Session N06: Molecular Machines II

Global conformational changes and their allosteric control by ubiquitin regulate protein degradation of the 26S proteasome molecular machine

The 26S proteasome is the main ATP-dependent protease in eukaryotic cells, responsible for protein homeostasis and quality control. It degrades ubiquitin-tagged proteins in several well-coordinated steps that include ubiquitin binding, substrate engagement, de-ubiquitination, mechanical unfolding, translocation into an internal chamber, and proteolytic cleavage. Our recently developed single-molecule FRET measurements, relying on the fluorescent labeling of incorporated unnatural amino acids, provide unprecedented insights into the conformational changes of the proteasome and the progression of individual substrates through the ATPase motor. We found that ubiquitin binding to an allosteric trigger affects the proteasome switching between engagement-competent and processing-competent states, and accelerates motor engagement of the substrate for faster, more efficient degradation. Furthermore, mutational studies revealed important details about the mechanisms of degradation initiation and the role of individual ATPase subunits in substrate sensing and inducing the conformational switch for multi-step processing. These studies indicate how the proteasome may utilize the "ubiquitin code" to prioritize substrates in a complex and crowded cellular environment.    

Asymmetric Conformational Dynamics of the ClpB Disaggregase Nanomachine

The requisite remodeling of the protein structure in the degradation and disaggregation pathway is controlled by powerful AAA+ (ATPases Associated with diverse cellular Activities) motor proteins, which transduce chemical energy to mechanical energy. Repetitive application of mechanical force by these machines assists disassembly of misfolded substrate proteins or aggregates and promotes translocation of the polypeptide chain through a narrow pore. We performed molecular dynamics simulations of the ClpB disaggregation machine, a AAA+ motor with a double ring hexameric structure that contains two nucleotide binding domains per protomer. We probed factors underlying asymmetric ring dynamics of ClpB using both supervised and unsupervised machine learning approaches. Our results indicate that the ClpB hexamer is stabilized by a network of cross-protomer electrostatic interactions located in the central pore. Our community network clustering approach reveals strong collaboration between neighboring protomers. Predicted relaxation times of the pore loops dynamics correspond to tens to hundreds of microseconds on the millisecond experimental timescales, in excellent agreement with the relaxation times obtained using single-molecule FRET experiments.   

A General Bayesian Framework for Learning Molecular Kinetics From Single Photon Arrivals.

We present a new unifying Bayesian framework for analyzing conformational dynamics of molecules from single photon traces generated by multi-color Forster Resonance Energy Transfer (FRET) setups under continuous and pulsed illumination. In this framework, the likelihood function for single photon traces is derived using a second-order hidden Markov model (HMM). This allows us to improve upon the  presently available methods in five ways: 1) fast and slow dynamics can now be learned at the same time, 2) time scales faster than the instrument response function (IRF)  and detector dead times can now be estimated, 3) an unknown number of states can be learned through nonparamteric modeling, 4) likelihood function can be now be coveniently computed for different types experimental setups, and 5) parameters of interest are now reported with uncertainties. We apply this framework to study the dynamics of TRBP-sRNA complex using two- and three-color FRET labels. From the single photon arrivals, we learn the relevant lifetimes/transition rates and the number of conformational states involved. The length of RNA is also varied to investigate the discrete motion of the TRBP protein along the RNA.   

Withdrawn

Some simple RNA viruses have a surprising property: the complementary strand of their genetic code lacks stop codons, and could code for a protein. 

I shall discuss some observations on the structure of the genetic code which explain how this 'ambigrammatic' property can evolve, and which can distinguish different explanations of why it offers a reproductive advantage.

A technique called ribosome profiling shows that the ambigrammatic viral RNA becomes covered by a string of ribosomes, which do not detach from its end. I speculate that this provides a mechanism to hide viral RNA from host cell defences, which may be utilised by more complex viruses. 

This talk reports collaboration with Hanna Retallack, Gytis Dudas, David Yllanes, Greg Huber, Amy Kistler and Joe DeRisi.   

