Membrane Remodeling During Viral Infection, Parasite Invasion, and Apoptosis; Components And Kinetics In Exocytosis

Joshua Zimmerberg, MD, PhD, Head, Section on Cellular and Membrane Biophysics
Paul S. Blank, PhD, Staff Scientist
Svetlana Glushakova, PhD, Staff Scientist
Vadim A. Frolov, PhD, Senior Research Fellow
Andrea Fera, PhD, Postdoctoral Fellow
Vladimir A. Lizunov, MS, Visiting Fellow
Julia Mazar, PhD, Visiting Fellow
Elena M. Mekhedov, MS, Visting Fellow
Pavel Bashkirov, MS, Guest Researcher
Alexander Chanturiya, PhD, Guest Researcher
Jane E. Farrington, MS, Guest Researcher
Glen Humphrey, PhD, Guest Researcher
Ivan Polozov, PhD, Guest Researcher
Shurong Yin, PhD, Guest Researcher
Gulcin Pekkurnaz, MS, Predoctoral Fellow
Anna Shnyrova, MS, Predoctoral Fellow
Ludmila Bezrukov, MS, Contractor

We study membrane mechanics, intracellular molecules, membranes, viruses, organelles, and cells in order to understand viral and parasite infection, exocytosis, and apoptosis. This year, we report on four projects. For malaria, we study the mechanisms by which proteases take part in the release of parasitic cells from erythrocytes, leading to our suggestion for a protease inhibitor therapy for malaria. Second, in enveloped viral disease, we discovered how lipids forming the envelope of the influenza virus gel at cooler temperatures, suggesting that the gel protects the virus during airborne transmission and explaining the winter-time occurrence of flu epidemics. Third, using electrophysiology and fluorescence microscopy, we found that cholesterol plays a 2-fold role in viral fusion mediated by the hemagglutinin of influenza: as a lipid curvature agent helping hemifusion and as a specific widener of the fusion pore. Fourth, we investigate the theoretical effect of protein wetting to determine its role in protein domain formation in cell membranes. We find that wetting can provide important features seen in vivo.

Irreversible effect of cysteine protease inhibitors on the release of malaria parasites from infected erythrocytes

Glushakova, Hama,¹ Blank, Humphrey, Kapnik,² Mazar, Zimmerberg; in collaboration with Hohmann-Marriott

Given the absence of an effective vaccine for malaria and Plasmodium falciparum’s resistance to many drugs, there is interest in developing inhibitors that target cysteine and serine proteases. Such drug inhibitors are known to interfere with the parasitic asexual life cycle by trapping maturing parasites in clusters and diminishing de novo infection of erythrocytes. A deeper understanding of the role of proteases in the pathophysiology of malaria is critically needed; despite decades of research, controversy still surrounds the proteases’ function in the malarial erythrocyte cycle, particularly in parasite release. Two recent reports identified two parasite proteases that may mediate the cascade of final cycle events—the cysteine protease dipeptidyl peptidase 3 (DPAP3) and the subtilisin-family serine protease PfSUB1—although the precise time and place of the proteases’ action in the release process remains enigmatic.

The release of parasites appears to be a two-step process. Parasites are released first from the erythrocyte’s parasitophorous vacuole within the erythrocyte and then from the erythrocyte. After breaching the parasitophorous vacuolar membrane and the erythrocyte membrane, the parasites are free to invade fresh erythrocytes. While the release of parasites is crucial for the parasite’s life cycle, the mechanism of release remains unknown, and only limited technologies exist for its analysis. Imaging of the release process in vitro shows that the process occurs rapidly and is highly sensitive to a variety of conditions, thus limiting possible experimentation in living cells at the end of the erythrocyte cycle. Wide-spectrum cysteine- and cysteine/serine-protease inhibitors such as E-64, bADA, and leupeptin (the last alone or in combination with chymostatin and antipain) block the cycle. Even though such treatment does not interfere with parasite maturation and thus permits analysis of the accumulated parasite clusters, it blocks rupture of one of the two membranes surrounding the parasites. Given disagreement as to the origin of the limiting membrane that preserves the remaining cluster, the order of membrane rupture during parasite release remains subject to debate.

We developed a new approach for the differential labeling of the membrane of live infected erythrocytes as well as a quantitative parasite release assay, coupled with morphological analysis of live infected cells undergoing a cycle transition. Using confocal microscopy of living cells and electron microscopy of high-pressure–frozen and freeze-substituted cells to study inactivation of malaria parasite culture by cysteine protease inhibition, we confirmed that cysteine proteases likely regulate the opening of the erythrocyte membrane, causing release of malaria parasites. Inhibition of cysteine proteases within the last few minutes of cycle does not affect rupture of the parasitophorous vacuole but irreversibly blocks the subsequent rupture of the host cell membrane, thereby locking in resident parasites, which die within a few hours of captivity. In contrast to the findings of Salmon and colleagues, both reversible and irreversible cysteine protease inhibitors irreversibly inactivate clustered parasites. Such irreversible inactivation of mature parasites inside host cells makes plasmodial cysteine proteases attractive candidates for antimalarials, as parasite-specific cysteine protease inhibitors may significantly augment multitarget drug cocktails.

