sign in sign up
<<
Home , Ion channel

Research Article 2155 Targeting of voltage-gated channel isoforms to distinct surface microdomains

Kristen M. S. O’Connell1 and Michael M. Tamkun1,2,* 1Department of Biomedical Sciences and 2Department of and Molecular , Colorado State University, Ft Collins, CO 80523, USA *Author for correspondence (e-mail: [email protected])

Accepted 23 February 2005 Journal of Cell Science 118, 2155-2166 Published by The Company of Biologists 2005 doi:10.1242/jcs.02348

Summary Voltage-gated potassium (Kv) channels regulate action diffused throughout the cell surface. Additionally, PA- potential duration in nerve and muscle; therefore changes GFP-Kv2.1 moved into regions of the not in the number and location of surface channels can adjacent to the original photoactivation ROI. Sucrose profoundly influence electrical excitability. To investigate density gradient analysis indicated that half of Kv2.1 is trafficking of Kv2.1, 1.4 and 1.3 within the plasma part of a large, macromolecular complex while Kv1.4 membrane, we combined the expression of fluorescent sediments as predicted for the tetrameric channel complex. -tagged Kv channels with live cell confocal imaging. Disruption of membrane by cyclodextrin Kv2.1 exhibited a clustered distribution in HEK cells minimally altered Kv2.1 mobility (Mf=0.32±0.03), but similar to that seen in hippocampal neurons, whereas significantly increased surface cluster size by at least Kv1.4 and Kv1.3 were evenly distributed over the plasma fourfold. By comparison, the mobility of Kv1.4 decreased membrane. Using FRAP, surface Kv2.1 displayed limited following cholesterol depletion with no change in surface mobility; approximately 40% of the fluorescence recovered distribution. The mobility of Kv1.3 was slightly increased within 20 minutes of photobleach (Mf=0.41±0.04). following cyclodextrin treatment. These results indicate Recovery occurred not by diffusion from adjacent that (1) Kv2.1, Kv1.4 and Kv1.3 exist in distinct membrane but probably by transport of nascent channel compartments that exhibit different trafficking properties, from within the cell. By contrast, the Kv1 family members (2) membrane cholesterol levels differentially modulate the Kv1.4 and Kv1.3 were highly mobile, both showing trafficking and localization of Kv channels and (3) Kv2.1 approximately 80% recovery (Kv 1.4 Mf=0.78±0.07; Kv1.3 expressed in HEK cells exhibits a surface distribution Mf=0.78±0.04; without correction for photobleach); unlike similar to that seen in native cells. Journal of Cell Science Kv2.1, recovery was consistent with diffusion of channel from membrane adjacent to the bleach region. Studies using PA-GFP-tagged channels were consistent with the Supplementary material available online at FRAP results. Following photoactivation of a small region http://jcs.biologists.org/cgi/content/full/118/10/2155/DC1 of plasma membrane PA-GFP-Kv2.1 remained restricted to the photoactivation ROI, while PA-GFP-Kv1.4 rapidly Key words: Kv channels, Membrane trafficking, rafts

Introduction interact with PDZ such as PSD-95 via binding channel regulation is an important determinant of electrical domains within the C-terminus (Kim et al., 1995). However, excitability in the nervous and cardiovascular systems, skeletal such motifs are not present in all channels. For example, the muscle, gastrointestinal tract and uterus. Voltage-gated C-terminus of Kv2.1 plays a role in the clustering of Kv2.1 in potassium (Kv) channels play a key role in setting the resting hippocampal neurons (Lim et al., 2000), despite the absence of and shaping any known interaction motifs, such as PDZ binding domains in these functionally diverse systems. Kv channels comprise (Scannevin et al., 1996). A role for the N- and C-termini in the largest and most diverse class of voltage-gated ion channels channel trafficking has been shown for the differential and most cells express multiple channel types belonging to one trafficking of inward rectifier K (KIR) channel isoforms or more subfamilies. Structurally, Kv channel isoforms share a (Bendahhou et al., 2003; Ma et al., 2002). conserved core of six transmembrane domains, which includes Kv channels often show a highly specific subcellular the voltage sensor and the ion conducting pore. However, the localization; for example, the A-type channel Kv1.4 is found intracellular N- and C-termini are divergent, even in Kv only along axonal membranes (Sheng et al., 1992), whereas the channel isoforms with nearly identical functional properties, delayed rectifier Kv2.1 is found only on the soma and proximal giving rise to speculation that these domains regulate dendrites (Lim et al., 2000). Such compartmentalization of trafficking and localization in a manner unique to each channel. functionally distinct channels is significant, as it permits spatial Consistent with these differences, some Kv1 family channels regulation of electrical excitability as well as localizing signal bind α-actinin at the N-terminus (Maruoka et al., 2000) and transduction molecules near their ion channel substrates. We 2156 Journal of Cell Science 118 (10)

