AMPA Receptor/TARP Stoichiometry Visualized by Single-Molecule Subunit Counting

AMPA receptor/TARP stoichiometry visualized by single-molecule subunit counting

Peter Hastiea,1, Maximilian H. Ulbricha,b,c,1, Hui-Li Wanga,d,e,f, Ryan J. Aranta, Anthony G. Laua,d,e, Zhenjie Zhanga,d,e, Ehud Y. Isacoffa,g,h,2, and Lu Chena,d,e,g,2

aDepartment of Molecular and Cell Biology, and gHelen Wills Neuroscience Institute, University of California, Berkeley, CA 94720; bBIOSS Centre for Biological Signalling Studies, and cInstitute of Physiology II, Albert-Ludwigs-Universität Freiburg, 79104 Freiburg, Germany; dDepartment of Psychiatry and Behavioral Sciences, and eStanford Institute of Neuro-Innovation and Translational Neuroscience, Stanford University, Stanford, CA 94305-5453; fSchool of Biotechnology and Food Engineering, Hefei University of Technology, Hefei, Anhui 230009, China; and hMaterial Science and Physical Bioscience Divisions, Lawrence Berkeley National Laboratory, Berkeley, CA 94720

Edited* by David Julius, University of California, San Francisco, CA, and approved February 15, 2013 (received for review November 1, 2012)

Members of the transmembrane AMPA receptor-regulatory protein magnitude of modulation of AMPA-Rs has been shown to be (TARP) family modulate AMPA receptor (AMPA-R) trafficking and proportional to the level of TARP expression, suggesting that function. AMPA-Rs consist of four pore-forming subunits. Previous complex stoichiometry is not fixed (20). One recent study examined studies show that TARPs are an integral part of the AMPA-R the stoichiometry between AMPA-Rs and γ-2(alsocalledstarga- complex, acting as accessory subunits for mature receptors in vivo. zin), and found that AMPA-R/γ-2 complex under overexpression The TARP/AMPA-R stoichiometry was previously measured indi- conditions has a variable stoichiometry (one to four γ-2 per com- rectly and found to be variable and dependent on TARP expression plex) but that one γ-2 unit was sufficient to modulate AMPA-R level, with at most four TARPs associated with each AMPA-R activity. It was also observed that in neurons γ-2 has a fixed stoi- complex. Here, we use a single-molecule technique in live cells that chiometry on AMPA-Rs (29). Another study showed that in- selectively images proteins located in the plasma membrane to creasing TARP expression level increases kainate efficacy (the directly count the number of TARPs associated with each AMPA-R amplitude of kainate-evoked currents as a fraction of glutamate- complex. Although individual GFP-tagged TARP subunits are ob- evoked current amplitude), with low kainate efficacy for AMPA-R served as freely diffusing fluorescent spots on the surface of Xenopus alone, intermediate efficacy when two of the four AMPA-R sub- NEUROSCIENCE laevis oocytes when expressed alone, coexpression with AMPA-R– units are fused to a TARP, and maximal kainite efficacy when all mCherry immobilizes the stargazin-GFP spots at sites of AMPA-R– four AMPA-R subunits are fused to a TARP or when the free mCherry, consistent with complex formation. We determined the AMPA-R is coexpressed with very high levels of free TARP (30). number of TARP molecules associated with each AMPA-R by counting We examined the stoichiometry of TARP/AMPA-R complexes bleaching steps for three different TARP family members: γ-2, γ-3, directly using a single-molecule method in total internal reflection and γ-4. We confirm that the TARP/AMPA-R stoichiometry de- fluorescence (TIRF) microscopy (31). We find that green fluo- pends on TARP expression level and discover that the maximum rescent protein (GFP)-tagged TARP subunits diffuse freely in the SCIENCES number of TARPs per AMPA-R complex falls into two categories: membrane of Xenopus oocytes when expressed alone, whereas up to four γ-2 or γ-3 subunits, but rarely above two for γ-4 subunit. fluorescently tagged AMPA-R subunits expressed alone are im- APPLIED PHYSICAL This unexpected AMPA-R/TARP stoichiometry difference has immobile. When they are coexpressed, the TARP subunits colocalize portant implications for the assembly and function of TARP/ with the AMPA-Rs and become immobile. We examined four AMPA-R complexes. TARP family members, γ-2, γ-3, γ-4, and γ-8 (γ-8 function was affected by tagging and therefore not included in the analysis). We single-molecule counting | TIRF | glutamate receptors | stargazin find that the TARP/AMPA-R stoichiometry depends on TARP expression level, as shown earlier. Strikingly, the maximum number of TARPs bound to each AMPA-R complex falls into two lutamate is the main excitatory neurotransmitter in the γ γ γ Gmammalian CNS. Most fast excitatory synaptic transmission categories: up to four -2 or -3 subunits, but only up to two -4 in the brain is mediated by AMPA receptors (AMPA-Rs). Four subunits. AMPA-R subunits, GluA1–4, contribute to the heterotetrameric Results assemblies of the AMPA-R (1–4). The localization of AMPA-Rs to the postsynaptic membrane is regulated by a large number of Counting GFP-Labeled AMPA-R Subunits with TIRF Microscopy. In proteins through multiple mechanisms and plays important roles neurons, most GluA1 subunits of AMPA-Rs assemble with in synaptic plasticity (5–9). GluA2 into GluA1/GluA2 heteromers. When expressed alone in Transmembrane AMPA-R regulatory proteins (TARPs) rep- heterologous systems, however, functional homomeric receptors resent a family of AMPA-R regulatory proteins that are tightly can be formed by either subunit, which provides a useful system associated with AMPA-Rs, and can be considered auxiliary for studying AMPA-R structure and function in vitro. Taking AMPA-R subunits (10–12). TARPs regulate AMPA-R function advantage of this property of the AMPA-Rs, we first validated by several mechanisms. They mediate the efficient cell surface our single-molecule approach in Xenopus laevis oocytes, our expression of AMPA-Rs, modulate gating, affect agonist efficacy, and even attenuate intracellular polyamine block of calcium- permeable AMPA-Rs (13–22). The effect of stargazin (γ-2) and Author contributions: P.H., M.H.U., and L.C. designed research; P.H., M.H.U., H.-L.W., other TARPs on GluA1 is not neuron specific and can be accu- R.J.A., and A.G.L. performed research; M.H.U., Z.Z., and L.C. contributed new reagents/ rately mimicked in nonneuronal mammalian cells (18, 23–25), as analytic tools; P.H., M.H.U., H.-L.W., R.J.A., A.G.L., and Z.Z. analyzed data; and E.Y.I. and well as Xenopus oocytes (13, 14, 17, 26, 27), suggesting that the L.C. wrote the paper. basic mechanisms for TARP/AMPA-R interaction and TARP- The authors declare no conflict of interest. mediated regulation of AMPA-R trafficking are preserved in *This Direct Submission article had a prearranged editor. nonneuronal cells. 1P.H. and M.H.U. contributed equally to this work. Although much is known about the interaction domains on the 2To whom correspondence may be addressed. E-mail: [email protected] or ehud@ classical TARPs (γ-2, γ-3, γ-4, and γ-8) and AMPA-Rs that me- berkeley.edu. diate assembly and modulation (13, 23, 24, 27, 28), far less is known This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. about the stoichiometry of the TARP/AMPA-R complexes. The 1073/pnas.1218765110/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1218765110 PNAS | March 26, 2013 | vol. 110 | no. 13 | 5163–5168 Downloaded by guest on September 25, 2021 chosen expression system for the study, by applying it to the ABimmobile stoichiometry of homotetrameric AMPA-Rs. mobile Our single-molecule technique is based on the photobleaching 200 co-localizing

