Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2018 Nov 22;597(1):29–40. doi: 10.1113/JP274113

Dynamic structural rearrangements and functional regulation of voltage‐sensing phosphatase

Souhei Sakata 1,✉, Yasushi Okamura 2,✉
PMCID: PMC6312414  PMID: 30311949

Abstract

The voltage‐sensing phosphatase (VSP) consists of a voltage sensor domain (VSD) and a cytoplasmic catalytic region. The latter contains a phosphatase domain and a C2 domain, showing remarkable similarity to the tumour suppressor enzyme PTEN. In VSP, membrane depolarization induces a conformational change in the VSD, which activates the phosphoinositide phosphatase. The final outcome in VSP is enzymatic activity in the cytoplasmic region, unlike in voltage‐gated ion channels where conformational change of the transmembrane pore is induced by the VSD. Therefore, it is crucial to detect structural change in the cytoplasmic catalytic region to gain insights into the operating mechanisms of VSP. This review summarizes a recent study in which a method of genetic incorporation of a non‐canonical amino acid, Anap, was used to detect dynamic membrane voltage‐controlled rearrangements of the structure of the catalytic region of sea squirt VSP (Ci‐VSP). Upon membrane depolarization, both the phosphatase domain and the C2 domain move in a similar time frame, suggesting that the two regions are coupled to each other. Measurement of Förster resonance energy transfer (FRET) between Anap introduced into the C2 domain of Ci‐VSP and dipicrylamine in the cell membrane suggested no large movement of the enzyme towards the membrane. Fluorescence changes in Anap induced by different membrane potentials indicate the presence of multiple conformations of the active enzyme.

Keywords: phosphoinositide, membrane potential, voltage sensor, amber suppressor tRNA

Introduction

Transduction of an electrical signal into a chemical one is an essential mechanism underlying many physiological functions, including synaptic transmission in neurons, hormone secretion from endocrine cells, and contraction of muscle cells. All those events are triggered through interplay among multiple proteins leading to an increase in intracellular calcium ions as a result of the activation of voltage‐gated ion channels (VGICs) in the plasma membrane. Fourteen years ago, bioinformatics analysis of the genome of Ciona intestinalis (sea squirt) led to the identification of a gene encoding a voltage‐sensing phosphatase (Ci‐VSP), which contains a voltage sensor domain (VSD) similar to that of VGICs but lacks a pore gate domain (Murata et al. 2005). Instead, VSP contains a cytoplasmic region that is similar to the well‐known tumour suppressor phosphatase PTEN, which dephosphorylates PIP3, a key signalling phosphoinositide (PI), into PI(4,5)P2, thereby antagonizing PI3 kinase (Fig. 1). PTEN plays a critical role in the regulation of cell signalling for cell survival, cell proliferation and energy metabolism during many biological and pathological events (Maehama et al. 2001; Okamura & Dixon, 2011; Worby & Dixon, 2014). Within VSP, membrane depolarization‐induced motion of the VSD activates the cytoplasmic enzyme region, resulting in dephosphorylation of PIP3, PI(4,5)P2 or PI(3,4)P2. That in turn alters the activities of various ion channels (Suh & Hille, 2005, 2008; Falkenburger et al. 2010) and transporters (Thornell & Bevensee, 2015), and may also induce changes in cell morphology (Yamaguchi et al. 2014). VSP thus provides a unique example of a membrane protein that performs electrochemical transduction as the single protein. VSP is conserved from marine invertebrates to humans, where it is expressed in testis, epithelium and neural tissue (Okamura et al. 2018).

Figure 1. Comparison of domain architectures of VSP, PTEN and voltage‐gated ion channels.

Figure 1

Open in a new tab

The voltage sensor domain (VSD) consists of four transmembrane helices, S1–S4, with a signature pattern of alignment of multiple positively charged residues on S4.

The structure of the VSD in VSP shows remarkable similarity to the VSD in VGICs, i.e. it consists of four transmembrane helices among which the fourth, S4, contains a signature sequence consisting of series of positively charged residues with two intervening hydrophobic residues. X‐ray crystallographic structures of the isolated VSD truncated before the cytoplasmic PTEN‐like region show that the multiple positively charged residues (R1–R4) in S4 form salt bridges with acidic residues in the other helices (Li et al. 2014). Moreover, when transferred to viral potassium channels, the VSD of Ci‐VSP confers voltage sensitivity to the channels (Arrigoni et al. 2013), confirming that the VSD mechanism is functionally well conserved between VGICs and VSP. The cytoplasmic catalytic region of VSP shows strong similarity to PTEN in that it shares a similar domain architecture consisting of a phosphatase domain (PD) and a C2 domain (C2D), though VSP lacks the C‐terminal disordered region for enzyme regulation found in PTEN. The PD of VSP forms a substrate binding region that consists mainly of α‐helices and flexible loop regions. The C2D contains β‐sheets and loop regions that orient towards the membrane. X‐ray crystallographic analyses of the isolated cytoplasmic region of Ci‐VSP show that the substrate binding pocket is formed not only by residues of the PD, but also by a part of the C2D, which is unlike PTEN (Matsuda et al. 2011; Liu et al. 2012). In addition, the substrate specificity of VSP differs slightly from that of PTEN. Whereas PTEN exclusively cleaves the 3‐phosphate from PI(3,4,5)P3 and PI(3,4)P2, VSP shows 5‐phosphate phosphatase activity towards PI(4,5)P2 and PI(3,4,5)P3, as well as weak but substantial 3‐phosphate phosphatase activity toward PI(3,4,5)P3 (Castle et al. 2015; Keum et al. 2016) and PI(3,4)P2 (Kurokawa et al. 2012; Liu et al. 2012). The VSD and PD are connected by a 20‐amino acid linker (VSD‐PD linker) consisting of two major regions: one that is unique to VSP and a more C‐terminal segment that is conserved between VSP and PTEN and contains multiple positively charged residues. Studies examining the effects of deleting or mutating the linker have shown that this region is essential for coupling the activities of the VSD and enzyme (Murata et al. 2005; Villalba‐Galea et al. 2009; Kohout et al. 2010).

