SCDB2010

G

Y

R

Mo

Ja

b

a

AA

KMPVVB

C

1

socr

i9

1d

ARTICLE IN PRESSModel

SCDB-1105; No. of Pages 7

Seminars in Cell & Developmental Biology xxx (2010) xxx–xxx

Contents lists available at ScienceDirect

Seminars in Cell & Developmental Biology

journa l homepage: www.e lsev ier .com/ locate /semcdb

eview

ass spectrometry-based phosphoproteomics reveals multisite phosphorylationn mammalian brain voltage-gated sodium and potassium channels

e-Hyun Baeka,1, Oscar Cerdaa,1, James S. Trimmera,b,∗

Department of Neurobiology, Physiology and Behavior, University of California, Davis, CA 95616-8519, United StatesDepartment of Physiology and Membrane Biology, University of California, Davis, CA 95616-8519, United States

r t i c l e i n f o

rticle history:

a b s t r a c t

Voltage-gated sodium and potassium channels underlie electrical activity of neurons, and are dynamically
vailable online xxx
eywords:ass spectrometry

hosphoproteomicsGSC (Nav channel)

regulated by diverse cell signaling pathways that ultimately exert their effects by altering the phosphory-lation state of channel subunits. Recent mass spectrometric-based studies have led to a new appreciationof the extent and nature of phosphorylation of these ion channels in mammalian brain. This has allowedfor new insights into how neurons dynamically regulate the localization, activity and expression throughmultisite ion channel phosphorylation.

GKC (Kv channel)rain ion channel

© 2010 Elsevier Ltd. All rights reserved.

ontents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 002. Strategies to identify phosphorylation on brain VGSCs and VGKCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

2.1. Complexity and dynamics of the brain proteome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 002.2. Sample preparation for phosphoproteomic analysis of brain ion channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

2.2.1. Antibody-dependent approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 002.2.2. Antibody-independent approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

3. Proteomic analyses of brain ion channel phosphorylation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.1. Voltage-gated sodium channels (VGSCs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.2. Voltage-gated potassium channels (VGKCs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

4. Useful phosphoproteomic technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 005. What comes next? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Appendix A. Supplementary data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

. Introduction channels exist as supramolecular protein complexes composed ofpore-forming transmembrane principal or � subunits, auxiliary or

Please cite this article in press as: Baek J-H, et al. Mass spectrometry-based pbrain voltage-gated sodium and potassium channels. Semin Cell Dev Biol (

Ion fluxes through voltage-dependent ion channels that areelectively permeable to sodium and potassium ions underlie muchf the electrical signaling in neurons. As such, voltage-gated sodiumhannel (VGSC) and potassium channel (VGKC) activity is highlyegulated in both excitable and non-excitable cells. These ion

∗ Corresponding author at: Department of Neurobiology, Physiology and Behav-or, University of California, 196 Briggs Hall, One Shields Avenue, Davis, CA5616-8519, United States. Tel.: +1 530 754 6075; fax: +1 530 754 6079.

E-mail address: [email protected] (J.S. Trimmer).1 These authors contributed equally to this work.

084-9521/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.oi:10.1016/j.semcdb.2010.09.009

regulatory � subunits [1], and a diverse array of interacting pro-teins [2]. VGSCs mediate the rapid influx of Na+ that underliesaction potential initiation and propagation, dendritic excitability,and many other aspects of neuronal excitability. These channelsconsist of a highly glycosylated 24 transmembrane segment � sub-unit of ≈260 kDa, associated with smaller auxiliary single passtransmembrane Nav� subunits [3]. VGKCs are composed of four

hosphoproteomics reveals multisite phosphorylation on mammalian2010), doi:10.1016/j.semcdb.2010.09.009

pore-forming and voltage-sensing � subunits, each of which hassix transmembrane segments S1–S6, extensive cytoplasmic N- andC-termini, and intracellular linkers between transmembrane seg-ments S2–S3 and S4–S5 [4]. VGKCs typically contain additionalsingle-pass transmembrane or cytoplasmic auxiliary subunits [1].

dx.doi.org/10.1016/j.semcdb.2010.09.009


http://www.sciencedirect.com/science/journal/10849521

http://www.elsevier.com/locate/semcdb

mailto:[email protected]


ARTICLE IN PRESSG Model

YSCDB-1105; No. of Pages 7

2 J.-H. Baek et al. / Seminars in Cell & Developmental Biology xxx (2010) xxx–xxx

F nd VGt rget prs n: FAs

alovpmcnV

t2oprmppippn[va(aaorimfs

2a

2

e

ig. 1. Flowchart for identification/characterization of phosphorylation in VGSCs aext for details). (A) Conventional antibody-dependent method for purification of taample using a phosphospecific antibody (filter aided antibody capturing and elutioamples in the absence of an antibody (FASP: filter aided sample preparation).

Diverse posttranslational events acting on components of VGSCnd VGKC complexes can dynamically regulate the expression,ocalization and function of these ion channels. While numer-us non-covalent mechanisms such as sensing of transmembraneoltage, ligand binding, and interaction with other proteins playrominent roles in regulating VGSCs and VGKCs, direct covalentodification of components of these multiprotein ion channel

omplexes represents a widely used and potent mechanism foreurons to achieve dynamic and reversible changes in VGSC andGKC function, and impact their contribution to neuronal signaling.

