Maya Topf

Protein structure fitting and refinement guided by cryo-EM density

For many macromolecular assemblies, both a cryo-electron microscopy map and atomic structures of its component proteins are available. Here we describe a method for fitting and refining a component structure within its map at intermediate resolution (<15 A). The atomic positions are optimized with respect to a scoring function that includes the crosscorrelation coefficient between the structure and the map as well as stereochemical and nonbonded interaction terms. A heuristic optimization that relies on a Monte Carlo search, a conjugate-gradients minimization, and simulated annealing molecular dynamics is applied to a series of subdivisions of the structure into progressively smaller rigid bodies. The method was tested on 15 proteins of known structure with 13 simulated maps and 3 experimentally determined maps. At approximately 10 A resolution, Calpha rmsd between the initial and final structures was reduced on average by approximately 53%. The method is automated and can refine both experimental and predicted atomic structures.

Integrating diverse data for structure determination of macromolecular assemblies.

To understand the cell, we need to determine the macromolecular assembly structures, which may consist of tens to hundreds of components. First, we review the varied experimental data that characterize the assemblies at several levels of resolution. We then describe computational methods for generating the structures using these data. To maximize completeness, resolution, accuracy, precision, and efficiency of the structure determination, a computational approach is required that uses spatial information from a variety of experimental methods. We propose such an approach, defined by its three main components: a hierarchical representation of the assembly, a scoring function consisting of spatial restraints derived from experimental data, and an optimization method that generates structures consistent with the data. This approach is illustrated by determining the configuration of the 456 proteins in the nuclear pore complex (NPC) from bakers yeast. With these tools, we are poised to integrate structural information gathered at multiple levels of the biological hierarchy--from atoms to cells--into a common framework.

The genetics of hyperekplexia: more than startle!

Hyperekplexia is characterised by neonatal hypertonia and an exaggerated startle reflex in response to acoustic or tactile stimuli. Genetic analysis of this disorder has revealed mutations in genes for several postsynaptic proteins involved in glycinergic neurotransmission, including the glycine receptor (GlyR) alpha1 and beta subunits, gephyrin and collybistin. However, new research suggests that mutations in the gene encoding the presynaptic glycine transporter GlyT2 are a second major cause of human hyperekplexia, as well as congenital muscular dystonia type 2 (CMD2) in cattle. These findings raise the intriguing possibility that both presynaptic and postsynaptic causes of disease might also exist in related disorders, such as idiopathic generalised epilepsies, where mutations in inhibitory GABA(A) receptor subunit genes have already been identified.

Comprehensive Molecular Structure of the Eukaryotic Ribosome

Despite the emergence of a large number of X-ray crystallographic models of the bacterial 70S ribosome over the past decade, an accurate atomic model of the eukaryotic 80S ribosome is still not available. Eukaryotic ribosomes possess more ribosomal proteins and ribosomal RNA than do bacterial ribosomes, which are implicated in extraribosomal functions in the eukaryotic cells. By combining cryo-EM with RNA and protein homology modeling, we obtained an atomic model of the yeast 80S ribosome complete with all ribosomal RNA expansion segments and all ribosomal proteins for which a structural homolog can be identified. Mutation or deletion of 80S ribosomal proteins can abrogate maturation of the ribosome, leading to several human diseases. We have localized one such protein unique to eukaryotes, rpS19e, whose mutations are associated with Diamond-Blackfan anemia in humans. Additionally, we characterize crucial interactions between the dynamic stalk base of the ribosome with eukaryotic elongation factor 2.

Structure of the mammalian 80S ribosome at 8.7 A resolution.

In this paper, we present a structure of the mammalian ribosome determined at approximately 8.7 A resolution by electron cryomicroscopy and single-particle methods. A model of the ribosome was created by docking homology models of subunit rRNAs and conserved proteins into the density map. We then modeled expansion segments in the subunit rRNAs and found unclaimed density for approximately 20 proteins. In general, many conserved proteins and novel proteins interact with expansion segments to form an integrated framework that may stabilize the mature ribosome. Our structure provides a snapshot of the mammalian ribosome at the beginning of translation and lends support to current models in which large movements of the small subunit and L1 stalk occur during tRNA translocation. Finally, details are presented for intersubunit bridges that are specific to the eukaryotic ribosome. We suggest that these bridges may help reset the conformation of the ribosome to prepare for the next cycle of chain elongation.

TUBA1A mutations cause wide spectrum lissencephaly (smooth brain) and suggest that multiple neuronal migration pathways converge on alpha tubulins

We previously showed that mutations in LIS1 and DCX account for ∼85% of patients with the classic form of lissencephaly (LIS). Some rare forms of LIS are associated with a disproportionately small cerebellum, referred to as lissencephaly with cerebellar hypoplasia (LCH). Tubulin alpha1A (TUBA1A), encoding a critical structural subunit of microtubules, has recently been implicated in LIS. Here, we screen the largest cohort of unexplained LIS patients examined to date to determine: (i) the frequency of TUBA1A mutations in patients with lissencephaly, (ii) the spectrum of phenotypes associated with TUBA1A mutations and (iii) the functional consequences of different TUBA1A mutations on microtubule function. We identified novel and recurrent TUBA1A mutations in ∼1% of children with classic LIS and in ∼30% of children with LCH, making this the first major gene associated with the rare LCH phenotype. We also unexpectedly found a TUBA1A mutation in one child with agenesis of the corpus callosum and cerebellar hypoplasia without LIS. Thus, our data demonstrate a wider spectrum of phenotypes than previously reported and allow us to propose new recommendations for clinical testing. We also provide cellular and structural data suggesting that LIS-associated mutations of TUBA1A operate via diverse mechanisms that include disruption of binding sites for microtubule-associated proteins (MAPs).