The molecular design of the VSP has two notable features that contrast with that of VGICs. First, a single VSD regulates a single catalytic region in VSP. This is in sharp contrast to VGICs, where four homologous units, either as repeats within the same subunit (as in eukaryotic Nav and Cav channels) or as separate subunits (as in Kv and HCN channels), form an ion‐permeation path, and multiple VSDs act in concert to regulate the central pore. The concerted action of multiple VSDs in regulating the opening of a single pore underlies the sharp voltage‐dependent changes in ion flux through VGICs; e.g., voltage‐gated sodium channels increase their conductance 10‐fold with a change in membrane potential of a few millivolts (Hille, 2001). Second, within VSP, the voltage‐evoked motion of the VSD ultimately leads to a final readout in a cytoplasmic structure where PI binds to an active centre and phosphate is subsequently removed from the inositol ring. By contrast the final outcome in VGICs is a conformational change in a transmembrane structure, the pore gate. Important questions that arise from this difference are how does the VSD regulate the cytoplasmic enzyme region of VSP and what coupling mechanisms are shared between VGICs and VSP? Answering these questions may provide insight into the mechanisms underlying other voltage‐evoked signals that are not accompanied by an ion flux such as the mechanical coupling between Cav1.1 and the ryanodine receptor for excitation‐contraction coupling in skeletal muscle or the calcium‐independent, voltage‐dependent secretion of neurotransmitters (Ci‐VDS) from neurons (Bannister & Beam, 2013; Chai et al. 2017).

The VSD is tightly coupled to the enzyme in VSP

The coupling mechanisms (Arrigoni et al. 2013) within VSP have been studied by comparing the motion of the VSD with the enzyme activity. VSD motion can be quantified by measuring sensing currents, which correspond to the gating currents of VGICs, or by using voltage clamp fluorometry, which detects changes in a fluorescent indicator‐conjugated amino acid residue introduced through mutagenesis. Readouts of voltage‐dependent enzyme activity can be quantified by detecting the level of PIs using two methods: electrophysiological detection of PI(4,5)P2‐dependent activities of ion channels and detection of the signal from a fluorescent protein fused to a PI‐recognizing protein module. Advantages of the former are the relatively rapid speed of the readout and the ability to use a conventional electrophysiology setup, such as the two‐electrode voltage clamp with Xenopus oocytes or patch clamp. With the latter, PI(4,5)P2 and other PIs can be quantified using fluorescent protein sensors with protein modules specific to each PI species. The studies involved in measuring the voltage‐dependent phosphatase activity of VSP have been reviewed in detail elsewhere (Okamura et al. 2018). Quantification of VSP enzyme activities in living cells has led to two major conclusions: (1) VSD motion is tightly coupled to the enzyme over a wide range of membrane potentials, including unphysiologically large depolarizations of more than 100 mV, and (2) activation of the enzyme following the motion of the VSD is surprisingly rapid.

The enzyme activity of VSP is graded depending on the extent of the motion of the VSD. In an experiment where enzyme activity was measured at different membrane potentials using a mutant zebrafish VSP in which the VSD was stabilized at an intermediate activation state (Sakata & Okamura, 2014), enzyme activity varied biphasically in parallel with the biphasic voltage‐dependent changes in VSD motion. An analysis of the voltage‐evoked enzyme activities of several Ci‐VSP VSD mutants performed using Förster resonance energy transfer (FRET)‐based rapid PI sensors under the control of the membrane potential provided evidence that the cytoplasmic enzyme region of Ci‐VSP has two distinct states, possibly with biased substrate preferences (Grimm & Isacoff, 2016). On the other hand, a different conclusion was reached in another study, where all four subreactions (PI(3,4,5)P3 → PI(4,5)P2, PIP3 → PI(3,4)P2, PIP2 → PI4P, PI(3,4)P2 → PI4P) were analysed using fluorescent FRET sensors of the PIs (Keum et al. 2016). A mathematical model of a simple two‐state transition of the enzyme could account for the complex profiles of PIs during voltage‐evoked VSP enzyme activity without the presumption of distinct voltage‐dependencies among the four subreactions. To understand in detail the coupling mechanisms within VSP, one needs to monitor the structural changes in the cytoplasmic enzyme region directly. However, the structural changes in the cytoplasmic enzyme region have been difficult to examine in living cells.