It is likely that cysteine proteases suddenly enter the erythrocyte cytoplasm upon disruption of the parasitophorous vacuole membrane. Once active in the compartment facing the plasma membrane of the erythrocyte, the proteases disrupt the plasma membrane. In the presence of protease inhibitor, parasites degrade and die within clusters. It is unlikely that E-64 has a direct toxic effect on parasites because clustered parasites locked inside erythrocytes in E-64–containing medium have approximately the same life span as parasites in normal medium in vitro. We believe that E-64 is deleterious to parasites by fatally prolonging the otherwise transient cytoplasmic “cluster stage” of parasites that have broken through the vacuolar membrane. Given the short life span of mature parasites, prolongation of the cluster stage leads to irreversible blockage of the plasmodial erythrocyte cycle. The withdrawal of irreversible and reversible cysteine protease inhibitors does not restart parasite release from clusters as previously suggested but rather permits schizonts to release parasites in the future. Thus, a dosing schedule, as informed by our studies, may be important for testing protease inhibitors as antimalarial drugs. Hopes for antimalarial protease inhibitor drugs are augmented by the finding that relatively low concentrations of calpeptin, a reversible inhibitor, had a potent inhibitory effect on parasite release.

Chen SS, Fitzgerald W, Zimmerberg J, Kleinman HK, Margolis L. Cell-cell and cell-extracellular matrix interactions regulate embryonic stem cell differentiation. Stem Cells 2007;25:553-561.
Glushakova S, Yin D, Gartner N, Zimmerberg J. Quantification of malaria parasite release from infected erythrocytes: inhibition by protein-free media. Malar J 2007;6:61-66.

Progressive ordering with decreasing temperature of the phospholipids of influenza virus

Farrington, Fera, Blank, Frolov, Bezrukov, Polozov, Zimmerberg; in collaboration with Gawrisch, Hess, Reese

Membranes of most enveloped viruses form by budding out a highly select subset of plasma membrane components from the plasma membranes of the viruses’ host cells. Compared with the plasma membrane, the envelope of influenza contains higher amounts of both cholesterol and glycosphingolipids, lipids known to partition into the liquid-ordered phase (l_o) and characterized by extended hydrocarbon chains with a lower gauche-trans isomerization than the liquid disordered (l_d) phase, but with similar rotational and translational mobility.

Using NMR,³ we found that phospholipids in viral membranes form ordered lipid phases over a wide range of physiologically relevant temperatures. We have evidence that both l_o and gel (s_o) phases co-exist with the l_d phase. At 37°C, lipids with broadened spectra (the sum of l_o and s_o phases) represent a minute fraction of the membrane, whereas, at 4°C, almost the entire membrane is in the ordered state. These data rule out the hypothesis that the influenza envelope is created entirely from an ordered lipid domain. We observed co-existence of ordered and disordered lipid domains in both the intact viral envelope and liposomes prepared from viral lipids. A reduction in cholesterol reversibly increases gel-phase lipids. It is remarkable that, despite the wide temperature dependence of the fraction of ordered domains, the extracted lipids exhibit the same fraction of ordered domains as the intact virus at any given temperature, suggesting that the properties of membrane lipids primarily determine the fraction of ordered domains. Thus, proteins may play only a limited role in the phase behavior of lipids in the intact viral envelope, although they may affect line tension at boundaries between co-existing phases.

Three independent lines of evidence support the notion of solid ordered domains at ambient temperatures. First, in liposomes formed from viral lipids, gel lipids were confined to areas smaller than the size of the liposomes and exhibited the characteristics of solid-ordered domains; this finding is similar to our earlier observations for gel phase–fluid phase co-existence in SOPC-POPE⁴ lipid mixtures. Second, the ragged edges seen by fluorescence in both supported viral lipid bilayers and giant unilamellar vesicles suggest that crystalline packing shapes domains. Third, we assigned methylene resonances wider than 1.5 kHz (3 ppm) to the solid-ordered lipid phase. The presence of such broad components is visible in the first-order spinning side band of the virus spectrum even at temperatures higher than ambient. While we may not entirely exclude the possibility that protein resonances contribute to those spectra, the rapid increase in their intensity with decreasing temperature confirms that they are dominated by contributions from lipids. Taken together, the above evidence strongly supports the presence of s_o. We speculate that s_o phase formation is related to the high phosphatidylethanolamine (PE) content of viral envelopes; PE is a lipid with low affinity for cholesterol and high gel-fluid phase transition temperatures.