Subnanometer-resolution electron cryomicroscopy-based domain models for the cytoplasmic region of skeletal muscle RyR channel

The skeletal muscle Ca2+ release channel (RyR1), a homotetramer, regulates the release of Ca2+ from the sarcoplasmic reticulum to initiate muscle contraction. In this work, we have delineated the RyR1 monomer boundaries in a subnanometer-resolution electron cryomicroscopy (cryo-EM) density map. In the cytoplasmic region of each RyR1 monomer, 36 α-helices and 7 β-sheets can be resolved. A β-sheet was also identified close to the membrane-spanning region that resembles the cytoplasmic pore structures of inward rectifier K+ channels. Three structural folds, generated for amino acids 12–565 using comparative modeling and cryo-EM density fitting, localize close to regions implicated in communication with the voltage sensor in the transverse tubules. Eleven of the 15 disease-related residues for these domains are mapped to the surface of these models. Four disease-related residues are found in a basin at the interfaces of these regions, creating a pocket in which the immunophilin FKBP12 can fit. Taken together, these results provide a structural context for both channel gating and the consequences of certain malignant hyperthermia and central core disease-associated mutations in RyR1.

ATP-Triggered Conformational Changes Delineate Substrate-Binding and -Folding Mechanics of the Groel Chaperonin.

Summary The chaperonin GroEL assists the folding of nascent or stress-denatured polypeptides by actions of binding and encapsulation. ATP binding initiates a series of conformational changes triggering the association of the cochaperonin GroES, followed by further large movements that eject the substrate polypeptide from hydrophobic binding sites into a GroES-capped, hydrophilic folding chamber. We used cryo-electron microscopy, statistical analysis, and flexible fitting to resolve a set of distinct GroEL-ATP conformations that can be ordered into a trajectory of domain rotation and elevation. The initial conformations are likely to be the ones that capture polypeptide substrate. Then the binding domains extend radially to separate from each other but maintain their binding surfaces facing the cavity, potentially exerting mechanical force upon kinetically trapped, misfolded substrates. The extended conformation also provides a potential docking site for GroES, to trigger the final, 100° domain rotation constituting the “power stroke” that ejects substrate into the folding chamber.

PREDICT modeling and in‐silico screening for G‐protein coupled receptors

G‐protein coupled receptors (GPCRs) are a major group of drug targets for which only one x‐ray structure is known (the nondrugable rhodopsin), limiting the application of structure‐based drug discovery to GPCRs. In this paper we present the details of PREDICT, a new algorithmic approach for modeling the 3D structure of GPCRs without relying on homology to rhodopsin. PREDICT, which focuses on the transmembrane domain of GPCRs, starts from the primary sequence of the receptor, simultaneously optimizing multiple ‘decoy’ conformations of the protein in order to find its most stable structure, culminating in a virtual receptor‐ligand complex. In this paper we present a comprehensive analysis of three PREDICT models for the dopamine D2, neurokinin NK1, and neuropeptide Y Y1 receptors. A shorter discussion of the CCR3 receptor model is also included. All models were found to be in good agreement with a large body of experimental data. The quality of the PREDICT models, at least for drug discovery purposes, was evaluated by their successful utilization in in‐silico screening. Virtual screening using all three PREDICT models yielded enrichment factors 9‐fold to 44‐fold better than random screening. Namely, the PREDICT models can be used to identify active small‐molecule ligands embedded in large compound libraries with an efficiency comparable to that obtained using crystal structures for non‐GPCR targets. Proteins 2004.

Inferential optimization for simultaneous fitting of multiple components into a CryoEM map of their assembly.

Models of macromolecular assemblies are essential for a mechanistic description of cellular processes. Such models are increasingly obtained by fitting atomic-resolution structures of components into a density map of the whole assembly. Yet, current density-fitting techniques are frequently insufficient for an unambiguous determination of the positions and orientations of all components. Here, we describe MultiFit, a method used for simultaneously fitting atomic structures of components into their assembly density map at resolutions as low as 25 A. The component positions and orientations are optimized with respect to a scoring function that includes the quality-of-fit of components in the map, the protrusion of components from the map envelope, and the shape complementarity between pairs of components. The scoring function is optimized by our exact inference optimizer DOMINO (Discrete Optimization of Multiple INteracting Objects) that efficiently finds the global minimum in a discrete sampling space. MultiFit was benchmarked on seven assemblies of known structure, consisting of up to seven proteins each. The input atomic structures of the components were obtained from the Protein Data Bank, as well as by comparative modeling based on a 16-99% sequence identity to a template structure. A near-native configuration was usually found as the top-scoring model. Therefore, MultiFit can provide initial configurations for further refinement of many multicomponent assembly structures described by electron microscopy.