Genetically encoded fluorescent unnatural amino acid

Fluorescent probes have been used to uncover conformational changes within ion channels (Mannuzzu et al. 1996; Cha & Bezanilla, 1997). With this method, channels in the plasma membrane are labelled through application of a fluorescent dye (e.g. rhodamine or an Alexa fluor family member) conjugated with a maleimide group. Site specificity is ensured by mutating the target site to cysteine. If a channel operation is accompanied by a change in the local environment where the probe was introduced, this conformational change is reported as a change of fluorescence intensity that is sensitive to the environment around the fluorophore. This method has been used effectively to examine the operation of ion channels. However, it has two limitations. One is that the fluorophores commonly used for these studies are larger (>700 Da) than the side‐chains of amino acids (≤200 Da). This raises the possibility that introduction of the fluorophore will alter the local protein structure such that environmental changes in the local region around the target residue are not faithfully reported. The second is that intracellular targets cannot be selectively labelled. Although we are able to introduce the probe into the cytoplasm through microinjection, site‐direct labelling may not occur because there are many cysteine‐containing proteins in the cytoplasm that compete with the protein of interest in the membrane. Consequently, it has not been possible to use this method to examine conformational changes within the cytoplasmic catalytic domain of Ci‐VSP.

Another approach entails protein modification through genetic incorporation of artificially synthesized amino acid derivatives (unnatural amino acids). The group led by Drs. Lester and Dougherty has developed a method that uses nonsense codon suppression (Nowak et al. 1995; Nowak et al. 1998). Here, two kinds of RNAs are introduced into a Xenopus oocyte (Fig. 2 A): mRNA encoding a protein of interest and containing an amber stop codon (UAG) at the target position of the protein and a suppressor tRNA containing a CUA codon that is aminoacylated with a synthesized unnatural amino acid. The tRNA‐amino acid complex directs the mRNA's amber codon during protein synthesis. Using this method for site‐directed unnatural amino acid incorporation, they revealed the gating mechanisms of 5‐HT3 receptors (Lummis et al. 2005), as well as the role of the cation‐π interaction between the membrane‐spanning helices of G‐protein coupled receptors (Torrice et al. 2009). This method can also be applied to introduce a fluorescent probe. If a fluorescent artificial amino acid were synthesized, it could be incorporated anywhere within a protein by mutating the target site to a TAG codon. However, no fluorescent amino acid is available, and the technical challenge of in vitro aminoacylation of tRNA makes it impractical for most of ion channel researchers.

Figure 2. Schematic views of genetic incorporation of unnatural amino acids.

Figure 2

Open in a new tab

A, tRNA chemically aminoacylated with an unnatural amino acid is introduced into cells along with mRNA containing the amber stop codon. The unnatural amino acid is incorporated into the target site encoded by the amber stop codon. B, tRNA and aminoacyl‐tRNA synthetase (aaRS) are expressed in cells by introducing a plasmid expression vector into the nucleus. One day later, the unnatural amino acid (Anap) and mRNA containing the amber stop codon are co‐injected into the cell. The tRNA is aminoacylated with Anap by the aaRS in the cell, and the complex is targeted to the mRNA.

Dr Schultz's group overcame that limitation by exploiting another approach to genetic incorporation of an unnatural amino acid. With this method, the tRNA is aminoacylated in vivo by introducing three components into cells: a desired unnatural amino acid, a tRNA, and an aminoacyl‐tRNA synthetase (aaRS). The aaRS catalyses the binding between the tRNA and the unnatural amino acid within cells, after which the tRNA‐amino acid complex is sent to the ribosome to be incorporated into a polypeptide chain. To ensure specific incorporation of the unnatural amino acid at the targeted position, the following requirements must be met: (1) the RNA must be constructed such that it is not recognized by endogenous aminoacyl‐tRNA synthetases during translation;(2) the aminocyl‐tRNA synthetase acylates only the exogenous tRNA and no endogenous tRNAs; and (3) only the desired unnatural amino acid is bound to the exogenous tRNA. In short, new tRNA/aminoacyl‐tRNA synthetase pairs that are independent of those for the 20 endogenous amino acids must be developed. To establish an orthogonal tRNA/aminoacyl‐tRNA synthetase pair, Dr Schultz's group performed mutation screenings of natural tRNAs and aaRSs, successfully adding a variety of unnatural amino acids to the genetic codes of E. coli., yeast and mammalian cells. These were used to reveal protein interactions and structural changes, as well as biochemical and cellular functions within living cells (Chin et al. 2002; Wang et al. 2003, 2006b; Mori & Ito, 2006). They also studied genetic incorporation of a fluorescent unnatural amino acid, 3‐(6‐acetylnaphthalen‐2‐ylamino)‐2‐aminopropanoic acid (Anap) (Fig. 2 B), which is related to the dye 6‐propionyl‐2‐(N,N‐dimethyl)‐aminonaphthalene (prodan), an environmentally sensitive fluorophore. Because its molecular weight is comparable to that of a natural amino acid, tryptophan, Anap is a better probe than the commonly used maleimide conjugates, which have bulkier structures. This group first reported an orthogonal tRNA/aminoacyl‐tRNA synthetase pair for Anap that works in E. coli. and yeast (Summerer et al. 2006; Wang et al. 2006a) and then subsequently achieved Anap incorporation into mammalian cells (Chatterjee et al. 2013). In addition, Kalstrup & Blunck (2013) showed that Anap can be incorporated into Kv channels in Xenopus oocytes and that conformational changes in the channel are reported as changes in Anap fluorescence intensity. This method opened a new path to site‐directed fluorescent labelling of the cytoplasmic portions of proteins.