of GFP tags fused to the protein of interest (31). This technique eve 100 involves counting irreversible steps of photobleaching of GFP in Stg + GluA1 #nts 0 areas of membrane in intact oocytes where the protein of interest 0 1234567 is expressed at a low enough density that individual complexes can fl be resolved as single uorescent spots. Under such conditions, the GluA1 alone Stg alone 200 number of bleaching steps reflects the number of GFP tags and e v 100 Stg alone thus the number of protein subunits. The high sensitivity and low e fl #nts 0 background necessary for the observation of single uorescent 0 1234567 proteins were achieved by using TIRF microscopy, where a laser beam is reflected at the coverslip/sample interface and excitation 100 is restricted to the plasma membrane of the cell. 50 We fused GFP to the C terminus of GluA1 (GluA1-GFP) and Stg + GluK2 Stg + GluA1 Stg + GluK2 # events fi 0 con rmed the functionality of the GluA1-GFP construct by two- 0 1234567 electrode voltage clamping. At 12–24 h after injection of 25 ng of distance from origin (pixels) RNA encoding GluA1-GFP per cell, we mechanically removed the CD vitelline membrane of several cells and placed them on a coverslip. co-localizing TARP alone Upon illumination with 488-nm laser light in TIRF, single mole- 70 mobile ) TARP + GluA1 cules of GluA1-GFP appeared as bright immobile fluorescent spots 60 (no 70

%(n 60 50 on a dark background (Fig. 1A). The immobility of the AMPA-Rs is i 50 40 reminiscent of what is seen in cyclic nucleotide-gated channels and oi 40

30 fract % NMDA receptors and may reﬂect tethering to the cytoskeleleton car 30 20 li 20 through C-terminal protein-binding motifs (31, 32). Typically, the f) 10 10 ﬂ

uorescence intensity decreased in several discrete steps, indicating 0 mob e 0

B 0.1

1.0

2.5

gt photobleaching of the individual GFP tags (Fig. 1 ). K

0.75

0.25 We counted the bleaching steps from a total of 568 spots in 14 ul

x Stg + 25 GluA1 0 movies from different cells injected with GluA1-GFP. Most of the 5