Phosphorylation constitutes the most common covalent post-ranslational modification in eukaryotes [5], with (as of early009) up to 25,000 described phosphorylation sites (phosphosites)n 7000 human proteins, out of an estimated 500,000 potentialhosphosites that exist in a cellular proteome [6]. In neurons,eversible activity-dependent phosphorylation represents a majorechanism of dynamic regulation of synaptic development [7],

otentiation, depression and homeostatic plasticity [8], throughhosphorylation of a large number of synaptic proteins includ-

ng ligand-gated ion channels [9]. Neurons also exhibit cellularlasticity at the level of intrinsic excitability, achieved throughhosphorylation of components of VGSC [10] and VGKC [11] chan-el complexes, which localize in distinct neuronal compartments10,12,13]. As opposed to the classical approaches of in vivo or initro radiolabeling with 32P, peptide mapping and/or sequencing,nd site-directed mutagenesis (e.g., [14,15]), mass spectrometryMS)-based phosphoproteomic techniques have recently emergeds the primary tool for the identification phosphorylation on VGSCnd VGKC subunits, and have revealed an unanticipated extentf multisite phosphorylation on both VGSCs and VGKCs. Here weeview the current state of how applying such approaches hasmpacted our view of the extent and nature of phosphorylation of

ammalian neuronal VGSC and VGKC subunits, and perspectivesor future research into these important determinants of neuronalignaling.

. Strategies to identify phosphorylation on brain VGSCsnd VGKCs


.1. Complexity and dynamics of the brain proteome

Recent innovations in proteomics and bioinformatics havexpanded our knowledge of the mammalian brain proteome to

KCs. Antibody-dependent (A, B) and -independent (C) approaches are shown (seeoteins from a complex sample. (B) Enrichment of phosphopeptides from a complex

CE). (C) High-throughput strategy for phosphorylation identification from complex

include ≈10,000 different proteins [16]. It remains a major chal-lenge to overcome the high complexity of the brain proteome,and the dynamic nature of reversible phosphorylation, to definethe VGSC and VGKC phosphoproteome. To understand how thesignaling networks present in neuronal cells impinge on the expres-sion, localization and function of these ion channels, the need forcomprehensive information on VGSC and VGKC phosphosites isparamount. MS-based proteomic approaches represent a power-ful approach to define the VGSC and VGKC phosphoproteome, acritical first step in understanding how dynamic changes in phos-phorylation at specific sites impacts dynamic plasticity in neuronalelectrical activity. Proteomic analyses of a variety of VGSCs andVGKCs, whose polypeptide sequences have been deduced frommolecular cloning and genomic studies, have led to the identifi-cation of a large and rapidly expanding set of identified in vivophosphosites. Here, we review these studies and discuss practicalaspects and implication of phosphosite identification of brain VGSCand VGKC proteins.

2.2. Sample preparation for phosphoproteomic analysis of brainion channels

2.2.1. Antibody-dependent approachesGiven the high complexity of the brain proteome, and the highly

variable expression levels of different VGSCs and VGKCs, enrich-ment for specific channel subunits can greatly benefit attempts at acomprehensive analysis of the extent and nature of their phospho-rylation. Antibody based approaches represent a powerful tool todirectly isolate, as target antigens, specific ion channel � subunits,auxiliary subunits or interacting proteins for subsequent analysisof phosphosites. A suitable antibody can allow for the simple yetspecific immunopurification of the target VGSC or VGKC subunitdirectly from samples with a high degree of proteomic complex-ity (e.g., crude brain preparations) as shown in Fig. 1 (Path A).Following further fractionation by preparative one-dimensionalSDS-PAGE, VGSC or VGKC subunits can then be subjected to MSanalysis in a sample with a limited proteomic complexity. Whenexcised bands of immunopurified proteins are digested in gel slices,


unmodified peptides, as well as less abundant and more difficult todetect phosphopeptides, present in the sample in femtomole levels,can be detected in conventional nano-liquid chromatography–MS(LC–MS) without any enrichment steps [17,18]. Further enrichmentof phosphopeptides by simple methods such as immobilized metal


ING

Y

evelo

at

rpaotiK

ttbricsoswtp

ppptfoitppTi

2

aVwdstpoos

ihaaigyua[risiOf

to define the expression and localization Nav1.1, Nav1.2, Nav1.3and Nav1.6 in mammalian brain [30]. Two distinct monoclonalantibodies, one specific for Nav1.2, and one with pan-VGSC speci-ficity, were used in parallel immunopurification and MS analysesof rat brain VGSC phosphorylation [17]. These studies identified fif-

Table 1Phosphorylation sites identified on VGSC and VGKC � subunits by MS.

� Subunit Total N-terminal Linker C-terminal

Nav1.1/Scn1a 11 0 10 1Nav1.2/Scn2a1 29 1 24 4Nav1.6/Scn8a 5 0 5 0Nav1.9/Scn9a 4 0 4 0

Kv1.1/Kcna1 1 1 0 0Kv1.2/Kcna2 8 0 0 8Kv1.4/Kcna4 3 3 0 0Kv1.5/Kcna5 1 0 0 1Kv1.6/Kcna6 2 2 0 0

Kv2.1/Kcnb1 23 2 0 21Kv2.2/Kcnb2 7 0 0 7

Kv3.1/Kcnc1 5 2 0 3Kv3.2/Kcnc2 4 0 0 4Kv3.3/Kcnc3 5 0 0 5Kv3.4/Kcnc4 1 0 0 1

Kv4.1/Kcnd1 2 0 0 2Kv4.2/Kcnd2 5 1 0 4Kv4.3/Kcnd3 1 1 0 0

Kv7.2/Kcnq2 16 1 1 14Kv7.3/Kcnq3 6 1 1 4Kv7.5/Kcnq5 2 0 0 2

Slo1 BK/Kcnma1 36 2 2 32

Table 2Phosphorylation sites identified on VGSC and VGKC auxiliary subunits by MS.

Auxiliary subunit Total N-terminal Core C-terminal

Nav�2/Scn2b 2 0 0 2Nav�3/Scn3b 1 0 0 1

ARTICLEModel


J.-H. Baek et al. / Seminars in Cell & D

ffinity chromatography (IMAC) can lead to identification of addi-ional phosphosites not detected in unenriched samples [17].