previously reported that Kv channels differentially target to Live cell confocal imaging distinct lipid raft microdomains within the plasma membrane HEK cells expressing fluorescently tagged Kv channels were imaged (Martens et al., 2000; Martens et al., 2001), indicating that using a Zeiss LSM 510 Meta or Olympus FV1000 confocal protein-lipid interactions must also be considered as a potential microscope. For FRAP experiments on the Zeiss LSM510 Meta, YFP mechanism of Kv channel localization. Furthermore, many was excited with the 514 nm line of an Ar laser and emission collected pathways, including those known to using a Long Pass 530 emission filter. A 63 1.4 NA oil immersion modulate Kv channels, are often found in lipid rafts, including objective was used for imaging with the pinhole diameter set for 1 kinases, nitric oxide synthase, GPI-linked proteins and others Airy Unit. YFP was photobleached by 40-50 bleach iterations over the bleach region of interest (ROI) with the 514 nm laser set to 100% (Martens et al., 2004; O’Connell et al., 2004). transmission. Pre- and postbleach images were acquired at 1% laser From a clinical standpoint, resulting in power every 15 seconds for a total of 80 scans. The average mistrafficked ion channels are responsible for human disease. fluorescence intensity within the bleach ROI was normalized to the For example, the most common in cystic fibrosis, prebleach intensity for each time point and recovery kinetics ∆F508 in the cystic fibrosis transmembrane conductance determined by exponential fitting of the average data: regulator (CFTR), results in a mistrafficked channel (Powell n − τ = − ( t/ i) and Zeitlin, 2002). Mutations in the human ether-a-go-go f(x) ∑ Ai(1 e ) , (1) related (HERG) K+ channel result in trafficking defects that are i=1 a common disease mechanism in the LQT2 form of Long QT where Ai is the amplitude of each component, t is time and τi is the syndrome (Delisle et al., 2004). Despite this clinical relevance, time constant of each component. The mobile fraction was calculated the mechanisms by which ion channels are trafficked remain as: poorly understood. Few studies have examined channel (Fϱ−F ) trafficking in live cells (see Burke et al., 1999), with most using M = 0 , (2) f − fixed preparations or purely biochemical approaches. Thus, it (Fi F0) is essential that we expand our basic understanding of Kv where F is the fluorescence intensity at the end of the FRAP channel localization in live cells. To this end, we examined the experiment, Fi is the initial fluorescence intensity prior to bleach and trafficking of three Kv channel isoforms. Kv2.1, Kv1.4 and F0 is the intensity immediately postbleach. Cumulative photobleach Kv1.3 exist in distinct membrane compartments that traffic via outside of the FRAP ROI was typically 10-15% of the prebleach distinct mechanisms in HEK 293 cells and are differentially intensity and did not exceed 15% for the experiments shown here. sensitive to the perturbation of raft lipid by depletion of This bleach was not corrected for in the data presented. We chose to membrane cholesterol. In addition, the behavior of Kv2.1 in analyze recovery in this manner since the Kv channels being these cells establishes HEK293 cells as a meaningful model examined here clearly localize to different compartments with system in which to explore the mechanisms underlying different mobilities and subcellular distribution. Calculating mobile trafficking and cell surface localization of this delayed rectifier fractions and using exponentials to approximate Kv channel photobleach recovery provides us with a relatively model-independent channel. means of comparing different Kv channel behaviors. Some FRAP studies were done using GFP-tagged channels on an

Journal of Cell Science Olympus FluoView1000 confocal microscope. A circular region of Materials and Methods interest was photobleached in tornado scan mode with a 405 nm diode Preparation of cDNAs and cell transfection laser at 35% transmission for 1 second using the SIM scanner of the Standard PCR techniques were used to construct FV1000. The SIM scanner was synchronized with the main scanner Kv2.1HA and Kv1.4myc. To make Kv2.1HA, the HA epitope during bleach and acquisition. Following bleach, imaging was sequence YPYDVPDYA was inserted in tandem after Gly217 of the performed by raster scanning with the 488 nm line of a 40 mW Ar rat Kv2.1 primary sequence, which places the epitope in the laser at 1% transmission and the variable bandpass filter set at 505- extracellular S1-S2 loop. To make Kv1.4myc, the myc epitope 530 nm emission. A 60, 1.4 NA oil immersion objective was used sequence EQKLISEEDL was inserted in tandem after Gly430 of the with the pinhole set to 1 Airy Unit. Images were acquired every 10 human Kv1.4 primary sequence, which places the epitope in the seconds for 250 scans (41.6 minutes), except for some GFP-Kv1.4 extracellular S3-S4 loop. For both constructs, glycines were inserted FRAP experiments where images were acquired for 200 scans at the before and after the epitope sequence. Kv2.1HA, Kv1.4myc and maximum frame rate for 512512 resolution (1 frame every 1.1 Kv1.3 were also fused in frame to the C-terminus of fluorescent seconds). proteins using the pECFP, pEYFP or pEGFP-C1 vectors (Clontech, For photoactivation experiments, photoactivatable GFP-tagged Palo Alto, CA) or PA-GFP-C1 (gift from J. Lippincott-Schwartz, channels were photoactivated by raster scanning a region of interest NTH, Bethesda, MD). with a 405 nm laser at 10% transmission for 500 milliseconds using For imaging and immunostaining, vectors containing fluorescently the SIM scanner. Cells were cotransfected with monomeric RFP tagged channels were transfected into HEK293 cells (ATCC, to identify expressing cells. Following photoactivation, GFP Manassas, VA) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) fluorescence was imaged as described with images acquired every 10 at 200 ng per 60 mm dish. For live cell imaging, HEK cells were grown seconds for 250 scans. in 60 mm culture dishes and, 24 hours post-transfection, were passed All offline image analysis was done using Zeiss LSM510 v3.2 at a 1:4 dilution onto collagen coated glass bottom 35 mm dishes software or Olympus FV1000 software and Sigmaplot 8.0. All data (MatTek, Ashland, MA). For some experiments, HEK cells were are presented as mean±s.e.m. unless otherwise indicated. electroporated using a BioRad Genepulser Xcell (BioRad Laboratories, Hercules CA). Electroporation was done using a single pulse to 110 V for 25 milliseconds in a 0.2 cm gap cuvette. Cells were Depletion of membrane cholesterol subsequently imaged within 48-72 hours following transfection or For cholesterol depletion, cells were washed once with serum-free electroporation. All FRAP experiments were performed in phenol-red DMEM, then incubated in 2% 2-hydroxypropyl-β-cyclodextrin and serum-free media to decrease nonspecific fluorescence intensity. (Sigma, St Louis, MO) in serum-free DMEM for 1.5 hours at 37°C Kv channel membrane trafficking 2157

