Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Biochemistry. Author manuscript; available in PMC 2010 September 1.
Published in final edited form as:
PMCID: PMC2774806

Calcium-dependent lateral organization in phosphatidylinositol (4,5) bisphosphate (PIP2)- and cholesterol-containing monolayers


Biological membrane function depends in part on the local regulation of lipid composition. The spatial heterogeneity of membrane lipids has been extensively explored in the context of cholesterol and phospholipid acyl-chain dependent domain formation, but the effects of lipid headgroups and soluble factors in lateral lipid organization are less clear. In this contribution, the effects of divalent calcium ions on domain formation in monolayers containing phosphatidylinositol (4,5) bisphosphate (PIP2), a polyanionic, multi-functional lipid of the cytosolic leaflet of the plasma bilayer, are reported. In binary monolayers of PIP2 mixed with zwitterionic lipids, calcium induced a rapid, PIP2-dependent surface pressure drop, with the concomitant formation of laterally-segregated, PIP2-rich domains. The effect was dependent on headgroup multivalency, as lowered pH suppressed the surface pressure effect and domain formation. In accord with previous observations, inclusion of cholesterol in lipid mixtures induced coexistence of two liquid phases. Phase separation strongly segregated PIP2 to the cholesterol-poor phase, suggesting a role for cholesterol-dependent lipid demixing in regulating PIP2 localization and local concentration. Similar to binary mixtures, subphase calcium induced contraction of ternary cholesterol-containing monolayers; however, in these mixtures calcium induced an unexpected, PIP2- and multivalency-dependent decrease in the miscibility phase transition surface pressure resulting in rapid dissolution of the domains. This result emphasizes the likely critical role of subphase factors and lipid headgroup specificity in the formation and stability of cholesterol-dependent domains in cellular plasma membranes.

Membrane lipids, especially polyphosphoinositides such as phosphatidylinositol 4,5-bisphosphate (PIP2), are important effectors in many cellular processes including apoptosis (1), inflammation (2), motility (3) and proliferation (4, 5). In addition to the concentration and functionality of individual lipids, the spatial organization of lipids affects how they interact with proteins at the membrane/cytosol interface. The effects of acyl chain structure and cholesterol content (6, 7) have been extensively studied as determinants of lateral organization in the context of the cholesterol-dependent formation of coexisting liquid phases. Headgroup-mediated lipid interactions and the influence of cytosolic factors have been less thoroughly investigated, but they may be important for lipids with unusually large electrostatic charge such as polyphosphoinositides (8).

Many eukaryotic cell proteins involved in cytoskeletal remodeling (3, 9), membrane trafficking (10, 11), transmembrane permeability (12, 13) and mitogenesis (14, 15) are regulated in vitro by phosphatidylinositol (4,5) bisphosphate (PIP2), a highly anionic lipid distributed largely in the inner leaflet of the plasma membrane and within the nucleus. When PIP2-regulated proteins were first reported (16), a plausible hypothesis was that these proteins function in a manner analogous to the specific binding of a soluble regulator to a protein active site. However, the list of PIP2-regulated proteins of various functions has risen to the point that the cellular concentration of PIP2 ligands is greater than the cellular concentration of PIP2 (3, 17). Also inconsistent with a simple mass action binding equilibrium between PIP2 and its protein ligands is the fact that cellular PIP2 distribution or turnover changes much more than total PIP2 levels (18, 19) during activations of cells where various PIP2-regulated processes are triggered. These and other findings suggest that lipid distribution and lateral organization, rather than global concentration, determine where and when the lipid binds its target protein.

A potentially important feature of PIP2 signaling is the varying functionality of PIP2 within regions of the membrane with different lateral organization, i.e. different roles for single, freely-diffusing PIP2 molecules, small transient clusters of PIP2, and large stable PIP2 aggregates. A central tenet of this hypothesis is that multiple pools of PIP2 are spatially, structurally and functionally distinct, but rapidly convert between the different organization states depending on membrane proximal factors and the local lipid environment. Evidence for this hypothesis includes the discovery of fractions of PIP2 in the plasma membrane that are inaccessible to a PIP2-binding PH domain (20) as well as differential turnover of labeled PIP2 probes (19). Consistent with the existence of spatially segregated pools of PIP2, immunofluorescence microscopy studies of various PIP2-binding domains (21, 22) and antibodies (23, 24) have confirmed the non-homogeneous distribution of PIP2 in the plasma membrane. A protein-based mechanism for PIP2 oligomerization has been proposed, suggesting that PIP2 lateral heterogeneity is a function of electrostatic interactions between several neighboring lipids and an unstructured polybasic protein domain, such as found in the MARCKS protein (21, 25-27). In this model, the protein is required for PIP2 clustering and serves to prevent interaction of PIP2 with other potential protein targets until the sequestering proteins dissociate from PIP2 as the result of signal-dependent post-translation modification such as phosphorylation by protein kinase C (28).