The antibody-dependent approach has been used in a number ofecent studies employing MS-based approaches to identify phos-hosites on brain VGSC [17] and VGKC [18–20] � subunits. Thispproach has also been applied to identification of phosphositesn co-purifying interacting proteins, such as in the MS-based iden-ification of phosphosites on the auxiliary Kv�2 subunit, followingts copurification from brain with antibodies directed against thev1.2 � subunit [21].

An alternative antibody-based approach is to use antibodieshat recognize an entire family of VGSC or VGKC proteins. A par-icularly useful set of such “pan” antibodies, targeting VGSCs, haseen raised by immunizing animals with the cytoplasmic linkeregion between domains III and IV (i.e., the ID III–IV linker) thats absolutely conserved among all vertebrate Nav channels. Poly-lonal [22] and monoclonal [23] antibodies raised against thisegment have proven extremely useful in labeling VGSC � subunitsn immunoblots and by immunocytochemistry. The pan-VGSC �ubunit monoclonal antibody K58/35 was recently used in concertith a Nav1.2-specific monoclonal antibody in immunopurifica-

ion of mammalian brain VGSCs for MS-based studies of in vivohosphorylation sites [17].

An even broader antibody-based approach uses anti-hosphotyrosine (anti-pTyr) antibodies to isolate and identifyTyr-containing proteins, or from tryptic digests, pTyr-containinghosphopeptides. Enrichment can be performed on materialhat has already been enriched by other means (e.g., subcellularractionation, immunopurification with specific antibodies, etc.),n membrane fractions, or on whole brain extracts (Path B in Fig. 1)n large-scale proteomic studies. The latter approach was usedo begin to define the extent and nature of mouse brain tyrosinehosphorylation, and led to identification of 414 unique pTyrhosphosites, including four sites on VGSC and VGKC subunits.ogether, antibody-dependent approaches have yielded valuablensights into the brain VGSC and VGKC phosphoproteome.

.2.2. Antibody-independent approachesHigh throughput, antibody-independent approaches represent

powerful approach to obtain important information on VGSC andGKC phosphorylation in the absence of suitable antibodies, orhen one aims to obtain a more global view of phosphorylation thatoes not focus on individual channel subunits. A number of recenttudies have effectively used antibody-independent approacheshat incorporate various fractionation procedures, as well as phos-hoprotein/phosphopeptide enrichment methods, to reduce theverall complexity of the sample. This has allowed for detectionf relatively low abundant VGSC and VGKC phosphopeptides. Suchtrategies are outlined in Fig. 1 as Path C.

A number of key recent studies have provided important newnformation on the VGSC and VGKC phosphoproteome, withoutaving this as a specific focus per se. A global phosphoproteomicnalysis of mouse brain proteins present in synaptic membranesnd/or synpatosomes yielded 1367 unique in vivo phosphosites,ncluding a number of VGSCs and VGKCs [24]. A more recentlobal analysis of the phosphoproteome of a mouse brain extractielded over 12,000 phosphosites on≈4600 brain phosphoproteins,sing an antibody-independent approach for pSer and pThr sites,nd an antibody-dependent approach for less abundant pTyr sites24]. The complexity of whole brain proteins and peptides waseduced by multiple fractionation steps (Path C in Fig. 1), includ-


ng strong anion exchange (SAX), strong cation exchange (SCX),ize exclusion chromatography (SEC), and immobilized metal affin-ty chromatography (IMAC) and TiO2 phosphopeptide enrichment.ne item of note from these studies is that the overall ratio obtained

or pSer:pThr:pTyr phosphosites in these mouse brain samples

PRESSpmental Biology xxx (2010) xxx–xxx 3

was 85:14:1 [24] and 83:15:2 [25], comparable to those (86:12:2)obtained from previous studies on a human cell line [26].

Together, the recent application of broader antibody-independent high-throughput approaches, in combination with amore focused antibody-dependent approach, has proved advan-tageous in identifying an unexpected wealth of phosphosites onbrain VGSCs and VGKCs.

3. Proteomic analyses of brain ion channel phosphorylation

In this section we review available data on phosphorylationof brain VGSCs and VGKCs, focusing on data obtained in recentantibody-dependent and independent MS-based studies. Note thatthis synopsis is assuredly not comprehensive, as there are likelyother high-throughput studies that have yielded data leadingto identification of phosphosites that we have not included inour analyses, which focuses on data from five recent antibody-independent studies [24,25,27–29], and data obtained from recentantibody-dependent approaches (summarized in Tables 1 and 2,and detailed in the Supplemental Table).

3.1. Voltage-gated sodium channels (VGSCs)

Antibodies recognizing specific VGSC � subunits have been used


Kv�1/Kcnab1 2 0 1 1Kv�2/Kcnab2 8 5 2 1

KChIP2/Kcnip2 3 3 0 0KChIP3/Kcnip3 1 1 0 0


ING

Y

4 evelo

tTau(Cipgapat

sswsrbfilamsodn

oatfiNTiufo

3

piilveuKfi([tcwayThtnpps

ARTICLEModel



een phosphosites on Nav1.2, and three on Nav1.1 (Supplementalable). Note that two sites (S468, S610) were only observed afterdditional IMAC phosphopeptide enrichment following immunop-rification [17]. Phosphorylation on the cytoplasmic ID I–II linker11 sites identified) is known to modulate Nav1.2 gating [10]. The-terminus (three sites identified) is the site of G protein bind-

ng, which also impacts Nav1.2 gating [31], however, whetherhosphorylation affects G protein binding has not been investi-ated. The Nav1.1 phosphosites, while in peptides unique to Nav1.1,re in sequence alignments in positions identical to three of thehosphosites found on the Nav1.2 ID I–II linker [17], suggestingconservation of VGSC modulation through phosphorylation at

hese sites.A recent antibody-independent global phosphoproteomics

tudy on mouse brain [25] identified 45 phosphosites on VGSC �ubunits, and 3 phosphosites on Nav� subunits. Eight phosphositesere identified on the Nav1.1 ID I–II linker, including two of the

ites (S551 and S607) identified in the antibody-dependent study ofat brain Nav1.1 [17], suggesting conservation of phosphorylationetween mouse and rat Nav1.1. A total of twenty sites were identi-ed on Nav1.2, including nine of the twelve sites within the ID I–II

inker identified in the antibody-dependent study of rat brain [17],gain suggesting strong conservation of phosphorylation betweenouse and rat Nav1.2 orthologs. Note that previous biochemical

tudies (reviewed in [10]) had also demonstrated phosphorylationn the Nav1.2 ID I–II linker at S579. Together, these antibody-ependent and -independent studies together yielded a wealth ofew phosphosites on Nav1.1 (11 sites) and Nav1.2 (29 sites).