(Martens et al., 2000). Following cyclodextrin treatment, cells were linear sucrose gradient (containing 150 mM NaCl, 10 mM HEPES, washed once with DMEM containing 10% FBS and imaged within 2 pH 7.4, 1 mM CaCl2, 10 mM MgCl2 and 0.1% Triton X-100) poured hours. Imaging was performed in phenol red- and serum-free media. in a SW 50.1Ti Beckman rotor tube. The gradient was centrifuged for Cluster sizes in control and cholesterol-depleted cells were measured 2 hours at 180,000 g. Four hundred microliter fractions were collected by randomly selecting six clusters from the basal optical section and starting from the top of the gradient and aliquots analyzed by western measuring the area of each cluster. No change in Kv2.1 distribution analysis using an anti-GFP monoclonal antibody (Chemicon, was seen in cells incubated in serum-free media with no cyclodextrin Temecula, CA) as previously described (Martens et al., 2001). The (data not shown). amount of channel present in each fraction was quantitated using a BioChemi CCD camera system from UVP (Upland, CA). Immunostaining The detection of cell surface YFP-Kv2.1HA and YFP-Kv1.4myc was Results performed by incubating live cells with either anti-HA monoclonal (1:500, Sigma) (St Louis, MO) or anti-myc monoclonal (1:500, Upstate, Localization of Kv channels in HEK cells Waltham, MA) antibodies diluted in MEM + 10% horse serum for 1 The studies undertaken in our present work were initiated to hour at room temperature. Following antibody incubation, cells were understand the differential dynamics of Kv channel isoform washed twice with MEM+10% HS and then fixed with 4% trafficking in live cells and thus lay the foundation for a paraformaldehyde. Alexa 594-conjugated goat anti-mouse secondary detailed study of the of Kv channel intracellular (Molecular Probes, Eugene, OR) was used for secondary detection. trafficking. Because we are interested in the dynamic Fluorescence signals were collected using a Zeiss LSM 510 Meta trafficking of these channels, we added fluorescent tags (CFP, confocal microscope. YFP fluorescence was imaged using 514 nm excitation and a LP530 emission filter. Alexa594 was imaged using 543 YFP or GFP) to the N-terminus of each channel to permit nm excitation and BP 585-615 emission filters. The YFP and Alexa594 visualization in living cells. Additionally, epitope tags were signals were acquired using multitrack mode and no crosstalk between inserted into extracellular loops of Kv2.1 (HA) and Kv1.4 YFP and Alexa594 was observed. Three-dimensional imaging of cells (myc) to permit detection of those channels present on the was done by optically sectioning the cells with the pinhole optimized surface membrane. Both YFP-Kv2.1HA and YFP-Kv1.4myc at 1 Airy unit on each channel. Zeiss LSM510 v.3.2 software was used generate nanoampere currents when expressed in HEK cells to reconstruct and two-dimensional deconvolve the images. and display a subcellular distribution identical to their untagged counterparts as detected with anti-channel antibodies Sucrose density gradient centrifugation (data not shown), confirming that the addition of the N- terminal and loop tags does not interfere with channel biology. Two 100 mm dishes containing HEK cells at 50% confluency were transfected with channel expressing vectors using Lipofectamine 2000 YFP-Kv1.4myc does display less inactivation than the non- as described above and incubated for 18-24 hours. All subsequent YFP-tagged channel, most probably because the N-terminal steps were performed at 4°C. The media were replaced with 1.5 ml of YFP alters ball and chain inactivation (Burke et al., 1999). PBS containing 20 mg/ml pefabloc and the cells were then scraped To examine subcellular expression patterns, plasmid from the dish with a rubber policeman. Following centrifugation of the cell

Journal of Cell Science suspension at 1000 g the cells were resuspended in 0.1 ml of PBS/pefabloc and diluted 1:2 with 2% Triton X-100 and 60 mM octylglucoside in PBS. The extract was then homogenized with a glass-glass dounce and incubated for 30 minutes at 4°C. Three hundred and fifty microliters of this detergent extract was then layered on top of a 4.6 ml 5-20%

Fig. 1. Expression of YFP-tagged Kv channels in HEK cells. Confocal images of YFP-Kv2.1HA (A), YFP-Kv1.4myc (B) and YFP-Kv1.3 (C). Surface channels were detected by incubation of live cells with either anti-HA (to detect Kv2.1HA) or anti-myc (to detect Kv1.4myc) antibodies before fixation and Alexa 594- labeled secondary antibody addition as described in Materials and Methods. Colour coding is as follows: red, immunofluorescence surface labeling of the extracellular HA or myc epitopes; green, YFP fluorescence. YFP-Kv1.3 was not epitope tagged, therefore only YFP fluorescence is shown. Optical sections correspond to the relative middle of each cell. Bars, 10 µm. 2158 Journal of Cell Science 118 (10)

Fig. 2. CFP-Kv2.1HA and YFP-Kv1.4myc do not colocalize when co-expressed. Confocal image of a live HEK cell expressing both CFP- Kv2.1HA and YFP- Kv1.4myc. (A) The CFP fluorescence, (B) the YFP and (C) the overlay of the CFP and YFP signals. CFP is pseudocolored green and YFP is pseudocolored red. (D) Plot of CFP and YFP pixel intensity verses cell surface length within the boxed region shown in C. Note that while both channels are expressed in the same regions, the intensity profiles indicate that any true colocalization is unlikely to occur. Journal of Cell Science