Genetic incorporation of Anap revealed the voltage‐dependent conformational changes in the catalytic domain of Ci‐VSP

To study the structural changes in the cytoplasmic region of Ci‐VSP, we applied genetic incorporation of Anap in a Xenopus oocyte expression system. A plasmid encoding the orthogonal tRNA and aaRS for Anap was injected into the nucleus of oocytes. One day later, Ci‐VSP mRNA in which the target site was mutated to a UAG codon was co‐injected with Anap (Fig. 2 B). We used two band‐pass emission filters: one passed fluorescence between 420 and 460 nm, while the other passed fluorescence between 460 and 510 nm, and the respective fluorescences were detected using a photomultiplier tube. We first incorporated Anap into several sites within the ‘gating loop’ of the PD, which an earlier crystallographic study suggested would move upon voltage sensor activation (Liu et al. 2012). Many sites within the gating loop showed voltage‐dependent fluorescence changes (Fig. 3 A). These observations showed directly for the first time that the catalytic domain changes its conformation in association with voltage sensor movement. The kinetics of the change in fluorescence from F401Anap recorded through the 420–460 nm filter were as rapid as the kinetics of the sensing currents (Fig. 3 A). The latency of the fluorescence change following the voltage step was a few milliseconds (Sakata et al. 2016). It has been suggested that the C2D plays a key role in keeping the catalytic domain beneath the plasma membrane. To examine the conformational change in the C2D, we incorporated Anap into the loops of the C2D and recorded the voltage‐dependent fluorescence changes (Fig. 3 B). Our findings indicate that the C2D also changes its conformation or its distance from the membrane.

Figure 3. Voltage‐dependent changes in the fluorescence of Anap incorporated into the catalytic domain of Ci‐VSP.

Figure 3

Open in a new tab

A, voltage‐dependent changes in Anap fluorescence and its time constant. Anap was incorporated into F401. B, representative changes in Anap fluorescence evoked by 160‐mV test pulses. Anap was incorporated into the 515 loop and the Cα2 loop. The 515 loop and the Cα2 loop are the loop structures in the C2 domain, which includes amino acid residues from S513 to R520 and K553 to I559, respectively. The figure is reproduced from Sakata et al. (2016).

While characterizing the Anap fluorescence changes at numerous sites within the C2D, we made another important finding in an Anap‐incorporating construct, K555 (K555Anap), in which Anap was incorporated into the Cα2 loop of the C2D. We compared the voltage dependence of enzyme‐active and ‐inactive (C363S) forms of K555Anap Ci‐VSP and noticed that the fluorescence intensity of the enzyme‐inactive form increased as the membrane voltage increased, whereas the fluorescence intensity of the enzyme‐active form decreased until around 50 mV and then started to increase as the membrane voltage became higher (Fig. 4 A). This difference between the enzyme active form and the enzyme‐defective form may imply that the PI level in the membrane affects the voltage dependency of the fluorescence change. We compared the voltage‐dependent fluorescence changes in the enzyme‐active Ci‐VSP form using two pulse protocols that cause different PI levels. In one protocol, repetitive pulses were administered with short (1.2 s) intervals where PI level in the membrane gradually declines. In another protocol, long (60 s) pulse intervals were used where the PI level recovers when the membrane potential is held at −60 mV to silence Ci‐VSP during the pulse interval. The F‐V relationship of enzyme‐active K555Anap Ci‐VSP recorded using the short interval was fitted by the sum of two components with Boltzmann functions, while that recorded using the long interval was fitted by a single component (Fig. 4 B). Therefore, the conformation of the catalytic domain as probed by K555Anap depends on the substrate availability.

Figure 4. Voltage‐dependent changes in fluorescence from Anap incorporated in the C2 domain differ between enzyme‐active and enzyme‐inactive forms of Ci‐VSP.

Figure 4

Open in a new tab

A, traces show the change in fluorescence from Anap incorporated at K555 of the enzyme‐ active form of Ci‐VSP. B, the F‐V relationship of the enzyme‐active form of K555Anap Ci‐VSP recorded using short (1.2 s) and long (60 s) pulse intervals. Data are shown as the means ± SD (n = 4 and 9 for fluorescence measured using 1.2‐s and 60‐s test pulse intervals, respectively). C, pulse protocol used for the prepulse experiment shown in D. D, prepulse‐dependent changes in the kinetics of K555Anap (left panel) and Y522A/K555Anap (right panel) Ci‐VSP. The top panel shows the timing of the depolarization. Arrows show the kinetics that depend on a depolarizing prepulse. The left bottom panel depicts the fluorescence change in the absence of the pre‐pulse, recorded 1 min after measurement in the presence of the pre‐pulse, which is shown in the second panel from the bottom. Note that the kinetics indicated by the arrow in the second panel from the top were restored by clamping the membrane at −60 mV for about 1 min (left bottom panel). E, structure around the active centre and the Cα2 loop (PDB ID; 3V0H) (Liu et al. 2012). The Cα2 loop is shown in cyan. IP3 is located at the active centre. Dotted lines indicate a hydrogen bond. Panels A, C, D and E are reproduced from Sakata et al. (2016).