An alternative or complementary mechanism for formation of lateral PIP2 aggregates in lamellar membranes is suggested by evidence that PIP2 headgroups, unlike nearly all other phospholipids (PL), can form hydrogen-bonded networks (29, 30). Additionally, it has been suggested that divalent cations can not only reduce the electrostatic repulsion between the anionic PIP2 headgroups but also act as bridges between two adjacent lipids (29), as has been recently observed in model liposomes where macromolecular aggregates of PIP2 were induced by both Ca2+ and Mg2+ (48). Macroscopic polycation-mediated domain formation has been observed in monolayers containing synthetic anionic lipids (31, 32), but these observations have been made with singly-charged lipids that contain two saturated acyl chains (DPPS/DPPA), both factors that would be expected to affect lateral organization.

In the present study, a lipid monolayer system is used to investigate the formation and electrostatic mechanisms underlying calcium-induced PIP2 domains and the effect of PIP2 clustering on phase coexistence in cholesterol-containing lipid mixtures. PIP2 clustering is shown to depend on the multivalency of the counterion, as well as the high charge density of the lipid headgroup. Additionally, calcium induces a significant reduction in the phase coexistence surface pressure in PIP2-containing monolayers. These effects demonstrate that lateral organization of lipids can be modulated by cytosolic factors on several length scales, and suggests that cellular control of concentrations of ions such as Ca2+ can be used to remodel lipid organization and thereby impact multiple cell signaling pathways.

Experimental Procedures


Bovine liver L-α-phosphatidylinositol-4,5-bisphosphate (PIP2), cholesterol, 1-stearoyl-2-oleoyl phosphatidylcholine (SOPC), 1-stearoyl-2-oleoyl phosphatidylserine (SOPS), 1-oleoyl-2-NBD-phosphatidylserine, and 1,2-dioleoyl-3-phosphoethanolamine-N-lissamine rhodamine B sulfonyl from Avanti (Alabaster, AL) and long-chain analogs of PIP2 (Bodipy FL-PIP2 and NBD-PIP2) from Echelon Inc. (Salt Lake City, UT) were obtained as organic solutions and stored at -20°C. The concentrations of the lipid solutions were confirmed initially with phosphate analysis following acid digestion of organic components (33) and subsequently by comparing to the measured area per lipid molecule. Subphase reagents (HEPES, EDTA, CaCl2, and NaCl) were purchased from Sigma (St. Louis, MO).

Monolayer visualization and manipulation

30 mL of subphase solution (10 mM HEPES, 1 μM EDTA, pH 7.5 in 18.2 MΩ ddH2O) were added to a MicroTroughX Langmuir trough (Kibron Inc. Helsinki, Finland). For varying pH experiments, the buffer was 3.3 mM sodium phosphate, 3.3 mM sodium citrate and 3.3 mM glycine. To avoid oxidation-induced effects, 5 mM DTT was added to the subphase for the fluorescence visualization of cholesterol-containing monolayers. No effect of the various buffers at the same pH and ionic strength was observed. Approximately 7 nmol of pre-mixed lipid solutions were deposited slowly on the subphase surface and the lipids were compressed at 15 Å2/molecule/min to the desired surface pressure by the barriers of the trough using a microstepping motor. The monolayer surface pressure was monitored with a surface probe using the Wilhelmy method (33) and the FilmWare software package (Kibron).

Subphase calcium was introduced after the monolayers were compressed to 20mN/m (unless otherwise noted) by injection from behind one of the barriers (i.e. without disturbing the monolayer) with a gel-loading pipette tip (Sigma). The subphase was then mixed by withdrawing and expelling 300 μL of the subphase from behind each barrier 5×. Uniform mixing was verified using a soluble dye, and control experiments showed no lasting effect of mixing on the resultant surface pressure (fluctuations due to subphase movement during mixing are observed as spikes in Fig. 3B).

Figure 3
Subphase calcium induces immediate, reversible, pH-dependent surface pressure decrease in PIP2 containing monolayers

Monolayers were imaged by epifluorescence microscopy using an inverted microscope (Leica Microsystems, Wetzlar, Germany) with the appropriate filter sets.

PIP2 binding markers

cDNA of a GST chimera of the PH domain from rhPLCδ1 (gift from Dr. Tobias Baumgart) was expressed in XL1-blue E. coli and purified on glutathione-functionalized Sepharose beads using the provided protocol (Sigma). The purified GST-PH was fluorescently labeled by reaction with fluorescein isothiocyanate (FITC) for 30 minutes at room temperature. The reaction time was carefully controlled because FITC reacts with solvent-accessible primary amines on the surface of proteins, typically lysines and arginines, the same residues known to make up the PIP2 binding pocket of PH-PLC (34). For this reason, the labeling ratio of FITC:PH domain was minimized to maintain PIP2-binding activity and measured by optical absorbance of protein and label to be ~1:1. The PIP2-binding-peptide (PBP10) based on the sequence of the PIP2 binding site of the actin-binding protein gelsolin functionalized with a rhodamine B fluorescent group was synthesized previously (35).


