Mixing Water, Transducing Energy, and Shaping Membranes

Feb 11, 2016 - Department of Mechanical and Aerospace Engineering, University of ..... mixing of membrane components and uniform molecular...
0 downloads 0 Views 2MB Size
Invited Feature Article pubs.acs.org/Langmuir

Mixing Water, Transducing Energy, and Shaping Membranes: Autonomously Self-Regulating Giant Vesicles James C. S. Ho,† Padmini Rangamani,‡ Bo Liedberg,† and Atul N. Parikh*,†,§ †

Centre for Biomimetic Sensor Science, School of Materials Science & Engineering, Nanyang Technological University, Singapore 637553 ‡ Department of Mechanical and Aerospace Engineering, University of CaliforniaSan Diego, La Jolla, California 92093, United States §

Departments of Biomedical Engineering and Chemical Engineering & Materials Science, University of CaliforniaDavis, Davis, California 95616, United States ABSTRACT: Giant lipid vesicles are topologically closed compartments bounded by semipermeable flexible shells, which isolate femto- to picoliter quantities of the aqueous core from the surrounding bulk. Although water equilibrates readily across vesicular walls (10−2−10−3 cm3 cm−2 s−1), the passive permeation of solutes is strongly hindered. Furthermore, because of their large volume compressibility (∼109−1010 N m−2) and area expansion (102−103 mN m−1) moduli, coupled with low bending rigidities (10−19 N m), vesicular shells bend readily but resist volume compression and tolerate only a limited area expansion (∼5%). Consequently, vesicles experiencing solute concentration gradients dissipate the available chemical energy through the osmotic movement of water, producing dramatic shape transformations driven by surface-area−volume changes and sustained by the incompressibility of water and the flexible membrane interface. Upon immersion in a hypertonic bath, an increased surface-area−volume ratio promotes large-scale morphological remodeling, reducing symmetry and stabilizing unusual shapes determined, at equilibrium, by the minimal bending-energy configurations. By contrast, when subjected to a hypotonic bath, walls of giant vesicles lose their thermal undulation, accumulate mechanical tension, and, beyond a threshold swelling, exhibit remarkable oscillatory swell−burst cycles, with the latter characterized by damped, periodic oscillations in vesicle size, membrane tension, and phase behavior. This cyclical pattern of the osmotic influx of water, pressure, membrane tension, pore formation, and solute efflux suggests quasi-homeostatic self-regulatory behavior allowing vesicular compartments produced from simple molecular components, namely, water, osmolytes, and lipids, to sense and regulate their microenvironment in a negative feedback loop.

1. INTRODUCTION A solute, excluded from or confined within a spatial “compartment” embedded in an aqueous continuum, creates a gradient in the chemical potential of water. This in turn prompts the directed flow of water pushing it into the solute-laden compartment and out of the solute-starved one, enabled by the rectification of the solute particle’s Brownian motion resulting in the gain in solute entropy and dissipation of the chemical energy of the concentration gradient.1 This universal, nonspecific entropic force or “osmotic pressure” acting at the semipermeable boundary of a compartment has significant ramifications for biomolecules. By steric exclusion, preferential hydration, and crowding, osmotic pressure can drive conformational change in proteins, gate membrane channels, and mediate enzyme actions, giving the living entity a global mechanism for regulating protein activity (Figure 1) simply by modulating physical interactions among the macromolecular “compartment”, solute, and water.1,2 Beyond single macromolecules, the consequences of osmotic forces become even more drastic when the compartment itself is a topologically closed, supramolecular structure delimited by a © XXXX American Chemical Society

structurally deformable, elastic boundary. From simple phospholipid vesicles to living cells, such compartments abound in living systems isolating femto- to picoliter quantities of the aqueous phase from the surrounding bulk. Their flexible boundaries, whose structural motif is essentially defined by a hydrophobically assembled phospholipid bilayer,3 meets the key requirement of semipermeability for the creation of osmotic force. Although the hydrophobic core of the bilayer hinders the permeation of both water and solute molecules, water exchanges across vesicular walls with sufficiently high permeabilities (10−2−10−3 cm3 cm−2 s−1)4 to allow equilibration over time. By contrast, the passive permeation of solutes across the membrane is much more strongly hindered4 because of their size and/or charge, characterized by orders of magnitude lower permeability values. Thus, over sufficiently long time scales, vesicular boundaries act as semipermeable barriers allowing the free exchange of water but confining the solute. Received: December 8, 2015 Revised: January 27, 2016

A

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

surrounding milieu and the accumulation of “compatible solutes” or osmoprotectants genetically or biochemically, high cytoplamic concentrations can be attained without disrupting normal cellular functions.9 This adaptive strategy against osmotic stress is evolutionarily well conserved across bacteria, archaea, and eukarya. Typical examples of osmoprotectants include lowmolecular-weight metabolic products such as polyhydric alcohols (polyols; e.g., glycerol, trehalose, and sucrose), free amino acids and amino acid derivatives (proline, taurine, and 13-alanine), and urea and methylamines (trimethylamine-N-oxide (TMAO), betaine, and sarcosine).10 In the absence of the availability of osmoprotectants, such as in metabolically simple archeae halobacteria, cells rapidly accumulate inorganic (e.g., K+) ions. In either case, the accumulation of osmolytes in the cytoplasm serves to reduce the osmotic imbalance, allowing the cell to restore hydration and thereby normal function.9 Osmotic downshifts, on the other hand, tend to increase the cell volume through the influx of water. This osmotic stress, as a result, strains the membrane and alters the cytoskeleton networks,11 converting the osmotic energy to mechanical energy until the cell reaches a new mechanical equilibrium. By opening up channels of the mechanosensitive kind (e.g., MscL), which allows the efficient efflux of cytoplasmic solutes, the cells can then achieve cell-volume regulation, avoid lysis, and regain normal functionality.12 Extensive reviews of the effects of osmotic pressure changes and cellular osmoregulatory mechanisms capture these issues in the literature.13,14 The movement of water accompanying an osmotic stress, involving the rapid influx or efflux across cellular compartments, is intrinsically intertwined with the membrane’s structural (i.e., elastic properties) and compositional (i.e., chemical heterogeneity) degrees of freedom. Because of their large volume compressibility (∼109−1010 N m−2) and area expansion moduli (102−103 mN m−1), coupled with low bending rigidities (10−19 Nm), vesicular boundaries bend readily but resist volume compression and tolerate only a limited area expansion (∼5%).15 During an osmotic upshift, an increased surface-area−volume ratio in topologically closed membrane compartments induces global shape transformations in flaccid membranes, altering local curvatures, fluctuations, and long-range lateral diffusivities. Similarly, during an osmotic downshift, increased membrane tension irons out thermal undulations, drives lateral redistributions of membrane components, and beyond a threshold tension induces membrane poration.5,6 Moreover, in living cellular compartments, besides changes in the membrane area and mechanical tension, osmotic shifts tend to induce additional changes in physical properties including cell volume, turgor pressure in plant cells, effective solute concentrations, ionic strength, and cytoplasmic macromolecular crowding as well as changes in cytoskeleton−membrane interactions, all of which act to modulate membrane structure and organization.7,8 Many of these membrane deformations and reorganizations have important implications for cellular osmoregulation, which are only beginning to be appreciated. Indeed, several bacterial mechanosensitive channels, which effect water efflux under osmotic downshifts, have been suggested to become “allosterically” activated (such as by a conformational change) through the transduction of mechanical tension generated in the surrounding lipid matrix of the membrane.12 Understanding how cellular capacities for osmosensing and osmoregulation affect (and are affected by) membrane properties continues to be an exciting area for inquiry and research.

Figure 1. Schematic representation of different modes of physical modulation by osmotic forces, between macromolecular compartments, solutes, and water, on topologically open and closed macromolecular systems. On the molecular level, osmotic stresses can be produced in cavities by the steric exclusion of solutes or on molecular surfaces by the preferential hydration of exposed macromolecular surfaces. The image is reproduced from Rand et al.2 (copyright 2000, Springer).