In addition to their voltage dependence, the kinetics of the fluorescence differed between enzyme‐active and ‐inactive K555Anap Ci‐VSP (Fig. 4 A). This difference in kinetics reflects differences in substrate availability. The kinetics of Anap fluorescence was compared between two conditions of PI(4,5)P2/PI(3,4)P2/PI(3,4,5)P3. These PI substrates were depleted by Ci‐VSP's own enzyme activity using a long depolarizing pre‐pulse (100 mV for 3 s) (Fig. 4 C and D). In the absence of the pre‐pulse, a transient decrease in fluorescence was seen just after the large increase evoked by each test pulse (arrows in Fig. 4 D), but that transient decrease was not observed when a pre‐pulse was applied.

The catalytic active centre is in the PD, but K555, where we observed the difference in the fluorescent signal, is in the C2D. So why is K555Anap fluorescence affected by substrate availability? Crystallographic studies have shown that Y522 bridges IP3 in the active centre and K553 in the Cα2 loop through hydrogen bonding (Fig. 4 E; Liu et al. 2012). When Y522 was mutated to alanine, the F‐V relationship became independent of the pulse interval, and the observed hump in the K555Anap fluorescence disappeared in both the presence and absence of the depolarizing pre‐pulse (Fig. 4 D). Thus, substrate binding at the active centre affects the conformation of the C2D through hydrogen bonding via Y522.

FRET analysis indicates that the catalytic domain stays beneath the plasma membrane in both the resting and activated state of the voltage sensor

Experiments using Anap demonstrated that upon activation of the VSD, the Ci‐VSP PD changes its conformation to accomplish its catalytic activity (Sakata et al. 2016). Because PIs, the substrates of Ci‐VSP, are a membrane component, it would make sense for the voltage sensor to regulate enzyme activity by controlling the distance between the PD and the plasma membrane. Whether the observed conformational change affected the distance between the PD and the membrane was tested using FRET analysis. In this case, dipicrylamine (DPA) served as the FRET acceptor for Anap because its absorption spectrum overlaps the fluorescence spectra of Anap (Fig 5 A; Chanda et al. 2005a). The membrane of Xenopus oocytes expressing S513Anap Ci‐VSP was stained using DPA. The S513 site was chosen for Anap incorporation because the fluorescence is unaffected by membrane potential changes (Fig. 3 B). It is known that DPA translocates to the outer and inner leaflets of the membrane upon hyper‐ and depolarization, respectively (Fig. 5 B; Chanda et al. 2005a, b ). Moreover, according to the charge‐voltage (Q‐V) relationship, DPA moves mainly at voltages lower than 0 mV, whereas the voltage sensor of Ci‐VSP is activated mainly at voltages higher than 0 mV (Fig. 5 B). If DPA works as a FRET acceptor, the Anap fluorescence should be increased upon hyperpolarization from 0 mV to −120 mV because the DPA moves towards the outer leaflet while the catalytic domain stays beneath the membrane (Fig. 5 C). On the other hand, if the catalytic domain moves up upon membrane depolarization from 0 mV to 160 mV, the fluorescence should be decreased (Fig. 5 C). If it does not move upon depolarization, the fluorescence should be unchanged (Fig. 5 C). We observed an increase in fluorescence upon hyperpolarization to −120 mV, and a slight decrease in the fluorescence was observed on depolarization to 160 mV (Fig. 5 D). Then to determine whether the decrease in fluorescence indicates movement of the catalytic domain, we performed the FRET experiment using D129R/S513Anap/C363S Ci‐VSP, in which the voltage sensor is immobilized (Tsutsui et al. 2013). We observed a slight decrease in fluorescence, just as with the wild‐type voltage sensor (Fig. 5 E and G), which means the catalytic domain is not brought closer to the plasma membrane upon voltage sensor activation. This suggests that the conformational change probably plays a more direct role in the phosphatase activity rather than changing the distance from the plasma membrane.

Figure 5. Test of VSD motion‐evoked changes in the distance between the C2D and plasma membrane based on FRET between Anap and DPA.

Figure 5

Open in a new tab

A, absorption spectrum of DPA (black curve) and fluorescence spectra of Anap. B, charge‐voltage relationships of DPA and Ci‐VSP. The inset shows the transient current recored from DPA‐stained oocytes. C, strategy for examining the movement of the catalytic domain. D and E, the changes fluorescence from C363S/S513Anap in wild‐type Ci‐VSP (D) and from D129R/C363S/S513Anap in a voltage‐sensor immobilized Ci‐VSP mutant (E). Insets show the fluorescence changes in the absence of DPA at −120 mV. Vertical and horizontal bars in the insets indicate 0.25% and 0.2 s, respectively. G, the voltage‐dependent changes of Anap fluorescence in the presence of DPA. Data are reproduced from Sakata et al. (2016).

Multiple conformations of the activated enzyme suggested by the voltage‐dependent changes in Anap fluorescence

In nearly all cases, the F‐V relationship was fitted by a single component with a Boltzmann function, like the Q‐V relationship of the voltage sensor. As shown in Fig. 4 A, however, the fluorescence from the enzyme‐active form of K555Anap Ci‐VSP shows bidirectional changes: it decreases at lower voltages, around 50 mV, but begins to increase as the membrane voltage gets higher. A F‐V relationship fitted by the sum of two components with a Boltzmann function is also observed with enzyme‐active K553Anap Ci‐VSP and with K558Anap, located in the Cα2 loop (Sakata et al. 2016), or R520Anap, located in the 515 loop of Ci‐VSP (Sakata et al. 2016). The voltage at which the increase and decrease in the fluorescence inverted was around 50 mV in all four constructs, implying that the catalytic domain has an additional conformation that is distinct from both the inactivated and fully‐activated states of the enzyme. With R520Anap incorporation, two‐step voltage dependence was found with both the enzyme‐active and ‐inactive forms of Ci‐VSP, suggesting that the third conformation does not reflect substrate availability. Given that the phosphatase activity of Ci‐VSP is graded depending on the degree of voltage sensor activation (Sakata & Okamura, 2014), this conformation may be associated with a partially activated state of the voltage sensor.