The effects of calcium ions added to the subphase of preformed unsaturated PL monolayers containing PIP2 were measured using a Langmuir film balance technique and epifluorescence microscopy. The lateral organization of PIP2-containing monolayers was visualized by the inclusion of an acyl-chain labeled fluorescent PIP2 analog (1% NBD-PIP2). Addition of Ca2+ underneath a PIP2-containing monolayer induced the formation of small, bright domains enriched in fluorescent PIP2 (Fig. 1A) that persisted to physiological values of surface pressure (π ~ 32 mN/m). Coincident with domain formation, a drop in monolayer surface pressure ([product]) was observed, with the magnitude of Δ[product] upon calcium addition directly related to PIP2 fraction (Fig. 1B). The linear plot (R2 = 0.995) of Δ[product] and [PIP2] passes through the origin suggesting that the observed effects are present even at the small PIP2 mol fractions that would occur in cell membranes assuming random distribution of all lipids. Similarly, the abundance of bright domains increased significantly with increasing PIP2 fraction (Fig. 1A) with this effect also persisting down to lower PIP2 concentrations that are physiologically relevant.

Figure 1
Calcium-induced change in surface pressure and domain formation linearly depends on PIP2 fraction

Calcium-induced PIP2-enriched domains

At pH 7.5, the domains appeared quickly (< 10 mins) as abundant, sub-micron bright spots on a dark background, and coarsened into micron-scale ribbons after several hours. Other shapes - from larger dots to several-micrometer circles - were observed as a function of pH (Fig. 2A). Once formed, PIP2-enriched domains persisted through a significant range of monolayer surface pressures ([product] = 20 – 40 mN/m) and for the entire durations of experiments (up to 24 hours).

Figure 2
Divalent calcium induces microscopic PIP2 domains in mixed lipid monolayers

Domains were only observed under conditions at which multivalent lipids were present in the monolayer (PIP2 at pH 7.5, 6, and 4.5). PIP2-containing monolayers at low pH (3.0) and monolayers where PIP2 was replaced by the monoanionic lipid SOPS both showed small changes in surface pressure upon calcium addition (Fig. 3), but were fluorescently homogeneous (Fig. 2A).

To ensure that the bright domains observed following calcium injection were PIP2-enriched membrane structures as opposed to membrane defects, surface contaminants, or dye-dependent artifacts, a different fluorescent PIP2 analog (BodipyFL-PIP2) was visualized in parallel with a non-specific membrane dye (rhodamine-labeled SOPE; rho-SOPE). The distribution of BodipyFL-PIP2 mirrored the punctate domains of NBD-PIP2, whereas no lateral heterogeneity was observed in the distribution of rho-SOPE (Fig. 2B), confirming that the domains were indeed PIP2-enriched lateral domains.

Calcium-induced PIP2-dependent pressure drop

Concomitant with the observed domains, mixed monolayers of SOPC:PIP2 (3:1) show reduced surface pressures across a large range of molecular areas in the presence of subphase divalent cations (Fig. 3A). This effect is divalent cation specific, as a much greater reduction was observed with Ca2+, as compared to Mg2+ Similarly, injection of 1 mM CaCl2 to the buffered subphase (pH 7.5) underneath monolayers of the same composition induced an immediate and substantial contraction of the monolayer, evidenced by ~25% reduction in surface pressure ([product]) (Fig. 3B, top). The time scale of the observed surface pressure decrease was not measurable in our experiments, but was shorter than the ~10 secs required for Ca2+ introduction and mixing. This contraction was reversible, as chelation of Ca2+ by excess EDTA led to an equivalent increase in [product] (Fig. 3B, bottom; spikes indicate transient changes in subphase volume induced by mixing). The concentration dependence of the calcium-induced monolayer contraction at pH 7.5 (Fig. 3C, circles) showed that the effect persisted to 1 μM Ca2+ and could be fit by a saturable binding model with a dissociation constant of Kd ~ 3 μM. A similar magnitude of monolayer contraction and formation of domains were observed at increased salt concentrations (150mM NaCl), although the calcium level required was ~10-fold higher (data not shown) likely due to the screening of electrostatic interactions by subphase counterions.

Whereas the magnitude of the contraction was dependent on the protonation state of the PIP2 in the monolayer (as controlled by subphase pH), the dissociation constant was nearly pH independent (Fig. 3C). The exception to this observation was at pH 3.0, where PIP2 would be expected to be monovalent (8) and no significant change in surface pressure was observed upon addition of Ca2+ up to 1 mM (Fig. 3C – diamonds). Similarly, no change in surface pressure occurred when calcium was added to monolayers including the monovalent lipid, SOPS, or without charged lipid (Fig. 3D).