Applying an osmotic force, such as by the mere transfer of vesicles to an osmotically imbalanced bath, enriched or depleted in solute concentration, triggers a response that acts to reduce the osmotic pressure difference across the closed semipermeable membrane by the influx (or efflux) of water depending on the direction of the solute concentration gradient and the osmotic force. Thus, upon transfer to a hypotonic environment depleted of solute concentration, water permeates the solute-laden vesicles, swelling the compartment, until the internal Laplace pressure compensates for the osmotic pressure, decreasing its surface-area−volume ratio.5 Similarly, the efflux of compartmentalized water from vesicles embedded in hypertonic media increases the surface area for the encapsulated volume.6 These osmotically prompted perturbations of surface-area− volume ratios of closed compartments have profound effects on live cells. In the constant struggle with their environment, freeliving cells (e.g., bacterial cells) experiencing a sudden spike or drop in the number of dissolved molecules in water, such as due to high or fluctuating salinity, desiccation, or freezing, are confronted with significant osmotic stresses.7 If left unchecked, these environmental perturbations would result in an instantaneous flow of water out of the cell as the ambient aqueous phase becomes more concentrated (an osmotic upshift) and into the cell as the surrounding aqueous phase becomes more dilute (an osmotic downshift). The former will cause the cell to shrink and dehydrate while the latter will result in swelling, potentially inducing rupture and cell death. To avoid these catastrophic outcomes, most bacterial, plant, and animal cells have evolved mechanisms, conserved and often highly sophisticated, allowing them to regulate their water content in response to osmotic assaults imposed by variations in their local microenvironments or global environmental conditions.8 Cells accomplish these osmoregulatory activities, as components of their normal cellular homeostasis, primarily by altering the compositions of the cytoplasm such as by accumulating or releasing electrolytes and small organic solutes but also by altering chemical compositions and physical properties of their membrane shells and/or cytoplasmic cores.8 Faced with an osmotic upshift through increased concentrations of solute in the B

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

In what follows, we draw from a selection of recent efforts, including some of our own, which highlights how model membranes respond to applied osmotic forces, seemingly selfregulating themselves and altering their surroundings in a quasihomeostatic fashion. Utilizing giant unilamellar vesicles (GUVs) as simple but molecularly tailored models for cellular compartments, these studies, taken cumulatively, delineate membranemediated biophysical mechanisms, which are activated during the osmotic activity of water across topologically closed compartments, producing molecular and morphological remodeling, together characterizing their remarkable capacity for autonomous self-regulation against an osmotic assault. They thus provide a framework for understanding not only the relationships between the osmotic activity of water and membrane compartments but also broad types of membrane−water interactions. The references and attributions here are not comprehensive but should serve as a guide to the literature.

2. GIANT VESICLES One of the simplest, topologically closed cell-like compartments allowing independent control of the properties of the encapsulated aqueous core and the surrounding aqueous phase is a phospholipid vesicle. Formed through spontaneous self-assembly of amphiphilic lipids, vesicles are bounded by a 3−5-nm-thick spherical shell consisting of two apposed monomolecular leaflets of amphiphiles, isolating the encapsulated aqueous core from the surrounding bulk. They were first discovered serendipitously by Bangham in 1964 during investigations of phospholipids.15 Although the bilayer shells are spontaneously formed through the hydrophobic interactions3 between lipids and the aqueous phase, a monodisperse population of vesicles is seldom obtained without an extraneous input of energy: spontaneously formed vesicles through uncontrolled hydration are invariably entropically determined polydisperse, multilamellar dispersions. This is because the elastic properties of the bilayer membrane are determined by the local bending energy per unit area, Eb/A = κ/2(C1 + C2 − C0)2, where C1 and C2 are local principle radii of curvature (m−1); C0 is the spontaneous membrane curvature; and κ represents the mean bending rigidity.16 Thus, the free-energy cost of forming a simple, symmetric spherical vesicle (C1 = C2 = 1/R, C0 = 0, A = 4πR2 where R is the vesicle radius) becomes a sizeindependent energy penalty, Eb = 8πκ, which for a typical lipid κ = 20KbT becomes a significant, size-independent energy penalty of ∼500KbT, where KbT (∼0.6 kcal mol−1) represents the thermal energy. A variety of well-established techniques including mechanical extrusion, sonication, and electroformation, through an energy input, reproducibly transform the spontaneous multilamellar lipid dispersions in water into welldefined monodisperse vesicles, broadly categorized as small (50−100 nm), large (100−1000 nm), and giant (1−50 um) unilamellar vesicles. Cell-sized giant unilamellar vesicles (GUVs)17 (Figure 2) are ideally suited for unraveling the roles that the membrane plays during osmoregulation for a variety of important reasons. First, the ability to prepare vesicular compartments using purified natural/synthetic lipids enables a molecularly tailored selection of membrane components (Figure 2). Moreover, GUVs can be prepared using non-native amphiphilic block copolymers,18−20 which qualitatively recapitulate physical processes of lipid vesicles, but their significantly lower water permeabilities (1−10 vs 10−100 μm/s for lipids) and higher bending rigidities (40−460KbT vs 10−30KbT for lipids) offer further insights into

Figure 2. Cell-sized giant unilamellar vesicles (GUVs) offer several possibilities to address the role of membranes in various contexts. The main advantages include (i) the structural and compositional tailorability of the membranes, (ii) the isolation of the encapsulated aqueous milieu, allowing control over the external and internal solution properties, (iii) the reconstitution of membrane proteins, allowing one to address the effects of membrane properties on protein function, and (iv) the reconstitution of intracellular macromolecular crowding, mimicking cytoplasmic aqueous-phase properties.

the roles of membrane material properties in vesicle osmoregulation.21−24 The importance of this is obvious in understanding cellular osmoregulation as the membranes of different organisms and organelles are composed of a highly diverse repertoire of lipids, which undergo significant compositional adaptations in response to osmotic stresses.25 This ability to tailor membrane chemical compositions in model membranes then allows one to map the relationships between osmotic stress and the membrane’s chemical makeup toward appreciating membrane lipid homeostasis in free living cells.25 Second, many important osmosensory and osmoregulatory membrane proteins can be purified (or isolated in native membrane fragments) and subsequently reconstituted, allowing one to insert minimal cellular osmoregulatory machines into synthetic GUVs.7 This ability is important for developing an understanding of the effects of membrane properties on the conformations and functions of osmoregulatory, mechanosensitive channels. Moreover, the ability to reconstitute membrane proteins (Figure 2) into membranes of defined compositions also allows one to decipher how any spatial localization and crowding of proteins affect membrane-phase separation26 and thus the osmotic response. Third, GUVs allow one to isolate (and independently adjust) the properties of the encapsulated aqueous phase milieu (in addition to the external ambient phase) (Figure 2), thus affording control over the internal and external solute compositions and concentrations. Fourth, the semipermeable property also makes it possible to mimic intracellular macromolecular crowding27 (Figure 2) such as by introducing the so-called crowding agents (e.g., poly(ethylene glycol), dextran), thus reconstituting the properties of the cytoplasmic aqueous phase (i.e., crowding, phase separation, viscosity, and reduced diffusivities of solutes) in GUVs.28 Taken together, the C

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

tension, rupture and pore formation become energetically favorable, lysing the GUVs at lateral tensions corresponding to ∼30−40 mN m−1.30−32 Consistent with Laplace’s law (σ = ΔΠ(r/2), where σ is the lateral membrane tension, Π is the osmotic pressure difference, and r is the vesicle radius), vesicle size is inversely proportional to the osmotic differential, with small vesicles able to tolerate greater residual osmotic differentials.30,32 Previous theoretical work and experiments have shown that an osmotically tense vesicle does not lyse catastrophically; rather, it follows a stepwise sequence of events.33 The vesicles first swell progressively and then undergo a transition to a stretching stage, at which point the vesicles transform into an inflated sphere, prior to vesicle rupture. The action of this pressure-release valve is not all or nothing. During each membrane rupture event, the valve (or hole) opens for a short period of time, releasing only a fraction of the intravesicular solute (and water) before the bilayer reseals, leaving the vesicle hyperosmotic with a lower osmotic differential34 (Figure 3). This then prompts subsequent events

structural and compositional tailorability of the GUV membranes and the encapsulated aqueous phase then makes them elegant models of cellular compartments, allowing one to recapitulate the diverse conditions that characterize variations in the cytosolic milieu and membrane compositions that occur during cellular osmoregulation.