DNAs encoding YFP-tagged Kv2.1, Kv1.4 and Kv1.3 were localization of either channel; CFP-Kv2.1HA still exhibited a transfected into HEK293 cells and examined by confocal punctate distribution (Fig. 2A), while YFP-Kv1.4myc fluorescence microscopy as shown in Fig. 1. Kv2.1 and Kv1.4 remained evenly distributed over the entire cell surface (Fig. isoforms were readily detected on the cell surface by the 2B). Moreover, despite the apparent overlap between Kv2.1 immunostaining of live cells with either anti-HA (Fig. 1A) or and Kv1.4, shown in Fig. 2C, detailed pixel-by-pixel analysis anti-myc (Fig. 1B) antibodies. Comparison of this surface (Fig. 2D) suggests that they are unlikely to truly colocalize. signal to YFP fluorescence indicated that both isoforms traffic Kv2.1 and Kv1.4 must reside in distinct raft domains. efficiently to the surface, with only a small fraction of the In neurons, Kv2.1 displays a punctate expression pattern and channel in an intracellular compartment. Kv1.3, another tends to preferentially accumulate on the bottom surface of the member of the Kv1 family, is closely related to Kv1.4, so for cell body (Lim et al., 2000). This expression pattern was comparison, we expressed YFP-tagged Kv1.3 in HEK cells, reproduced well in the HEK cells (Fig. 1A and Fig. 3A). Fig. where it exhibited a distribution similar to that of YFP- 3A (top) shows an optical section of the basal surface of Kv1.4myc. However, Kv1.3 tended to accumulate in an a representative YFP-Kv2.1HA-expressing cell, clearly intracellular compartment to a greater extent than either YFP- illustrating the clustered distribution of Kv2.1 in HEK cells. Kv2.1HA or YFP-Kv1.4myc (Fig. 1C). An x-z image of the same cell is presented in Fig. 3A (bottom), Kv2.1 and Kv1.4 do not form heteromeric channels when illustrating the preferential distribution of Kv2.1 on the basal coexpressed, and given their distinct cell surface distributions, cell membrane. By comparison, YFP-Kv1.4myc is evenly it seemed unlikely that the two channels would reside in similar distributed over the entire bottom surface of the cell, and shows plasma membrane compartments. However, both Kv2.1 and no apparent polarization between top and bottom (Fig. 3B). Kv1.4 are present in noncaveolar lipid raft domains (Martens et al., 2000; Wong and Schlichter, 2004). To address the possibility that Kv2.1 and Kv1.4 could reside in the same raft Mobility of Kv2.1HA compartment, we coexpressed CFP-tagged Kv2.1HA and Because Kv2.1 formed punctate surface structures that may YFP-tagged Kv1.4myc in HEK cells. Coexpression of CFP- be indicative of channel anchored to the cytoskeleton, we Kv2.1HA and YFP-Kv1.4myc did not alter the subcellular investigated the mobility of the channel on the plasma Kv channel membrane trafficking 2159 Journal of Cell Science Fig. 3. YFP-Kv2.1HA and YFP-Kv1.4myc exhibit a differential polarization in HEK cells. Confocal images of YFP fluorescence in live cells were taken of the bottom surface of expressing cells (top panels in both A and B). Three-dimensional confocal stacks were used to generate X-Z images of each cell, which were two- dimensional deconvolved using Zeiss LSM510 v3.2 software (bottom panel, A and B). Kv2.1 clearly has a preference for the basal µ Fig. 4. YFP-Kv2.1HA exhibits limited mobility after photobleach. surface, whereas Kv1.4 is evenly distributed. Bars, 10 m. (A) Images from a FRAP time series in a YFP-Kv2.1HA-expressing HEK cell. The region highlighted by the white box was bleached and membrane using two complementary approaches: photobleach recovery in that region was followed by scanning every 15 seconds (FRAP) and photoactivation. For photobleach experiments, for 20 minutes (see Materials and Methods). The inset shows a small region of surface membrane of a YFP-Kv2.1HA- enlargement of the bleach region and shows that recovery of YFP fluorescence does not occur via lateral diffusion of YFP-Kv2.1 from expressing cell was bleached using the 514 nm laser line and adjacent nonbleached membrane. (B) Average recovery after recovery monitored by scanning once every 15 seconds for ~21 photobleach for YFP-Kv2.1HA. Fluorescence at each time point minutes. Fluorescence recovery within the bleach region was after bleach (bleach=time 0) was normalized to the prebleach normalized to the pre-bleach intensity and the data fit by Eqn intensity and averaged for all cells (n=25). The solid line through the 1 to obtain recovery time constants (τ) (see Materials and data is a two-exponential fit to the average data (see Materials and Methods). Following photobleach, YFP-Kv2.1 exhibited low Methods). Fit parameters: τ1=36.1 seconds, A1=0.13; τ2=308 µ recovery, with an average mobile fraction (Mf) of 0.41±0.04 seconds, A2=0.24, Mf=0.41±0.04. Bar, 10 m. (Fig. 4A,B). Even if a 10-15% cumulative photobleach during the recovery process is taken into account, no more than 50- 55% of the channel probably recovers during the 20 minute (Fig. 4B). Further evidence for a more complex recovery recovery period. Additionally, recovery was not well fit by a scheme is that YFP fluorescence recovered evenly across the single exponential, suggesting that YFP-Kv2.1 mobility is bleached ROI (see Fig. 4, insets and Fig. S1 in supplementary not a diffusion-limited process as would be expected if material) as opposed to first recovering at the ROI border. If fluorescence recovery was due to lateral membrane diffusion YFP-Kv2.1 fluorescence was due to the lateral diffusion of 2160 Journal of Cell Science 118 (10)

Fig. 5. Use of photoactivatable-GFP to monitor Kv2.1 mobility. HEK cells were transfected with PA-GFP-Kv2.1HA and mRFP to mark transfected cells (mRFP not shown). Before activation, PA-GFP-Kv2.1HA-expressing cells displayed no GFP fluorescence (Preactivation). Photoactivation was via illumination of an ROI (red box) using a 405 nm laser for 500 milliseconds at 10-15% attenuation. This protocol resulted in bright green fluorescence with minimal photobleach. After activation, GFP was excited by 488 nm laser light (emission bandpass set at 505-530 nm) and monitored by scanning every 5 seconds for 250 scans (~41 minutes). At 640 seconds post-photoactivation, green fluorescence is apparent on the cell membrane opposite to the activation ROI (red arrow). Note that even 2500 seconds following activation, most of the fluorescence is still restricted to the photoactivation ROI, consistent with the low recovery observed with FRAP. Bar, 10 µm.

nonbleached channels from adjacent membrane, recovery channels are not anchored at the membrane and traffic through Journal of Cell Science would occur initially at the edge of the ROI. Furthermore, in an intracellular pathway rather than diffusing though the no experiment did fluorescence recover completely, and the membrane. slow component of the recovery kinetics has a relatively long time constant (τslow=4.9 minutes, compared with 36 seconds for the fast component). We were only able to follow recovery Mobility of Kv1.4myc and Kv1.3 for ~20 minutes due to cumulative photobleaching and cell While Kv2.1 resides in distinct surface puncta, Kv1.4 and movement, and not all cells reached steady state during this Kv1.3 are both evenly distributed over the surface membrane time frame, introducing some error in the long time constant. in HEK cells (Fig. 1B,C). However, like Kv2.1, both of these While FRAP is useful for determining the mobility of a channels are also present in lipid raft domains, raising the protein, it provides no information on protein trafficking question of whether raft association alone is sufficient to pathways. However, the recent development of immobilize channels in the plasma membrane. We therefore photoactivatable GFP, which becomes fluorescent only after also used photobleach and photoactivation to investigate the irradiation by 405-413 nm light (Patterson and Lippincott- mobility of these channels. As shown in Fig. 6, following Schwartz, 2002), provides a means of following a specific photobleach, both channels recovered to a greater extent than photolabeled protein. The restricted lateral mobility of Kv2.1 Kv2.1, with Kv1.4 and Kv1.3 having an Mf of 0.78±0.07 and is clearly illustrated through the use of photoactivatable GFP- 0.78±0.04, respectively. Because these data have not been tagged Kv2.1. A region of interest along the cell membrane corrected for 10-15% photobleach during recovery, the was photoactivated (see Materials and Methods) and the measured recovery is at least 90% and in reality probably movement of GFP fluorescence monitored by scanning once complete. Unlike Kv2.1, Kv1.4 recovery was well fit by a every 10 seconds for ~41 minutes (see Materials and Methods). single exponential (τ=133 seconds). Similarly Kv1.3 As seen in Fig. 5, even 2500 seconds after photoactivation, at fluorescence recovery was also well fit with a single time least half the GFP fluorescence is restricted to the ROI, constant of 118 seconds. Unlike the recovery observed with indicating that this Kv2.1 is probably anchored in place on the Kv2.1, Kv1.4 appeared to recover via lateral diffusion of cell surface. Interestingly, approximately 10 minutes after nonbleached channel from adjacent membrane (Fig. 6 and Fig. photoactivation, GFP fluorescence can be seen on the opposite S2 in supplementary material). YFP-Kv1.3 also appears to membrane from the activation ROI, suggesting that some recover by lateral diffusion, similar to YFP-Kv1.4 (data not Kv channel membrane trafficking 2161