PIP2-mediated calcium effects in cholesterol-containing monolayers

Inclusion of cholesterol in PL monolayers results in large circular domains indicative of fluid-fluid phase coexistence that can be readily visualized by inclusion of a small concentration of a tracer lipid with a non-unity phase partition coefficient (e.g. Fig. 4E - tracer lipid is 0.1% rho-SOPE). To determine the phase partitioning of PIP2 in these monolayers, the localization of a variety of PIP2 tracers was assayed with respect to PL-rich phase markers (rho-SOPE and NBD-PC). The different tracers were chosen to control for potential artifacts of each label. Comparison of fluorescent images of two distinct PIP2 analogs (NBD-PIP2, Fig. 4A; BodipyFL-PIP2, Fig. 4B), a fluorescently-labeled PIP2 binding protein domain (FITC-PLC δ1, Fig. 4C), and a rhodamine-labeled PIP2-binding peptide corresponding to the PIP2-binding domain of gelsolin (PBP10; Fig. 4D) with known PL-rich phase markers (rho-SOPE and NBD-PC, Fig. 4E-H) clearly demonstrate the preferential partitioning of PIP2 into the Chol-poor phase of these biphasic monolayers.

Figure 4
PIP2 co-localizes with PL-rich phase markers in fluid-fluid coexistence monolayers

In addition to the PIP2 segregating effect of cholesterol described above, PIP2 induced a change in the miscibility transition pressure of cholesterol-containing monolayers that was strongly dependent on subphase calcium concentration. As shown in Fig. 5A and extensive previous work (6, 36), cholesterol-containing PL monolayers undergo a surface pressure-dependent transition from coexisting fluid phases at low pressure to a single homogeneous phase at a composition-dependent miscibility transition pressure ([product]t). Although [product]t has been previously demonstrated to be strongly dependent on cholesterol mole fraction (7), PL acyl chain composition, and physical parameters like temperature and applied electric fields (37), there has been little evidence for a role of subphase factors in modulating the stability of fluid phase coexistence. Fig. 5 shows that addition of 1 mM Ca2+ lowers the [product]t of a DChol:PIP2:SOPC (1.4:1:1.6) monolayer from 6.1 to 3.0 mN/m. The same effect was demonstrated by injecting Ca2+ underneath a preformed biphasic monolayer, which led to dissolution of domains, concomitant with the expected pressure drop for PIP2 containing monolayers described in Figs. 1 and and33 (Fig. 5B). Although the pressure drop and dissolution of domains were simultaneous, they are not causative, as a decrease in pressure would be expected to stabilize phase coexistence (38), as opposed to the observed monolayer homogenization.

Figure 5
Subphase calcium decreases monolayer [product]T in cholesterol containing monolayers

Recalling the multivalency requirement of the calcium-induced pressure drop and PIP2-domain formation described in Figs. 1--3,3, the calcium-induced transition pressure effect was also dependent on both the presence and multivalency of PIP2 in the monolayer. When no PIP2 was present in the monolayer, essentially no effect of subphase calcium was observed on 35% DChol monolayers (Fig. 6A and B). Inclusion of PIP2 led to a >60% decrease of [product]t (from 6.8 to 2.4 mN/m) upon addition of 1 mM calcium (Fig. 6A and B). [product]t in the presence of PIP2 (without calcium) was significantly higher than that of a binary SOPC/DChol mixture (Fig. 6A) and the addition of Ca2+ restored [product]t to the PIP2-free level. The presence of monovalent anionic lipids (PIP2 at pH 3 or SOPS at pH 7.5) had no significant effect either on the calcium-free [product]t, compared to PC/DChol monolayers, nor on the Δ[product]t[product]t = [product]t -[product]t,Ca) following Ca2+ addition. Finally, the dependence of these phenomena on cholesterol and PIP2 fraction was evaluated. The Ca2+-induced decrease in [product]t was not greatly affected by varying [DChol] from 35% to 50% (Fig. 6C), but varying the amount of PIP2 in the monolayer had a significant effect. Δ[product]t was directly related to [PIP2] with the effect persisting down to <10% PIP2, and the linear relationship (R2 = 0.952) and zero intercept imply that the observed effects are present at PIP2 concentrations too low to measure with this method (Fig. 6D).

Figure 6
Presence of subphase Ca2+ affects [product]T only in monolayers containing highly charged lipids


The data presented here demonstrate the multiple and varied effects of subphase calcium on PIP2-containing phospholipid monolayers. These effects include significant calcium-induced condensation of PIP2/SOPC monolayers, as evidenced by the reduction of monolayer surface pressure, and the concomitant formation of sub-micron to micron-scale PIP2-enriched domains. Experiments with phase separated cholesterol-containing monolayers showed PIP2 to be strongly enriched in the Chol-depleted phase, in which the addition of subphase calcium induced a significant reduction of the miscibility transition surface pressure. These effects suggest a role for changes in membrane structure as a possible effector of calcium signaling in biological systems.