3. OSMOTIC DIFFERENTIALS Osmotic pressure refers to the difference in osmotic pressure (ΔΠ) of the aqueous solution on the “inside” and the “outside” of the GUV compartment. It is perhaps best expressed in terms of the osmotic potential (Π, atm) of an aqueous solution, which is proportional to its water activity and is determined by the activities (but not the identities) of all of its solutes ⎛ RT ⎞ Π=⎜ ⎟ ln a w ⎝ Vw ⎠

where R is the gas constant (0.082054 L atm mol−1 K−1), T is the temperature (kelvin, K), and Vw is the partial molar volume of water (0.01801 L mol−1). By definition, the activity of pure water (aw) is 1.0, which corresponds to zero osmotic potential for pure water. As solutes are added, aw generally decreases toward zero and therefore the osmotic potential associated with solutions becomes negative. Absent preferential interactions between the solute particles (and those between the solute and the membrane bilayer) in the limit of dilute solution exhibiting ideal behavior, the osmotic potential reduces to the ideal gas law: Π = −RTcs where cs represents the total molar concentration of the solute.29 Applied to vesicular compartments embedded in solutions containing osmotically active solutes (rather than pure water), the expression for the osmotic potential gradient generalizes to ΔΠ = RT(cs−inside − cs−outside). This then indicates that when a solution is separated from another by a vesicular membrane that is permeable to water but impermeable to the solutes, water will move into the vesicle from the region of higher to lower osmotic potential, down the activity (or concentration) gradient. The physical origin of water movement and the osmotic pressure itself resides in the rectification of the Brownian motion arising from the balance of forces that are generated at the membrane interface as a result of the semipermeable nature of the membrane. Simply put, the impermeable compartmental wall exerts a repulsive force on the solute particles, which through viscous coupling and momentum exchange is also transmitted to the solvent. The net effect of this Brownian motion rectifying force is to produce a steep concentration gradient and pressure drop in the vicinity of the wall within the solute-laden compartment, which provides a driving force for the solvent flow from the adjoining solute-deprived compartment. Thermodynamically, the useful work of the directed movement of water is compensated for by the loss of order (or gain of entropy) through solute dilution and the dissipation of energy associated with the chemical concentration gradient.1 3.1. Osmotic Downshift. GUVs immersed in a hypotonic environment experience an osmotic downshift. The resulting relaxation process serves to reduce the osmotic pressure difference across the closed semipermeable membrane by an influx of water, increasing its volume−surface-area ratio. An immediate consequence of the osmotic influx of water in vesicles is the suppression of membrane undulations and subsequent buildup of lateral membrane tension due to changes in the balance of forces within the bilayer producing high-energy states (compared to isotonic relaxes vesicles).30 Beyond a threshold

Figure 3. Transient pores in giant vesicles and in an in silico experiment in response to an osmotic downshift. (A) Appearance of a transient pore in a DOPC vesicle of initial radius Ri. (a). A membrane pore forms when the membrane tension reaches a critical value. The pore size reaches its maximum very rapidly (b), thereafter decreasing slowly until complete resealing (c−f). The white bar corresponds to 10 μm. The vesicle contains 20 mol % cholesterol. (B) Theoretical model of the behavior of vesicle radius r as a function of time t upon transfer from a sucrose/ glucose solution to an iso-osmolar glycerol solution. (C) Snapshots of the cross-sectional slices of the rupture and healing processes of a vesicle from dissipative particle dynamics (DPD) simulations. The vesicle before rupture (dimensionless DPD time step = 0) and right after rupture (time step = 5000) is demonstrated. At DPD time steps of 5000, 12 500, and 30 000, a pore can be clearly seen. The pore reseals completely at a DPD time step of 49 500. Images are reproduced with permission from Karatekin et al.34 (copyright 2003, Elsevier), Peterlin and Arrigler35 (copyright 2008, Elsevier), and Lin et al.37 (copyright 2012, Royal Society of Chemistry). D

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

POPC), and cholesterol. Depending on the composition and temperature, these mixtures are known to form a uniform single phase or exhibit microscopic phase separation, the latter characterized by two coexisting liquid phases: a dense phase enriched in SM and Ch designated as the Lo (liquid-ordered) phase and a second, less dense Ld (liquid-disordered) phase consisting primarily of POPC. Subjecting GUVs prepared from these mixtures to a hypotonic bath, we found that the nominal swell−burst cycles involving size pulsation are also accompanied by corresponding periodic oscillations in membrane texture: with each swell and burst phase of the swell−burst oscillation, phaseseparated and mixed states of the membrane follow. In addition to the periodic oscillation in membrane phase separation, we also find that a pore opens transiently during the phase-separated swollen membrane phase, whose subsequent closure heralds the mixing of membrane components and uniform molecular distributions.38 The pore (or valve), we find, remains open for a short period of time, typically less than 1 s, releasing some of the internal solutes and decreasing the net osmotic pressure differential in a stepwise manner. The process repeats itself several times, producing a pulsating, breathing pattern in the size and molecular texture of the vesicle (Figure 5). A more detailed characterization reveals that a highly coordinated interplay between a number of distinct elementary physical mechanisms appears to operate synchronously, producing an osmoregulatory response in giant vesicles. The mechanistic steps involved are described briefly below. 3.1.1. Osmotically Triggered Water Influx. A trans-bilayer osmotic differential, imposed on osmotically equilibrated isotonic GUVs such as by dilution of the extravesicular dispersion medium, triggers an osmotic relaxation process that acts to reduce the pressure difference across the semipermeable membrane by an influx of water.32 The retention of osmotic pressure and the buildup of membrane tension iron out thermal undulations. In the initial isotonic environment, vesicles are flaccid, with

of water influx, vesicle swelling, and rupture until sufficient intravesicular solute has been lost and the membrane is able to withstand the residual sublytic osmotic pressure without collapsing.7 Thus, GUVs in hypotonic media exhibit characteristic oscillations in their sizes, characterized by alternating modulations of vesicular volume and mechanical tension, prompted by repeated cycles of swelling and bursting.35−37 More recently, we and others have shown that the long-lived transient swell−burst cycles becomes coupled with the membrane’s compositional degrees of freedom38,39 (Figure 4).

Figure 4. Schematic of a GUV subjected to an osmotic differential. (Left panel) GUV immersed in an osmotically balanced isotonic bath exhibits flaccid topology and an optically homogeneous phase. (Right panel) Dilution of the extravesicular bath by water subjects the GUV to a hypotonic bath, which renders the flaccid vesicle stiff and replaces the uniform dye distribution by a heterogeneous pattern characterized by a microscopic domain pattern of coexisting Ld and Lo phases. Solute is rendered as white particles; the membrane, pink; and the domain pattern, pink and purple. Images are reproduced with permission from Oglecka et al.38 (copyright 2014, eLife Sciences Publications).

To explore the relationships between the out-of-plane osmotic flow and in-plane membrane composition, we used ternary lipid mixtures consisting of a saturated component (sphingomyelin, SM), an unsaturated component (1-palmitoyl-2-oleoyl-sn-1-glycerol,

Figure 5. Schematic representations of physical mechanisms and changes in vesicular membrane properties during the osmotic downshift. (Left panel) (a) GUV in an isotonic medium exhibiting flaccid morphology. (b, c) Immersion in a hypotonic bath initiates an osmotically triggered influx of water rendering the GUV tense. (d−f) The optically uniform vesicular surface breaks up into a pattern of microscopic domains, which grow by collision and coalescence. (g) Transient appearance of a microscopic pore (∼0.3−0.5 s lifetime), enabling solute efflux and tension relaxation, which drive pore closure, producing closed GUVs with a reduced osmotic differential and homogeneous surface. Steps b−g repeat until the sublytic solute concentration differential is reached and the Laplace tension in the membrane is able to compensate for the residual osmotic pressure. (Right panel, top) Fluorescent images of the oscillatory phase separation (corresponding to steps d−f on the left panel) and (right panel, bottom) transient appearance of a pore (corresponding to step g on the left panel). Images are reproduced with permission from Oglecka et al.38 (copyright 2014, eLife Sciences Publications). E

