Open Access is an initiative that aims to make scientific research freely available to all. To date our community has made over 100 million downloads. It’s based on principles of collaboration, unobstructed discovery, and, most importantly, scientific progression. As PhD students, we found it difficult to access the research we needed, so we decided to create a new Open Access publisher that levels the playing field for scientists across the world. How? By making research easy to access, and puts the academic needs of the researchers before the business interests of publishers.
We are a community of more than 103,000 authors and editors from 3,291 institutions spanning 160 countries, including Nobel Prize winners and some of the world’s most-cited researchers. Publishing on IntechOpen allows authors to earn citations and find new collaborators, meaning more people see your work not only from your own field of study, but from other related fields too.
To purchase hard copies of this book, please contact the representative in India:
CBS Publishers & Distributors Pvt. Ltd.
www.cbspd.com
|
customercare@cbspd.com
Eukaryotic cells contain phospholipids and nonphospholipids. The latter lack phosphodiester groups in their head group regions. Lipid-dependent gating of voltage-gated ion channels represents a steady-state energetic effect of nonphospholipids in favoring the resting state of voltage-sensor domains (VSDs) of the channels. It suggests adaptation of ion channels to lipid compositions in their native niche and significant roles of low-to-intermediate affinity lipid-binding sites at the channels. The nonphospholipids include glycoglycerolipids, glycosphingolipids, ceramides, cholesterol or cholesterol esters, diacylglycerol (DAG), fatty acids, cation lipids, etc. Change in relative ratios of phospholipids to nonphospholipids can shift the energetic levels of the VSDs and the gating of these channels, which in turn may alter excitability in certain cells. It is expected that reduced relative abundance of nonphospholipids / phospholipids in plasma membranes may change resting transmembrane potential or gating transitions of voltage-gated Na or K channels. The net results will be a change in action potential firing at least in certain areas of an excitable cell. Such changes in the central nervous system (CNS) are anticipated to affect brain functions and contribute to early-onset neurological phenotypes observed in patients carrying lipid metabolic defects. We will describe the basics of lipid-dependent gating and review its projected links to phenotypes of monogenic lipid metabolic defects and related changes of lipid composition in cell membranes as well as altered neuronal excitability in CNS. However, lack of high-resolution techniques to measure lipid composition around individual channels in cell membranes has been limiting the studies of direct connections between lipid redistribution caused by metabolic defects and altered ion channel activities. Potential solutions will be described for future studies.
Department of Physiology and Biophysics, University at Buffalo, Buffalo, NY, USA
Department of Medicinal Chemistry, University of Florida, Gainesville, FL, USA
Laboratory of Molecular Physiology and Biophysics, Hauptman-Woodward Medical Research Institute, Buffalo, NY, USA
Felix Chin
Department of Physiology and Biophysics, University at Buffalo, Buffalo, NY, USA
Perelman School of Medicine, University of Pennsylvania
*Address all correspondence to: qxjiang@ufl.edu;, qiuxingj@buffalo.edu;, jiangq9992003@gmail.com
1. Introduction
All living cells are thermodynamically off-equilibrium. As open systems, their identities and existence are demarcated by their plasma membranes, and they exchange materials and energy through these membranes [1]. Lipid metabolism basically covers exchange of lipid molecules or their constituents between a cell and its environment or between an organism and its surroundings as well as lipid distribution (or redistribution) and composition within a cell or an organism [2]. At the level of individual cells, lipids can be distributed among the plasma membranes, the intracellular membranes, and the nonlamellar-structured lipid-containing particles inside or outside cells [3]. Lipid metabolic defects can change lipid composition and/or distribution inside or outside a cell and may exert significant effects on cell physiology. This chapter will focus on possible effects of lipid metabolic defects on activities of voltage-gated ion channels at the molecular and cellular levels, which are probably part of the basis for organismal level phenotypes observed in human patients [4].
All cells have specific resting transmembrane electrostatic potentials, which require voltage-gated ion channels (VGICs), often a voltage-gated potassium (Kv) channel, as a feedback pathway, to be stabilized at certain levels under well-defined ionic conditions [5]. The transmembrane potential serves as a driving force or energy source for transporting various molecules and ions across the cell membranes. Inside a human body, all cells contain high K+ and low Na+ inside and face high Na+ and low K+ outside (an exception is the cochlear inner hair cells whose top ends meet with high-K+ endolymph) [6]. The ionic differences are often maintained by Na+-pump at the expense of ATP. Ion channels allow specific ions to flow down their electrochemical gradients and dissipate the chemical energy [5, 7]. For these channels to function well, lipid molecules in cell membranes must maintain stable ion gradients by separating the inside and outside milieus of a cell reliably with negligible shunting pathways, due to hydrophobicity in the membrane core and the high energy barriers for hydrated ions to cross. These channels therefore must reside in lipidic environments to be physiologically relevant. Their interactions with the lipid molecules are presumably a critical factor for shaping their functions [4, 8]. The same are expected to be true for other transmembrane proteins that have been evolved to function in specific lipid environments in their native niches.
VGICs belong to a superfamily of integral membrane proteins that have distinct voltage-sensor domains (VSDs) responsible for detecting changes in transmembrane electrostatic potential [5]. Except voltage-gated proton channels whose proton-conducting pores are within their VSDs, all known VGICs are made of four subunits / domains that together form a central pore surrounded by four VSDs in the periphery. For these channels, each subunit/domain contains a pore region and a VSD. Each pore region has two transmembrane helices and a pore loop, and each contributes to one quarter of an ion-conducting pore [9]. Each VSD contains four transmembrane segments (S1–S4) that are folded into a four-helix bundle, and harbors intrinsic positive charges, which are mostly carried in the extracellular half of the fourth segment (S4), physically relocate in response to change in transmembrane electric field and are called gating charges [10]. A VSD switches from a resting state to an activated state through sequential translocation of the fixed gating-charge-contributing residues (mainly in S4) from the intracellular side of the transmembrane electric field to the extracellular side [11, 12, 13, 14, 15]. The physical movement of typical four residues in 3 to 4 helical turns has been extensively studied. Presumably, five sequential states are required to account for moving all four positively-charged residues from the intracellular side to the extracellular side of the transmembrane electric field [16]. Crystallographic data have so far observed local rearrangements around one or two residues, making the general model of four-step motions reasonably well accepted [11]. However, this type of model may still be a simplification and may need modifications for a particular VSD [7]. We will not delve into such subtle details because a general mechanistic two-state model is often sufficient for our thermodynamic and kinetic considerations [7, 17].