The reduction of surface pressure of anionic monolayers by subphase multivalent cations has been previously reported (39, 40), and is consistent with the substantial PIP2-dependent decreases in surface pressure upon injection of Ca2+ beneath preformed monolayers of polyunsaturated PIP2 and SOPC (Fig. 1 and and3).3). The magnitude of the decrease in these mixed monolayers was roughly proportional to the calcium-induced contraction observed for pure PIP2 monolayers (29), and was attributed to aggregation (41), neutralization (42) and dehydration (43) of the inositol headgroups by direct electrostatic interaction between the cations and the anionic phosphomonoesters of PIP2 (42). Ca2+ had a much stronger condensing effect than Mg2+ (Fig. 3A), suggesting a specific affinity between the bisphosphorylated headgroup of PIP2 and Ca2+. Although the magnitude of monolayer contraction was strongly dependent on subphase pH (and subsequently, PIP2 charge), we measured a pH-independent binding affinity (Kd ~ 3 μM; Fig. 3C) of the calcium ions to the monolayers, suggesting surface charge independent binding between calcium and PIP2. Both of these observations correspond well to previous measurements of calcium binding to anionic phospholipids and bacterial polysaccharides in monolayers (44).

Both ion and lipid headgroup multivalency is required for the observed monolayer contraction. Neither monolayers of pure zwitterionic PC nor those containing 25% of the monovalent lipid PS showed any significant decrease of surface pressure upon addition of calcium, and addition of monovalent cations did not induce contraction of PIP2 containing monolayers, as previously shown (29). Similarly, monolayers containing PIP2 showed no effect of Ca2+ on surface pressure at low pH, where PIP2 would not be expected to be polyvalent (although experimental pKa values for PIP2 below pH 4.0 are unavailable, they have been estimated from PA) (8). These observations are consistent with a recent analysis showing that monolayers containing monovalent lipids have charge spacing that is insufficient to induce significant surface pressure effects (8) as well as previous experiments showing that the binding of basic peptides to membranes is dependent on the presence of polyvalent lipids (45).

While the interactions between divalent cations and anionic lipids have been investigated, the effect of calcium on PL organization in planar systems has not been clearly elucidated. Previous studies with phospholipid monolayers have indicated that the addition of calcium results in the formation of domains enriched in anionic lipid (40), but only recently has phase separation been microscopically visualized (32, 46, 47). Older results describe a mechanism for electrostatic-induced PL reorganization, but recent studies have opened new questions concerning the role of surface pressure (46). Here it is shown that polyanionic domains of unsaturated PIP2 can be induced with micromolar calcium concentrations and maintained through a large range of surface pressures. This result is confirmed by recent observations of PIP2-rich domains in GUVs (48), and coupled with the multivalent requirement for both the lipid and cation species suggests that the mechanism for domain formation was not dependent on acyl chain packing/ordering but rather electrostatic interactions between the cation and lipid headgroup.

The evidence that calcium can alter both the molecular density and the lateral organization of PIP2- and cholesterol-containing membranes suggests a role for membrane structure as a possible effector of calcium signaling. It has been previously shown that the interactions between PIP2 and unstructured polybasic protein domains (such as those of MARCKS and GAP-43) can sequester PIP2, and that those interactions are modulated by Ca2+/calmodulin (49). One implication of the results presented here is that in addition to releasing PIP2 from its polybasic ‘sinks’, calcium ion flux would induce a rearrangement of the membrane and may itself sequester PIP2 in stable, multimolecular aggregates.

Both the change in monolayer surface pressure and the abundance of PIP2-enriched domains were functions of the molar fraction of PIP2, while the small and punctate appearance of the domains was composition-independent (Fig. 1). Notably, PIP2 domains were observed down to 8 mol%, which – while still significantly greater than the 1% typically estimated for the unstimulated plasma membrane concentration of PIP2 (50) – suggests that calcium-induced PIP2 clustering may be important for the low PIP2 concentrations typically observed in cellular plasma membranes.

One possibility for modulation of local PIP2 concentration/functionality in the physiological context of the plasma membrane is by physical exclusion of this lipid from cholesterol-enriched membrane microdomains, e.g. lipid rafts. Such domains are associated with phase separation into coexisting fluid phases (6, 7) and segregate specific lipid components (51, 52). We observed cholesterol-dependent liquid-liquid coexistence in mixed monolayers of PIP2, SOPC, and cholesterol, and found that PIP2 partitions strongly into domains enriched in PL-rich phase markers (Fig. 4). Although not surprising in light of PIP2's highly unsaturated acyl chains and headgroup charge, this finding is in contrast to the enrichment of PIP2 in detergent-resistant membranes (22), which has led to the suggestion that PIP2 may have a preference for the more ordered environment of lipid rafts. This discrepancy might be explained by the binding of PIP2 to the GMC family of proteins (GAP43, MARCKS, and CAP23), which have been proposed to localize to lipid rafts (21).