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

It suggests that the lateral tension elevates the line tension between coexisting phases.48 Therefore, although membrane tension disfavors the nucleation of a new phase (by raising the energy barrier that must be met for the formation of critical nuclei), it can promote the coalescence of small pre-existing nanoscale domains driven by minimization of line tension between the Lo and Ld phases. Additional experiments using ternary lipid mixtures (DOPC, DPPC, and Chol), which have been thought not to produce nanodomains at temperatures above 20 °C,45 also produce oscillatory phase behavior. This then suggests that the osmotically generated tension alone might be insufficient to explain the observed osmotically induced isothermal phase transition and that the nonideality in mixing is likely a consequence of a combined effect of the pressure and tension. Additional work is needed to sort out the mechanism by which osmotic stress drives membrane phase separation. The tension- and pressure-mediated appearance of the phaseseparated state in an osmotically swollen membrane does not account for the oscillations in the domain pattern. The period of oscillation between optically uniform and phase-separated states increases with the passage of time. The cycle period, defined as the time elapsed between two consecutive instances of homogeneous fluorescence, increases 3- to 10-fold before reaching a nonoscillating quiescent state 60−120 min after the imposition of the osmotic differential. This temporal dynamics reveals that the osmotic differential and accompanying tension weaken with each cycle. This requires a separate mechanism for solute efflux, which reduces the osmotic driving force. 3.1.4. Appearance of a Short-Lived Transient Pore. This enables partial solute efflux, reducing osmotic pressure and membrane tension. According to classical nucleation theory, the cost (E) of creating a pore in a tense membrane is determined by the competition between membrane tensional energy (−πr2σ) and the line tension energy (+2πrϒ) at the edge of the pore. Thus, under conditions of sufficient membrane tension (dE/dr > 0), pores nucleate and grow, enabling solute efflux.35 Membrane lysis proceeds via cascades of pores during each cycle of the swell−burst sequence:34 during the swelling segment of each oscillation cycle, a single microscopic pore, several micrometers across, becomes visible under conditions of maximum swelling and the largest domain size, typically for a period not exceeding 1.0 s. Although domain formation is not required for pore formation, the probability of pore nucleation might be enhanced by surface defects, such as those present at the boundary between coexisting phases, since the energy required to open a pore (>40KbT) is considerably higher than the thermal activation energy.34 The long life spans (∼0.3−1.0 s) of the pores are likely supported by two opposing processes, namely, the osmotic influx of water and the leakage rate of solute through the pore.49 3.1.5. Pore Closure Resulting in Closed GUVs with a Reduced Osmotic Differential. Healing of the pore is promoted by the reduction in the net membrane area and partial loss of the encapsulated solutes, both of which reduce membrane tension, σ,34 and prompt the homogenization or mixing of membrane molecules. The time scale for this mixing−demixing oscillation or the cycle period, defined as the time elapsed between two consecutive instances of the mixed state, depends on the GUV size ranging from ∼10 s (for 10-μm-diameter GUVs) to ∼30 s (for 50-μm-diameter GUVs).38 During each membrane rupture event, only a fraction of the intravesicular solute is released before the bilayer reseals, leaving the vesicle hyperosmotic, albeit with a reduced osmotic differential. This then prompts subsequent

thermally undulating surface topology caused by bendingdominated shape fluctuations. As water enters, the GUVs swell, ironing out the undulations, generating lateral membrane tension,33 and rendering the vesicular boundary tense.40 At equilibrium, the lateral tension generated in the membrane then compensates for the osmotic pressure, consistent with Laplace’s law. 3.1.2. Appearance of Microscopic Domains in the Membrane. Membranes formed from multiple components can partition laterally into binary coexisting microscopic liquid phases or domains. A ternary mixture, namely, POPC, SM and Chol, in a molar ratio of 1:1:1 at 23 °C exhibits uniform phase behavior.41 Below a miscibility transition temperature, these lipid components coexist as a liquid-ordered (Lo) phase, which is a liquid phase with short-range order populated by SM and Chol, and a POPC-rich liquid-disordered (Ld) phase.41 The use of POPC is highly relevant due to its abundance in the plasma membrane. SM and Chol, on the other hand, in addition to their presence on the outer leaflet of the plasma membrane, lie in the unique geometry of SM to hydrogen bond with the 3-OH group of the cholesterol through an ester linkage, an amine, and hydroxyl groups.42 When subject to hypotonic conditions as described above, the optically homogeneous vesicles break up into surface patterns consisting of large microscopic rounded liquid-disordered domains bounded by a spatially continuous liquid-ordered region. The stochastic nature of the phase transformation is evident from the heterogeneous landscape of optically homogeneous vesicles and those decorated with microscopic domains. The time-dependent process of the appearance and disappearance of large microscopic domains repeats itself multiple times over several tens of minutes, ultimately producing a steady state characterized by a fixed microstructure of a rounded boundary. The oscillary phase separation behavior is fully reproducible for a variety of neutral osmolytes (including glycerol, glucose, lactose, galactose, dextran, sorbitol, and sucrose), GUV sizes (5−50 μm), osmotic gradient strengths (20−2000 mM), and a wide range of lipid compositions within the phase coexistence window.41 3.1.3. Coarsening of Domains. Under lateral tension and pressure differences, (1) the domains coarsen through collision and coalescence and (2) the appearance of the phase-separated state invariably coincides with the swollen, tense state of GUVs during the cyclical swell−burst processes. Although lateral tension and pressure differences influence the membrane phase behavior in our osmotically driven case, it is instructive to consider how each of the two factors individually affects the membrane phase behavior. A recent thermodynamic analysis and experiments examining the effects of mechanically generated tension reveal a lowering of the miscibility phase-transition temperature between the Lo and Ld phases with increasing tension (dT/dσ, ∼−1 K (mN m−1)−1).43,44 However, how this shift in transition temperature affects the membrane phase behavior and domain morphology is not obvious: several recent experimental studies suggest that even tension alone can stabilize complex domain morphologies.45,46 A theoretical model by Givli and Bhattacharya,47 explicitly introducing osmotic pressure contributions within the generalized Helfrich energy treatment, suggests that pressure can perturb the isothermal phase diagram, driving domain formation primarily by affecting the interaction between geometry and composition. An alternate explanation involves separate theoretical arguments that require pre-existing phase-separated domains in the optically homogeneous state. F

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

practically unlimited number of different shapes are accessible for deflated vesicles with excess membrane area. Experimentally, however, a large but finite number of distinct shape types have been observed52 (Figure 6). These include oblates, prolates,

cycles of water influx, vesicle swelling, and rupture, albeit under reduced osmotic differential, producing damped oscillation cycles between the mixed and the demixed state. With increasing cycle number, the cycle period climbs by as much as 5- to 10-fold from roughly 10−30 s to 100−120 s. The oscillatory sequence repeats until sufficient intravesicular solute is lost and the Laplace tension in the membrane is able to compensate for the residual osmotic pressure.38 The entire period of the oscillatory regime typically lasts for 60−120 min after the imposition of the osmotic differential. These oscillations in vesicle texture continue until sufficient intravesicular solute is lost and the Laplace tension in the membrane is able to compensate for the residual osmotic pressure.31 This seemingly autonomous vesicle response, in which an external osmotic perturbation is managed by a coordinated and cyclical sequence of physical mechanisms allowing vesicles to sense (by domain formation) and regulate (by solute efflux) their local environment, suggests a primitive form of quasi-homeostatic regulation in a synthetic material system. The observations also illustrate the coupling between the out-of-plane osmotic activity of water and the membrane’s in-plane compositional degrees of freedom, producing an exquisite response, and thus underscore the intrinsic coupling between the membrane phase and mechanical tension. The osmotically generated membrane tension and poration can produce additional interesting scenarios of plausible consequences for membrane growth. In a recent study, Chen et al.50 have shown that the entrapment of high concentrations of osmolytes (e.g., 1 M sucrose) within nanometer-scale protocell compartments (e.g., ∼100 nm fatty acid vesicles), mimicking the physical effect of counterions that accumulate during RNA replication, produces an osmotic pressure difference across the compartmental boundary that renders the vesicular membrane tense but does not lyse it. Subsequent uptake of additional membrane amphiphiles from surrounding relaxed vesicles, they show, then allows the osmolyte-loaded protocell to grow preferentially. A consequence of this competitive growth of RNA-replicating vesicles at the expense of empty vesicles is that they have a shorter cell cycle allowing them to dominate the population, a primitive form of Darwinian evolution through the osmotic activity of RNA-replicating vesicles.51 These observations thus suggest how the osmotic activity of membrane compartments can fuel growth and fitness and, in conjunction with budding and fission, allow for simple protocellular compartments to exhibit self-reproducing behaviors. 3.2. Osmotic Upshift. GUVs immersed in a hypertonic bath experience osmotic upshift due to the osmotic potential gradient (ΔΠ) across the membrane boundary, which acts to expel water from the vesicular compartment. Assuming that the initial shape of the vesicle is spherical, the expulsion of water causes the vesicle volume (V) to decrease, rendering it flaccid. A dominant effect of the availability of excess membrane area for enclosing the vesicle volume is the transformation of vesicle shapes. This can be described in terms of a dimensionless parameter termed the reduced volume

v=

Figure 6. Flaccid vesicle shapes formed as a result of the vesicle formation process or in response to the osmotic upshift. Phase contrast microscopy images. First row: characteristic oblate shapes exemplified by the cup-shape class (1−3) and the disc-shape class (4). Second row: characteristic prolate shapes exemplified by the cigar-shape (shape 5) and pear-shape (6−8) classes. Third row: codocyte (9), torocyte (10), starfish (11), and worm (12) shapes are characterized by their relatively small vesicle volume/membrane area ratios. The fourth row shows shapes characterized by narrow necks connecting nearly spherical vesicle parts (13−16). Images are reproduced with permission from Svetina and Zeks52 (copyright 2002, John Wiley and Sons).