MS-based phosphorylation site analysis has not been performedn other VGSC � subunits purified from brain. However, recentntibody-independent analysis of the mouse brain phosphopro-eome employing FASP sample preparation [25] yielded a total ofve Nav1.6 phosphosites, all in the ID I–II linker, and four sites onav1.9, two in the ID I–II linker, and two in the ID II–III linker.his study also yielded the first data on phosphorylation of aux-liary Nav� subunits. These data represent a huge advance in ournderstanding of molecular determinants of VGSC modulation, anduture studies are likely to yield important information on the rolef phosphorylation at these sites in regulating neuronal excitability.

.2. Voltage-gated potassium channels (VGKCs)

Phosphoproteomic analysis of VGKCs has also been accom-lished using both antibody-dependent [18–20,32] and antibody-

ndependent [24,25,27,29] approaches. Kv1 channels are predom-nantly localized in axons, with certain subunits present at highevels in presynaptic terminals, the juxtaparanodes of nodes of Ran-ier in myelinated axons, and axon initial segments [30]. With thexception of Kv1.2, antibody-dependent approaches have not beensed to address the extent and nature of phosphorylation of brainv1 channel subunits. MS analysis of Kv1.2 channels immunopuri-ed from rat, mouse and human brain identified four phosphositesS434, S440, S441, and S449) in heterologous expression systems20]. Functional characterization of these residues demonstratedhat phosphorylation at S440/S441 is specifically associated withell surface Kv1.2, while phosphorylation at S449 is associatedith newly synthesized, intracellular Kv1.2 [20]. These sites were

lso identified in antibody-independent phosphoproteome anal-ses from mouse brain, which identified additional phosphosites421, Y429, T433, and S447 [24,25,27,33]. Together, these studiesave provided a total of eight phosphosites on brain Kv1.2, all on


he cytoplasmic C-terminal tail. The role of many of these sites isot known, although mutation of the S440, S441 and S449 phos-hosites negatively impacts biosynthetic trafficking of Kv1.2 to thelasma membrane [20]. The location of these C-terminal phospho-ites in the Kv1.2 structure is also not known, as this domain is

PRESSpmental Biology xxx (2010) xxx–xxx

lacking in the fragment of Kv1.2 used in crystallographic studies[34].

While antibody-dependent approaches have not been usedin MS-based studies to characterize phosphorylation on otherKv1 family � subunits, recent antibody-independent phosphopro-teomics analyses [24,25,27,29] have provided some initial dataon their phosphorylation status. These studies have identified N-terminal phosphosites on Kv1.1 (1 site), Kv1.4 (3 sites), and Kv1.6(2 sites), and C-terminal phosphorylation of Kv1.5 (1 site). It isintriguing that unlike Kv1.2, so many other Kv1 family membersexhibit phosphorylation on the N-terminus. The phosphosites iden-tified on Kv1.4 are all located on the so-called “chain” segmentinvolved in the “ball and chain” mechanism of N-type inactivation,and one of these sites (S122) is required for the CamKII-dependentmodulation of Kv1.4 inactivation kinetics [35] that mediates firingfrequency-dependent inactivation of Kv1.4-containing channelsin presynaptic terminals [36]. The N-terminal phosphorylation ofKv1.6 is also intriguing as it occurs within the so-called NIP domainthat prevents N-type inactivation in Kv1.6-containing Kv1 channels[37], suggesting that, as in Kv1.4-containing channels, N-terminalphosphorylation of Kv1.6 may regulate channel inactivation. TheKv1 family associated auxiliary Kv�1 and Kv�2 subunits have alsobeen found phosphorylated in vivo in mouse brain in a number ofhigh throughput phosphoproteomic studies [24,25,27,29,33,38].

The delayed rectifier Kv channel Kv2.1 plays a highly condi-tional role in determining the firing of neuronal action potentials[39,40]. Kv2.1 channels are found in large plasma membrane clus-ters in somatodendritic regions of neurons, with a minor populationon axon initial segments [30]. Early studies suggested that Kv2.1is present in a highly phosphorylated state in mammalian brainand that phosphorylation determines the voltage-dependence ofchannel activity [41]. More recent studies revealed that changesin neuronal activity during in vivo epileptic and hypoxic episodes,and upon glutamate stimulation of cultured neurons, trigger Kv2.1dephosphorylation [40,42–44]. Dephosphorylation is mediated bythe phosphatase calcineurin, and leads to dispersion of Kv2.1clusters, and hyperpolarizing shifts in voltage-dependent acti-vation and inactivation of Kv2.1 channels [40,42–44]. Similarphosphorylation-dependent modulation is seen for recombinantKv2.1 expressed in heterologous cells [41,45,46].