From the angle of molecular composition of a bilayer membrane, lipid molecules in the two leaflets are arranged in juxtaposition to have hydrophobic tails from two leaflets face each other in the interior, achieve high stability and minimize trans-bilayer flipping of polar or charged lipid groups [7]. Laterally, lipids and proteins can form small functional units or domains, and lipid/protein composition may vary from one location to another in the plasma membrane of a cell. Across a cell membrane, the lipid composition of the two leaflets often differs from each other. The canonical fluid-mosaic model of a biomembrane suggests that a membrane protein complex and the few layers of lipid molecules surrounding it form a functional unit [1, 7, 8]. This same notion probably applies to the intracellular membranes as well, highlighting the importance of lipid-protein interactions in every biological membrane and the general proposal that a membrane protein evolves and becomes adapted to the lipid environment of its native niche in membrane [1].
Changes of lipid environments in the native niche of a membrane protein are thus expected, especially for those proteins whose conformational changes are particularly sensitive to lipids, to generate significant functional effects, which may alter cell physiology [17]. Lipid metabolic defects may cause changes in lipid uptake, degradation or synthesis of lipids, recycling and redistribution of lipids in plasma and organellar membranes, and/or removal of lipids from organelles or from cells through vesicles, protein binders or carriers, or enzymatic activities [6]. We will focus on the lipid metabolic effects on VGICs in the next sections, mainly because these channels have their VSDs at the lipid-protein interfaces and are particularly sensitive to their lipid environments [8, 18]. Our general theme is that the lipid-metabolic defects may change the so-called lipid-dependent gating of VGICs and bring about many of the neurological phenotypes or the functional and structural changes in organs or tissues outside the nervous systems [8].
2. Lipid metabolic defects and lipid distribution among cell membranes
There are at least five important processes for lipid metabolism and homeostasis (Figure 1) [19]: (a) lipid uptake and absorption/assimilation from diets, (b) ab initio synthesis of lipids from building blocks, (c) lipid recycling including transport from one organelle to another, release from lipid storage, conversion from one type to another by specific enzymes, (d) lipid degradation or removal from cells through exosomes or carrier protein complexes, and (e) storage of lipid molecules in tissues/organs, storage particles like lipid droplets or accumulation in lysosome-related compartments in cells. These processes are integrated and contribute to a complicated network in regulation of lipid homeostasis (Figure 1). We will not elaborate on these processes, but instead will focus on some of the key changes that will likely affect the voltage-gated ion channels [4].
Figure 1.
Lipid homeostasis at systems and cellular levels. The lipids can be taken into or removed from the cells, can be synthesized inside the cells, can be stored insider or outside the cells, and can be recycled or degraded in cells.
Based on chemical structures and their arrangements in bilayers, all lipid molecules in human cell membranes fall into two groups --- phospholipids and nonphospholipids, depending on whether a lipid has a phosphodiester group in its headgroup region [4] (Figure 2). Typically, phospholipids include glycerophospholipids, phosphosphingolipids and phosphono-sphingolipids (sphingomyelin as a prominent example), and others that contain a properly positioned phosphate group, such as cholesterol-phosphate, ceramide-1-phosphate, geranyl phosphate, isopentenyl phosphate, etc. [20]. The phospho-diester groups in a typical phospholipid bilayer are aligned to form two bands that have high charge density and participate in extensive H-bonding networks, making a prominent chemical feature that is missing in a bilayer made of nonphospholipids. It remains unclear and untested whether a sulfate group can replace a phosphodiester group in the right locations of a lipid bilayer, even though its chemistry is different and in general it cannot substitute for a phosphate in biochemical settings. As a demonstration of the importance of the chemical bonding structures at the phosphate bands, plasmalogens deserve mention here [21, 22]. A plasmalogen contains one normal ester-bonded (at the sn-2 position of the glycerol backbone) acyl chain and one ether-linked chain (at the sn-1 position). They make 5–20% of the phospholipids in most mammalian cell membranes and account for ∼18% of total phospholipid mass in humans [23]. Their vinyl-ether linkages are thought to scavenge reactive oxygen species (ROS). Their presence may introduce changes in biophysical properties of biological membranes, such as thickness, rigidity, fluidity, phase separation, fusion, etc. and may dictate their importance to myelin sheaths in nervous systems. Metabolic changes in plasmalogen abundance can alter CNS functions and have been proposed as an etiological factor of Alzheimer’s disease (AD) [23].
Figure 2.
Compositional difference between bilayers made of phospholipids and nonphospholipids contributes to lipid-dependent gating effects on a voltage-gated ion channel. The phosphate bands endow significant chemical properties that are unique to a phospholipid bilayer. A voltage-sensing domain (VSD) is favored energetically in a resting state when it is surrounded by nonphospholipids, and is stabilized in an activated state in the presence of surrounding phospholipids.
The nonphospholipids include a more diversified group, including members of fatty acids, glycerolipids, sphingolipids, sterol lipids (cholesterol and analogs), prenol lipids, saccharolipids and polyketides [20]. Among them, cholesterol (CHOL or Chol) and glycosphingolipids (such as gangliosides) are quite abundant in human cells, especially in cells of CNS. For example, a human adult brain accounts for ∼3% of body weight, but contains nearly 25% of total body cholesterol [24]. All cholesterol in the brain is synthesized de novo in astrocytes. Cholesterol is not only a structural component in the branched membrane systems of a neuronal cell in the brain, but also serves as a precursor for critical signaling molecules and hormones in the system [19]. Not surprisingly, metabolic changes in the abundance or distribution of cholesterol, or nonphospholipids in general, is important to the proper functions of the CNS [25].
