Robert C. Brewster

Hybrid Lipids as a Biological Surface-Active Component

Cell membranes contain small domains (on the order of nanometers in size, sometimes called rafts) of lipids whose hydrocarbon chains are more ordered than those of the surrounding bulk-phase lipids. Whether these domains are fluctuations, metastable, or thermodynamically stable, is still unclear. Here, we show theoretically how a lipid with one saturated hydrocarbon chain that prefers the ordered environment and one partially unsaturated chain that prefers the less ordered phase, can act as a line-active component. We present a unified model that treats the lipids in both the bulk and at the interface and show how they lower the line tension between domains, eventually driving it to zero at sufficiently large interaction strengths or at sufficiently low temperatures. In this limit, finite-sized domains stabilized by the packing of these hybrid lipids can form as equilibrium structures.

Tuning Promoter Strength through RNA Polymerase Binding Site Design in Escherichia coli

One of the paramount goals of synthetic biology is to have the ability to tune transcriptional networks to targeted levels of expression at will. As a step in that direction, we have constructed a set of unique binding sites for E. coli RNA Polymerase (RNAP) holoenzyme, designed using a model of sequence-dependent binding energy combined with a thermodynamic model of transcription to produce a targeted level of gene expression. This promoter set allows us to determine the correspondence between the absolute numbers of mRNA molecules or protein products and the predicted promoter binding energies measured in energy units. These binding sites adhere on average to the predicted level of gene expression over orders of magnitude in constitutive gene expression, to within a factor of in both protein and mRNA copy number. With these promoters in hand, we then place them under the regulatory control of a bacterial repressor and show that again there is a strict correspondence between the measured and predicted levels of expression, demonstrating the transferability of the promoters to an alternate regulatory context. In particular, our thermodynamic model predicts the expression from our promoters under a range of repressor concentrations between several per cell up to over per cell. After correcting the predicted polymerase binding strength using the data from the unregulated promoter, the thermodynamic model accurately predicts the expression for the simple repression strains to within . Demonstration of modular promoter design, where parts of the circuit (such as RNAP/TF binding strength and transcription factor copy number) can be independently chosen from a stock list and combined to give a predictable result, has important implications as an engineering tool for use in synthetic biology.

Line Active Hybrid Lipids Determine Domain Size in Phase Separation of Saturated and Unsaturated Lipids

A simple model of the line activity of a hybrid lipid (e.g., POPC) with one fully saturated chain and one partially unsaturated chain demonstrates that these lipids preferentially pack at curved interfaces between phase-separated saturated and unsaturated domains. We predict that the domain sizes typically range from tens to hundreds of nm, depending on molecular interactions and parameters such as molecular volume and area per headgroup in the bulk fluid phase. The role of cholesterol is taken into account by an effective change in the headgroup areas and the domain sizes are predicted to increase with cholesterol concentration.

Promoter architecture dictates cell-to-cell variability in gene expression

Variability in gene expression among genetically identical cells has emerged as a central preoccupation in the study of gene regulation; however, a divide exists between the predictions of molecular models of prokaryotic transcriptional regulation and genome-wide experimental studies suggesting that this variability is indifferent to the underlying regulatory architecture. We constructed a set of promoters in Escherichia coli in which promoter strength, transcription factor binding strength, and transcription factor copy numbers are systematically varied, and used messenger RNA (mRNA) fluorescence in situ hybridization to observe how these changes affected variability in gene expression. Our parameter-free models predicted the observed variability; hence, the molecular details of transcription dictate variability in mRNA expression, and transcriptional noise is specifically tunable and thus represents an evolutionarily accessible phenotypic parameter. Changes to regulatory DNA tune gene expression noise in a quantitatively predictable way. Promoters tune gene expression noise Although cells in a tissue are genetically identical and appear the same, they often exhibit variability in their patterns of gene expression. Organisms may need this to prepare for exposure to varying environmental stresses. Using the tools of synthetic biology, Jones et al. construct a wide range of E. coli promoters in which the key molecular parameters (such as protein binding and unbinding rates) are systematically varied and compare the resulting expression noise to parameter -free model predictions. This work demonstrates that expression noise is a tunable parameter, with different generegulatory architectures giving rise to different, but predictable, patterns of expression noise. Science, this issue p. 1533

Chain ordering of hybrid lipids can stabilize domains in saturated/hybrid/cholesterol lipid membranes

We use a liquid-crystal model to predict that hybrid lipids (lipids that have one saturated and one unsaturated tail) can stabilize line interfaces between domains in mixed membranes of saturated lipids, hybrid lipids, and cholesterol (SHC membranes). The model predicts the phase separation of SHC membranes with both parabolic and loop binodals depending on the cholesterol concentration, modeled via an effective pressure. In some cases, the hybrid lipids can reduce the line tension to zero in SHC membranes at temperatures that approach the critical temperature as the pressure is increased. The differences in the hybrid saturated tail conformational order in bulk and at the interface are responsible for the reduction of the line tension.

Plug flow and the breakdown of Bagnold scaling in cohesive granular flows

Cohesive granular media flowing down an inclined plane are studied by discrete element simulations. Previous work on cohesionless granular media demonstrated that within the steady flow regime where gravitational energy is balanced by dissipation arising from intergrain forces, the velocity profile in the flow direction scales with depth in a manner consistent with the predictions of Bagnold. Here we demonstrate that this Bagnold scaling does not hold for the analogous steady flows in cohesive granular media. We develop a generalization of the Bagnold constitutive relation to account for our observation and speculate as to the underlying physical mechanisms responsible for the different constitutive laws for cohesive and noncohesive granular media.

Rheology and contact lifetimes in dense granular flows

We study the rheology and distribution of interparticle contact lifetimes for gravity-driven, dense granular flows of noncohesive particles down an inclined plane using large-scale, three dimensional, granular dynamics simulations. Rather than observing a large number of long-lived contacts as might be expected for dense flows, brief binary collisions predominate. In the hard-particle limit, the rheology conforms to Bagnold scaling, where the shear stress is quadratic in the strain rate. As the particles are made softer, however, we find significant deviations from Bagnold rheology; the material flows more like a viscous fluid. We attribute this change in the collective rheology of the material to subtle changes in the contact lifetime distribution involving the increasing lifetime and number of the long-lived contacts in the softer particle systems.

Line active molecules promote inhomogeneous structures in membranes: theory, simulations and experiments.

We review recent theoretical efforts that predict how line-active molecules can promote lateral heterogeneities (or domains) in model membranes. This fundamental understanding may be relevant to membrane composition in living cells, where it is thought that small domains, called lipid rafts, are necessary for the cells to be functional. The theoretical work reviewed here ranges in scale from coarse grained continuum models to nearly atomistic models. The effect of line active molecules on domain sizes and shapes in the phase separated regime or on fluctuation length scales and lifetimes in the single phase, mixed regime, of the membrane is discussed. Recent experimental studies on model membranes that include line active molecules are also presented together with some comparisons with the theoretical predictions.

Long-range interaction between heterogeneously charged membranes.

Despite their neutrality, surfaces or membranes with equal amounts of positive and negative charge can exhibit long-range electrostatic interactions if the surface charge is heterogeneous; this can happen when the surface charges form finite-size domain structures. These domains can be formed in lipid membranes where the balance of the different ranges of strong but short-ranged hydrophobic interactions and longer-ranged electrostatic repulsion result in a finite, stable domain size. If the domain size is large enough, oppositely charged domains in two opposing surfaces or membranes can be strongly correlated by the electrostatic interactions; these correlations give rise to an attractive interaction of the two membranes or surfaces over separations on the order of the domain size. We use numerical simulations to demonstrate the existence of strong attractions at separations of tens of nanometers. Large line tensions result in larger domains but also increase the charge density within the domain. This promotes correlations and, as a result, increases the intermembrane attraction. On the other hand, increasing the salt concentration increases both the domain size and degree of domain anticorrelation, but the interactions are ultimately reduced due to increased screening. The result is a decrease in the net attraction as salt concentration is increased.

Single-molecule analysis of RAG-mediated V(D)J DNA cleavage

Significance V(D)J recombination is essential for assembling immunoglobulin and T-cell receptor genes. The lymphoid-specific RAG recombinase proteins (RAG1 and RAG2) perform DNA cleavage at recombination signal sequences (RSSs) during V(D)J recombination. Although RAG-mediated DNA cleavage has been analyzed in bulk, it has not at the single-molecule level. We used singly tethered DNA particle motion to examine the process in real time. We have found that RAG–RSS binding is reversible, with measurable dwell times before release. We also directly observed a reversible synaptic complex before RAG-mediated hairpin formation and measured its dwell time. These biochemical findings are compatible with a model for V(D)J recombination in vivo, in which the RAG recombinase often samples multiple partner RSSs before initiating the V(D)J recombination reaction. The recombination-activating gene products, RAG1 and RAG2, initiate V(D)J recombination during lymphocyte development by cleaving DNA adjacent to conserved recombination signal sequences (RSSs). The reaction involves DNA binding, synapsis, and cleavage at two RSSs located on the same DNA molecule and results in the assembly of antigen receptor genes. We have developed single-molecule assays to examine RSS binding by RAG1/2 and their cofactor high-mobility group-box protein 1 (HMGB1) as they proceed through the steps of this reaction. These assays allowed us to observe in real time the individual molecular events of RAG-mediated cleavage. As a result, we are able to measure the binding statistics (dwell times) and binding energies of the initial RAG binding events and characterize synapse formation at the single-molecule level, yielding insights into the distribution of dwell times in the paired complex and the propensity for cleavage on forming the synapse. Interestingly, we find that the synaptic complex has a mean lifetime of roughly 400 s and that its formation is readily reversible, with only ∼40% of observed synapses resulting in cleavage at consensus RSS binding sites.