Future directions

Because the outcome of VSP action derives from the activity of the cytoplasmic catalytic region, detailed analysis of the effects of structural changes within that region will be critical for understanding the molecular mechanisms of VSP. Our recent application of the Anap method to Ci‐VSP has proven to be a powerful tool for detecting stimulus‐induced motion of the cytoplasmic region, which has been difficult to study using conventional methods, such as approaches based on cysteine‐maleimide reactions. Anap has also been utilized as a reporter of structural changes in other membrane proteins (Zagotta et al. 2016; Soh et al. 2017; Dai et al. 2018). However, achieving further understanding of the mechanisms involved in VSP activity through the use of fluorescent unnatural amino acids will probably require overcoming several important issues. At the moment, Anap provides important information about sites that are mobile. However, it is not possible to precisely define the mechanisms underlying the stimulus‐induced fluorescence changes. Among the sites of Anap incorporation within the cytoplasmic region of Ci‐VSP, there were some that showed reciprocal changes in the two wavelength bands, while other sites showed changes in the same direction in the two bands. More detailed information about the chemical properties of Anap in relation to its environment (e.g. tension or electrostatics) are awaited. Analysis of the Anap spectra within live cells will be one way to gain insight into the mechanism underlying the fluorescence changes. However, the intensity of Anap fluorescence is much weaker than that of Alexa or Cyanine series fluorophores. Consequently, single‐molecule imaging of Anap‐labelled molecules is currently impractical.

Despite the limitations of the Anap method, our experiments employing voltage clamp fluorometry of Anap fluorescence incorporated into numerous individual sites within the catalytic region of Ci‐VSP has enabled several observations that could not be made using other methods. First, we were able to identify regions that are mobile upon induction of enzyme activity. One surprising finding is that, in addition to the PD, the C2D also changes its structure upon activation of the VSD. This indicates that the structural changes associated with enzyme activation go beyond local changes within the substrate binding region (such as ‘gating loop’) as previously proposed (Liu et al. 2012). Second, the kinetics of the voltage‐dependent motion of the cytoplasmic region are comparable to the kinetics of the motion of the VSD. The speed of the Anap‐reported motion is faster than previously thought, and the motion of the C2D was no slower than that of the PD, despite the fact that only the PD is directly connected to the VSD. This is consistent with the idea that the PD and C2D change their structure as a unit. Third, by measuring FRET using DPA as a quencher of Anap fluorescence, we found no detectable change in distance between the C2D and the plasma membrane. Fourth, when Anap was incorporated into the C2D, the fluorescence intensity turned out to be biphasic along the voltage axis, indicating the enzyme exists in multiple states during the course of its activation. These multiple states may correspond to transitions among distinct states with different substrate preferences, as was proposed by Grimm & Isacoff (2016), or to transitions among different states with distinct levels of enzyme activity, as we proposed (Sakata & Okamura, 2014). Future mutagenesis studies dissecting the distinct components of Anap fluorescence changes will help us understand the biophysical mechanisms mediating the coupling between the activities of the VSD and the enzyme in VSP.

Additional information

Competing interests

The authors have no conflicts of interests related to this paper.

Author contributions

Both authors wrote the manuscript and approved the final version.

Funding

This work is supported by Grants‐in‐Aid from the Japan Society for the Promotion of Science (JSPS) (JP21229003, JP25253016, JP16H02617 to Y.O.), Ministry of Education, Culture, Sports, Science, and Technology (MEXT) (JP24111529, JP26111712, JP15H05901 to Y.O. and JP25860163 to S.S.), and Core Research for Evolutional Science and Technology (CREST, JST) (JPMJCR14M3 to Y.O.).

Acknowledgements

We would like to thank Dr Fumihito Ono for encouragement and support.

Biographies

Souhei Sakata is an associate professor at the Department of Physiology in Osaka Medical College, Japan. He studied the biophysics of voltage‐sensing phosphatases when he was in the Okamura lab at Osaka University. He is currently working on the genetic incorporation of an unnatural amino acid in zebrafish.

graphic file with name TJP-597-29-g001.gif

Yasushi Okamura is a professor at the Department of Physiology, Graduate School of Medicine, Osaka University. He has been studying the biophysical mechanisms and physiological roles of non‐canonical voltage‐sensor containing proteins, including voltage‐gated proton channels and voltage‐sensing phosphatase, which were discovered using a bioinformatics approach.

Edited by: Ole Petersen & Florian Lesage

Contributor Information

Souhei Sakata, Email: sakatas@osaka-med.ac.jp.

Yasushi Okamura, Email: yokamura@phys2.med.osaka-u.ac.jp.