Metabolic defects in fatty acids, phospholipids or nonphospholipids often lead to changes in their relative abundance in circulation, tissues and organs, or in plasma membranes and intracellular membranes of individual cells, affecting both their structural roles, transport, distribution, recycling, and signaling functions [25, 26, 27, 28, 29, 30, 31, 32, 33, 34]. Many genetic defects may be lethal in early stages of life. For examples, the most prominent fatty acid oxidation disorder, medium-chain acyl-CoA dehydrogenase deficiency (MCADD), may cause dehydration, lethargy and hypoketotic hypoglycemia, and result in early onset liver dysfunction and brain edema, and thus a high mortality rate [27]. Phospholipid metabolic defects, such as the ones leading to high or low phosphatidylcholine (PC)/phosphatidyl-ethanolamine (PE) ratios, can affect the formation and stability of triacylglyceride (TAG)-rich lipoproteins and alter de novo lipid synthesis via sterol regulatory element binding protein (SREBP) pathway. Changes in PC/PE ratios are linked to liver failure, impaired liver regeneration, etc., and genetic mutations affecting PC/PE ratios may give rise to embryonic lethality [29]. Similarly, multiple monogenic defects in metabolism of non-fatty-acid nonphospholipids alter their distribution in cells. The level of severity of their phenotypes varies, and can still lead to obvious functional changes even when organ development and cell growth appear normal in infancy (Table 1). Notably, altered nonphospholipid abundance in cell membranes almost always leads to changes in excitable cells, and invariably is manifested in severe or life-threatening neurological phenotypes (Table 1).
1 in 90,000; mainly in infants, progressive and fatal.
Table 1.
Examples of metabolic defects in fatty acids, phospholipids and other nonphospholipids.
Based on these phenotypes, our central hypothesis is that these metabolic changes in the relative ratios of phospholipids and nonphospholipids in plasma membranes may change the activities of VGICs, and lead to phenotypical changes in nervous systems because of high abundance of these channels [4, 8, 18]. We will elaborate such connections in the next section first.
3. Lipid-dependent gating of voltage-gated ion channels (VGICs)
From the fluid mosaic model of a biomembrane, we proposed that a VGIC and a few layers of lipid molecules (called annular lipids here; Figure 2) right next to its transmembrane footprint together form a functional unit [1, 35]. We usually call the control of channel opening and closing as a gating process. Besides other factors, the gating energetics and physiological functions of a VGIC is also a function of lipid composition in the annular layer. As listed in Table 1, the metabolic defects of nonphospholipids (including fatty acids) may change the distribution and redistribution of these lipids in native cell membranes, whose effects may penetrate the annular layers of a VGIC that happens to be in the region and alter its functions. Altered VGIC activities may change the excitability of neurons and muscle cells in specific neural circuits or in muscles, leading to severe pathophysiological effects at tissues, organs, or organismal levels. To keep focused, we will discuss changes in relative abundance of nonphospholipids and phospholipids in plasma membranes where the VGICs contribute to key physiological events. We will not discuss the effects of lipid metabolic defects on the VGICs in intracellular membranes because of insufficient studies of their connections to lipid metabolic diseases so far [36].
Our studies in lipid-dependent gating of Kv channels led us to a more general concept of nonphospholipid-mediated gating effects on VGICs (Figure 3) [4, 17]. A typical depolarization-activated Kv channel undergoes transitions between different states in a voltage-dependent or -independent manner (Figure 3A) [37, 38]. The measured gating charge (Q) movement (due to changes in the voltage sensor domain) or conductance (G) change (due to the opening of the pore domain) can be presented as a function of transmembrane potential (Q-V or G-V in Figure 3B). From the general concepts in Figure 2, the nonphospholipids favor the resting (closed) state of a VGIC. A prediction is that the Q-V and/or G-V would shift to the right and may keep the same slope or become shallower due to procrastinated conformational changes in certain steps (purple trace in Figure 3B) [39, 40]. Indeed, when we constructed a bead-supported unilamellar membrane (bSUM) with precisely controlled lipid composition and channel orientation in bilayers and without any additional solvents (e.g., no decane), it was feasible to measure accurately the shifts of G-V curves in the presence of 4.0% or 9.6% cholesterol (Figure 3C) [41]. Due to decane islands and annuli in planar lipid bilayers and potential (or preferential) interaction (packing) of cholesterol with decane, it was difficult to control cholesterol concentration and phase behavior of planar bilayers when bilayer experiments were performed by introducing cholesterol or glycolipids [42]. The bSUMs were thus better when planer glass electrodes were utilized for electrophysiological recordings from channels in them, and so far, remain the only system that is suitable to control lipid composition and channel orientation at the same time. Our results agree very well with the predictions of lipid dependent gating (Figure 3C vs. B).
Figure 3.
Lipid-dependent gating directly alters properties of a VGIC. (A) A typical gating model for a VGIC that is depolarization-activated. C: closed state(s); Cx: closed state right before opening; O: open state; and I: inactivated state; (B) Changes of gating charge (Q) and conductance (G) as a function of transmembrane potential (Vm). The VSD switches from a resting (“down”) to an activated (“Up”) state, while the channel pore changes from the closed to the open state. The predicted change of the G-V or Q-V curve in the presence of nonphospholipids is a right shift (purple arrow and trace) with the same or a shallower slope; and (C) The measured changes of G-V curves of the KvAP channel in the presence of different lipids in bead-supported unilamellar membranes (bSUMs) made of different lipids (inserted cartoon). Those on the far left were from phospholipid membranes, while the red and orange curves were in DOPC bilayers doped with small amounts of cholesterol. From references [8, 18] with permissions following PubMed open access policy.