starfish, dumbbells, and pearls, often following well-defined pathways for shape changes.53 Two characteristic oblate shapes include cup (shapes 1−3) and disc (shape 4) shapes. Two typical prolate shapes are the cigar (shape 5) and pear shapes (shapes 6−8). For vesicles with lower volumes, shapes 9−12 are most frequently observed, and shapes 13−16 are those characterized by narrow necks, namely, dumbbells, and pearls.39,54 The finite range of distinct vesicle shapes that are experimentally observed in experiments modulating area− volume ratios reflects the selection of some symmetry characteristics. This in turn suggests that their formation may not be determined solely by the geometric criterion of reduced volume alone but rather by the fact that the membrane material properties play a role in determining vesicle shapes. Several different theoretical models have been advanced to predict the shape behavior of vesicles, which incorporate material elastic properties of the membrane bilayers. Two independent elastic deformational modes, namely, in-plane elasticity and bending rigidity, are relevant. Because lipid molecules comprising the membrane are two-dimensionally fluid, membranes do

V 4πR 0 3/3

where R0 corresponds to the radius of an equivalent sphere of area S = [4πR02]. From geometrical considerations alone, when the vesicle volume is reduced to below the maximum, V0 [= 4πR03/3] at a constant surface area S, v < 1, the shape deviates from the spherical motif assuming irregular shapes: a G

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

flip-flop rate).59 The generalized BC model or the area difference elasticity (ADE) model,60 which interpolates between the SC and BC models, describes the shape free-energy functional formally in terms of both the SC (first term) and the BC (second term) models

not exhibit shear elasticity.16 Moreover, the area expansion moduli (102−103 mN m−1) of typical membranes are large, and the bending rigidities (10−19 N m) are low. Thus, the vesicular shells respond to changing vesicle volume during the osmotic upshift by producing deformations and shape changes determined primarily by their bending rigidities and hence curvatures.30 Major models, which predict vesicle shapes under the condition of v < 1, consider the minimal bending-energy configurations of vesicles: an implicit assumption in these models is that the vesicles adopt shapes that correspond to the smallest possible value of the membrane bending energy55 (Figure 7).

FADE[S] =

κb 2

∮s

κ π dΑ(C1 + C2 − 2Co)2 + ̅ (ΔA − ΔA 0)2 2 AD2

in which C1 and C2 are the local principle radii of curvature (m−1), C0 is the spontaneous curvature of the membrane surface, and kb represents the mean local bending rigidity, a material property describing the SC term (first term). In the BC contribution (second term), A is the membrane area, D is the membrane thickness, and k̅ is the nonlocal bending rigidity. Thus, the minimal-energy vesicle shape can be determined for given reduced volume V and by an effective reduced area difference, which is determined by the spontaneous curvature of the SC model and by the curvature-induced area difference between the inner and outer monolayers from the BC model. By the minimization of the total curvature energy of vesicular compartments under the imposed constraints of constant area and volume, minimal-energy shapes can be obtained as functions of the enclosed volume and the area of the vesicle. Applying the SC model alone, Deuling and Helfrich obtained a rich catalog of shapes,61 which recapitulates the experimentally observed GUV shapes. Organizing the shapes into so-called phase diagrams as functions of reduced volume then allows one to predict shape transformations that occur when an osmotic stress is applied.62 Svetina and Zeks63 combined the bending elasticity of the membrane with the BC model and developed corresponding phase diagrams for vesicle shapes. The two models identify comparable shapes, although the order of the shape transitions and the paths of transition differ for the two models when the area−volume ratio (or osmotic pressure difference) is varied.64 The GUV response to osmotic deflation becomes more involved when additional structural and compositional complexities are introduced. For instance, GUVs prepared using mixtures of lipids, (e.g., egg−PC lecithin and phospholipid−cholesterol mixtures) have been observed to respond to osmotic upshifts by gradually shrinking their size without exhibiting any measurable distortion from the nominally spherical shape.6 The gradual shrinkage, in these cases, appears to be accompanied by an irreversible topological transition producing daughter vesicles that detach from the parent GUV. In many instances, the daughter vesicles remain attached to the mother vesicle through thin tethers. These observations support the notion that budding and vesiculation might be driven by the generation of spontaneous membrane curvature through the nonhomogeneous distribution of lipids. In another study, Döbereiner et al.65 examined the morphological behavior of vesicles consisting of lipid−cholesterol mixtures under conditions of increasing area−volume ratio by heating and through osmotic effects. They too find budding and repeated fission, which they attribute to a combination of liquid/ gel domain formation, the asymmetric transverse distribution of lipids, and coupling of the spontaneous curvature of the membrane to the local lipid composition, all consequences of the membrane’s compositional heterogeneity. Using a phaseseparating ternary system, Yanagisawa and co-workers66 monitored the shape deformations of ternary vesicles undergoing phase separation under an osmotic upshift. They observe that lateral phase separation became coupled with shape changes that converged prolate, discocyte, and starfish vesicles into discocytes

Figure 7. Flaccid vesicle shape changes in response to the osmotic upshift and their volume−area relation depicted on a vesicle phase diagram in a (v, Δa) plane, where v and Δa are the reduced volume and the reduced monolayer area difference, respectively. The vesicle shapes are color coded along with representative shapes from experiments, and the two solid lines indicate the stomatocyte−discocyte transition (S) and the location of the cigar shape (C). Images are reproduced with permission from Sakashita et al.56 (copyright 2012, Royal Society of Chemistry).

In these models, the membrane is regarded as a two-dimensional surface embedded in three-dimensional space.56 The spontaneous curvature (SC) model16,55 explicitly accounts for the local elastic energy penalty due to any deviation of the local mean curvature from the preferred curvature (C0, where 2C0 is the spontaneous curvature) of the membrane. The SC model then formally accounts for the local elastic bending energy in terms of the local bending rigidity (k) and reflects the membrane asymmetry due to differences in the chemical composition (or the immediate chemical environment) of the two constituent leaflets. The alternate bilayer couple (BC) model57 considers the global elastic energy due to bending in terms of the departure, albeit fixed, of the geometrical area difference, ΔA, between the two monolayers from its equilibrium value ΔA0. This then becomes the basis for the nonlocal bending rigidity. In this approach, the two monolayers are considered to be coupled at a fixed distance, but they are not allowed to exchange lipid molecules. This then leads to an additional constraint which can be incorporated into the continuum model. In addition to the differences in chemical environment surrounding the two monolayers (see above), an area difference can also originate from (1) the asymmetric insertion of molecules;57 (2) differences in the thermal expansivities of the two monolayers;58 and (3) the increased flip-flop of molecules due to changes in the lateral membrane tension (e.g., during osmotic deflation, the flow of water through the membrane can lead to a greatly enhanced H

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

changes in vesicles embedding a dense, viscoelastic agarose gel with those containing sucrose. They found that the presence of gels within the vesicles, above a threshold concentration, suppresses large-scale deformations and produces unique sets of morphologies characterized by the appearance at the vesicle surface of facets, bumps, and spikes for low-moduli gels and wrinkles for higher-moduli gels reminiscent of echinocytic shapes of red blood cells.69 The latter erythrocyte shape is well described by combining the area difference elasticity model with the elasticity of the spectrin network anchored to the plasma membrane.69 These findings then suggest prospects for resolving a significant controversy surrounding how cell volume regulation is effected in cellular compartments. Specifically, by allowing the incorporation of gel-like poroelastic material in the lumen of the GUVs, these complex compartments promise to resolve whether the osmotic stress is distributed throughout the cell volume and not confined to the membrane cortex alone.11 These and other recent advances summarized above in designing complex, giant vesicles promise new insights geared toward a more complete understanding of how the internal structure of the encapsulated phase in vesicles, which mimic cytoplasmic contents, affects vesicular osmotic responses.