References

  1. Arrigoni C, Schroeder I, Romani G, Van Etten JL, Thiel G & Moroni A (2013). The voltage‐sensing domain of a phosphatase gates the pore of a potassium channel. J Gen Physiol 141, 389–395. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bannister RA & Beam KG (2013). CaV1.1: The atypical prototypical voltage‐gated Ca2+ channel. Biochim Biophys Acta 1828, 1587–1597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Castle PM, Zolman KD & Kohout SC (2015). Voltage‐sensing phosphatase modulation by a C2 domain. Front Pharmacol 6, 63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Cha A & Bezanilla F (1997). Characterizing voltage‐dependent conformational changes in the Shaker K+ channel with fluorescence. Neuron 19, 1127–1140. [DOI] [PubMed] [Google Scholar]
  5. Chai Z, Wang C, Huang R, Wang Y, Zhang X, Wu Q, Wang Y, Wu X, Zheng L, Zhang C, Guo W, Xiong W, Ding J, Zhu F & Zhou Z (2017). CaV2.2 gates calcium‐independent but voltage‐dependent secretion in mammalian sensory neurons. Neuron 96, 1317—1326, e1314. [DOI] [PubMed] [Google Scholar]
  6. Chanda B, Asamoah OK, Blunck R, Roux B & Bezanilla F (2005a). Gating charge displacement in voltage‐gated ion channels involves limited transmembrane movement. Nature 436, 852–856. [DOI] [PubMed] [Google Scholar]
  7. Chanda B, Blunck R, Faria LC, Schweizer FE, Mody I & Bezanilla F (2005b). A hybrid approach to measuring electrical activity in genetically specified neurons. Nat Neurosci 8, 1619–1626. [DOI] [PubMed] [Google Scholar]
  8. Chatterjee A, Guo J, Lee HS & Schultz PG (2013). A genetically encoded fluorescent probe in mammalian cells. J Am Chem Soc 135, 12540–12543. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chin JW, Santoro SW, Martin AB, King DS, Wang L & Schultz PG (2002). Addition of p‐azido‐L‐phenylalanine to the genetic code of Escherichia coli . J Am Chem Soc 124, 9026–9027. [DOI] [PubMed] [Google Scholar]
  10. Dai G, James ZM & Zagotta WN (2018). Dynamic rearrangement of the intrinsic ligand regulates KCNH potassium channels. J Gen Physiol 150, 625–635. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Falkenburger BH, Jensen JB & Hille B (2010). Kinetics of PIP2 metabolism and KCNQ2/3 channel regulation studied with a voltage‐sensitive phosphatase in living cells. J Gen Physiol 135, 99–114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Grimm SS & Isacoff EY (2016). Allosteric substrate switching in a voltage‐sensing lipid phosphatase. Nat Chem Biol 12, 261–267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hille B (2001). Ion Channels of Excitable Membranes. Sinauer Associates, Inc, Sunderland, MA, USA. [Google Scholar]
  14. Kalstrup T & Blunck R (2013). Dynamics of internal pore opening in KV channels probed by a fluorescent unnatural amino acid. Proc Natl Acad Sci U S A 110, 8272–8277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Keum D, Kruse M, Kim DI, Hille B & Suh BC (2016). Phosphoinositide 5‐ and 3‐phosphatase activities of a voltage‐sensing phosphatase in living cells show identical voltage dependence. Proc Natl Acad Sci U S A 113, E3686–3695. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kohout SC, Bell SC, Liu L, Xu Q, Minor DL Jr & Isacoff EY (2010). Electrochemical coupling in the voltage‐dependent phosphatase Ci‐VSP. Nat Chem Biol 6, 369–375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kurokawa T, Takasuga S, Sakata S, Yamaguchi S, Horie S, Homma KJ, Sasaki T & Okamura Y (2012). 3ʹ Phosphatase activity toward phosphatidylinositol 3,4‐bisphosphate [PI(3,4)P2] by voltage‐sensing phosphatase (VSP). Proc Natl Acad Sci U S A 109, 10089–10094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Li Q, Wanderling S, Paduch M, Medovoy D, Singharoy A, McGreevy R, Villalba‐Galea CA, Hulse RE, Roux B, Schulten K, Kossiakoff A & Perozo E (2014). Structural mechanism of voltage‐dependent gating in an isolated voltage‐sensing domain. Nat Struct Mol Biol 21, 244–252. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Liu L, Kohout SC, Xu Q, Muller S, Kimberlin CR, Isacoff EY & Minor DL Jr (2012). A glutamate switch controls voltage‐sensitive phosphatase function. Nat Struct Mol Biol 19, 633–641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lummis SC, Beene DL, Lee LW, Lester HA, Broadhurst RW & Dougherty DA (2005). Cis‐trans isomerization at a proline opens the pore of a neurotransmitter‐gated ion channel. Nature 438, 248–252. [DOI] [PubMed] [Google Scholar]
  21. Maehama T, Taylor GS & Dixon JE (2001). PTEN and myotubularin: novel phosphoinositide phosphatases. Annu Rev Biochem 70, 247–279. [DOI] [PubMed] [Google Scholar]
  22. Mannuzzu LM, Moronne MM & Isacoff EY (1996). Direct physical measure of conformational rearrangement underlying potassium channel gating. Science (New York, NY) 271, 213–216. [DOI] [PubMed] [Google Scholar]