Similar physiological changes have been reported for Nav1.9 channels in the sensory neurons and Kv3 channels in acutely isolated dopaminergic neurons [43, 44], suggesting a more generic nature of the nonphospholipid-dependent gating effects on voltage-gated ion channels. When more experimental data published in literature were studied, a generally agreed conclusion is that loading cell membranes by methyl-beta-cyclodextrin-cholesterol causes consistent right shifts of all VGICs (Kv, Nav and Cav channels) in different cells [45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56], whereas divergent or conflicting data were reported after cholesterol depletion by methyl-beta-cyclodextrin [17]. We know that strong depletion may cause significant changes to membrane integrity and submembrane structures inside cells. It is therefore reasonable to put more trust in the cholesterol-loading data and be very cautious of the results from cholesterol depletion in cell membranes. It is hence reasonable for us to conclude that the published data in literature agree with the predictions in Figure 3B and our hypothesis in Figure 2.
Thermodynamically, the free energy, ΔG (n, T, V, P, Q, Vm, μP, μS, μL), for a VGIC in a lipid membrane can be treated as a function of molecular quantities (n), temperature (T), volume (V), pressure (P), charge (Q), transmembrane electrostatic potential (Vm)and chemical potentials of the protein (μP), solutes/solvents (μS), and lipids (μL) [4, 8, 17]. The charge movement across a gradient of electrostatic potential leads to a free energy change of Q *ΔVm or its integral over time. The G-V curves did not change steepness significantly when cholesterol was added, suggesting a similar or the same gating model with similar apparent gating charges (Q ∼ 2.0–3.0 e0). A right shift by 200 mV (as in Figure 3C) is equivalent to a minimal energy change of 9.2 kcal/mol in stabilizing the closed states, slightly more than the hydrolytic energy of one ATP.
For VGICs, the chemical potential changes from the lipid bindings are manifested indirectly through energetic compensation from the gating charge movement. If we consider that an average number of lipid-binding sites is <m>, and there is an average change in affinity of the binding sites. If we use kd,1 and kd,2 to represent geometrically averaged apparent dissociation constants of lipids at equivalent binding sites in the closed and open states, respectively and consider no mutual interference between them, a simplified expression of free energy change from the lipid-binding sites is - <m > RT ln (kd,1/kd,2). A typical VGIC allows ∼120 lipid-binding sites in the first layer of annular lipids around it. To make for the estimated free energy change from G-C curves, that is ∼9.2 kcal/mol, ∼10 lipid-binding sites are sufficient if each on average contributes ∼1.5 RT, which equates to a kd,2/kd,1 ∼ 4.5. A small change in relative affinity of a group of binding sites hence suffices to cause a major shift in G-V (and Q-V) of a VGIC. In other words, the changes in affinity can happen to those of low, intermediate or high-affinities, but are not limited to the high-affinity binding sites because the latter usually play a critical structural role in stabilizing the 3D structure of a functional membrane protein.
The lipid-dependent gating is expected to impact on both Nav and Kv channels. For a neuron that fires action potentials (APs) using the Nav and Kv channels, the nonphospholipid-induced gating effect will make it more difficult to open Nav channels, which may result in a larger number of Nav channels needed to allow the cell to reach the threshold for firing action potential. Meanwhile, the lipid-dependent gating of Kv channels will slow down the repolarization of the transmembrane potential. The net effects may include a shift in resting membrane potential, a shift in the threshold for AP firing, a decrease in the AP firing frequency, elongation of the AP duration, or a decrease in AP overshoots, etc. Consistent with these predictions (to the opposite direction due to cholesterol decrease), in cultured dorsal root ganglion (DRG) neurons, cholesterol-dependent gating effects on the Nav1.9 channels were found to be the root cause when local cholesterol decrease in the neurons resulted from carrageenan injection caused hyperexcitability and enhanced pain sensation. As an example of the physiological data, in Figure 4A, decrease of thresholds happened when cholesterol content was decreased and AP firing frequencies were increased after carrageenan cocktail treatment (Figure 4B and C at t + 12 min). Notably, in Figure 4B and C (t = 0 min), the neurons preloaded with cholesterol (MβCD/Chol 20 mM) showed an AP threshold for current injection of 200 pA, much higher than the control 75 pA, consistent with the predictions of cholesterol-dependent gating.
Figure 4.
Physiological effects of cholesterol-dependent gating of Nav1.9 channels in dorsal root ganglion (DRG) cells. (A) Carrageenan cocktail decreases the threshold for AP firing in DRG neurons, concurrent with decreased cholesterol content in the cells. MβCD/Chol (20 mM) treatment decreased the effects; (B) Control neurons had a lower threshold for AP firing, and only 75 pA current injection was needed to elicit an AP (blue arrow; t = 0 in B); and (C) Neurons pre-treated with MβCD/Chol for 10 minutes showed much higher threshold for AP firing (200 pA; blue arrow), due to cholesterol-dependent gating effects on Nav1.9 channels. In both cases the cocktail treatment (t + 12 min) increased the AP firing frequencies (but more pronounced in B than in C), and shortened the AP duration (especially in C). Adapted from reference [43] with permissions under the free access policy of EMBO press.
Similar to what was reported by the Padilla group, my group in collaboration with Dr. Yuqing Li’s lab at the University of Florida raised an NPC1-I1061T-knockin model mouse [57] and discovered that the AP firing frequency in triphasic cerebellar Purkinje neurons was ∼30% lower in the mutant than that in the wild-type (QXJ, unpublished observations). The decreased cholesterol-content likely activated Kv channel more significantly than the Nav or Cav channels in the Purkinje neurons, leading to a marked inhibitory effect. More detailed studies are needed to examine the channels in these Purkinje neurons to define the lipid dependent gating effects better.
5. Computational methods for low-to-intermediate-affinity binding sites of non-phospholipids on membrane proteins
Our theoretical analysis in the last section suggested a likely scenario for the lipid-dependent gating to function in a cell membrane, which contains abundant content of nonphospholipids. It relies on changes of ten or so low- to-intermediate-affinity binding sites because high lipid content would saturate all high-affinity binding sites continuously. However, it is difficult to study these non-structural lipids because of possible heterogeneity in the bound poses and/or slightly larger volumes of such binding sites to accommodate different poses, making these almost impossible to pin down accurately using conventional structural methods.