Fig. 6. YFP-Kv1.4myc and YFP-Kv1.3 are freely mobile in the plasma membrane. (A) Images from FRAP times series in a YFP-Kv1.4myc- expressing HEK cell. The region highlighted in the white box was bleached and recovery in that region followed by scanning every 15 seconds Journal of Cell Science for a total of 80 scans (see Materials and Methods). (B) Average recovery after photobleach for YFP-Kv1.4myc. Fluorescence at each time point was normalized to prebleach intensity and averaged for all cells (n=16). The solid line through the data is a single exponential fit to the average data (see Materials and Methods). Fit parameters: τ=133.0 seconds, A=0.52; Mf=0.78±0.07. (C) Average recovery following photobleach for YFP-Kv1.3 (n=31). Curve and fit was generated as for YFP-Kv1.4 with the solid line representing a single exponential fit. Fit parameters: τ=117.6 seconds, A=0.56; Mf=0.78±0.04. Bar, 10 µm.

shown). The high degree of mobility exhibited by Kv1.4 is glucoside to fully disrupt lipid raft domains, and the entire dramatically illustrated by photoactivation of PA-GFP-Kv1.4, extract, including nuclear material, was then loaded onto a 5- as shown in Fig. 7. As for PA-GFP-Kv2.1, a region of cell 20% linear sucrose density gradient. As shown in Fig. 8, membrane was photoactivated by brief scanning with a 405 approximately 50% of the YFP-Kv2.1HA migrated only nm laser and movement of GFP fluorescence monitored every slightly heavier than the predicted tetrameric molecular 10 seconds for approximately 41 minutes. Unlike Kv2.1, PA- weight of 520,000, while the remaining channel sedimented GFP-Kv1.4 rapidly diffuses out of the ROI, with green to the bottom of the sucrose gradient. Because the gradient fluorescence apparent in membrane proximal to the ROI was centrifuged for only 2 hours at 180,000 g, the Kv2.1 at within 1 minute (Fig. 7, 69 seconds). Approximately 4 the bottom of the tube represents a very large macromolecular minutes after photoactivation, green channel appears to have complex, easily exceeding a molecular weight of several diffused throughout the entire cell membrane (Fig. 7, 259 million. By contrast, 95% of the Kv1.4 channel was found at seconds). the top of the sucrose gradient near the expected molecular weight of 480,000 for the tetrameric complex. The transferrin and caveolin remained, as expected, near the top of Sucrose density gradient analysis of Kv2.1 and Kv1.4 the gradient (data not shown). Lack of Kv1.4 association with Because nearly half of the surface Kv2.1 channel is immobile a large macromolecular complex is consistent with its but most of Kv1.4 is readily diffusible, we asked whether a apparent high degree of mobility following FRAP. It is also difference in macromolecular complex assembly could be tempting to speculate that the mobile fraction of Kv2.1 is detected between these two channel isoforms. Cells were represented by the Kv2.1 protein at the top of the sucrose homogenized in the presence of Triton X-100 and octyl- gradient. 2162 Journal of Cell Science 118 (10)

Fig. 7. Use of photoactivatable-GFP to observe diffusion of Kv1.4. Photoactivation and imaging was performed as described in Fig. 5. No fluorescence is apparent before photoactivation of PA-GFP-Kv1.4myc; following illumination by 405 nm light, there is bright green fluorescence in the ROI. Unlike PA-GFP-Kv2.1, ~1 minute after activation, there is GFP signal apparent throughout the cell membrane, consistent with rapid diffusion of Kv1.4 throughout the plasmalemma. Approximately 13 minutes after photoactivation, PA-GFP-Kv1.4 is µ Journal of Cell Science evenly distributed over the entire cell membrane. Bar, 10 m.

Effects of cholesterol depletion on Kv channel discrete clusters (Fig. 9). Panel A shows the typical pattern in localization and mobility control cells (cluster size equals 1.6±0.2 µm2), whereas panel All three Kv channels studied here can be isolated in B illustrates the altered distribution following cholesterol cholesterol-rich lipid raft domains (Bock et al., 2003; Martens depletion (cluster area=4.4±0.6 µm2). The Kv2.1 cluster size et al., 2000; Wong and Schlichter, 2004). We have previously following cholesterol depletion is an underestimate, as shown that treatment of cells stably expressing Kv2.1 with 2% the boundaries of the clusters become less distinct after 2-hydroxypropyl-cyclodextrin for 1 hour alters Kv2.1 channel cyclodextrin treatment (see Fig. 9B) and only those clusters function, shifting the voltage-dependence of inactivation to whose margins could be clearly identified were chosen for more hyperpolarized potentials (Martens et al., 2000). analysis. By contrast, the surface distribution of both YFP- Therefore, we next asked whether Kv channel localization or Kv1.4 and YFP-Kv1.3, which are already evenly distributed mobility was altered following cholesterol depletion with over the cell membrane, was unaffected by cholesterol cyclodextrin. Interestingly, cholesterol depletion dramatically depletion (data not shown). altered Kv2.1 cluster size. Following cyclodextrin treatment, Given the dramatic increase in Kv2.1 cluster size following Kv2.1 redistributed into large patches, as opposed to small cholesterol depletion, we next used FRAP to determine