An antibody-dependent approach was used to purify recombi-nant Kv2.1 protein from heterologous cells, and endogenous Kv2.1protein from rat brain [19,47]. Subsequent MS analysis led to theinitial identification of 21 Ser and two Thr residues as in vivophosphosites [19]. All phosphosites except two were located onthe cytoplasmic C-terminus. Stable Isotope Labeling with Aminoacids in cell Culture (SILAC) allowed for quantitative MS analysisof the subset of sites (9/16 sites identified in the original analy-sis) targeted during calcineurin-mediated dephosphorylation [19].Subsequent high throughput analyses on mouse brain [24,25,27,29]have yielded seven additional Kv2.1 phosphosites (one N-terminaland six C-terminal), as well as providing corroborating evidencefor five of the sixteen sites identified in the antibody-dependentanalysis of rat brain Kv2.1. The recent antibody independent analy-sis of mouse brain phosphoproteome revealed seven phosphositeson the Kv2.2 C-terminus; the functional significance of these newphosphosites has not yet been determined. Two Kv2.2 phosphosites(S488 and S507) are present in sequence alignments in positionsidentical to two of the phosphosites (S484 and S503) that regulateKv2.1 gating.

KCNQ/Kv7 � subunits form the VGKCs corresponding to the


M-type current. These channels localize in axon initial segmentsand nodes of Ranvier throughout the nervous system [30], modu-late neuronal excitability, bursting, and neurotransmitter release,and are subject to Gq/11 modulation [48]. One prominent form ofthese channels that is also highly regulated by phosphorylation


ING

Y

evelo

iaeTapc[soP[rKKa

taoVmemufnopwtpcustsbssmfcrtwttipsat

4

ispnctt

ARTICLEModel



s formed by Kv7.2/Kv7.3 heteromers [49–51]. Phosphoproteomicnalysis of recombinant human Kv7.2 and Kv7.3 expressed in het-rologous cells revealed phosphorylation at T217 (in Kv7.2) and246 (in Kv7.3), both localized within the cytoplasmic S4-S5 linker,nd S579 in the C-terminus of Kv7.3 [32]. High throughput phos-hoproteomic studies have identified phosphorylation at a siteorresponding to human T579 on the mouse ortholog of Kv7.325,29]. These studies and others [24,27] identified numerous otherites, primarily on the Kv7.2 and Kv7.3 C-termini. Interestingly, onef these sites (i.e. S52) had been previously identified as crucial toKA-mediated activation of Kv7.2 channels in heterologous cells52]. A number of Kv7.2 phosphosites cluster near the C-terminalegion proposed to mediate PIP2 binding [53], which is required forv7 channel function [48]. How phosphorylation at these sites onv7.2, and the others identified on Kv7 subunits in mouse brain,ffect PIP2 binding and other aspects of Kv7 function, is not known.

BK channels are composed of tetramers of a large � subunithat is encoded by a single gene, termed Slo1, whose transcriptsre subjected to extensive alternative splicing to generate numer-us BK channel isoforms [54]. BK channels are unique amongGKCs in that under physiological conditions they require bothembrane depolarization and increased [Ca2+]i to activate. The

xtended C-terminal region, a substrate for phosphorylation, is aajor determinant of BK channel function, as both a direct reg-

lator of gating via calcium binding, and as an organizing centeror numerous protein–protein interactions that influence BK chan-el function [55,56]. A recent study employing immunopurificationf rat brain BK channels, followed by analysis of in vivo phos-hosites using MS analysis, identified 30 Ser/Thr phosphosites, ofhich 4 were ambiguous [17]. Mutation of phosphosite S891 in

he 1209 a.a. rat BK sequence Swiss Prot Q62976-2 causes a hyper-olarizing shift in the voltage-dependent activation curve of thehannel, suggesting a role for phosphorylation at this site in mod-lating BK channel gating. Interestingly, phosphorylation at thisame site (the S924 residue in mouse Q08460-1 and the S885 posi-ion in mouse Q08460-2) was also identified in analyses of theynaptosomal [38] and global [25] phosphoproteomes of mouserain. Importantly, different BK channel phosphosites can maketoichiometrically distinct contributions to function, with certainites exhibiting dominant and other recessive effects [57]. As suchultisite BK channel phosphorylation can have diverse effects on

unction. Moreover, the effects of phosphorylation on BK channelsan be impacted by changes in BK channel sequences outside of theegion containing the phosphosites themselves, through alterna-ive splicing of BK channel mRNAs, to generate alternative isoformsith different sensitivities to phosphorylation-dependent modula-

ion [58]. Recent high throughput proteomic analyses have allowedhe identification of a small number of additional phosphositesn BK channels [27,29,38]. However, changes in the in vivo phos-horylation state of BK channels at specific sites during neuronalignaling events that regulated BK channel expression, localizationnd function, have not yet been observed and remain an importantopic for future experiments.

. Useful phosphoproteomic technologies

The sections above detail recent studies that, through theirncreased coverage of VGSC and VGKC sequences, and enhancedensitivity, have allowed for identification of a wealth of novel


hosphosites. Advances in LC/MS-based phosphoproteomic tech-ologies, and in sample preparation, phosphopeptide enrichment,hemical tagging, etc. have and will continue to impact studies ofhe VGSC and VGKC phosphoproteome. Here, we discuss aspects ofhese methods and their impact on phosphoproteome analysis.


Filter-aided sample preparation (FASP), combines the advan-tages of in-gel and in-solution protein digestion, and was developedspecifically for MS-based proteomics [59]. Analyses of the mousebrain phosphoproteome employing FASP led to the remarkableidentification of ≈5000 phosphosites on ≈2000 proteins in a singleLC–MS experiment on 1 mg of total mouse brain protein, repre-senting an unprecedented depth of phosphoproteome coverage[25]. Much of the data reviewed above came from this singlestudy in which the FASP technique was applied to analyses ofthe mouse brain phosphoproteome. Based on the success of thisstudy, FASP-based analyses will likely form the core of many futurephosphoproteome analyses, especially those targeting membraneproteins such as VGSCs and VGKCs.