4. OSMOTIC EFFECTS IN MAMMALIAN CELLS Although most mammalian cells are bathed in extracellular fluids of tightly regulated total solute concentrations, cells encounter osmotic imbalances due to rapid fluctuations in intracellular solute concentrations and often to a considerable degree during disease states.10 One of the most extensively studied cases of osmotic deformation is that of human red blood cells (hRBCs) studied in the context of osmotic hemolysis:55 placed in hypotonic media, hRBCs swell, abandoning their characteristic biconcave shape, becoming spherical and losing their intracellular hemoglobin, thereby producing erythrocyte ghosts. A peculiar property of red blood cells is that the normal biconcave human red blood cell with an ∼90 fL volume and a surface area of 140 μm2 possesses an excess surface area of 40% compared to a sphere of the same volume. This excess volume allows hRBCs to undergo dramatic shape changes without increasing their area, needed for their passage through the narrow capillaries of the microvasculature, often with cross sections smaller than their own effective diameter.70 Thus, the swelling of hRBCs under hemolytic osmotic conditions occurs largely under an essentially constant membrane area, revealed only by the approach of the spherical shape. Although the details of red blood cell biomechanics are significantly more complicated than those of free membranes of vesicles primarily because of the involvement of the spectrin network, it is notable that hemolysis also proceeds in a stepwise manner71 (Figure 9). Specifically, hRBCs undergo hemolysis through repeated cycles of swelling and bursting or jumps via the periodic opening of pores and hemoglobin leakage, characterizing the late-stage hemolysis72,73 in a striking parallel to the response of lipid vesicles (see above). Equally notable is the fact that this pulsewise lysis of red blood cells was speculated by Koslov and Markin49 in their theoretical treatment of the osmotic lysis of lipid vesicles. Another interesting consequence of the osmotic activity of a cell is its potential for promoting cell motility. A prominent example includes intra- and extravasations of metastatic cancer cells. To maneuver the architecturally complex and physically restricting three-dimensional space spanning the extracellular matrix, stroma, and blood vessels, which are subjected to a range

Figure 8. Time evolution of shape deformations induced by osmotic stress, coupled with phase separation, of multicomponent vesicles. (A) Outside (A−C, top panels) and inside (A−C, bottom panels) budding of a sphere (A), a discocyte (B), and a tube (C) in response to osmotic stress. Scale bars are 5 μm. Images are reproduced with permission from Yanagisawa et al.66 (copyright 2008, American Physical Society).

(Figure 8). Moreover, in late stages, they find that domains bud vectorially, inward or outward depending on the amount of available excess area. All of the observations above can be reconciled in terms of theoretical frameworks above for understanding equilibrium shapes for compositionally complex vesicles. Here, in addition to bending energy, the minimization of membrane energy requires a consideration of additional contributions from the free energy for lateral phase separation and the line tension.67 The next level of complexity introduced into synthetic vesicles involves the recapitulation of the internal structure of the vesicle volume. Unlike synthetic vesicles, which typically employ dilute aqueous solutions, the properties of cellular cytoplasm, the cytoskeleton network, and macromolecular crowding in the cellular interior all contribute to the cellular response to the osmotic challenge. Studies of the osmotic response of complex GUVs, which incorporate these physical properties of the cytoplasm, are sparse. In one such study, Viallat, Dalous, and Abkarian68 compared osmotically induced shape I

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

compositional redistributions, and ultimately recapitulate at least some of the behaviors that confer extraordinary responsiveness to living cells to changes in the properties of their environment. In this feature article, we have revisited the fundamental physical−chemical processes that determine the vesicular response to osmotic perturbations, most of which are already well understood within the framework of membrane mechanics. Contextualizing their relative roles in determining vesicle osmoregulation has allowed us to organize the cumulative evidence and progress in understanding the osmotic responses of membrane vesicles. It has also offered a glimpse into challenges and opportunities for future work, which we have highlighted within the chapter. The ability to engineer these complex vesicular compartments now promises a clearer molecular-level mechanistic understanding of the relationships between membrane deformations including changes in the membrane shape, curvature, tension, fluidity, and lateral phase separation and functioning of osmoregulatory machines, such as mechanosensitive channels.12 It should also help to unravel mechanisms of more involved consequences of osmotic activities, including recent demonstrations that reveal how a polarized distribution of channels in the cell membrane, creating a net inflow of water (and ions) at the cell leading edge and a net outflow of water (and ions) at the trailing edge, propels the cell, leading to the migration of cells, which is a fundamental physiological process of broad relevance.74 In a parallel vein, an understanding of osmosensing and osmoregulating capacities of simple vesicular compartments made of amphiphiles alone may help us to understand how the early evolution of the cellular design of living systems might have used these simple and pervasive perturbations to enable the stability, growth, fusion, division, and motility of early protocells at the dawn of life.50,75

Figure 9. Repetitive “jumps” and “cloud ejection” of a red blood cell during hypotonic lysis. (A) Stills of a video recording showing prehemolytic swelling (a, b) and hemolysis (c, d). (a) The three cells are in isotonic saline. (b) When exposed to a hypotonic perfusion of water, the same cells undergo prehemolytic swelling and adopt a spherical shape. (c) Superposition of three images of the three cells. The background renders black, and the superimposed cells in the first two images are semitransparent to highlight the cell movement. Cell 3 moved from its lower-right position first upward and then sideway to the left, thus pushing cell 2 sideways. Between (c) and (d), cell 2 made a movement and pushed cell 3 into the streaming layer of the medium, where it is washed away. Cells 1 and 2 both moved, as revealed by the increased distance in (d). Scale bar, 10 μm. (B) Stills from a video recording show the cloud appearance (arrowhead) from red blood cells during hemolysis. Images are reproduced from Zade-Oppen 2002 (copyright 2002, John Wiley and Sons).

of mechanical forces and altered interstitial fluid pressure, cancer cells are rendered motile through cell shape changes, which accompany cell-volume regulation. In a study reported recently,74 cancer cell migration has been proposed to be driven by osmotic forces, the so-called osmotic engine model. Specifically, on the basis of the observations of polarized distributions of Na+/H+ pumps and aquaporins, the authors suggest that the directed water permeation, involving the inflow of water (and ions) at the leading edge and net outflow at the trailing edge, propels the cell. These (and others not discussed here) multifaceted effects of osmotic forces thus support the notion that the nonspecific osmotic forces are not passive consequences of cellular activity. Rather they might provide the living cell a generic mechanism for performing many of life’s essential functions.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. Biographies

5. OUTLOOK AND PERSPECTIVES The ability to engineer ever more complex giant unilamellar vesicles has led to tremendous progress in the understanding of the roles of membrane’s structural and compositional degrees of freedom in aiding the fundamental processes of cellular homeostasis such as osmoregulation. From a biological perspective, work at this interface of membrane biophysics and cell biology has yielded a clearer mechanistic picture of how membrane physical−chemical attributes play a role in (1) facilitating the sensing of osmotic irregularities, (2) activating osmoregulatory mechanosensitive channels, and (3) driving cellular shape changes that occur in response to the osmotic challenge. Viewed from a physical science perspective, model vesicular membranes offer a means to study stimuli-responsive, dynamic, and adaptive behaviors at soft and flexible interfaces. They also allow the abstraction of fundamental principles and suggest practical means to develop cell-like synthetic model compartments that exhibit complex shapes and molecular textures, undergo well-defined shape transformations and lateral

James C. S. Ho is a research fellow in the School of Materials Science & Engineering at Nanyang Technological University (NTU). He received his B.Sc. degree in biological sciences from NTU (2009) and Ph.D. degree in medical science (2014) from Lund University. His research interest lies at the interface of cell biology, structural biology, and lipid membrane dynamics and associated processes, with a focus on osmoregulation in artificial vesicles and cellular systems, biosensing platforms, and alternative cancer cell death mechanisms. J

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

Atul N. Parikh is a professor of biomedical engineering and of chemical engineering and materials science at the University of California Davis. Since 2012, he has also served as a visiting professor in the School of Materials Science & Engineering at NTU. He received his B. Chem. Eng. (1987) degree from the University of Bombay (UDCT) and his Ph.D. degree (2004) from the Department of Materials Science & Engineering at The Pennsylvania State University. Earlier, he was a postdoctoral scholar and then a technical staff member in the Chemical Science and Bioscience divisions at Los Alamos National Laboratory from 1996 to 2001. His current research includes fundamental studies of dynamic self-assembly, active interfaces, and physical compartmentalization in soft and living material systems.



ACKNOWLEDGMENTS We are grateful to Kamila Oglȩcka, Rachel Kraut, Jeremy Sanborn, Doug Gettel, Sean Hong, Morgan Chabanon, and Madhavan Nallani, whose work and collaboration on the topic shaped this article. This work is supported by a grant from the Biomolecular Materials Program, Division of Materials Science and Engineering, Basic Energy Sciences, U.S. Department of Energy under award DE-FG02-04ER46173 (A.N.P.). J.C.S.H. acknowledges the support from the Provost Office, Nanyang Technological University. A.N.P. and B.L. acknowledge additional support from Nanyang Technological University through the Centre for Biomimetic Sensor Science.