In addition to its condensing and domain-inducing effect on PIP2/PC monolayers, divalent calcium also lowered the miscibility transition pressure of cholesterol-containing phase separated monolayers (Figs. 5 and and6).6). Liquid-liquid phase coexistence in cholesterol-containing monolayers depends on surface pressure, and PIP2-containing monolayers had higher average miscibility transition pressures ([product]t) than SOPC/cholesterol monolayers, with the effect persisting to less than 10 mol% PIP2. The PIP2-dependent increase of [product]t was attributed to a disordering effect of PIP2's polyunsaturated acyl chains and large headgroup on the PIP2-rich disordered phase (Fig. 4). Strikingly, subphase calcium completely abrogated the increased [product]t in PIP2-containing monolayers, lowering the transition surface pressure by 50%. This reduction of [product]t led to dissolution of domains following introduction of subphase calcium, which was an unexpected result as the decrease in monolayer surface pressure observed after addition of calcium would be expected to stabilize domain coexistence. The reduction of [product]t was dependent on lipid multivalency, as neither PS-containing monolayers, nor those with PIP2 at low pH showed any change in [product]t with subphase calcium (Fig. 6B). These findings suggests that the electrostatic sequestration and condensation of PIP2 by divalent calcium results in tight association of PIP2 molecules, increasing the molecular packing and ordering the more disordered phase such that phase miscibility is achieved at lower surface pressures.

Although there has been significant investigation into the dependence of miscibility transition pressure on membrane composition (6, 7, 36), to our knowledge, this is the first demonstration of the modulation of fluid-fluid phase miscibility by strictly soluble subphase factors. This result suggests a role for lipid-interacting small molecules (e.g. metal ions, polyamines, charged or amphiphilic peptides) in regulating the stability of fluid phase coexistence, which has been postulated to be a principle involved in the formation of lipid rafts (53). More specifically, the significant calcium-dependent modulation of [product]t observed in our experiments implies a role for calcium-signaling in the formation/stability of lipid rafts which have been implicated in a variety of critical cellular processes (54).


The dependence of lateral lipid heterogeneity on electrostatic interactions between charged lipid headgroups and soluble factors plays an important biological role but has been relatively overlooked compared to the organizing effects of cholesterol, acyl chain composition, and temperature. Here, the lateral organization of the highly anionic lipid PIP2 was shown to be sensitive to small changes in calcium ion concentration. Introduction of subphase calcium induced the formation of PIP2-rich domains that were laterally segregated from net neutral lipids in binary monolayers. The formation of calcium-induced domains required that the lipid be multivalently anionic, as monovalent PIP2 (at pH 3) and PS did not form domains. There was no evidence that Ca2+-induced domain formation required a critical concentration of membrane PIP2 that might be considered non-physiological and the linear dependence of Δ[product] and domain abundance on [PIP2] suggested that domains form even at mole fractions of PIP2 where their clustering would produce surface pressure changes too small to measure. In addition to altering the lateral organization of PIP2 in binary monolayers, divalent calcium increased the miscibility of coexisting domains in cholesterol-containing monolayers, in a PIP2-dependent fashion. This result may have biological relevance as it demonstrates the ability of soluble factors to induce lipid reorganization in the plane of the membrane, suggesting the possibility that calcium fluxes can affect formation of lipid rafts in PIP2-containing plasma membranes.


The authors would like to acknowledge funding from NSF-MRSEC grant 05-20020 and NIH grant R01HL067286.