22 spots had either two or three bleaching steps, with a minority 0.25 S + bleaching in one or four steps (Fig. 1C, red bars). Rarely, we ob- Fig. 2. Movement of TARPs with and without coexpression of GluA1 and served ﬁve bleaching steps, which could be accounted for by the GluK2. (A) GluA1-GFP was immobile, and γ-2–GFP [γ-2 = stargazin (Stg)] was rare instance of two receptors within a distance below the dif- mobile. Binding to GluA1 immobilized Stg. GluK2 does not bind Stg and did fraction limit. As we had already shown in earlier work, the not immobilize Stg. The red crosses in the lower panels mark the AMPA-R positions. (Scale bar, 250 nm.) (B) Histogram of distances from initial positions that Stg-GFP molecules travel until they photobleach, alone (n = 1,192 spots), or coexpressed with GluA1 (n = 584) or GluK2 (n = 457). Mobile fraction A B GluA1-GFP shaded light, fraction colocalizing with AMPA-Rs shaded dark. (C) Fractions of 10 mobile spots and spots colocalizing with AMPA-Rs for different amounts of 8 RNA injected (all values in nanograms) for Stg alone or with GluA1 or GluK2 (n = 6 6–11 movies per condition). (D) Mobile fractions of four TARPS (γ-2, γ-3, γ-4, 4 and γ-8) and γ-1, which does not bind to GluA1, alone or coexpressed with 2 GluA1 (n = 3–8 movies per condition). All error bars indicate SEM. 0

).u.a( 0 5 10 15 20 25 10 occurrence of spots with fewer than four bleaching steps can be 8 accounted for by 20% of the GFP tags being nonﬂuorescent (31).

ytisnetni 6 4 The observed distribution of bleaching steps for GluA1-GFP 2 closely resembles the predicted binomial distribution for homo- 0 tetramers with an 80% probability of the GFP being fluorescent 0 5 10 15 20 25 (Fig. 1C, blue bars). 6 4 TARP Mobility in the Absence and Presence of AMPA-Rs. We began C with an examination of the first TARP identified, Stargazin (γ-2). 2 200 To visualize γ-2 behavior in the membrane, we expressed the GFP-

orescence 0 γ γ –

sto tagged -2 ( -2 GFP) alone and imaged at high speed using TIRF

ulF 0 5 10 15 20 25 150 microscopy. In contrast to the immobile AMPA-Rs, a large frac-

p 3 γ – A s# 100 tion of the -2 GFP moved laterally in the membrane (Fig. 2 ). 2 We predicted that TARPs may become immobile once they bind 50 1 to the AMPA-Rs, because these are immobile on their own. To test this prediction, we coexpressed γ-2–GFP and GluA1 tagged 0 0 fl 12345 0 5 10 15 20 25 with the red uorescent protein mCherry (GluA1-mCherry) and Bleaching steps sequentially imaged first the red fluorescence from the immobile Time (s) GluA1-mCherry to identify the location of AMPA-Rs, and then the green fluorescence from γ-2–GFP. We obtained the trajecto- Fig. 1. Single-molecule subunit counting of GFP-tagged GluA1. (A) Single γ – A molecules of GluA1-GFP in a Xenopus oocyte membrane patch appear as ries of the -2 GFP spots (Fig. 2 ) using an automated tracking bright spots under 488-nm illumination. The circles mark spots used in program and determined for each individual spot the maximum bleaching steps statistics. (Scale bar, 2 μm.) (B) Intensity from example spots displacement from its starting position, which we used as a crite- with four, three, two, and one bleaching steps. The green arrows mark rion for mobility, and the distance to the closest GluA1-mCherry fluorescence intensity levels. (C) Histogram of bleaching steps for GluA1-GFP spot, which defined whether or not the γ-2–GFP was colocalized (red) and fit with 64% probability of GFP to be fluorescent (blue) (a total of with the GluA1-mCherry. When the two proteins were coex- 568 spots from 14 experiments was analyzed). pressed, a large fraction of γ-2–GFP spots colocalized with the