Fig. 8. YFP-Kv2.1 and YFP-Kv1.4 sediment differently on a sucrose density gradient. Whole-cell homogenates from HEK cells expressing either YFP-Kv2.1HA (top) or YFP-Kv1.4myc (bottom) were sedimented through a 5-20% sucrose density gradient (see Materials and Methods) to estimate channel-containing complex size. Kv2.1 and Kv1.4-containing gradient fractions were identified via western blot analysis using anti-GFP antibody. Thyroglobulin (660 kDa) was used as a size standard, giving a symmetrical peak in fraction 4. P represents the pellet material collected from the bottom of the tube. Kv channel membrane trafficking 2163 whether channel mobility was also affected. As illustrated in Fig. 10A, YFP-Kv2.1 fluorescence recovers slightly less than control following cyclodextrin treatment (Mf=0.32±0.03); however, the kinetics of recovery were altered, as recovery appears to be slowed over the early part of the recovery curve. Recovery could not be fit by a single exponential, requiring a two exponential fit with time constants of 51 seconds and 2890 seconds as compared with 36 seconds and 307 seconds under control conditions. The fractional recovery of Kv1.4 was decreased also following cholesterol depletion but to a greater extent than Kv2.1 (Mf=0.58±0.07). By contrast, the kinetics of recovery were only minimally affected (Fig. 10B), with only a slight increase in the time constant (τ=196 seconds). This decreased recovery is not a common feature of Kv1 family members, as Kv1.3 recovery was slightly increased following cyclodextrin Fig. 9. Depletion of membrane cholesterol alters Kv2.1 cluster size. Cells were cholesterol depleted by incubation with 2% 2-hydroxypropyl-β-cyclodextrin for 1.5 treatment (Mf=0.89±0.06 compared with 0.78±0.04 under control conditions). hours at 37°C, then imaged within 2 hours (see Materials and Methods). Cluster size was determined by confocal imaging of the basal surface of expressing cells and Furthermore, similar to what was observed measurement of six randomly selected puncta from each cell. The average cluster size for for Kv2.1, the recovery kinetics were control cells (A) was 1.6±0.2 µm2 (n=102) and 4.4±0.6 µm2 (n=132)* (*=P<0.001) dramatically altered for Kv1.3. Following following cholesterol depletion (B). Note that only discrete aggregates were measured as cholesterol depletion, a single exponential opposed to the large plaques in B, therefore the noted size increase is probably an fit no longer adequately fits the data and two underestimate. Bars, 5 µm. exponentials are required (Fig. 10C). Taken together, these FRAP data indicate that Kv2.1, 1.4 and 1.3 are differentially affected by cholesterol the effects of cholesterol depletion reported here support some depletion as expected if these three channels traffic through role for lipid rafts in Kv channel localization and trafficking. distinct membrane compartments.

Journal of Cell Science Implications of different channel mobilities Discussion At least half of the expressed Kv2.1 had a very limited mobility The degree to which cells go to segregate ion channels into in HEK cells as evidenced by FRAP recovery. Channels discrete compartments (e.g. Kv2.1 on the soma, Kv4.2 in that did recover appeared to do so from an intracellular dendrites and Kv1.4 in ) is striking. However, little is compartment, as no apparent diffusion of channel from known about how these channels reach their ultimate neighboring membrane was observed (see Fig. 4 insets and Fig. destination in native tissues, or how the highly specific S1 in supplementary material). The existence of two distinct localization affects channel function and cellular excitability. Kv2.1 channel populations is supported by the sucrose density Surprisingly, we also know little about how these channels are gradient centrifugation experiment of Fig. 8 in which ~50% of trafficked to the cell surface of HEK 293 cells, the most Kv2.1 sediments with a mass of at least several million while common system used for the heterologous expression of ion the rest sedimented as expected for a Kv channel tetramer. It channels. We show here that even in the simplified HEK 293 is tempting to correlate the high molecular weight pool with system, Kv2.1, Kv1.4 and Kv1.3 show differential trafficking the immobile fraction observed during FRAP, as low mobility and distinct sensitivities to membrane cholesterol depletion. is expected if the channel is part of a large protein complex. These channels share extensive sequence in the Consistent with this idea, Kv1.4 was freely mobile in the HEK transmembrane domains (particularly Kv1.4 and Kv1.3) but cells and 95% of this channel was found in the lighter fractions are highly divergent within the N- and C-termini, supporting of the sucrose density gradient (Fig. 8). Despite this argument, the hypothesis that it is these unique domains that play a it should be kept in mind that while channel association with primary role in intracellular trafficking and/or immobilization the cytoskeleton could result in the heavy fraction of Kv2.1, it on the cell surface. Indeed, several groups have implicated is also plausible that Kv2.1 is interacting with a large these domains in either channel trafficking or localization for cytoskeletal motor, making it both heavy and highly mobile. both Kv1 and Kv2 family channels (Antonucci et al., 2001; Li However, Kv2.1 remains associated with insoluble elements et al., 2000; Lim et al., 2000; Misonou et al., 2004; Zhu et al., following Triton X-100 extraction of live cells (data not 2003). Kv channel N- and C-termini most probably drive shown), which is highly suggestive of cytoskeletal attachment. trafficking through different subcellular pathways via It was recently reported that TrpC channels are rapidly interaction with unique sets of endogenous proteins. However, recruited to the membrane from a submembrane compartment 2164 Journal of Cell Science 118 (10) that a subsurface pool of Kv2.1 exists, even in HEK cells, thus providing a means of rapidly delivering Kv2.1 to the surface. We have attempted to detect such an intracellular pool of Kv2.1 just below the cell surface. For example, YFP-Kv2.1-HA channel truly on the surface, in theory, can be distinguished from subsurface channel by comparing YFP fluorescence with that derived from fluorescent antibody labeling of the external HA epitope. Despite several attempts, we have been unable to consistently detect a significant YFP signal in the absence of a cell surface derived fluorescence signal (data not shown). However, channels present in a submembrane compartment or ER near the membrane may not be discernable from surface membrane at the resolution of confocal microscopy.