Withdrawn

The simplest viruses are built from a protein shell, the capsid, assembled around its genome, which can be RNA or DNA. The focus of this talk is on RNA viruses that can form in vitro from their constituent components. It is widely accepted that electrostatic interactions between the capsid proteins and the genome is the main driving force for the self-assembly of simple single-stranded RNA viruses. In this talk, we investigate by means of computer simulations and scaling theory the impact of spatial localization of the RNA binding domains on the stability of viral shells. We obtain the optimal length of the genome as a function of capsid radius and find different scaling laws depending on the level of uniformity of the charge distribution.   

Ionic propulsion in a non-motile dinoflagellate.

For hundreds of years, a subset of phytoplankton cells deemed 'bladder type' have been observed to occupy niches in nutrient poor, low light ocean habitats.  Defining features of these organisms include their large volume, as well as their ability to transit vertical distances of 10's of meters with no cilia or flagella. Therefore, an obvious question exists; how do cells lacking mechanical propulsion orient themselves in the water column? Without mechanical means, how do they change their position?  Fundamentally, since cytoplasmic density of cells is often 5-10% heavier than water, these eukaryotic cells evolved in the ocean to integrate environmental signals such as light, pressure, salinity, and nutrients to control their buoyancy in the water column with regulated ion flux and concomitant volume changes. Using high-resolution vertical tracking microscopy along with molecular perturbations we describe a coordinated order of magnitude volume increase that takes place in under 15 minutes to rapidly decrease sedimentation speed after cell division. We use electron and light sheet microscopy along with modeling to describe how the cellular design principles behind Pyrocystis's cellular morphology enable it to cope with rapid inflation while conserving biochemical function.   

Quantitative in vivo measurement of fork velocity by deep sequencing

DNA is the substrate for many competing biological processes in vivo. Recent experiments have revealed that transcription is subject to conflicts which stall the replication fork and predict that the replication fork velocity may vary significantly over the genome. Here, we demonstrate a novel approach to measuring the locus-dependent in vivo replication fork velocity by deep sequencing and test a number of mechanisms for replication slowdown, including transcription-replication conflicts. We expect this approach to be widely applicable to other biological systems.   

Accessibility Patterns and Folding Frustration in Long Telomeric Overhangs

We present single molecule experimental and computational modeling studies investigating accessibility and folding landscape of human telomeric overhangs that contain 4-28 repeats of GGGTTA (G-Tract) sequence. These sequences reach physiologically relevant lengths and can accommodate 1-7 G-quadruplex (GQ) structures. Using FRET-PAINT, we probed the distribution of telomeric sites that are accessible external agents via a short imager strand that transiently binds to G-Tracts that are not folded into GQ. We report accessibility patterns that periodically change with overhang length and interpret these patterns in terms of their implications for folding frustration. Overhangs that have multiples of four G-Tracts (12, 16, 20…), demonstrate maximum folding frustration, while those with two additional G-Tracts, (14, 18, 22…), have minimal frustration. We also developed a computational model that suggests positive folding cooperativity and lower folding stability at 5´-end, compared to other regions, are required for persistence of such folding frustration and accessibility patterns.   

Insertions and deletions affect RNA-protein interactions through RNA secondary structure

RNA binding proteins are crucial for every step of the RNA life cycle and are major contributors to post-transcriptional regulation. Since binding of a protein competes with the internal base pairing of an RNA, the binding affinity of an RNA binding protein is sensitive to the overall RNA structure. We have previously shown that single nucleotide polymorphisms have the ability to significantly affect protein binding even 50 nucleotides away from the protein binding site through modifications to the RNA secondary structure. Here, we show that insertions and deletions in an RNA have an even stronger effect on protein-RNA interactions in their vicinity and that this effect grows with the length of the insertion or deletion. By studying the occurrence of natural insertion and deletion variants near protein binding sites, we also demonstrate that this effect appears to be under natural selection.   