5164 | www.pnas.org/cgi/doi/10.1073/pnas.1218765110 Hastie et al. Downloaded by guest on September 25, 2021 GluA1-mCherry spots, and the fraction of mobile γ-2–GFP spots that GFP-γ-2, γ-3, and γ-4 are suitable for the stoichiometry decreased (Fig. 2 A and B). In contrast, a kainate receptor subunit analysis, but that, although we observed clear colocalization of GluK2 (tagged with mCherry), which is structurally similar to γ-8–GFP and GluA1-mcherry on the surface membrane of AMPA-Rs, but does not interact with γ-2 (26), did not reduce the oocytes, we could not rule out the possibility that the interference movement of γ-2–GFP or colocalize with γ-2–GFP molecules (Fig. with function by the GFP tag was accompanied by (or even due to) 2 A and B). an alteration in stoichiometry. For this reason, although we could To maximize counting of γ-2–GFP that are associated with determine the stoichiometry of γ-8 (Fig. S2), we left it out of our AMPA-Rs, we first determined the expression conditions at main analysis and interpretation and focused on γ-2, γ-3, and γ-4. which most of the γ-2–GFP becomes immobilized by varying the To determine whether the fluorescent tags on GluA1 disturb the ratio of GluA1-mCherry to γ-2–GFP expression. We changed the TARP–GluA1 interaction, we repeated a subset of the experi- amount of injected γ-2–GFP RNA between 0.1 and 2.5 ng per ments with untagged GluA1. In the first experiment, we performed cell while keeping the amount of GluA1-mCherry RNA constant single-molecule subunit counting on oocytes in which we coex- at 25 ng. At injection levels of 0.25 ng or less of γ-2–GFP RNA, pressed GFP-labeled γ-2 with untagged GluA1 (Fig. S3A). Similar the fraction of immobile γ-2–GFP molecules was above 90%. to the previous results from oocytes coexpressing GluA1-mCherry With increasing amounts of γ-2–GFP RNA in the injection mix, and γ-2–GFP, we observed up to four GFP bleaching steps from the fraction of mobile γ-2–GFP spots increased and the fraction the immobile fluorescent spots, suggesting that up to four of γ-2–GFP spots that colocalized with GluA1-mCherry de- γ-2–GFPs can assemble with an untagged GluA1 receptor. When creased (Fig. 2C). we coexpressed untagged GluA1 with γ-8–GFP, most of the im- We next extended our analysis of colocalization and immobili- mobile fluorescent spots had one bleaching step and up to 25% of zation to other TARPs that are known to interact with and mod- spots had two bleaching steps (Fig. S3B). Very few spots had three ulate the functions of AMPA-Rs (17, 19, 20, 33). We chose γ-3, γ-4, bleaching steps, and none had more than three. The near absence and γ-8, which are functionally similar to γ-2. We also included γ-1 of spots with more than two bleaching steps suggests that a max- as a negative control because it does not interact with GluA1 and imum of two γ-8–GFPs can assemble with an untagged GluA1 bears only low homology to γ-2 (16, 26, 34). Based on our obser- receptor, as observed in the experiments with GluA1-mCherry. vation that a large excess (>100:1) of GluA1-mCherry RNA over The only difference between the experiments with the tagged γ-2–GFP RNA resulted in an almost complete immobilization of versus the untagged GluA1 was a higher occurrence of spots with γ-2–GFP, we injected GFP-tagged γ-1, γ-3, γ-4, and γ-8, either only one bleaching step in the untagged GluA1. This can be ex- alone or together with at least 100-fold excess of GluA1-mCherry. plained by γ-8–GFP monomers that are not associated with GluA1 γ – Similar to what we observed with -2 GFP, the fraction of mobile but are immobile. In this experiment with untagged GluA1, these NEUROSCIENCE γ-3–GFP, γ-4–GFP, or γ-8–GFP spots strongly decreased when free γ-8–GFPs cannot be distinguished from the GluA1-associated GluA1-mCherry was coexpressed. In contrast, γ-1–GFP was not γ-8–GFPs because of the lack of a fluorescence tag on GluA1. In immobilized by GluR1-mCherry (Fig. 2D). contrast, in the experiments with GluA1-mCherry, free γ-8–GFPs Taken together, these results support previous findings of that were not associated with GluA1 were excluded from the sta- specific interactions between the classical TARPs (γ-2, γ-3, γ-4, tistics because they did not colocalize with the red fluorescence and γ-8) and the GluA1 subunit of AMPA-Rs (13, 17, 26, 27, 33) from GluA1-mCherry. and show that these interactions, which are strong enough to We next examined coimmunoprecipitation efficiency between permit biochemical copurification (11), are stable enough in the GFP-tagged TARPs (γ-2 or γ-8) and GluA1 and asked whether SCIENCES