The influence of membrane cholesterol on Kv channel trafficking and localization One of the most striking effects of cholesterol depletion was the dramatic redistribution of Kv2.1 from small clusters into large patches (Fig. 9). It is unknown what causes the assembly of Kv2.1 into small clusters, other than the long C-terminus seems to play a role. Expression of a C-terminal truncation in hippocampal neurons results in a diffuse pattern of expression of Kv2.1 as does de- of the C-terminus (Lim et al., 2000; Misonou et al., 2004). Because the precise molecular machinery involved in Kv2.1 clustering remains unknown, it is not clear whether depletion of membrane cholesterol in HEK cells is mechanistically linked to the effects of C-terminal truncation or de-phosphorylation. However, Kv2.1 is lipid raft-associated, and these membrane microdomains are thought to be platforms for phosphorylation pathways and intracellular trafficking. Given the dramatic effect cholesterol depletion had on Kv2.1 localization, the more subtle effects on Kv2.1 mobility were surprising. While overall recovery at 20 minutes was decreased Journal of Cell Science by 20%, the time course of recovery was significantly slowed, Fig. 10. Effect of cholesterol depletion on FRAP kinetics for Kv2.1, with the long time constant now approaching 1 hour (Fig. 10). Kv1.4 and Kv1.3. HEK cells were cholesterol depleted by incubation Because 20 minutes is insufficient time to reach steady state, with 2% 2-hydroxypropyl-β-cyclodextrin (CD) and FRAP measured as described in Materials and Methods. Average recovery after it is likely that recovery would have continued, eventually photobleach for YFP-Kv2.1HA (A), YFP-Kv1.4myc (B) and YFP- reaching the control value had it been possible to follow Kv1.3 (C). The dashed line in each plot represents the FRAP under recovery for long enough. As cholesterol depletion is expected control conditions for each channel, and solid lines are exponential to alter raft structure, it is reasonable to assume that rafts play fits to the average data following cholesterol depletion. Fit some role in Kv2.1 localization and/or trafficking. Perhaps parameters – Kv2.1 (n=8): τ1=51.0 seconds, A1=0.12; τ2=2889.1 cholesterol-depleted submembrane vesicles are not as mobile τ seconds, A2=0.54; Mf=0.32±0.03. Kv1.4 (n=9): =195.7 seconds, as their native counterparts. Interestingly, while Kv1.4 and τ A=0.39; Mf=0.58±0.07. Kv1.3 (n=18): 1=39.2 seconds, A1=0.34; Kv1.3 are also lipid raft-associated, their surface distribution τ 2=535.7 seconds, A2=0.42; Mf=0.89±0.06. was not altered following cholesterol depletion (data not shown). One potential artifact is that cholesterol depletion is (Bezzerides et al., 2004), and similar mechanisms may exist expected to change membrane fluidity and the overall health for inward rectifier K+ channels (Ma et al., 2002). If such a of the cells, potentially affecting Kv2.1 trafficking. If Kv2.1 is mechanism exists for Kv2.1 trafficking, it could explain why part of a macromolecular complex, then a decrease in Kv2.1 exhibits two different mobilities. Perhaps the mobile membrane fluidity following cyclodextrin treatment could Kv2.1 fraction represents such a submembrane, vesicular conceivably slow the mobility of a large complex. There are compartment while the immobile fraction represents several arguments against this, however. First, Kv2.1 recovery cytoskeleton-linked channel on the cell surface. Electron is not due to lateral membrane diffusion. Second, the Kv1.4 microscopy studies have shown that surface Kv2.1 clusters are mobile fraction is decreased, but with minimal effects on localized near membrane- (ER) recovery kinetics. Third, Kv1.3 recovery is also significantly junctions in hippocampal and cortical neurons and subsurface slowed, despite an increase in the mobile fraction. Together Kv2.1 clusters have been detected in spinal motoneurons (Du these data suggest that protein-lipid interactions may play a et al., 1998; Muennich and Fyffe, 2004). Thus, it is plausible role in ion channel trafficking and that the different effects of Kv channel membrane trafficking 2165 cholesterol depletion observed with the three channels under alters Kv2.1 clustering, but not the localization of Kv1.4 or 1.3, study here may be due to differential targeting to distinct raft its remains likely that Kv isoform association with distinct lipid domains. raft domains is involved channel trafficking and localization. The use of HEK cells is potentially problematic, as the Kv channels studied here are normally expressed in nerve and Use of photoactivatable-GFP to follow Kv channel muscle and almost certainly interact with tissue-specific trafficking proteins not present in HEK cells. However, Kv2.1 retains its While photobleach techniques such as FRAP are a well- neuronal expression pattern in HEK cells, indicating that much established means of determining protein mobility, FRAP of the machinery required for this distribution is common suffers from inherent limitations, such as a low signal-to-noise between the two cell types. Therefore, HEK cells are a suitable ratio. Furthermore, FRAP provides little to no information system for studying Kv2.1 mobility and will be invaluable in about the origin or mechanism behind fluorescence recovery identifying protein partners for Kv2.1 by biochemical and protein mobility. For protein trafficking studies, it is more purifications, which are not feasible in a primary culture advantageous to selectively photolabel the protein of interest system. and visualize its movement through the cell. The recent introduction of photoactivatable fluorescent proteins, such as We thank Patrick D. Sarmiere for discussion and critical reading of the PA-GFP used in this study, represents a significant advance this manuscript and Roger Tsien for providing the monomeric RFP for live-cell imaging of protein trafficking, allowing the construct. This work was supported by grants from the National monitoring of channel movement in real time in a way that Institutes of Health to M.M.T. (HL49330 and NS41542) and an NIH real-time imaging with conventional fluorescent proteins NRSA to K.M.S.O. (HL77056). cannot. One consideration that must be kept in mind is the inherent References limitations on the excitation focal plane imposed by single Antonucci, D. E., Lim, S. T., Vassanelli, S. and Trimmer, J. S. (2001). photon excitation. The point spread function of any single Dynamic localization and clustering of dendritic Kv2.1 voltage-dependent photon laser will result in the bleach or activation of a potassium channels in developing hippocampal neurons. Neuroscience 108, substantial volume of the cell (Fig. S3 in supplementary 69-81. material). This is of course also a consideration during FRAP, Bendahhou, S., Donaldson, M. R., Plaster, N. M., Tristani-Firouzi, M., Fu, confounding precise measurements of recovery, since bleach Y. H. and Ptacek, L. J. (2003). Defective Kir2.1 trafficking underlies Andersen-Tawil syndrome. J. Biol. Chem. 278, 51779- and recovery occur in three dimensions. For this reason, we 51785. chose to quantitate recovery by measuring the time constant of Bezzerides, V. J., Ramsey, I. S., Kotecha, S., Greka, A. and Clapham, D. recovery rather than calculating a diffusion constant, which E. (2004). Rapid vesicular translocation and insertion of TRP channels. Nat. would have required knowing the volume of the bleach region. Cell Biol. 6, 709-720. For future work with PA-GFP, it will be necessary to consider Bock, J., Szabo, I., Gamper, N., Adams, C. and Gulbins, E. (2003). Ceramide inhibits the potassium channel Kv1.3 by the formation of that the photolabeled channels exist in a three-dimensional membrane platforms. Biochem. Biophys. Res. Commun. 305, 890-897. volume rather than a two-dimensional plane. Nevertheless, PA-