Computational analysis has been used to make some predictions that may be tested experimentally (for examples, Figure 5) [7]. For cholesterol, the consensus binding motifs predicted in the past carried uncertainties. Docking and energy minimization was used to identify 20 possible binding sites (limited to 20; Figure 5A), eight at the periphery and 12 near the interfaces between the pore domain and the VSDs. The same analysis of Kv7.1 led to 16 binding sites to the periphery of four VSDs and four at the pore-VSD interfaces. Experimental testing of these sites is still lacking.
Figure 5.
Computational analysis of action sites for lipid-dependent gating. (A) Computational docking suggested multiple cholesterol-binding sites in the intra- (green balls) and extra-cellular (blue) leaflets around a Kv1.2 (PDB: 2R9R) structure, leading to 20 possible sites. The channel is viewed from the extracellular side. Adapted from reference [58] with permissions following the PubMed open access policy; and (B) Cholesterol-binding densities (replicas 1–4) around an A2AR (PDB: 4EIY; only helices showed in gray) predicted from molecular dynamics simulation, indicating at least 8 binding sites. The cholesterol resolved in the crystal structure is in cyan and the structure is viewed from the extracellular side. Adapted from reference [59] with permissions following the PubMed open access policy.
In a similar situation, structural studies of G-protein coupled receptors (GPCRs) revealed only limited number of cholesterol-binding sites, but it is known that cholesterol is critically important for physiological functions of various GPCRs. Bovine rhodopsin, for example, encounters a high cholesterol content at an 8:1 molar ratio that are important for ensuring high stability of the protein in membrane [60]. Molecular dynamics (MD) simulation in GPCRs has predicted multiple binding sites (Figure 5B) that were not detected by structural studies and are not usually agreed on among different proteins. A majority of the predicted binding sites are likely not conserved and may not be well occupied for structural identification [61]. MD simulation may uncover more cholesterol binding events right next to each receptor [59, 62] (Figure 2). These studies highlight the difference between structural lipids of high binding affinity and those loose and dynamic binding sites with lower affinity but may be the major contributors to the lipid-dependent gating.
6. Better technologies desired for studying lipid-dependent gating
Technically, we are rather limited in studying lipid-dependent gating even in a small patch of cell membranes because of complicated lipid environments in face of phase separation and domain organization, lipid heterogeneity and asymmetry in two leaflets, membrane-matrix or membrane-cytoskeleton interactions, etc. The dynamic occupancy of the low-to-intermediate affinity lipid-binding sites, which were predicted to be chiefly responsible for lipid-dependent gating, becomes a roadblock against the structural / biophysical studies of their effects on different VGICs in both reconstituted and native cell membranes. Ideally, we would like to use high-resolution real-time methods to analyze the lipid composition or distribution within the annulus or annuli of one or a few VGICs that are in a uniform lipid environment. With available technologies, single molecule detection of nonphospholipids around a VGIC with sufficient spatial resolution is almost impossible in the presence of a sea of lipids in a biomembrane. In a recently-published book chapter [1], QXJ discussed the possible development of single molecule (ion) detection in mass spectrometry, surface-enhanced Raman spectrometry [63], nanoscale secondary ion mass spectrometry (NanoSIMS) [64], and high-resolution atomic force microscopy (AFM). These methods will need significant developments to achieve sufficient resolutions and differentiate nonphospholipids from phospholipids around individual VGICs in membranes. Here we will describe a few directions that may practically enrich our understandings of lipid-dependent gating in the near future.
First, even though the binding poses and positions of lipids as predicted from computational analysis often carry a lot of uncertainty, many of them can be tested in functional studies in order to identify a prominent one or a few that are functionally relevant and allow testable predictions to be formulated. The sites around the Kv channels (such as those in Figure 5A) may be suitable targets. Alternatively, zero-length cross-linkers can be introduced and pin down the low affinity cholesterol-binding sites around a VGIC [65]. These sites can be further verified by functional studies. Single molecular fluorescence resonance energy transfer (smFRET) may be another method that is suitable for probing dynamic lipid-protein interactions within the annular layer of a VGIC when suitable probes can be introduced into the proteins to serve as sensors.
Second, genetic methods may be suitable to probe the co-evolution of VGICs in contrastingly different lipid environments. For example, it is reported that some strains of bacteria isolated from Lake Matano with phosphate concentration < 50 nM, can grow without added phosphorus, and replace nearly 100% of membrane lipids with amino- or glycolipids, like what was reported for photoautotrophs and heterotrophs [66]. Because these strains have evolved to survive in a phosphorus-limited native niche, they can make a good model system to assess how the VGICs have evolved to deal with the selection pressure. The amino-glycolipids in the acidic environment in these bacteria must be positively charged so that a VGIC would face a very different environment for voltage-gating than a phospholipid-rich membrane.
Third, the potential of cryo-electron tomography has been gradually improved for high-resolution imaging [67]. The major bottleneck still resides with the limited dose of electrons used per exposure. For unilamellar biomembranes that are of 4–6 nm in thickness, it might be feasible to introduce graphene films to both sides so that the individual membranes can tolerate a much higher dose of electrons for imaging [68, 69]. If the dose rate can improve by 10–100 folds, the tomographic resolution will be augmented significantly. If so, it might become feasible to locate individual lipid molecules directly or achieve the minimum to distinguish phospholipids from nonphospholipids within the annular layer of a particular VGIC.
Fourth, from a physiological angle, it is interesting to ask if targeted delivery of nonphospholipid analogs may be used for treating the life-threatening neurological diseases suffered by human patients carrying lipid metabolic defects [70, 71]. Biopsy fibroblasts from these patients can be isolated and reprogrammed into neurons in order to test these analogs before their clinical usage will be pursued. These analogs may be easier to deliver and can be constructed as biodegradable and will satisfy current shortage of suitable clinical measures for these patients and extend their lifespans for better quality of lives [72].