Based on the strong chelating properties of these metals withnegatively charged phosphopeptides, immobilizing Fe3+ (in immo-bilized metal affinity chromatography or IMAC), or TiO2 on beadsallows for effective enrichment of phosphopeptides from highcomplexity mixtures of digested peptides, and these methodshave achieved wide use for phospho-protein/peptide enrichment[60]. Hydrophilic interaction chromatography (HILIC) can also bevery advantageous for phosphopeptide enrichment, due to thehydrophilic nature of phosphopeptides. When HILIC is used priorto IMAC, the selectivity of the metal affinity resin for phospho-peptides versus other acidic peptides is improved to greater than95% [61,62]. High-resolution MS instruments, such as Orbitrap andLTQ-FT mass spectrometers, can help reduce false positives in phos-phopeptide identification, due to the low part-per-million (ppm)precursor ion mass accuracy. As a tagging method for quantita-tive analysis, iTRAQ and mTRAQ (Isobaric and multiple tags forrelative and absolute quantification) are useful tools [63] for quan-tifying changes in phosphorylation at specific VGSC and VGKCphosphosites, and how this relates to their dynamic regulation inmammalian brain.

5. What comes next?

Generation of antibodies against many of the principal and aux-iliary subunits of VGSCs and VGKCs has allowed not only for theirdetection on immunoblots and brain sections in analyses of theirexpression and localization, but has also greatly facilitated theirpurification. This provides a major opportunity for obtaining VGSCsand VGKCs in sufficient quantities and purity for comprehensiveMS-based phosphoproteomic studies. However, validated antibod-ies suitable for purification are not available for every VGSC andVGKC subunit expressed in brain. Moreover, each purification ofthe dozens of VGSC and VGKC subunits represents a large under-taking. Improvements in high-throughput proteomic methods arelikely to allow for increased insights into the extent and naturethe brain VGSC and VGKC phosphoproteome; whether they areable to replace analyses based on target protein purification andin themselves allow for a comprehensive data set remains to beseen.

Since changes in phosphorylation status are dynamic and highlyregulated processes, the determination of different mechanisms ofVGSC and VGKC phosphorylation related to neurodegenerative dis-eases (e.g. Alzheimer’s Disease, Parkinson’s Disease and MultipleSclerosis) and psychiatric/psychological disorders (e.g. addiction,depression, autism and schizophrenia) remain a major focus ofongoing research. Generation of phosphospecific antibodies againstspecific brain VGSC and VGKC phosphosites may provide effective


tools for understanding the role of phosphorylation in the patho-physiology of these diseases, and in some cases, as diagnostic toolsfor the diseases themselves. In this context, improvements in gen-eration of specific antibodies useful for VGSC and VGKC purification,and in general membrane protein purification methods as they


ING

Y

6 evelo

atpqomfrhbirVps

Mrnitbas

A

mwN

A

t

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

ARTICLEModel



re applied to high-throughput MS approaches are fundamentalo increasing our knowledge of the VGSC and VGKC phospho-roteome. Further proteomic method development may allow foruantitative comparisons of VGSCs and VGKCs in small amountsf material from normal and diseased brain, and allow the deter-ination of specific changes associated with disease. The need

or normal and diseased brain tissue and detailed clinical histo-ies underscores the importance of high quality tissue banks, andighlights the importance of a close working relationship betweenasic and clinical researchers. Another major focus is determin-

ng the role of phosphorylation at these newly identified sites inegulating the expression, localization and function of neuronalGSCs and VGKCs, by careful analyses of channel isoforms withhosphorylation-eliminating or -mimicking mutations at specificites.

The power of sensitive antibody-dependent and independentS-based analyses of the VGSC and VGKC phosphoproteome rep-

esents a new world of possibilities for integrative studies ineuroscience. It is important to remember that the biological signif-

cance of any identified phosphosites still relies on careful analyseshrough classical methodologies such as immunohistochemistry,iochemistry, site-directed mutagenesis, and electrophysiology,nd the generation of animal models expressing VGSC and VGKCubunits with altered phosphorylation state.

cknowledgments

We thank Ms. Ashleigh Evans for critical comments on theanuscript, and Dr. Ry Tweedie-Cullen for providing data sets. Theork in our laboratory described here was supported by NIH grantsS38343, NS42225, NS64428 and NS64957.

ppendix A. Supplementary data

Supplementary data associated with this article can be found, inhe online version, at doi:10.1016/j.semcdb.2010.09.009.

eferences

[1] Pongs O, Schwarz JR. Ancillary subunits associated with voltage-dependent K+channels. Physiol Rev 2010;90:755–96.

[2] Dai S, Hall DD, Hell JW. Supramolecular assemblies and localized regulation ofvoltage-gated ion channels. Physiol Rev 2009;89:411–52.

[3] Catterall WA, Goldin AL, Waxman SG. International Union of Pharmacology.XLVII. Nomenclature and structure-function relationships of voltage-gatedsodium channels. Pharmacol Rev 2005;57:397–409.

[4] Swartz KJ. Towards a structural view of gating in potassium channels. Nat RevNeurosci 2004;5:905–16.

[5] Cohen P. The role of protein phosphorylation in human health and disease. TheSir Hans Krebs Medal Lecture. Eur J Biochem 2001;268:5001–10.

[6] Lemeer S, Heck AJ. The phosphoproteomics data explosion. Curr Opin ChemBiol 2009;13:414–20.

[7] Saneyoshi T, Fortin DA, Soderling TR. Regulation of spine and synapse formationby activity-dependent intracellular signaling pathways. Curr Opin Neurobiol2010;20:108–15.

[8] Turrigiano GG. The self-tuning neuron: synaptic scaling of excitatory synapses.Cell 2008;135:422–35.

[9] Collins MO, Grant SG. Supramolecular signalling complexes in the nervoussystem. Subcell Biochem 2007;43:185–207.

10] Cantrell AR, Catterall WA. Neuromodulation of Na+ channels: an unexpectedform of cellular plasticity. Nat Rev Neurosci 2001;2:397–407.