membranes of live cells to persist for at least tens of seconds. the C-terminal mCherry tag on GluA1 could affect the interac- APPLIED PHYSICAL tion (Fig. S4). The pull-down efficiency for mCherry-GluA1 was Counting Bleaching Steps of TARPs Bound to AMPA-R. Previous 0.62 ± 0.12 for γ-2 (n = 4 independent experiments) and 0.72 ± studies have shown that TARPs directly interact with AMPA-Rs 0.37 for γ-8 (n = 4 independent experiments) when normalized (11, 13, 27) and regulate both their localization and gating (14–20, to untagged GluRA1. This modestly lower efficiency of coim- 22). The stoichiometry of TARP/AMPA-R complexes was re- munoprecipitation fell within the range of variability for pull- cently deduced indirectly from functional population assays to be down efficiency of GluA1 by the GluA1 antibody (1.00 ± 0.33, a maximum of four TARPs per AMPA-R complex (30). Having n = 4fortheγ-2 experiments; 1.00 ± 0.37, n = 4fortheγ-8 established that TARPs remain bound to immobile AMPA-Rs for experiments), as well as the range of variability of pull-down the duration of imaging, we decided to directly count the number efficiency of mCherry-GluA1 by the GluA1 antibody (1.00 ± of TARP subunits present at individual AMPA-R complex using 0.17, n = 4fortheγ-2 experiments; 1.00 ± 0.23, n = 4fortheγ-8 our single-molecule photobleaching assay. experiments), all calculated from the data from the same ex- The AMPA-R/TARP stoichiometry deduced from our single- periments (P > 0.5). molecule photobleaching assay hinges on one important as- Having established that mCherry-tagging on GluA1 and GFP- sumption—the GFP tag on the TARPs does not interfere with its tagging on TARPs (with the exception of γ-8) do not seem to in- function. In addition, we also prefer to have GluA1 tagged with terfere with their function, we proceeded to the single-molecule mCherry to count the photobleaching steps of only the GFP- photobleaching experiments. GFP-tagged TARPs were coex- TARPs associated with AMPA-Rs. Therefore, we next evaluated pressed with GluA1-mCherry, and green TARP-GFP spots that whether tagging alters the properties of the interaction between were colocalized with red GluA1-mCherry spots were analyzed TARPs and GluA1. (Fig. 3 A and B). For γ-2 and γ-3, we mainly observed spots with We first evaluated whether tagging alters the modulatory effect one or two bleaching steps at low TARP expression, and as the of TARPs on the function of GluA1. Untagged GluA1 cRNAs amount of γ-2 or γ-3 RNA increased (with the amount of GluA1- were injected into oocytes either alone or together with cRNAs mCherry RNA held constant), the distribution of bleaching steps encoding either the native untagged TARPs or the GFP-tagged shifted toward three and four bleaching steps (Fig. 3C). The be- TARPs, and glutamate-evoked currents were measured 1 d after havior of γ-4 was very different from that of γ-2 and γ-3. At low injection using two-electrode voltage-clamp recording. We tested TARP expression levels, most γ-4 spots had one bleaching step, the untagged and tagged versions of all four TARPs. As shown a few with two steps, and even fewer with three bleaching steps earlier (26, 33), in all of these cases, the expression of untagged (Fig. 3C). At higher γ-4 expression, the fraction of spots with two version of TARPs significantly increased the glutamate-evoked steps increased to almost 30%, but the occurrence of spots with currents (Fig. S1). However, whereas the GFP-tagged γ-2, γ-3, three or four bleaching steps stayed at a very low level (<2%). This and γ-4 enhanced the glutamate-evoked currents to a similar observation of low bleaching step numbers was in stark contrast to degree as did the untagged γ-2, γ-3, and γ-4, the GFP-tagged γ-8 γ-2 and γ-3, where the shift toward higher occurrence of two failed to significantly increase glutamate-evoked current, in- bleaching steps was always accompanied by an increase of events dicating compromised function (Fig. S1). Thus, we are confident with three or four bleaching steps.

Hastie et al. PNAS | March 26, 2013 | vol. 110 | no. 13 | 5165 Downloaded by guest on September 25, 2021 A C 100 0.10ng 0.15ng 80 0.25ng 0.75ng 60 1.00ng 40 2.50ng

fraction (%) 20 0 12345

GluA1-mCh Stg-GFP 100 0.05ng B 0.10ng 80 0.15ng 6 0.25ng 60 Fig. 3. Bleaching steps of GFP-labeled TARPs 0.50ng 4 bound to AMPA-R. (A) Overlay of red image with 40 1.25ng GluA1-mCherry spots and green image of Stg-GFP

2 fraction (%) 20 spots (Left) and Stg-GFP image with circles showing 0 spots with one to four bleaching steps (Right). (a.u.) 0 0 10 20 30 40 12345 (Scale bar, 2 μm.) (B) Examples of intensity traces from GluA1-mCherry plus Stg-GFP with four (Up- GluA1-mCh Stg-GFP 100 0.10ng per) and three (Lower) GFP bleaching steps. The red 6 80 0.25ng bar marks illumination with 593 nm (excites

Intensity 0.75ng mCherry), and the green bar marks illumination 4 60 2.50ng with 488 nm (excites GFP). The green arrows mark 4.00ng ﬂuorescence intensity levels. (C) Distributions of 2 40

fraction (%) bleaching steps from three TARPs for concen- 20 0 trations as indicated (n = 2–7 movies per condition 0 10 20 30 40 0 with 292 ± 29 spots each, except for 0.5 ng of γ-3 12345 bleaching steps with only 1 movie and 148 spots). All error bars Time (s) indicate SEM.

Correction for Undercounting Due to Nonfluorescent GFP. To de- model gives a better fit; for γ-4, one cannot distinguish between termine the actual numbers of TARP subunits bound to the the models. AMPA-R from the distribution of bleaching steps, we needed to correct for the underestimation of the numbers of GFPs present Dependence of Occupancy on TARP:AMPA-R Ratio. To assess the in each complex due to the 20% of GFP tags that are typically dependence of occupancy on TARP expression, we calculated p nonfluorescent (31, 32, 35). We corrected for this undercounting across the series of experiments that used different levels of to obtain the distribution of TARP subunits for each of the ex- TARP RNA. Occupancy was plotted semilogarithmically against pression levels (Fig. S5). An examination of these distributions the ratio of the number of TARP subunits to the number of showed that γ-2 and γ-3 reached four TARP subunits per AMPA-Rs in the field of view (Fig. 4D and Fig. S7)(Materials AMPA-R, consistent with the receptor having four identical and Methods). Under the assumption that there are four binding γ γ GluA1 subunits that provide four TARP binding sites. However, sites for each of the TARPs, -2 and -3 were seen to increase γ γ monotonically toward an individual binding-site occupancy of although four -2 and -3 subunits were often found, the majority γ of observations were of three or fewer per receptor. The ten- 1.0, whereas -4 only slowly increased and did not rise above an dency to have a submaximal number of TARPs per complex was individual binding-site occupancy of 0.4. In contrast, under the γ assumption of a two–binding-site model, the binding curve of γ-4 much more pronounced for -4, which rarely had three TARP γ γ subunits per receptor and almost never had four. increased at a similar steepness to what was seen for -2 and -3, and reached a maximum occupancy of 0.64. Occupancy of TARP Binding Sites at the AMPA-R. Having seen that Discussion the number of TARPs per complex often was less than four, we set out to calculate TARP occupancy, i.e., the fraction of the recep- We used a direct subunit-counting approach to explore TARP/ tor’s binding sites that was occupied. For illustration of the dif- AMPA-R stoichiometry under conditions in which TARPs were free to associate with GluA1. By imaging coexpressed GFP-TARP ferent behavior of the four TARPs, we use the highest TARP subunits and mCherry-GluA1 subunits via TIRF microscopy, we expression levels from the corrected count distributions in Fig. S1. – fi were able to monitor the TARP GluA1 interaction via the im- By calculating the least-squares t of the TARP subunit dis- mobilization of otherwise highly mobile TARPs at sites of GluA1. tributions to a binomial distribution assuming four possible bind- p fi Our observation of TARP mobility suggests that TARPs do not ing sites, we determined the occupancy (Fig. 4; ts for all interact with scaffolding or cytoskeletal proteins in oocytes, but expression conditions in Fig. S6). does not say how mobile TARPs would be in neurons where For γ-2 and γ-3, the fits closely matched the observed dis- A B TARPs can interact with postsynaptic density-95 (PSD95) (36). tributions (Fig. 4 and ) and gave estimates of high occupancy The stability of these interactions and TARP mobility has not been p = p = ( γ-2 0.77 and γ-3 0.74) for the highest RNA injections. We determined. AMPA-R mobility has been studied within synapses fi γ C also obtained a good t for -4 (Fig. 4 ), but the occupancy was using single-molecule tracking (over much longer time periods p = fi much lower ( γ-4 0.33). We also performed a least-square tof than our 20-s bouts of observation) and found to be complex, with each TARP distribution by assuming two binding sites per re- highly mobile and less mobile pools of AMPA-Rs that are subject ceptor. The sum of the residuals representing the discrepancy to interchange and whose presence could explain at least some between the data and the estimates from the fit for γ-4 was about aspects of recovery from desensitization experiments (37–39). The equal for the two– and four–binding-site models, but for γ-2 and immobilization of AMPA-Rs appears likely to reflect several γ-3 the sum of the residuals was approximately sevenfold and parameters, including anchoring of the AMPA-Rs and TARPs by approximately fourfold larger for the model with two binding PSD-95/Discs large/zona occludens-1 (PDZ) domain-containing sites, respectively. Thus, for γ-2 and γ-3, the four–binding-site proteins [e.g., to glutamate receptor-interacting protein (GRIP)

5166 | www.pnas.org/cgi/doi/10.1073/pnas.1218765110 Hastie et al. Downloaded by guest on September 25, 2021 ABfour identical subunits provides four γ-2 or γ-3 TARP association sites. 120 80 By contrast, we found that the number of TARPs per receptor 100 p=0.77 60 p = 0.74 rarely exceeded two for the γ-4 isoform, and that the slope of the 80 dependence of occupancy on the ratio of TARP-to-GluA1 ex- 40 60 pression was considerably lower for γ-4 than it was for γ-2 and 40 # events # events 20 γ 20 -3. The frequency distributions still followed a binomial func- 0 0 tion with four possible binding sites, but could be equally well 123 4 123 4 ﬁtted by a more parsimonious two–binding-site model. The be- bound TARP subunits bound TARP subunits havior of γ-4 thus signiﬁcantly deviated from that of γ-2 and γ-3, suggesting a fundamentally different interaction mode. C experiment fit One concern about the unexpected behavior of γ-4 is that GFP 60 tagging may affect its function. However, we think this is highly γ γ 50 p = 0.33 unlikely. Previous studies (29, 30, 40) had shown for -2 and -8 40 that there is a dose-dependent effect of TARPs on AMPA-R 30 properties, including the amplitude of glutamate-evoked respon-

# events 20 ses—i.e., the size of glutamate-evoked responses is signiﬁcantly 10 bigger in four-TARP complex than in two-TARP complexes. We 0 123 4 have shown that the tagged γ-4 enhances the glutamate-evoked bound TARP subunits response ∼10-fold, similar to the effect of wild-type, untagged γ-4, indicating that tagging does not interfere with either the normal D association of γ-4 with AMPA-Rs or the modulation of AMPA-Rs by γ-4. It is worth pointing out, moreover, that the ∼10-fold boost 0.8 we see with tagged γ-4 is similar to values reported in neurons (33). Thus, our electrophysiological assessment provides strong evi- 0.6 dence for a normal function of GFP-tagged γ-4, permitting us to conclude that the 2:1 stoichiometry is likely to be correct. 0.4 4 binding In hippocampal pyramidal neurons, the pharmacological proﬁle

Occupancy fi sites of AMPA-Rs (measured as relative ef cacy of kainite and gluta- NEUROSCIENCE 0.2 2 binding mate as agonists) most closely resembles that of a receptor complex sites formed by the heterologous expression of a fusion protein in which 0.0 the γ-8 subunit is attached to the GluA1 subunit—a situation that is 0.5 1 2 4 expected to force a stoichiometry of four γ-8 subunits per tetrameric TARP : AMPA-R ratio receptor (30). This seemingly straightforward conclusion may be complicated by several issues. First, AMPA-Rs interact with other Fig. 4. Fit of binding-site occupancy. The probability of each TARP binding proteins, including cytosine-knot AMPA-R–modulating protein 44 fi SCIENCES site of the AMPA-R to be occupied by a TARP had been tted (blue) to the (CKAMP44) and cornichons (41, 42); and both AMPA-Rs and their distribution of bound TARP subunits (red) as calculated from the experi- = – auxiliary subunits, including TARPs, are subject to posttranslational APPLIED PHYSICAL ments with the highest amounts of injected RNA (n 4 7 per condition). modification—factors that could affect the relative efficacy of kai- Lower concentrations are in Fig. S1.(A) γ-2 and (B) γ-3 have high occupancy around 0.8, whereas (C) γ-4 has a lower occupancy around 0.3. Four equiv- nate vs. glutamate in neurons. Second, it was unclear whether the alent binding sites were assumed. (D) Occupancy p as a function of TARP- actions of TARPs are equivalent for tethered and nontethered to-AMPA-R ratio for all three TARPs using the model with four binding sites conditions at each of the stoichiometries. Third, the macroscopic (solid lines) and for γ-4 with the two–binding-site model (dashed lines). analysis of glutamate-evoked responses could not determine whether the summed current reflected pure populations of AMPA- Rs with a single stoichiometry, either in HEK cells where the fusion and PSD95, respectively], and protein crowding. In this sense, the may not be 100% preserved because of protease activities, or in oocyte may recapitulate only one aspect of the complex immobi- neurons where receptors may be loaded with different numbers of lization of AMPA-Rs produced by their anchoring, possibly to TARPs. This does not diminish the importance of those studies, the cytoskeleton. which were the first to address the question of TARP/AMPA-R By counting steps of irreversible photobleaching of GFP- stoichiometry, but it prevented them from being definitive. Our labeled TARPs in single GluA1 receptor complexes, we de- study addresses the same question with a completely different ap- termined the distribution of TARP subunits at hundreds of proach that avoids these pitfalls and has the benefit of determining individual GluA1 receptors on the cell surface. Counting of bleach- association by three distinct assays (immobilization, colocalization, ing steps was done across a wide range of ratios of TARP-to- and single-molecule counting), each of which reveals the properties GluA1 expression, allowing us to observe the saturating level of dozens of individual complexes. Our results both confirm earlier of binding sites in the GluA1 receptor complex by TARPs. interpretation and provide an unexpected observation, namely that We examined four different TARPs: γ-2, γ-3, γ-4, and γ-8 using TARP/AMPA-R associations may differ among TARPs. Although single-molecule imaging. Three of these TARPs—γ-2, γ-3, and γ-2 and γ-3 follow the established 4:1 stoichiometry, γ-4 does not, γ-4—provided interpretable results because we could show in preferring 2:1, suggesting that not all TARPs are created equal. control experiments that the GFP-tagged versions of these Although our results suggest that γ-8 may be compromised by TARPs were fully capable of normally regulating AMPA-Rs. GFP tagging, the stoichiometry of the γ-8/AMPA-R complex we Consistent with earlier work (33), each of these TARPs asso- observed is similar to that of the γ-4/AMPA-R complex, suggesting ciated with GluA1. The TARP–GluA1 associations were stable that a 2:1 TARP/AMPAR-R stoichiometry may be shared by both for the duration of our experiments (up to 40 s), consistent TARP isoforms. Directly tethering γ-8 to the receptor subunit with with high affinity binding. Strikingly, we found different max- a short linker creates a high γ-8 (perhaps unnaturally high) density imal occupancies for thedifferentTARPs.Forγ-2 and γ-3, we that could allow a third and fourth γ-8 subunit to be added to the counted up to four TARPs bound to each GluA1 receptor, receptor complex. Alternatively, association of AMPA-Rs with four which agrees with previous studies using neuronal tissues γ-8 subunits may occur during association of free receptor with free (29, 30) and, importantly, validates our approach. Our analysis TARPs in certain locations in the cell (i.e., synapses) where the γ-8 of the relationship of occupancy to the TARP-to-GluA1 ratio subunit could cluster due to interactions with scaffold proteins. was consistent with the expectation that a receptor made of Such densities are higher than those that could be tested in our