At physiological temperature and higher, MAS⁵ NMR provides little evidence for ordered lipid, and, even though we continue to assay for fusion to cell membranes, we have not been able to detect ordered lipid in the host or target membrane. Thus, phase properties of complex, biologically relevant lipid mixtures need to be studied at the appropriate temperatures. An issue of particular interest is why the viral envelope lipid composition is set to be near an apparent lipid phase boundary with respect to temperature. It may be instructive to consider that the lipid composition of the influenza virus differs dramatically from that of vesicular stomatitis virus (VSV), another virus that buds from the same cells. Both enveloped viruses bud from the plasma membrane, albeit from different locations. The difference in location relates to the viruses’ mode of host-to-host transmission—apical budding for aerial transmission of influenza and serosal budding for transmission through animal bite for VSV (and for rabies, the other virus of the Rhabdoviridea pathogenic genus). A virus would be at room temperature during aerial transmission but not during transmission by animal bite; the ordered phases we documented here may be important for stability. Indeed, a recent report shows that airborne transmission of influenza virus by guinea pigs increases at lower temperatures, a result predicted by our proposal that the progressive ordering of lipids with decreasing temperature is important during the low-temperature stages of the influenza life cycle.

Hess ST, Gould TJ, Gudheti MV, Maas SA, Mills KD, Zimmerberg J. Dynamic clustered distribution of hemagglutinin resolved at 40 nm in living cell membranes discriminates between raft theories. Proc Natl Acad Sci USA 2007;104:17370-17375.
Polozov IV, Bezrukov L, Gawrisch K, Zimmerberg J. Progressive ordering with decreasing temperature of the phospholipids of influenza virus. Nat Chem Biol 2008;4:248-255.
Shnyrova AV, Ayllon J, Mikhalyov II , Villar E, Zimmerberg J, Frolov VA. Vesicle formation by self-assembly of membrane-bound matrix proteins into a fluidlike budding domain. J Cell Biol 2007;179:627-633.

Cholesterol promotes hemifusion and pore widening in membrane fusion induced by influenza hemagglutinin

Biswas,⁶ Shnyrova, Frolov, Yin, Pekkurnaz, Zimmerberg

Successful infection by influenza virus requires the envelope spike protein hemagglutinin (HA) both to catalyze fusion between the viral envelope and the target cell’s intracellular endosomal membrane and to create a pore large enough to release the viral genome. Experiments on and theory of lipid composition in relationship to membrane monolayer curvature stress strongly suggest that membrane lipids play a role in this critical event. Recently, attention has focused on the role of membrane phase behavior and membrane microdomains in the lateral distribution, sorting, and interactions of lipids with membrane proteins in general and with viral envelope glycoproteins in particular. Cholesterol is a major and vital constituent of eukaryotic cell membranes. Its unique structure—a small hydrophilic head group and rigid, hydrophobic fused rings—favors preferential association with saturated acyl-chain lipids and sphingolipids to form liquid-ordered microdomains (termed lipid rafts) in phospholipid bilayer membranes of the appropriate composition. Lipid rafts are hypothesized to exist in the cell plasma membrane at specialized sites where proteins concentrate, creating favorable associations with the ordered, cholesterol-rich environment.

Cholesterol regulates specific intermediates of membrane fusion catalyzed by the influenza virus protein HA. The extent of early lipid transfer in Sf9 cells expressing HA (HAS cells) is similar to that previously observed in mammalian cell systems, but the initial fusion pore is small and pore expansion is stunted. A 3-fold increase in cellular cholesterol leads to (1) faster lipid dye transfer kinetics, (2) increased aqueous dye transfer kinetics and extent, and (3) an increase in the rate of pore conductance. The cholesterol-dependent increase in fusion efficiency requires an intact HA transmembrane domain (TMD) and optimal pH. Overall, our results support the hypothesis that host-cell cholesterol acts at two stages in membrane fusion: at an early, lipidic stage before fusion pore opening and at a later stage during fusion pore expansion.

We are interested in how the physical properties of cholesterol influence fusion. We have shown that cholesterol promotes both lipid transfer (hemifusion) and fusion pore expansion in the cell-cell membrane fusion mediated by influenza HA . We hypothesize that cholesterol promotes fusion pore expansion by virtue of its negative intrinsic curvature and the same specific cholesterol/lipid/HA interactions that mediate (1) the 1 to 10 nm scale clustering of HA in the plane of the membrane, (2) the mobility of HA in fibroblasts, and (3) the phase behavior of the influenza envelope. These structural forces act at the fusion stage of viral invasion to facilitate fusion pore widening.