Fifth, super-resolution fluorescence microscopy (srFM) that can reach 1.0 nm or better in resolution may be suitable for imaging nonphospholipids directly if proper labeling can be done. For example, MINFLUX nanoscopy shows the promise to localize individual fluorophores with ∼1.0 nm resolution [73]. With a fluorophore on a VGIC and labels on the nonphospholipids, a system may be constructed for determining the distance of nearest nonphospholipids to the channel. If the tetrameric nature of the channel can be used to orient it in membrane, it may even be feasible to separate the nonphospholipids at the periphery of the VSDs from those closer to the pore domain (as in Figure 5A). Proper labels still await development.
It is for sure a big challenge, if not an impossible task, to exhaust feasible avenues for the near future. The exciting perspective is that with new methods, it will for the first time become feasible to probe the importance of annular lipids to membrane protein functions with experimental evidence, far better than the qualitative predictions from a macroscopic view of the fluid mosaic model [35].
Lipid-dependent gating of VGICs and metabolic defects in fatty acids and other categories of nonphospholipids are probably tightly connected. The underlying natural driving force appears the co-evolution of voltage-gated ion channels with the phospholipid-rich biomembranes in all three-kingdoms of life. The thermodynamic principles are hinged on the collective interactions at a dozen or so lipid-binding sites of low-to-intermediate affinity with varying occupancy. New methods are still desired to directly dissect the annular lipids around individual channels for a physical test of the lipid-dependent gating hypothesis.
We would like to thank Dr. Roderick MacKinnon for supporting QXJ to pursue the direction of lipid-dependent gating as an independent investigator. The work in the Jiang laboratory in lipid-dependent gating and other programs has been supported by different funding agencies, including NIH (R21GM131231, R01GM111367, R01GM093271, & R01GM088745), AHA (12IRG9400019), CF Foundation (JIANG15G0), Welch Foundation (I-1684), CPRIT (RP120474) and intramural funds at the UT Southwestern Medical Center, the University of Florida and the Hauptman-Woodward Medical Research Institute (HWI). We should also mention that the earlier efforts (before year 2014) of using electron crystallography to elucidate the structural basis of lipid-dependent effects on Kv channels were mainly invested by the Tamir Gonen lab because of limited access of high-end cryo-EM and lack of expertise in analyzing 2D crystals in the Jiang lab. We thank Dr. Gonen and Dr. Liang Shi for transferring the 2D crystal project to the Gonen lab and investing their time in crystal optimization, data collection and analysis in face of technical difficulties and pitfalls in different areas. Only till recently (after year 2019), the Jiang lab was able to gain enough cryo-EM time in order to achieve near-atomic resolutions of Kv channels in lipid environments. We thank Dr. Gaya Yadav, now at TAMU, for his persistent efforts in this area.
1.Jiang QX. Introudction: New tools for challenges in membrane biology. In: Jiang QX, editor. New Techniques for Studying Biomembranes. Methods in Signal Transduction. 1st ed. Boca Raton, FL: CRC Press; 2020. pp. 1-8
2.Wigge C, Stefanovic A, Radjainia M. The rapidly evolving role of cryo-EM in drug design. Drug Discovery Today: Technologies. 2021. pp. 91-102
3.Alberts B, Johnson A, Lewis J, Raff M, Roberts K, Walter P. Molecular Biology of the Cell. 5th ed. New York: Galand Science; 2007
4.Jiang QX. Lipid-dependent gating of ion channels. In: Catala A, editor. Protein-Lipid Interactions: Perspectives, Techniques and Challenges. U.S.: Nova Science Publishers, Inc.; 2018. p. 196
5.Hille B. Ion Channels of Excitable Membranes. 3rd ed. Sunderland, MA, USA: Sinauer Associates, Inc.; 2001
6.Iwasa J, Marshall W. Karp’s Cell and Molecular Biology. 8th ed. USA: Wiley; 2016. p. 740
7.Jiang QX. High-resolution structures of K(+) channels. Handbook of Experimental Pharmacology. 2021;267:51-81
8.Jiang QX. Cholesterol-dependent gating effects on ion channels. Advances in Experimental Medicine and Biology. 2019;1115:167-190
9.MacKinnon R. Potassium channels. FEBS Letters. 2003;555(1):62-65
10.Sigworth FJ. Voltage gating of ion channels. Quarterly Reviews of Biophysics. 1994;27(1):1-40
11.Guo J, Zeng W, Chen Q, Lee C, Chen L, Yang Y, et al. Structure of the voltage-gated two-pore channel TPC1 from Arabidopsis thaliana. Nature. 2016;531(7593):196-201
12.Kintzer AF, Stroud RM. Structure, inhibition and regulation of two-pore channel TPC1 from Arabidopsis thaliana. Nature. 2016;531(7593):258-262
13.Kintzer AF, Stroud RM. On the structure and mechanism of two-pore channels. The FEBS Journal. 2018;285(2):233-243
14.Payandeh J, Scheuer T, Zheng N, Catterall WA. The crystal structure of a voltage-gated sodium channel. Nature. 2011;475(7356):353-358
15.Payandeh J, Gamal El-Din TM, Scheuer T, Zheng N, Catterall WA. Crystal structure of a voltage-gated sodium channel in two potentially inactivated states. Nature. 2012;486(7401):135-139
16.Tao X, Lee A, Limapichat W, Dougherty DA, MacKinnon R. A gating charge transfer center in voltage sensors. Science. 2010;328:67-73