Padmini Rangamani is an assistant professor in mechanical engineering at the University of CaliforniaSan Diego. Previously, she was a UC Berkeley Chancellor’s Postdoctoral Fellow, where she worked on lipid bilayer mechanics (2010−2014). She obtained her Ph.D. in biological sciences from the Icahn School of Medicine at Mount Sinai (2010). She received her B.S. (2001) and M.S. (2005) in chemical engineering from Osmania University (Hyderabad, India) and Georgia Institute of Technology, respectively. Her research interests include the regulation of cell shape through membrane dynamics, actin cytoskeleton remodeling, and mathematical modeling of biological processes.



ABBREVIATIONS GUV, giant unilamellar vesicle; Lo, liquid-ordered; Ld, liquiddisordered



REFERENCES

(1) Nelson, P. C. Biological Physics: Energy, Information, Life. W. H. Freeman: New York, 2002. (2) Rand, R. P.; Parsegian, V. A.; Rau, D. C. Intracellular osmotic action. Cell. Mol. Life Sci. 2000, 57 (7), 1018−1032. (3) Chandler, D. Interfaces and the driving force of hydrophobic assembly. Nature 2005, 437 (7059), 640−647. (4) Paula, S.; Volkov, A. G.; VanHoek, A. N.; Haines, T. H.; Deamer, D. W. Permeation of protons, potassium ions, and small polar molecules through phospholipid bilayers as a function of membrane thickness. Biophys. J. 1996, 70 (1), 339−348. (5) Ertel, A.; Marangoni, A. G.; Marsh, J.; Hallett, F. R.; Wood, J. M. Mechanical-Properties Of Vesicles 0.1. Coordinated analyses of osmotic swelling and lysis. Biophys. J. 1993, 64 (2), 426−434. (6) Boroske, E.; Elwenspoek, M.; Helfrich, W. Osmotic shrinkage of giant egg-lecithin vesicles. Biophys. J. 1981, 34 (1), 95−109. (7) Wood, J. M. Osmosensing by bacteria: Signals and membranebased sensors. Microbiol. Mol. Biol. Rev. 1999, 63 (1), 230−262. (8) Wood, J. M. Bacterial Osmoregulation: A paradigm for the study of cellular homeostasis. Annu. Rev. Microbiol. 2011, 65, 215−238. (9) Kempf, B.; Bremer, E. Uptake and synthesis of compatible solutes as microbial stress responses to high-osmolality environments. Arch. Microbiol. 1998, 170 (5), 319−330. (10) Yancey, P. H.; Clark, M. E.; Hand, S. C.; Bowlus, R. D.; Somero, G. N. Living With Water-Stress - Evolution of osmolyte systems. Science 1982, 217 (4566), 1214−1222. (11) Sachs, F.; Sivaselvan, M. V. Cell volume control in three dimensions: Water movement without solute movement. J. Gen. Physiol. 2015, 145 (5), 373−380. (12) Kung, C. A possible unifying principle for mechanosensation. Nature 2005, 436 (7051), 647−654. (13) Poolman, B.; Spitzer, J. J.; Wood, J. A. Bacterial osmosensing: roles of membrane structure and electrostatics in lipid-protein and protein-protein interactions. Biochim. Biophys. Acta, Biomembr. 2004, 1666 (1−2), 88−104.

Bo Liedberg is a full professor of materials science, director of the Center for Biomimetic Sensor Science (CBSS), and Dean of the Interdisciplinary Graduate School (IGS) at Nanyang Technological University (NTU). He received his Ph.D. degree in applied physics (1986) from Linköping University. His basic research is primarily devoted to soft materials science including plasmonics, surface chemistry, self-assembly, and biomimetics. He is also interested in developing new biosensing tools in biology and medicine and exploiting novel transduction principles for biochemical sensing and biomedical diagnostics.

K

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

(14) Andersen, O. S. Perspectives on: The response to osmotic challenges. J. Gen. Physiol. 2015, 145 (5), 371−372. (15) Bangham, A. D.; Horne, R. W. Negative staining of phospholipids + their structural modification by-surface active agents as observed in electron microscope. J. Mol. Biol. 1964, 8 (5), 660. (16) Helfrich, W. Elastic Properties Of Lipid Bilayers - Theory And Possible Experiments. Z. Naturforsch. C 1973, 28, 693−703. (17) Walde, P.; Cosentino, K.; Engel, H.; Stano, P. Giant vesicles: preparations and applications. ChemBioChem 2010, 11 (7), 848−865. (18) Discher, B. M.; Won, Y. Y.; Ege, D. S.; Lee, J. C. M.; Bates, F. S.; Discher, D. E.; Hammer, D. A. Polymersomes: Tough vesicles made from diblock copolymers. Science 1999, 284 (5417), 1143−1146. (19) Mai, Y.; Eisenberg, A. Self-assembly of block copolymers. Chem. Soc. Rev. 2012, 41 (18), 5969−5985. (20) Chambon, P.; Blanazs, A.; Battaglia, G.; Armes, S. P. Facile synthesis of methacrylic abc triblock copolymer vesicles by raft aqueous dispersion polymerization. Macromolecules 2012, 45 (12), 5081−5090. (21) Salva, R.; Le Meins, J. F.; Sandre, O.; Brulet, A.; Schmutz, M.; Guenoun, P.; Lecommandoux, S. Polymersome shape transformation at the nanoscale. ACS Nano 2013, 7 (10), 9298−9311. (22) Yoshida, E. PH response behavior of giant vesicles comprised of amphiphilic poly(methacrylic acid)-block-poly(methyl methacrylaterandom-mathacrylic acid). Colloid Polym. Sci. 2015, 293 (2), 649−653. (23) Kim, K. T.; Zhu, J. H.; Meeuwissen, S. A.; Cornelissen, J.; Pochan, D. J.; Nolte, R. J. M.; van Hest, J. C. M. Polymersome stomatocytes: controlled shape transformation in polymer vesicles. J. Am. Chem. Soc. 2010, 132 (36), 12522−12524. (24) Meeuwissen, S. A.; Kim, K. T.; Chen, Y. C.; Pochan, D. J.; van Hest, J. C. M. Controlled shape transformation of polymersome stomatocytes. Angew. Chem., Int. Ed. 2011, 50 (31), 7070−7073. (25) Zhang, Y. M.; Rock, C. O. Membrane lipid homeostasis in bacteria. Nat. Rev. Microbiol. 2008, 6 (3), 222−233. (26) Scheve, C. S.; Gonzales, P. A.; Momin, N.; Stachowiak, J. C. Steric pressure between membrane-bound proteins opposes lipid phase separation. J. Am. Chem. Soc. 2013, 135 (4), 1185−1188. (27) Zimmerman, S. B.; Minton, A. P. Macromolecular crowding biochemical, biophysical, and physiological consequences. Annu. Rev. Biophys. Biomol. Struct. 1993, 22, 27−65. (28) Keating, C. D. Aqueous phase separation as a possible route to compartmentalization of biological molecules. Acc. Chem. Res. 2012, 45 (12), 2114−2124. (29) Csonka, L. N. Physiological and genetic responses of bacteria to osmotic-stress. Microbiol. Rev. 1989, 53 (1), 121−147. (30) Needham, D.; Nunn, R. S. Elastic-deformation and failure of lipid bilayer-membranes containing cholesterol. Biophys. J. 1990, 58 (4), 997−1009. (31) Ertel, A.; Marangoni, A. G.; Marsh, J.; Hallett, F. R.; Wood, J. M. Mechanical properties of vesicles. I. Coordinated analysis of osmotic swelling and lysis. Biophys. J. 1993, 64 (2), 426−434. (32) Mui, B. L.; Cullis, P. R.; Evans, E. A.; Madden, T. D. Osmotic properties of large unilamellar vesicles prepared by extrusion. Biophys. J. 1993, 64 (2), 443−453. (33) Peterlin, P.; Arrigler, V.; Haleva, E.; Diamant, H. Law of corresponding states for osmotic swelling of vesicles. Soft Matter 2012, 8 (7), 2185−2193. (34) Karatekin, E.; Sandre, O.; Guitouni, H.; Borghi, N.; Puech, P.-H.; Brochard-Wyart, F. Cascades of transient pores in giant vesicles: line tension and transport. Biophys. J. 2003, 84 (3), 1734−1749. (35) Peterlin, P.; Arrigler, V. Electroformation in a flow chamber with solution exchange as a means of preparation of flaccid giant vesicles. Colloids Surf., B 2008, 64 (1), 77−87. (36) Popescu, D.; Popescu, A. G. The working of a pulsatory liposome. J. Theor. Biol. 2008, 254 (3), 515−519. (37) Lin, C. M.; Wu, D. T.; Tsao, H. K.; Sheng, Y. J. Membrane properties of swollen vesicles: growth, rupture, and fusion. Soft Matter 2012, 8 (22), 6139−6150. (38) Oglęcka, K.; Rangamani, P.; Liedberg, B.; Kraut, R. S.; Parikh, A. N. Oscillatory phase separation in giant lipid vesicles induced by transmembrane osmotic differentials. eLife 2014, 3, e03695.

(39) Oglecka, K.; Sanborn, J.; Parikh, A. N.; Kraut, R. S., Osmotic gradients induce bio-reminiscent morphological transformations in giant unilamellar vesicles. Front. Physiol. 2012, 3.10.3389/ fphys.2012.00120 (40) Haleva, E.; Diamant, H. Critical swelling of particle-encapsulating vesicles. Phys. Rev. Lett. 2008, 101 (7), 078104. (41) Veatch, S. L.; Keller, S. L. Miscibility phase diagrams of giant vesicles containing sphingomyelin. Phys. Rev. Lett. 2005, 94 (14), 148101. (42) Barenholz, Y.; Thompson, T. E. Sphingomyelins in bilayers and biological membranes. Biochim. Biophys. Acta, Rev. Biomembr. 1980, 604, 129−158. (43) Portet, T.; Gordon, Sharona E.; Keller, Sarah L., Increasing membrane tension decreases miscibility temperatures; an experimental demonstration via micropipette aspiration. Biophys. J. 2012, 103 (8), L35−L37.10.1016/j.bpj.2012.08.061 (44) Uline, M. J.; Schick, M.; Szleifer, I., Phase behavior of lipid bilayers under tension. Biophys. J. 2012, 102 (3), 517−522.10.1016/ j.bpj.2011.12.050 (45) Hamada, T.; Kishimoto, Y.; Nagasaki, T.; Takagi, M. Lateral phase separation in tense membranes. Soft Matter 2011, 7 (19), 9061−9068. (46) Chen, D.; Santore, M. M. Large effect of membrane tension on the fluid−solid phase transitions of two-component phosphatidylcholine vesicles. Proc. Natl. Acad. Sci. U. S. A. 2014, 111 (1), 179−184. (47) Givli, S.; Giang, H.; Bhattacharya, K. Stability of multicomponent biological membranes. SIAM J. Appl. Math. 2012, 72 (2), 489−511. (48) Akimov, S. A.; Kuzmin, P. I.; Zimmerberg, J.; Cohen, F. S. Lateral tension increases the line tension between two domains in a lipid bilayer membrane. Phys. Rev. E 2007, 75 (1), 011919. (49) Koslov, M. M.; Markin, V. S. A theory of osmotic lysis of lipid vesicles. J. Theor. Biol. 1984, 109 (1), 17−39. (50) Chen, I. A.; Roberts, R. W.; Szostak, J. W. The emergence of competition between model protocells. Science 2004, 305 (5689), 1474−1476. (51) Schrum, J. P.; Zhu, T. F.; Szostak, J. W. The origins of cellular life. Cold Spring Harbor Perspect. Biol. 2010, 2 (9), 15. (52) Svetina, S.; Zeks, B. Shape behavior of lipid vesicles as the basis of some cellular processes. Anat. Rec. 2002, 268 (3), 215−225. (53) Hotani, H. Transformation pathways of liposomes. J. Mol. Biol. 1984, 178 (1), 113−120. (54) Sanborn, J.; Oglecka, K.; Kraut, R. S.; Parikh, A. N. Transient pearling and vesiculation of membrane tubes under osmotic gradients. Faraday Discuss. 2013, 161, 167−176. (55) Canham, P. B. Minimum energy of bending as a possible explanation of biconcave shape of human red blood cell. J. Theor. Biol. 1970, 26 (1), 61−76. (56) Sakashita, A.; Urakami, N.; Ziherl, P.; Imai, M. Three-dimensional analysis of lipid vesicle transformations. Soft Matter 2012, 8 (33), 8569− 8581. (57) Sheetz, M. P.; Singer, S. J. Biological-membranes as bilayer couples - molecular mechanism of drug-erythrocyte interactions. Proc. Natl. Acad. Sci. U. S. A. 1974, 71 (11), 4457−4461. (58) Kas, J.; Sackmann, E. Shape transitions and shape stability of giant phospholipid-vesicles in pure water induced by area-to-volume changes. Biophys. J. 1991, 60 (4), 825−844. (59) Bruckner, R. J.; Mansy, S. S.; Ricardo, A.; Mahadevan, L.; Szostak, J. W. Flip-flop-induced relaxation of bending energy: implications for membrane remodeling. Biophys. J. 2009, 97 (12), 3113−3122. (60) Miao, L.; Seifert, U.; Wortis, M.; Dobereiner, H. G. Budding transitions of fluid-bilayer vesicles - the effect of area-difference elasticity. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 1994, 49 (6), 5389−5407. (61) Deuling, H. J.; Helfrich, W. Curvature elasticity of fluid membranes - catalog of vesicle shapes. J. Phys. 1976, 37 (11), 1335− 1345. (62) Seifert, U. Configurations of fluid membranes and vesicles. Adv. Phys. 1997, 46 (1), 13−137. L

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

(63) Svetina, S.; Zeks, B. Membrane bending energy and shape determination of phospholipid-vesicles and red blood-cells. Eur. Biophys. J. 1989, 17 (2), 101−111. (64) Seifert, U.; Berndl, K.; Lipowsky, R. Shape transformations of vesicles - phase-diagram for spontaneous-curvature and bilayer-coupling models. Phys. Rev. A: At., Mol., Opt. Phys. 1991, 44 (2), 1182−1202. (65) Dobereiner, H. G.; Kas, J.; Noppl, D.; Sprenger, I.; Sackmann, E. Budding and fission of vesicles. Biophys. J. 1993, 65 (4), 1396−1403. (66) Yanagisawa, M.; Imai, M.; Taniguchi, T. Shape deformation of ternary vesicles coupled with phase separation. Phys. Rev. Lett. 2008, 100 (14), 4. (67) Julicher, F.; Lipowsky, R. Domain-induced budding of vesicles. Phys. Rev. Lett. 1993, 70 (19), 2964−2967. (68) Viallat, A.; Dalous, J.; Abkarian, M. Giant lipid vesicles filled with a gel: Shape instability induced by osmotic shrinkage. Biophys. J. 2004, 86 (4), 2179−2187. (69) Lim, H. W. G.; Wortis, M.; Mukhopadhyay, R. Stomatocytediscocyte-echinocyte sequence of the human red blood cell: Evidence for the bilayer-couple hypothesis from membrane mechanics. Proc. Natl. Acad. Sci. U. S. A. 2002, 99 (26), 16766−16769. (70) Mohandas, N.; Gallagher, P. G. Red cell membrane: past, present, and future. Blood 2008, 112 (10), 3939−3948. (71) Seeman, P. Transient holes in erythrocyte membrane during hypotonic hemolysis and stable holes in membrane after lysis by saponin and lysolecithin. J. Cell Biol. 1967, 32 (1), 55−70. (72) Pajic-Lijakovic, I. Erythrocytes under osmotic stress - modeling considerations. Prog. Biophys. Mol. Biol. 2015, 117 (1), 113−124. (73) Zade-Oppen, A. M. M. Repetitive cell ’jumps’ during hypotonic lysis of erythrocytes observed with a simple flow chamber. J. Microsc. 1998, 192, 54−62. (74) Stroka, K. M.; Jiang, H. Y.; Chen, S. H.; Tong, Z. Q.; Wirtz, D.; Sun, S. X.; Konstantopoulos, K. Water Permeation Drives Tumor Cell Migration in Confined Microenvironments. Cell 2014, 157 (3), 611− 623. (75) Szostak, J. W.; Bartel, D. P.; Luisi, P. L. Synthesizing life. Nature 2001, 409 (6818), 387−390.

M

DOI: 10.1021/acs.langmuir.5b04470 Langmuir XXXX, XXX, XXX−XXX