phosphatidylinositol (4,5) bisphosphate
1-stearoyl-2-oleoyl phosphatidylcholine


1. Esposti MD. Lipids, cardiolipin and apoptosis: a greasy license to kill. Cell Death Differ. 2002;9:234–236. [PubMed]
2. Lawrence T, Willoughby DA, Gilroy DW. Anti-inflammatory lipid mediators and insights into the resolution of inflammation. Nat Rev Immunol. 2002;2:787–795. [PubMed]
3. Yin HL, Janmey PA. Phosphoinositide regulation of the actin cytoskeleton. Annu Rev Physiol. 2003;65:761–789. [PubMed]
4. Eling TE, Glasgow WC. Cellular proliferation and lipid metabolism: importance of lipoxygenases in modulating epidermal growth factor-dependent mitogenesis. Cancer Metastasis Rev. 1994;13:397–410. [PubMed]
5. Czech MP. PIP2 and PIP3: complex roles at the cell surface. Cell. 2000;100:603–606. [PubMed]
6. McConnell HM, Vrljic M. Liquid-liquid immiscibility in membranes. Annu Rev Biophys Biomol Struct. 2003;32:469–492. [PubMed]
7. Keller SL, Anderson TG, McConnell HM. Miscibility critical pressures in monolayers of ternary lipid mixtures. Biophys J. 2000;79:2033–2042. [PubMed]
8. Levental I, Janmey PA, Cebers A. Electrostatic contribution to the surface pressure of charged monolayers containing polyphosphoinositides. Biophys J. 2008;95:1199–1205. [PubMed]
9. Lassing I, Lindberg U. Specific interaction between phosphatidylinositol 4,5-bisphosphate and profilactin. Nature. 1985;314:472–474. [PubMed]
10. Cremona O, De Camilli P. Phosphoinositides in membrane traffic at the synapse. J Cell Sci. 2001;114:1041–1052. [PubMed]
11. Martin TFJ. PI(4,5)P2 regulation of surface membrane traffic. Curr Op Cell Biol. 2001;13:493–499. [PubMed]
12. Hilgemann DW, Feng S, Nasuhoglu C. The complex and intriguing lives of PIP2 with ion channels and transporters. Sci STKE. 2001;2001:RE19. [PubMed]
13. Berridge MJ, Irvine RF. Inositol trisphosphate, a novel second messenger in cellular signal transduction. Nature. 1984;312:315–321. [PubMed]
14. Ho KK, Anderson AA, Rosivatz E, Lam EW, Woscholski R, Mann DJ. Identification of cyclin A2 as the downstream effector of the nuclear phosphatidylinositol 4,5-bisphosphate signaling network. J Biol Chem. 2008;283:5477–5485. [PubMed]
15. Uno I, Fukami K, Kato H, Takenawa T, Ishikawa T. Essential role for phosphatidylinositol 4,5-bisphosphate in yeast cell proliferation. Nature. 1988;333:188–190. [PubMed]
16. Janmey PA. Protein regulation by phosphatidylinositol lipids. Chem Biol. 1995;2:61–65. [PubMed]
17. Catimel B, Schieber C, Condron M, Patsiouras H, Connolly L, Catimel J, Nice EC, Burgess AW, Holmes AB. The PI(3,5)P2 and PI(4,5)P2 Interactomes. J Proteome Res. 2008;7:5295–5313. [PubMed]
18. Lassing I, Lindberg U. Evidence that the phosphatidylinositol cycle is linked to cell motility. Exp Cell Res. 1988;174:1–15. [PubMed]
19. Varnai P, Balla T. Visualization of phosphoinositides that bind pleckstrin homology domains: calcium- and agonist-induced dynamic changes and relationship to myo-[3H]inositol-labeled phosphoinositide pools. J Cell Biol. 1998;143:501–510. [PMC free article] [PubMed]
20. Balla T, Bondeva T, Várnai P. How accurately can we image inositol lipids in living cells? Trends Pharmacol Sci. 2000;21:238–241. [PubMed]
21. Laux T, Fukami K, Thelen M, Golub T, Frey D, Caroni P. GAP43, MARCKS, and CAP23 modulate PI(4,5)P(2) at plasmalemmal rafts, and regulate cell cortex actin dynamics through a common mechanism. J Cell Biol. 2000;149:1455–1472. [PMC free article] [PubMed]
22. Pike LJ, Miller JM. Cholesterol depletion delocalizes phosphatidylinositol bisphosphate and inhibits hormone-stimulated phosphatidylinositol turnover. J Biol Chem. 1998;273:22298–222304. [PubMed]
23. Varnai P, Lin X, Lee SB, Tuymetova G, Bondeva T, Spat A, Rhee SG, Hajnoczky G, Balla T. Inositol lipid binding and membrane localization of isolated pleckstrin homology (PH) domains. Studies on the PH domains of phospholipase C delta 1 and p130. J Biol Chem. 2002;277:27412–27422. [PubMed]
24. Huang S, Lifshitz L, Patki-Kamath V, Tuft R, Fogarty K, Czech MP. Phosphatidylinositol-4,5-bisphosphate-rich plasma membrane patches organize active zones of endocytosis and ruffling in cultured adipocytes. Mol Cell Biol. 2004;24:9102–9123. [PMC free article] [PubMed]
25. Wang J, Gambhir A, McLaughlin S, Murray D. A computational model for the electrostatic sequestration of PI(4,5)P2 by membrane-adsorbed basic peptides. Biophys J. 2004;86:1969–1986. [PubMed]
26. Zhang W, Crocker E, McLaughlin S, Smith SO. Binding of peptides with basic and aromatic residues to bilayer membranes: phenylalanine in the myristoylated alanine-rich C kinase substrate effector domain penetrates into the hydrophobic core of the bilayer. J Biol Chem. 2003;278:21459–21466. [PubMed]
27. McLaughlin S, Wang J, Gambhir A, Murray D. PIP(2) and proteins: interactions, organization, and information flow. Annu Rev Biophys Biomol Struct. 2002;31:151–175. [PubMed]