17.JIang QX, Levitan I. Principles of cholesterol regulation of ion channels. In: Dopico AN, Ba AM, editors. Cholesterol: From Chemistry and Biophysics to the Clinic. London, San Diego, Oxford, Cambridge: Elsevier Inc.; 2022. pp. 1021-1037
18.Zheng H, Liu W, Anderson LY, Jiang QX. Lipid-dependent gating of a voltage-gated potassium channel. Nature Communications. 2012;2:250
20.MAPS® L. The LIPID MAPS® Lipidomics Gatway 2022. Available from: https://www.lipidmaps.org/
21.Su XQ, Wang J, Sinclair AJ. Plasmalogens and Alzheimer's disease: A review. Lipids in Health and Disease. 2019;18(1):100
22.Bichenkov E, Ellingson JS. Temporal and quantitative expression of the myelin-associated lipids, ethanolamine plasmalogen, galactocerebroside, and sulfatide, in the differentiating CG-4 glial cell line. Neurochemical Research. 1999;24(12):1549-1556
23.Wood PL, Khan MA, Mankidy R, Smith T, Goodenowe DB. Plasmalogen deficit: A new and testable hypothesis for the etiology of Alzheimer’s disease. In: SDL M, editor. Alzheimer's Disease Pathogenesis - Core Concepts, Shifting Paradigms and Therapeutic Targets. 2011
24.Huang X, Sterling NW, Du G, Sun D, Stetter C, Kong L, et al. Brain cholesterol metabolism and Parkinson's disease. Movement Disorders. 2019;34(3):386-395
25.Cariati I, Masuelli L, Bei R, Tancredi V, Frank C, D'Arcangelo G. Neurodegeneration in Niemann-pick type C disease: An updated review on pharmacological and non-pharmacological approaches to counteract brain and cognitive impairment. International Journal of Molecular Sciences. 2021;22(12)
26.Shekhawat PS, Matern D, Strauss AW. Fetal fatty acid oxidation disorders, their effect on maternal health and neonatal outcome: Impact of expanded newborn screening on their diagnosis and management. Pediatric Research. 2005;57(5 Pt 2):78R-86R
27.Merritt JL, Norris M, Kanungo S. Fatty acid oxidation disorders. Annals of Translational Medicine. 2018;6(24):473
28.Weber FD, Wiesinger C, Forss-Petter S, Regelsberger G, Einwich A, Weber WH, et al. X-linked adrenoleukodystrophy: Very long-chain fatty acid metabolism is severely impaired in monocytes but not in lymphocytes. Human Molecular Genetics. 2014;23(10):2542-2550
29.van der Veen JN, Kennelly JP, Wan S, Vance JE, Vance DE, Jacobs RL. The critical role of phosphatidylcholine and phosphatidylethanolamine metabolism in health and disease. Biochimica et Biophysica Acta—Biomembranes. 2017;1859(9 Pt B):1558-1572
30.Hoflinger P, Hauser S, Yutuc E, Hengel H, Griffiths L, Radelfahr F, et al. Metabolic profiling in serum, cerebrospinal fluid, and brain of patients with cerebrotendinous xanthomatosis. Journal of Lipid Research. 2021;62:100078
31.Nie S, Chen G, Cao X, Zhang Y. Cerebrotendinous xanthomatosis: A comprehensive review of pathogenesis, clinical manifestations, diagnosis, and management. Orphanet Journal of Rare Diseases. 2014;9:179
32.Azevedo O, Cordeiro F, Gago MF, Miltenberger-Miltenyi G, Ferreira C, Sousa N, et al. Fabry disease and the heart: A comprehensive review. International Journal of Molecular Sciences. 2021;22(9)
33.Schuchman EH, Desnick RJ. Types a and B Niemann-pick disease. Molecular Genetics and Metabolism. 2017;120(1–2):27-33
34.Cachon-Gonzalez MB, Zaccariotto E, Cox TM. Genetics and therapies for GM2 Gangliosidosis. Current Gene Therapy. 2018;18(2):68-89
35.Nicolson GL. The fluid-mosaic model of membrane structure: Still relevant to understanding the structure, function and dynamics of biological membranes after more than 40 years. Biochimica et Biophysica Acta. 2014;1838(6):1451-1466
36.Cang C, Zhou Y, Navarro B, Seo YJ, Aranda K, Shi L, et al. mTOR regulates lysosomal ATP-sensitive two-pore Na(+) channels to adapt to metabolic state. Cell. 2013;152(4):778-790
37.Schoppa NE, Sigworth FJ. Activation of shaker potassium channels. III. An activation gating model for wild-type and V2 mutant channels. The Journal of General Physiology. 1998;111(2):313-342
38.Islas LD, Sigworth FJ. Electrostatics and the gating pore of shaker potassium channels. The Journal of General Physiology. 2001;117(1):69-89
39.Schoppa NE, Sigworth FJ. Activation of shaker potassium channels. II. Kinetics of the V2 mutant channel. The Journal of General Physiology. 1998;111(2):295-311
40.Schmidt D, Jiang QX, MacKinnon R. Phospholipids and the origin of cationic gating charges in voltage sensors. Nature. 2006;444(7120):775-779
41.Zheng H, Lee S, Llaguno MC, Jiang QX. bSUM: A bead-supported unilamellar membrane system facilitating unidirectional insertion of membrane proteins into giant vesicles. The Journal of General Physiology. 2016;147(1):77-93
42.Lee S, Zheng H, Shi L, Jiang QX. Reconstitution of a Kv channel into lipid membranes for structural and functional studies. Journal of Visualized Experiments. 2013;77:e50436
43.Amsalem M, Poilbout C, Ferracci G, Delmas P, Padilla F. Membrane cholesterol depletion as a trigger of Nav1.9 channel-mediated inflammatory pain. The EMBO Journal. 2018;37(8):e97349
44.Gantz SC, Bean BP. Cell-autonomous excitation of midbrain dopamine neurons by endocannabinoid-dependent lipid signaling. Neuron. 2017;93(6):1375-87 e2
45.Balajthy A, Somodi S, Petho Z, Peter M, Varga Z, Szabo GP, et al. 7DHC-induced changes of Kv1.3 operation contributes to modified T cell function in Smith-Lemli-Opitz syndrome. Pflügers Archiv. 2016;468(8):1403-1418