11] Schulz DJ, Temporal S, Barry DM, Garcia ML. Mechanisms of voltage-gatedion channel regulation: from gene expression to localization. Cell Mol Life Sci2008;65:2215–31.

12] Shah MM, Hammond RS, Hoffman DA. Dendritic ion channel trafficking andplasticity. Trends Neurosci 2010;33:307–16.

13] Cerda O, Trimmer JS. Analysis and functional implications of phosphorylation


of neuronal voltage-gated potassium channels. Neurosci Lett 2010;486:60–7.14] Costa MR, Casnellie JE, Catterall WA. Selective phosphorylation of the alpha

subunit of the sodium channel by cAMP-dependent protein kinase. J Biol Chem1982;257:7918–21.

15] Costa MR, Catterall WA. Phosphorylation of the alpha subunit of the sodiumchannel by protein kinase C. Cell Mol Neurobiol 1984;4:291–7.

[

PRESSpmental Biology xxx (2010) xxx–xxx

16] Wang H, Qian WJ, Chin MH, Petyuk VA, Barry RC, Liu T, et al. Characterizationof the mouse brain proteome using global proteomic analysis complementedwith cysteinyl-peptide enrichment. J Proteome Res 2006;5:361–9.

17] Berendt FJ, Park KS, Trimmer JS. Multisite phosphorylation of voltage-gatedsodium channel alpha subunits from rat brain. J Proteome Res 2010;9:1976–84.

18] Yan J, Olsen JV, Park KS, Li W, Bildl W, Schulte U, et al. Profiling the phospho-status of the BKCa channel alpha subunit in rat brain reveals unexpectedpatterns and complexity. Mol Cell Proteomics 2008;7:2188–98.

19] Park KS, Mohapatra DP, Misonou H, Trimmer JS. Graded regulation of the Kv2.1potassium channel by variable phosphorylation. Science 2006;313:976–9.

20] Yang JW, Vacher H, Park KS, Clark E, Trimmer JS. Trafficking-dependent phos-phorylation of Kv1.2 regulates voltage-gated potassium channel cell surfaceexpression. Proc Natl Acad Sci USA 2007;104:20055–60.

21] Vacher H, Clark E, Park KS, Yang JW, Trimmer JS. Regulation of Kv1 channelaxonal targeting by phosphorylation of Kvbeta2. Neuroscience Meeting PlannerSan Diego, CA: Society for Neuroscience, online 2007;44.13:I5.

22] Dugandzija-Novakovic S, Koszowski AG, Levinson SR, Shrager P. Clustering ofNa+ channels and node of Ranvier formation in remyelinating axons. J Neurosci1995;15:492–503.

23] Rasband MN, Peles E, Trimmer JS, Levinson SR, Lux SE, Shrager P. Dependenceof nodal sodium channel clustering on paranodal axoglial contact in the devel-oping CNS. J Neurosci 1999;19:7516–28.

24] Tweedie-Cullen RY, Reck JM, Mansuy IM. Comprehensive mapping of post-translational modifications on synaptic, nuclear, and histone proteins in theadult mouse brain. J Proteome Res 2009;8:4966–82.

25] Wisniewski JR, Nagaraj N, Zougman A, Gnad F, Mann M. Brain phosphopro-teome obtained by a FASP-based method reveals plasma membrane proteintopology. J Proteome Res 2010;9:3280–9.

26] Olsen JV, Blagoev B, Gnad F, Macek B, Kumar C, Mortensen P, et al. Global, invivo, and site-specific phosphorylation dynamics in signaling networks. Cell2006;127:635–48.

27] Munton RP, Tweedie-Cullen R, Livingstone-Zatchej M, Weinandy F, WaidelichM, Longo D, et al. Qualitative and quantitative analyses of protein phospho-rylation in naive and stimulated mouse synaptosomal preparations. Mol CellProteomics 2007;6:283–93.

28] Trinidad JC, Specht CG, Thalhammer A, Schoepfer R, Burlingame AL. Com-prehensive identification of phosphorylation sites in postsynaptic densitypreparations. Mol Cell Proteomics 2006;5:914–22.

29] Trinidad JC, Thalhammer A, Specht CG, Lynn AJ, Baker PR, Schoepfer R, et al.Quantitative analysis of synaptic phosphorylation and protein expression. MolCell Proteomics 2008;7:684–96.

30] Vacher H, Mohapatra DP, Trimmer JS. Localization and targeting of voltage-dependent ion channels in mammalian central neurons. Physiol Rev2008;88:1407–47.

31] Mantegazza M, Yu FH, Powell AJ, Clare JJ, Catterall WA, Scheuer T. Molecu-lar determinants for modulation of persistent sodium current by G-proteinbetagamma subunits. J Neurosci 2005;25:3341–9.

32] Surti TS, Huang L, Jan YN, Jan LY, Cooper EC. Identification by mass spectrometryand functional characterization of two phosphorylation sites of KCNQ2/KCNQ3channels. Proc Natl Acad Sci USA 2005;102:17828–33.

33] Ballif BA, Carey GR, Sunyaev SR, Gygi SP. Large-scale identification and evolu-tion indexing of tyrosine phosphorylation sites from murine brain. J ProteomeRes 2008;7:311–8.

34] Long SB, Campbell EB, Mackinnon R. Crystal structure of a mammalian voltage-dependent Shaker family K+ channel. Science 2005;309:897–903.

35] Roeper J, Lorra C, Pongs O. Frequency-dependent inactivation of mammalianA-type K+ channel KV1.4 regulated by Ca2+/calmodulin-dependent proteinkinase. J Neurosci 1997;17:3379–91.

36] Geiger JR, Jonas P. Dynamic control of presynaptic Ca(2+) inflow by fast-inactivating K(+) channels in hippocampal mossy fiber boutons. Neuron2000;28:927–39.