28. Hartwig JH, Thelen M, Resen A, Janmey PA, Nairn AC, Aderem A. MARCKS is an actin filament crosslinking protein regulated by protein kinase C and calcium-calmodulin. Nature. 1992;356:618–622. [PubMed]
29. Levental I, Cebers A, Janmey PA. Combined electrostatics and hydrogen bonding determine intermolecular interactions between polyphosphoinositides. J Am Chem Soc. 2008;130:9025–9030. [PMC free article] [PubMed]
30. Redfern DA, Gericke A. pH-dependent domain formation in phosphatidylinositol polyphosphate/phosphatidylcholine mixed vesicles. J Lipid Res. 2005;46:504–515. [PubMed]
31. Lamberson ER, Cambrea LR, Hovis JS. Controlling the charge and organization of anionic lipid bilayers: effect of monovalent and divalent ions. J Phys Chem B. 2007;111:13664–13667. [PubMed]
32. Ross M, Steinem C, Galla HJ, Janshoff A. Visualization of chemical and physical properties of calcium-induced domains in DPPC/DPPS Langmuir-Blodgett layers. Langmuir. 2001;17:2437–2445.
33. Kates M. Techniques of Lipidology. 2nd. Elsevier Science Publishers B.V.; Amsterdam: 1986.
34. Harlan JE, Yoon HS, Hajduk PJ, Fesik SW. Structural characterization of the interaction between a pleckstrin homology domain and phosphatidylinositol 4,5-bisphosphate. Biochemistry (Mosc) 1995;34:9859–9864. [PubMed]
35. Cunningham CC, Vegners R, Bucki R, Funaki M, Korde N, Hartwig JH, Stossel TP, Janmey PA. Cell permeant polyphosphoinositide-binding peptides that block cell motility and actin assembly. J Biol Chem. 2001;276:43390–43399. [PubMed]
36. Hagen JP, McConnell HM. Liquid-liquid immiscibility in lipid monolayers. BBA - Biomembranes. 1997;1329:7–11. [PubMed]
37. Radhakrishnan A, McConnell HM. Electric field effect on cholesterol-phospholipid complexes. Proc Natl Acad Sci U S A. 2000;97:1073–1078. [PubMed]
38. Heinrich MC, Levental I, Gelman H, Janmey PA, Baumgart T. Critical Exponents for Line Tension and Dipole Density Difference from Lipid Monolayer Domain Boundary Fluctuations. J Phys Chem B. 2008;112:8063–8068. [PubMed]
39. Shah DO, Schulman JH. Binding of metal ions to monolayers of lecithins, plasmalogen, cardiolipin, and dicetyl phosphate. J Lipid Res. 1965;6:341–349. [PubMed]
40. Papahadjopoulos D. Surface properties of acidic phospholipids: interaction of monolayers and hydrated liquid crystals with uni- and bi-valent metal ions. Biochim Biophys Acta. 1968;163:240–254. [PubMed]
41. Flanagan LA, Cunningham CC, Chen J, Prestwich GD, Kosik KS, Janmey PA. The structure of divalent cation-induced aggregates of PIP2 and their alteration by gelsolin and tau. Biophys J. 1997;73:1440–1447. [PubMed]
42. Toner M, Vaio G, McLaughlin A, McLaughlin S. Adsorption of cations to phosphatidylinositol 4,5-bisphosphate. Biochemistry (Mosc) 1988;27:7435–7443. [PubMed]
43. Flach CR, Brauner JW, Mendelsohn R. Calcium ion interactions with insoluble phospholipid monolayer films at the A/W interface. External reflection-absorption IR studies. Biophys J. 1993;65:1994–2001. [PubMed]
44. Hagge SO, Hammer MU, Wiese A, Seydel U, Gutsmann T. Calcium adsorption and displacement: characterization of lipid monolayers and their interaction with membrane-active peptides/proteins. BMC Biochem. 2006;7:15. [PMC free article] [PubMed]
45. Golebiewska U, Gambhir A, Hangyas-Mihalyne G, Zaitseva I, Radler J, McLaughlin S. Membrane-bound basic peptides sequester multivalent (PIP2), but not monovalent (PS), acidic lipids. Biophys J. 2006;91:588–599. [PubMed]
46. Sovago M, Wurpel GW, Smits M, Muller M, Bonn M. Calcium-induced phospholipid ordering depends on surface pressure. J Am Chem Soc. 2007;129:11079–11084. [PubMed]
47. Eklund KK, Vuorinen J, Mikkola J, Virtanen JA, Kinnunen PK. Ca2+-induced lateral phase separation in phosphatidic acid/phosphatidylcholine monolayers as revealed by fluorescence microscopy. Biochemistry (Mosc) 1988;27:3433–3437. [PubMed]
48. Carvalho K, Ramos L, Roy C, Picart C. Giant unilamellar vesicles containing phosphatidylinositol(4,5)bisphosphate: characterization and functionality. 2008;95:4348–4360. [PubMed]
49. McLaughlin S, Murray D. Plasma membrane phosphoinositide organization by protein electrostatics. Nature. 2005;438:605–611. [PubMed]
50. Ferrell JE, Jr, Huestis WH. Phosphoinositide metabolism and the morphology of human erythrocytes. J Cell Biol. 1984;98:1992–1998. [PMC free article] [PubMed]
51. Baumgart T, Hunt G, Farkas ER, Webb WW, Feigenson GW. Fluorescence probe partitioning between Lo/Ld phases in lipid membranes. BBA - Biomembranes. 2007;1768:2182–2194. [PMC free article] [PubMed]
52. Dietrich C, Volovyk ZN, Levi M, Thompson NL, Jacobson K. Partitioning of Thy-1, GM1, and cross-linked phospholipid analogs into lipid rafts reconstituted in supported model membrane monolayers. Proc Natl Acad Sci U S A. 2001;98:10642–10647. [PubMed]
53. Simons K, Vaz WL. Model systems, lipid rafts, and cell membranes. Annu Rev Biophys Biomol Struct. 2004;33:269–295. [PubMed]
54. Simons K, Toomre D. Lipid rafts and signal transduction. Nat Rev Mol Cell Biol. 2000;1:31–39. [PubMed]