46.Rudakova E, Wagner M, Frank M, Volk T. Localization of Kv4.2 and KChIP2 in lipid rafts and modulation of outward K+ currents by membrane cholesterol content in rat left ventricular myocytes. Pflügers Archiv. 2015;467(2):299-309
47.Heaps CL, Bowles DK, Sturek M, Laughlin MH, Parker JL. Enhanced L-type Ca2+ channel current density in coronary smooth muscle of exercise-trained pigs is compensated to limit myoplasmic free Ca2+ accumulation. The Journal of Physiology. 2000;528(Pt 3):435-445
48.Xia F, Gao X, Kwan E, Lam PP, Chan L, Sy K, et al. Disruption of pancreatic beta-cell lipid rafts modifies Kv2.1 channel gating and insulin exocytosis. The Journal of Biological Chemistry. 2004;279(23):24685-24691
49.Randich AM, Cuello LG, Wanderling SS, Perozo E. Biochemical and structural analysis of the hyperpolarization-activated K(+) channel MVP. Biochemistry. 2014;53(10):1627-1636
50.Li Q, Wanderling S, Sompornpisut P, Perozo E. Structural basis of lipid-driven conformational transitions in the KvAP voltage-sensing domain. Nature Structural & Molecular Biology. 2014;21(2):160-166
51.Ramu Y, Xu Y, Lu Z. Enzymatic activation of voltage-gated potassium channels. Nature. 2006;442(7103):696-699
52.Xu Y, Ramu Y, Lu Z. Removal of phospho-head groups of membrane lipids immobilizes voltage sensors of K+ channels. Nature. 2008;451(7180):826-829
53.Pouvreau S, Berthier C, Blaineau S, Amsellem J, Coronado R, Strube C. Membrane cholesterol modulates dihydropyridine receptor function in mice fetal skeletal muscle cells. The Journal of Physiology. 2004;555(Pt 2):365-381
54.Hajdu P, Varga Z, Pieri C, Panyi G, Gaspar R Jr. Cholesterol modifies the gating of Kv1.3 in human T lymphocytes. Pflügers Archiv. 2003;445(6):674-682
55.Bowles DK, Heaps CL, Turk JR, Maddali KK, Price EM. Hypercholesterolemia inhibits L-type calcium current in coronary macro-, not microcirculation. Journal of Applied Physiology. 1985;96(6):2240-2248
56.Chun YS, Shin S, Kim Y, Cho H, Park MK, Kim TW, et al. Cholesterol modulates ion channels via down-regulation of phosphatidylinositol 4,5-bisphosphate. Journal of Neurochemistry. 2010;112(5):1286-1294
57.Praggastis M, Tortelli B, Zhang J, Fujiwara H, Sidhu R, Chacko A, et al. A murine Niemann-pick C1 I1061T knock-in model recapitulates the pathological features of the most prevalent human disease allele. The Journal of Neuroscience. 2015;35(21):8091-8106
58.Lee AG. Interfacial binding sites for cholesterol on Kir, Kv, K2P, and related potassium channels. Biophysical Journal. 2020;119(1):35-47
59.Guixa-Gonzalez R, Albasanz JL, Rodriguez-Espigares I, Pastor M, Sanz F, Marti-Solano M, et al. Membrane cholesterol access into a G-protein-coupled receptor. Nature Communications. 2017;8:14505
60.Ruprecht JJ, Mielke T, Vogel R, Villa C, Schertler GF. Electron crystallography reveals the structure of metarhodopsin I. The EMBO Journal. 2004;23(18):3609-3620
61.Prasanna X, Sengupta D, Chattopadhyay A. Cholesterol-dependent conformational plasticity in GPCR dimers. Scientific Reports. 2016;6:31858
62.Gimpl G. Interaction of G protein coupled receptors and cholesterol. Chemistry and Physics of Lipids. 2016;199:61-73
63.Faried M, Suga K, Okamoto Y, Shameli K, Miyake M, Umakoshi H. Membrane surface-enhanced Raman spectroscopy for cholesterol-modified lipid systems: Effect of gold nanoparticle size. ACS Omega. 2019;4(9):13687-13695
64.Gorman BL, Kraft ML. High-resolution secondary ion mass spectrometry analysis of cell membranes. Analytical Chemistry. 2020;92(2):1645-1652
65.Ferraro NA, Cascio M. Cross-linking-mass spectrometry studies of cholesterol interactions with human alpha1 glycine receptor. Analytical Chemistry. 2018;90(4):2508-2516
66.Yao M, Elling FJ, Jones C, Nomosatryo S, Long CP, Crowe SA, et al. Heterotrophic bacteria from an extremely phosphate-poor lake have conditionally reduced phosphorus demand and utilize diverse sources of phosphorus. Environmental Microbiology. 2016;18(2):656-667
67.Schur FK. Toward high-resolution in situ structural biology with cryo-electron tomography and subtomogram averaging. Current Opinion in Structural Biology. 2019;58:1-9
68.Liu N, Zhang J, Chen Y, Liu C, Zhang X, Xu K, et al. Bioactive functionalized monolayer graphene for high-resolution Cryo-electron microscopy. Journal of the American Chemical Society. 2019;141(9):4016-4025
69.Llaguno MC, Xu H, Shi L, Huang N, Zhang H, Liu Q, et al. Chemically functionalized carbon films for single molecule imaging. Journal of Structural Biology. 2014;185(3):405-417
70.Immordino ML, Dosio F, Cattel L. Stealth liposomes: Review of the basic science, rationale, and clinical applications, existing and potential. International Journal of Nanomedicine. 2006;1(3):297-315
71.Rufino-Ramos D, Albuquerque PR, Carmona V, Perfeito R, Nobre RJ, Pereira de Almeida L. Extracellular vesicles: Novel promising delivery systems for therapy of brain diseases. Journal of Controlled Release. 2017;262:247-258
72.Marques ARA, Saftig P. Lysosomal storage disorders - challenges, concepts and avenues for therapy: Beyond rare diseases. Journal of Cell Science. 2019;132(2)
73.Eilers Y, Ta H, Gwosch KC, Balzarotti F, Hell SW. MINFLUX monitors rapid molecular jumps with superior spatiotemporal resolution. Proceedings of the National Academy of Sciences of the United States of America. 2018;115(24):6117-6122
Written By
Qiu-Xing Jiang and Felix Chin
Submitted: 28 July 2022Reviewed: 18 August 2022Published: 31 May 2023
Open access peer-reviewed chapter