37] Roeper J, Sewing S, Zhang Y, Sommer T, Wanner SG, Pongs O. NIP domainprevents N-type inactivation in voltage-gated potassium channels. Nature1998;391:390–3.

38] Collins MO, Yu L, Coba MP, Husi H, Campuzano I, Blackstock WP, et al. Pro-teomic analysis of in vivo phosphorylated synaptic proteins. J Biol Chem2005;280:5972–82.

39] Du J, Haak LL, Phillips-Tansey E, Russell JT, McBain CJ. Frequency-dependentregulation of rat hippocampal somato-dendritic excitability by the K+ channelsubunit Kv2.1. J Physiol 2000;522(Pt 1):19–31.

40] Mohapatra DP, Misonou H, Pan SJ, Held JE, Surmeier DJ, Trimmer JS. Regu-lation of intrinsic excitability in hippocampal neurons by activity-dependentmodulation of the KV2.1 potassium channel. Channels (Austin) 2009;3:46–56.

41] Murakoshi H, Shi G, Scannevin RH, Trimmer JS. Phosphorylation of the Kv2.1 K+channel alters voltage-dependent activation. Mol Pharmacol 1997;52:821–8.

42] Misonou H, Menegola M, Mohapatra DP, Guy LK, Park KS, Trimmer JS. Bidirec-tional activity-dependent regulation of neuronal ion channel phosphorylation.J Neurosci 2006;26:13505–14.

43] Misonou H, Mohapatra DP, Menegola M, Trimmer JS. Calcium- and metabolic


state-dependent modulation of the voltage-dependent Kv2.1 channel reg-ulates neuronal excitability in response to ischemia. J Neurosci 2005;25:11184–93.

44] Misonou H, Mohapatra DP, Park EW, Leung V, Zhen D, Misonou K, et al. Regula-tion of ion channel localization and phosphorylation by neuronal activity. NatNeurosci 2004;7:711–8.


http://dx.doi.org/10.1016/j.semcdb.2010.09.009

ING

Y

evelo

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[61] McNulty DE, Annan RS. Hydrophilic interaction chromatography reduces the

ARTICLEModel



45] Mohapatra DP, Trimmer JS. The Kv2.1C terminus can autonomously transferKv2.1-like phosphorylation-dependent localization, voltage-dependent gat-ing, and muscarinic modulation to diverse Kv channels. J Neurosci 2006;26:685–95.

46] Mohapatra DP, Siino DF, Trimmer JS. Interdomain cytoplasmic interactions gov-ern the intracellular trafficking, gating, and modulation of the kv2.1 channel. JNeurosci 2008;28:4982–94.

47] Park KS, Mohapatra DP, Trimmer JS. Proteomic analyses of Kv2.1 channel phos-phorylation sites determining cell background specific differences in function.Channels (Austin) 2007;1:59–61.

48] Hernandez CC, Zaika O, Tolstykh GP, Shapiro MS. Regulation of neural KCNQchannels: signalling pathways, structural motifs and functional implications. JPhysiol 2008;586:1811–21.

49] Hoshi N, Zhang JS, Omaki M, Takeuchi T, Yokoyama S, Wanaverbecq N, et al.AKAP150 signaling complex promotes suppression of the M-current by mus-carinic agonists. Nat Neurosci 2003;6:564–71.

50] Nakajo K, Kubo Y. Protein kinase C shifts the voltage dependence of KCNQ/Mchannels expressed in Xenopus oocytes. J Physiol 2005;569:59–74.

51] Gamper N, Stockand JD, Shapiro MS. Subunit-specific modulation of KCNQpotassium channels by Src tyrosine kinase. J Neurosci 2003;23:84–95.


52] Schroeder BC, Kubisch C, Stein V, Jentsch TJ. Moderate loss of function ofcyclic-AMP-modulated KCNQ2/KCNQ3 K+ channels causes epilepsy. Nature1998;396:687–90.

53] Hernandez CC, Zaika O, Shapiro MS. A carboxy-terminal inter-helix linker asthe site of phosphatidylinositol 4,5-bisphosphate action on Kv7 (M-type) K+channels. J Gen Physiol 2008;132:361–81.

[[


54] Fury M, Marx SO, Marks AR. Molecular BKology: the study of splicing and dicing.Sci STKE 2002;2002:PE12.

55] Lu R, Alioua A, Kumar Y, Eghbali M, Stefani E, Toro L. MaxiK channel partners:physiological impact. J Physiol 2006;570:65–72.

56] Levitan IB. Signaling protein complexes associated with neuronal ion channels.Nat Neurosci 2006;9:305–10.

57] Tian L, Coghill LS, McClafferty H, MacDonald SH, Antoni FA, Ruth P, et al. Distinctstoichiometry of BKCa channel tetramer phosphorylation specifies channelactivation and inhibition by cAMP-dependent protein kinase. Proc Natl AcadSci USA 2004;101:11897–902.

58] Chen L, Tian L, MacDonald SH, McClafferty H, Hammond MS, Huibant JM, et al.Functionally diverse complement of large conductance calcium- and voltage-activated potassium channel (BK) alpha-subunits generated from a single siteof splicing. J Biol Chem 2005;280:33599–609.

59] Wisniewski JR, Zougman A, Nagaraj N, Mann M. Universal sample preparationmethod for proteome analysis. Nat Methods 2009;6:359–62.

60] Zhao Y, Jensen ON. Modification-specific proteomics: strategies for charac-terization of post-translational modifications using enrichment techniques.Proteomics 2009;9:4632–41.


complexity of the phosphoproteome and improves global phosphopeptide iso-lation and detection. Mol Cell Proteomics 2008;7:971–80.

62] Tolstikov VV. Metabolic analysis. Methods Mol Biol 2009;544:343–53.63] White FM. Quantitative phosphoproteomic analysis of signaling network

dynamics. Curr Opin Biotechnol 2008;19:404–9.


SCDB2010

Documents

Transcript of SCDB2010