  23. Matsuda M, Takeshita K, Kurokawa T, Sakata S, Suzuki M, Yamashita E, Okamura Y & Nakagawa A (2011). Crystal structure of the cytoplasmic phosphatase and tensin homolog (PTEN)‐like region of Ciona intestinalis voltage‐sensing phosphatase provides insight into substrate specificity and redox regulation of the phosphoinositide phosphatase activity. J Biol Chem 286, 23368–23377. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Mori H & Ito K (2006). Different modes of SecY‐SecA interactions revealed by site‐directed in vivo photo‐cross‐linking. Proc Natl Acad Sci U S A 103, 16159–16164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Murata Y, Iwasaki H, Sasaki M, Inaba K & Okamura Y (2005). Phosphoinositide phosphatase activity coupled to an intrinsic voltage sensor. Nature 435, 1239–1243. [DOI] [PubMed] [Google Scholar]
  26. Nowak MW, Gallivan JP, Silverman SK, Labarca CG, Dougherty DA & Lester HA (1998). In vivo incorporation of unnatural amino acids into ion channels in Xenopus oocyte expression system. Methods Enzymol 293, 504–529. [DOI] [PubMed] [Google Scholar]
  27. Nowak MW, Kearney PC, Sampson JR, Saks ME, Labarca CG, Silverman SK, Zhong W, Thorson J, Abelson JN, Davidson N et al (1995). Nicotinic receptor binding site probed with unnatural amino acid incorporation in intact cells. Science (New York, NY) 268, 439–442. [DOI] [PubMed] [Google Scholar]
  28. Okamura Y & Dixon JE (2011). Voltage‐sensing phosphatase: its molecular relationship with PTEN. Physiology 26, 6–13. [DOI] [PubMed] [Google Scholar]
  29. Okamura Y, Kawanabe A & Kawai T (2018). Voltage‐sensing phosphatases: biophysics, physiology and molecular engineering. Physiol Rev 98, 2097–2131. [DOI] [PubMed] [Google Scholar]
  30. Sakata S, Jinno Y, Kawanabe A & Okamura Y (2016). Voltage‐dependent motion of the catalytic region of voltage‐sensing phosphatase monitored by a fluorescent amino acid. Proc Natl Acad Sci U S A 113, 7521–7526. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Sakata S & Okamura Y (2014). Phosphatase activity of the voltage‐sensing phosphatase, VSP, shows graded dependence on the extent of activation of the voltage sensor. J Physiol 592, 899–914. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Soh MS, Estrada‐Mondragon A, Durisic N, Keramidas A & Lynch JW (2017). Probing the structural mechanism of partial agonism in glycine receptors using the fluorescent artificial amino acid, ANAP. ACS Chem Biol 12, 805–813. [DOI] [PubMed] [Google Scholar]
  33. Suh BC & Hille B (2005). Regulation of ion channels by phosphatidylinositol 4,5‐bisphosphate. Curr Opin Neurobiol 15, 370–378. [DOI] [PubMed] [Google Scholar]
  34. Suh BC & Hille B (2008). PIP2 is a necessary cofactor for ion channel function: how and why? Annu Rev Biophys 37, 175–195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Summerer D, Chen S, Wu N, Deiters A, Chin JW & Schultz PG (2006). A genetically encoded fluorescent amino acid. Proc Natl Acad Sci U S A 103, 9785–9789. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Thornell IM & Bevensee MO (2015). Phosphatidylinositol 4,5‐bisphosphate degradation inhibits the Na+/bicarbonate cotransporter NBCe1‐B and ‐C variants expressed in Xenopus oocytes. J Physiol 593, 541–558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Torrice MM, Bower KS, Lester HA & Dougherty DA (2009). Probing the role of the cation‐pi interaction in the binding sites of GPCRs using unnatural amino acids. Proc Natl Acad Sci U S A 106, 11919–11924. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Tsutsui H, Jinno Y, Tomita A & Okamura Y (2013). Optically detected structural change in the N‐terminal region of the voltage‐sensor domain. Biophys J 105, 108–115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Villalba‐Galea CA, Miceli F, Taglialatela M & Bezanilla F (2009). Coupling between the voltage‐sensing and phosphatase domains of Ci‐VSP. J Gen Physiol 134, 5–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Wang J, Xie J & Schultz PG (2006a). A genetically encoded fluorescent amino acid. J Am Chem Soc 128, 8738–8739. [DOI] [PubMed] [Google Scholar]
  41. Wang L, Xie J & Schultz PG (2006b). Expanding the genetic code. Annu Rev Biophys Biomol Struct 35, 225–249. [DOI] [PubMed] [Google Scholar]
  42. Wang L, Zhang Z, Brock A & Schultz PG (2003). Addition of the keto functional group to the genetic code of Escherichia coli . Proc Natl Acad Sci U S A 100, 56–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Worby CA & Dixon JE (2014). PTEN. Annu Rev Biochem 83, 641–669. [DOI] [PubMed] [Google Scholar]
  44. Yamaguchi S, Kurokawa T, Taira I, Aoki N, Sakata S, Okamura Y & Homma KJ (2014). Potential role of voltage‐sensing phosphatases in regulation of cell structure through the production of PI(3,4)P2 . J Cell Physiol 229, 422–433. [DOI] [PubMed] [Google Scholar]
  45. Zagotta WN, Gordon MT, Senning EN, Munari MA & Gordon SE (2016). Measuring distances between TRPV1 and the plasma membrane using a noncanonical amino acid and transition metal ion FRET. J Gen Physiol 147, 201–216. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES