software tools for analyzing the obtained scattering data. Software that is specially adapted to handle membrane proteins inserted into lipid bilayers or nanodiscs. A SANS/SAXS based pilote study of the relatively simple protein-lipid complex constituted by the empty nanodiscs has recently demonstrated that our approach with simultaneous molecular constrained model fitting of SAXS and SANS data allows for unprecedented structural resolution of the nanodisc particles, including structural information about the lipid bilayer. We found that the empty nanodiscs have a significant elliptical shape and that lateral packing of the encapsulated ~10 nm lipid-bilayer discs is visibly modulated by the surrounding protein belt. 

Mining post translational modifications in the endomembrane of the model plant Arabidopsis thaliana
Joshua Heazlewood
Joint BioEnergy Institute and Physical Biosciences Division,Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA.
e-mail: jlheazlewood@lbl.gov

Over the past decade a plethora of online proteomics resources have been developed that reflect a number of large-scale studies. Unfortunately these resources exist independently, and lack a level of integration that makes for easy analysis. The Multinational Arabidopsis Steering Committee, Proteomics (MASCP) sought to address this issue through the development of a proteomics aggregation portal, MASCP Gator (http://gator.masc-proteomics.org/). The portal provides a summary of proteomics and protein information aggregated directly from ten online resources. Access to these extensive proteomics data resources enabled us to develop a bioinformatics technique to identify post-translational modifications.
In plants, the endomembrane plays a crucial role in plant cell wall biosynthesis as well as a source of numerous proteins with complex post-translational modifications. We have purified the Golgi apparatus using density centrifugation and charge based separation on a Free Flow Electrophoresis (FFE) system. We have employed chemical removal of glycosylation residues from purified Golgi preparations to explore the feasibility of the modification detection method. Virtually all large-scale proteomics analyses in Arabidopsis have identified proteins with unmodified peptides. Collectively, these data reveal modified regions of a protein as unmatched areas within a protein model. Through the chemical removal of glycosylation groups from Golgi isolated proteins and previously published work, we could demonstrate that these unmatched regions likely represent modified hotspots for targeted investigations and characterizations. The ability to locate and identify post translational modifications by mass spectrometry is an extremely challenging undertaking. We have developed a method to locate putative regions with modifications by exploiting mass spectral data in the public domain. The approach was validated with data from previous studies and on the Golgi apparatus from Arabidopsis, a source of highly modified proteins.