Journal of Cell Science Burke, N. A., Takimoto, K., Li, D., Han, W., Watkins, S. C. and Levitan, GFP is a powerful tool for imaging channel trafficking in real E. S. (1999). Distinct structural requirements for clustering and time, as evidenced by experiments such as those shown in Figs immobilization of K+ channels by PSD-95. J. Gen. Physiol. 113, 71-80. 5 and 7. We have been able to clearly visualize migration of Delisle, B. P., Anson, B. D., Rajamani, S. and January, C. T. (2004). Biology of cardiac : ion channel protein trafficking. Circ. Res. 94, 1418- Kv2.1 from the activation ROI to distal regions of the cell (Fig. 1428. 5), as well as visualize the rapid diffusion of Kv1.4 (Fig. 7). In Du, J., Tao-Cheng, J. H., Zerfas, P. and McBain, C. J. (1998). The K+ the future, photoactivation of channels early in the biosynthetic channel, Kv2.1, is apposed to astrocytic processes and is associated with pathway will allow us to follow channels on their way to the inhibitory postsynaptic membranes in hippocampal and cortical principal plasma membrane. neurons and inhibitory interneurons. Neuroscience 84, 37-48. Kim, E., Niethammer, M., Rothschild, A., Jan, Y. N. and Sheng, M. (1995). Clustering of -type K+ channels by interaction with a family of membrane-associated guanylate kinases. Nature 378, 85-88. Concluding remarks Li, D., Takimoto, K. and Levitan, E. S. (2000). Surface expression of Kv1 Despite extensive sequence and functional channels is governed by a C-terminal motif. J. Biol. Chem. 275, 11597- similarities, the Kv2.1, Kv1.4 and Kv1.3 channels display 11602. Lim, S. T., Antonucci, D. E., Scannevin, R. H. and Trimmer, J. S. (2000). unique plasma membrane expression in HEK 293 cells, distinct A novel targeting signal for proximal clustering of the Kv2.1 K+ channel in FRAP kinetics and different responses to cholesterol depletion. hippocampal neurons. Neuron 25, 385-397. These differences support the hypothesis that the variable N- Ma, D., Zerangue, N., Raab-Graham, K., Fried, S. R., Jan, Y. N. and Jan, and C-termini play a primary role in intracellular trafficking L. Y. (2002). Diverse trafficking patterns due to multiple traffic motifs in G and/or immobilization of the cell surface, even in HEK 293 protein-activated inwardly rectifying potassium channels from brain and . Neuron 33, 715-729. cells that normally express no Kv currents. The majority of Martens, J. R., Navarro-Polanco, R., Coppock, E. A., Nishiyama, A., Kv1.4 and Kv1.3 are highly mobile in the plane of the surface Parshley, L., Grobaski, T. D. and Tamkun, M. M. (2000). Differential membrane. By contrast, only half of Kv2.1 is mobile and targeting of Shaker-like potassium channels to lipid rafts. J. Biol. Chem. 275, channel truly on the cell surface is probably tethered to a very 7443-7446. large macromolecular complex. The FRAP kinetics that are Martens, J. R., Sakamoto, N., Sullivan, S. A., Grobaski, T. D. and Tamkun, M. M. (2001). Isoform-specific localization of voltage-gated K+ channels observed for Kv2.1 are most probably due to diffusion from to distinct lipid raft populations. Targeting of Kv1.5 to caveolae. J. Biol. inside the cell to the surface as opposed to lateral movement Chem. 276, 8409-8414. within the membrane. Since cholesterol depletion dramatically Martens, J. R., O’Connell, K. and Tamkun, M. (2004). Targeting of ion 2166 Journal of Cell Science 118 (10)

channels to membrane microdomains: localization of KV channels to lipid GFP for selective photolabeling of proteins and cells. Science 297, 1873- rafts. Trends Pharmacol. Sci. 25, 16-21. 1877. Maruoka, N. D., Steele, D. F., Au, B. P., Dan, P., Zhang, X., Moore, E. D. Powell, K. and Zeitlin, P. L. (2002). Therapeutic approaches to repair defects and Fedida, D. (2000). alpha-Actinin-2 couples to cardiac Kv1.5 channels, in deltaF508 CFTR folding and cellular targeting. Adv. Deliv. Rev. 54, regulating and channel localization in HEK cells. FEBS Lett. 1395-1408. 473, 188-194. Scannevin, R. H., Murakoshi, H., Rhodes, K. J. and Trimmer, J. S. (1996). Misonou, H., Mohapatra, D. P., Park, E. W., Leung, V., Zhen, D., Misonou, Identification of a cytoplasmic domain important in the polarized expression K., Anderson, A. E. and Trimmer, J. S. (2004). Regulation of ion channel and clustering of the Kv2.1 K+ channel. J. Cell Biol. 135, 1619-1632. localization and phosphorylation by neuronal activity. Nat. Neurosci. 7, 711- Sheng, M., Tsaur, M. L., Jan, Y. N. and Jan, L. Y. (1992). Subcellular 718. segregation of two A-type K+ channel proteins in rat central neurons. Muennich, E. A. and Fyffe, R. E. (2004). Focal aggregation of voltage-gated, Neuron 9, 271-284. Kv2.1 subunit-containing, potassium channels at synaptic sites in rat spinal Wong, W. and Schlichter, L. C. (2004). Differential recruitment of Kv1.4 and motoneurones. J. Physiol. 554, 673-685. Kv4.2 to lipid rafts by PSD-95. J. Biol. Chem. 279, 444-452. O’Connell, K. M., Martens, J. R. and Tamkun, M. M. (2004). Localization Zhu, J., Watanabe, I., Gomez, B. and Thornhill, W. B. (2003). Trafficking of ion channels to lipid Raft domains within the cardiovascular system. of Kv1.4 potassium channels: interdependence of a pore region determinant Trends Cardiovasc. Med. 14, 37-42. and a cytoplasmic C-terminal VXXSL determinant in regulating cell-surface Patterson, G. H. and Lippincott-Schwartz, J. (2002). A photoactivatable trafficking. Biochem. J. 375, 761-768. Journal of Cell Science