Conformational dynamics of the Grp94 molecular chaperone in protein remodeling

The Hsp90 family of molecular chaperones plays an important role in protein homeostasis under both physiological and stress conditions. Grp94, the endoplasmic reticulum Hsp90 paralog, is one of the most abundant lumenal proteins that assists in protein folding and maturation of a diverse group of substrate (client) proteins. The ATP-driven conformational changes of Grp94 are important for chaperone activity and protein remodeling. To investigate how conformational changes of Grp94 are coupled to protein remodeling, we performed all-atom biased molecular dynamics simulations and probed the effects of allostery imparted onto a model client protein. We observed that a combination of allostery and strong interactions between Grp94 and the client protein results in client protein unfolding. The client protein may be released after one ATP hydrolysis cycle. These simulations reveal new Grp94 residues that could be important for client remodeling and have confirmed client interactions with experimentally identified Grp94 residues. These studies are beginning to provide insight into active protein remodeling mechanisms, which may be more broadly applicable to other Hsp90 chaperone proteins.   

New paradigms for bacteria sensing the world: molecular mechanisms of mechanosensation

Bacteria can thrive in fluctuating and extreme environments. Thus, a central question in biology concerns the mechanisms by which bacteria can adapt to stresses and the strategies they use for that purpose. Regulation of the internal and external osmotic forces across the cell membrane is one of the most fundamental processes bacteria employ to survive changes in the media osmolarity. At the center of this homeostatic strategy is the gating of mechanosensitive (MS) channels embedded in the plasma membrane of bacteria. We study cell survival in environments fluctuating in osmolyte concentrations using Bacillus Subtilis as a model system. For channel-less mutants, we systematically modulate the media exchange rate and quantify its effects on cell lysis. We find that high exchange rates lead to membrane topological defects that increase bacteria death probability when tension is applied to the membrane. Furthermore, for mutants with different combinations of MS channels, we find that the activation mechanism is channel-dependent: for channels of large conductance (MscL), the activation is loading rate-dependent, whereas, for channels of small conductance (MscS), the activation mechanism is membrane tension dependent. Moreover, we develop a model for cell lysis that captures the complexity of lipid dynamics and integrate that into the experimentally observed strategies of MS gating. Our results set a basis for a new paradigm of MS activation that, to our knowledge, has not yet been described.   

Investigating structural stability of transmembrane Ligand-Appended Pillar[n]ene (LAP)

Artificial water channels are synthetic, transmembrane molecules inspired by aquaporins and offer a scalable solution for biological membrane applications. Matching the high-water permeability of the aquaporin channel while presenting a self-assembly structure that leads to properly aligned channels, ligand-appended pillar[n]ene (LAP) is a promising candidate for forward water treatment membranes. However, due to the variety of chemistries that can be appended the pillar[n]ene ring structure, the stability and selectivity of this channel are not yet fully characterized. Through atomistic molecular dynamics, we correlate the stability, permeability, and selectivity of appended peptides on the pillar[5,6]ene structures to guide experimental design and identify structural motifs that benefit water-ion selectivity and single-file water permeation. Our findings show a favorable parameter space for both channel structures and investigate the free-energy profile for ion, water, and channel insertion.   

Building an artificial neuron with simple hydrophobic nanopores: a one component memristor

The Hodgkin–Huxley(HH)[1] model for the squid giant axon membrane potential has been a classic reference in neurophysiology research for more than half a century. Ion-based neuromorphic systems are a developing reseach topic[2] and by using hydrophobic gating in ion-channels[2] as a building block and the HH model as a description of their behaviour we study a simple model hydrophobic nanopore and it's ability to generate neuron-like pulses.

We use Molecular Dynamics simulations to study hydrophobic gating in a cylindrical hydrophobic nanopore imersed in an electrolyte solution and its response to an external voltage. The rates at which the pore opens can be controlled by the voltage and we construct HH-like model for the conductivity of the pore by extracting them from MD.

This pore displays an hysteretical I-V making it a very simple one component memristor and by creating a simple HH-like circut we give insight into generation of neuron-like pulses using iontronic one component memristors.

[1] A. Hodgkin, A. Huxley, J Physiol. (1952)

[2] Robin et al., Science 373, 687–691 (2021)

[3] P. Aryal et al., J. Mol. Bio. (2015)