Biswas S, Yin S-R, Blank PS , Zimmerberg J. Cholesterol promotes hemifusion and pore widening in membrane fusion induced by influenza hemagglutinin. J Gen Physiol 2008;131:503-513.
Frolov VA, Zimmerberg J. Flexible scaffolding made of rigid BARs. Cell 2008;132:727-729.
Plonsky I, Kingsley DH, Rashtian A, Blank PS, Zimmerberg J. Initial size and dynamics of viral fusion pores are a function of the fusion protein mediating membrane fusion. Biol Cell 2008;100:377-386.
Shnyrova A, Frolov VA, Zimmerberg J. ER biogenesis: self-assembly of tubular topology by protein hairpins. Curr Biol 2008;18:R474-6.

Domain formation in membranes caused by lipid wetting of protein

Frolov, Zimmerberg; in collaboration with Akimov, Chizmadzhev, Cohen

The classic fluid mosaic model views the lipid environment of a plasma membrane as essentially homogeneous. Developments over the past decade, however, have increasingly shown that nonhomogeneities in membranes are central to biological functions that depend on protein-protein interactions within membranes. Given that lipid and protein interactions mediated by hydrophobic, van der Waals, electrostatic, and chemical forces cause some lipids and proteins to cluster into domains and others to repel, it is obvious that membrane non-uniformities must exist. Over 30 years ago, researchers demonstrated by electron spin resonance the existence of small spatial inhomogeneities. They found that “boundary lipids” surrounding a protein exhibit about a 10-fold lower hop time than lipids not associated with proteins. In our project, formation of rafts and other domains in cell membranes is considered as wetting of proteins by lipids, with the membrane modeled as a continuous elastic medium. Our approach yields the conditions necessary for a macroscopic wetting film to form and permits us to determine the film’s thickness.

Using a mean-field theory of liquid crystals as adapted to biomembranes, we calculated thermodynamic functions of wetting lipid films. We showed that either molecular or macroscopic films can form, depending on the values of parameters such as membrane thickness, hydrophobic height mismatches, spontaneous curvature of lipids, and protein radius. We demonstrated that a single protein of approximately 1 nm radius is not large enough to induce a local phase transition to form a protein-lipid raft but that a macroscopic wetting film can form around a lipid/protein aggregate of more than tens of nanometers in diameter. We have assumed that the lipids in such aggregates are in a liquid-ordered state, analogous to wetting of a solid surface. Calculations and simulations in the context of the capillary wave model of Mouritsen support the validity of our assumption. Wetting films that coat an aggregate could be particularly important because they facilitate the merger of domains. Moreover, a wetting film prevents a protein from leaving an aggregate, thereby promoting protein accumulation and clustering.

Akimov SA , Frolov VA, Kuzmin PI , Zimmerberg J, Chizmadzhev YA, Cohen FS. Domain formation in membranes caused by lipid wetting of protein. Phys Rev E Stat Nonlin Soft Matter Phys 2008;77:051901.
Akimov SA, Kuzmin PI, Zimmerberg J, Cohen FS. Lateral tension increases the line tension between two domains in a lipid bilayer membrane. Phys Rev E Stat Nonlin Soft Matter Phys 2007;75:011919.

¹Erinn Hama, BA, former Summer Student
²Elena M. Kapnik, MS, former Biologist
³nuclear magnetic resonance
⁴1-stearoyl-2-oleoyl-phosphatidylcholine–1-palmitoyl-2-oleoyl-phosphatidylethanolamine
⁵magic angle spinning
⁶Subrata Biswas, PhD, former Visiting Fellow, now at The Johns Hopkins University School of Medicine, Baltimore, MD

Collaborators

Sergey Akimov, MS, Frumkin Institute of Electrochemistry, Russian Academy of Sciences, Moscow, Russia
Yuri Chizmadzhev, PhD, Frumkin Institute of Electrochemistry, Russian Academy of Sciences, Moscow, Russia
Fredric S. Cohen, PhD, Rush Medical College, Chicago, IL
Samuel W. Cushman, PhD, Diabetes Branch, NIDDK, Bethesda, MD
Klaus Gawrisch, PhD, Laboratory of Membrane Biochemistry and Biophysics, NIAAA, Bethesda, MD
Samuel T. Hess, PhD, University of Maine, Orono, ME
Martin Hohmann-Marriott, PhD, Laboratory of Bioengineering and Physical Science, NIBIB, Bethesda, MD
Thomas S. Reese, MD, Laboratory of Neurobiology, NINDS, Bethesda, MD

For further information, contact zimmerbj@mail.nih.gov.