Hastie et al. PNAS | March 26, 2013 | vol. 110 | no. 13 | 5167 Downloaded by guest on September 25, 2021 experiments, where individual protein complexes need to be The single-molecule approach presented here provides a more spatially resolved. detailed view of the interactions between TARPs and AMPA-Rs Although the molecular basis of the differences in binding for the than is possible to obtain from ensemble measurements of the different TARPs remains to be elucidated, two possible explan- average readout of many AMPA-Rs. With the future development ations come to mind, which can be addressed in future studies. of more photostable red and blue fluorescent proteins that can be First, the receptor may have four equal TARP binding sites, but for used for single-molecule experiments, it should become possible to some TARPs, docking of one TARP may sterically hinder binding determine the subunit composition for two interacting partners at of another TARP molecule at a neighboring site, so that for ex- fi the same time. This will open the way for elucidating complexes ample only two TARPs may bind with high af nity at diagonally formed by mixtures of GluA1 and GluA2 receptor subunits with situated positions on the receptor. Allosteric effects of TARP TARPs, and the interaction between TARPs and other newly binding on receptor conformation could have the same result – without there being a direct steric interference. Alternatively, discovered AMPA-R interacting proteins such as cornichons (41) AMPA-Rs may not have four equal docking sites. Indeed, the re- and CKAMP44 (42). cent crystal structure of a homotetramer of GluA2 provided clues Materials and Methods to such a scenario. Whereas the membrane spanning portion of the receptor exhibits a fourfold symmetry, the extracellular domain Microscopy was performed as described in ref. 31. All microscopy data were breaks this symmetry by pairing ligand binding domains into a di- processed using custom MATLAB (Mathworks), Labview, or Mathematica mer of dimers and by complex domain swapping between subunits software. Other related experimental and analysis procedures are described (43). Interestingly, the differences in kainate efficacy between the in SI Materials and Methods. four TARPs depend on the first extracellular domain between transmembrane segments 1 and 2, where γ-4 differs from γ-2 and ACKNOWLEDGMENTS. We thank Sarah Bell for help with two-electrode γ-3 in possessing an additional proline-rich motif (19). Because the voltage clamp. The work was supported by National Institutes of Health Grants 1R01MH091193 and 1P50MH086403 (to L.C.), and R01NS35549 and extracellular domains of tetrameric AMPA-Rs exhibit a two-by-two 2PN2EY018241 (to E.Y.I.), by the Excellence Initiative of the German Federal symmetry (43), it is possible that there are only two binding sites for and State Governments (EXC 294) (M.H.U.), and by National Basic Research γ-4 or two kinds of binding sites with differing affinities. Program of China (973 Program) Grant 2012CB525003.

1. Hollmann M, Maron C, Heinemann S (1994) N-glycosylation site tagging suggests 23. Turetsky D, Garringer E, Patneau DK (2005) Stargazin modulates native AMPA re- a three transmembrane domain topology for the glutamate receptor GluR1. Neuron ceptor functional properties by two distinct mechanisms. J Neurosci 25(32): 13(6):1331–1343. 7438–7448. 2. Keinänen K, et al. (1990) A family of AMPA-selective glutamate receptors. Science 24. Bedoukian MA, Weeks AM, Partin KM (2006) Different domains of the AMPA re- – 249(4968):556 560. ceptor direct stargazin-mediated trafficking and stargazin-mediated modulation of 3. Mano I, Teichberg VI (1998) A tetrameric subunit stoichiometry for a glutamate re- kinetics. J Biol Chem 281(33):23908–23921. – ceptor-channel complex. Neuroreport 9(2):327 331. 25. Soto D, Coombs ID, Kelly L, Farrant M, Cull-Candy SG (2007) Stargazin attenuates 4. Rosenmund C, Stern-Bach Y, Stevens CF (1998) The tetrameric structure of a gluta- intracellular polyamine block of calcium-permeable AMPA receptors. Nat Neurosci mate receptor channel. Science 280(5369):1596–1599. 10(10):1260–1267. 5. Chen L, Tracy T, Nam CI (2007) Dynamics of postsynaptic glutamate receptor target- 26. Chen L, El-Husseini A, Tomita S, Bredt DS, Nicoll RA (2003) Stargazin differentially ing. Curr Opin Neurobiol 17(1):53–58. controls the trafficking of alpha-amino-3-hydroxyl-5-methyl-4-isoxazolepropionate 6. Elias GM, Nicoll RA (2007) Synaptic trafficking of glutamate receptors by MAGUK and kainate receptors. Mol Pharmacol 64(3):703–706. scaffolding proteins. Trends Cell Biol 17(7):343–352. 27. Tomita S, Fukata M, Nicoll RA, Bredt DS (2004) Dynamic interaction of stargazin-like 7. Greger IH, Esteban JA (2007) AMPA receptor biogenesis and trafficking. Curr Opin – Neurobiol 17(3):289–297. TARPs with cycling AMPA receptors at synapses. Science 303(5663):1508 1511. 8. Ziff EB (2007) TARPs and the AMPA receptor trafficking paradox. Neuron 53(5): 28. Tomita S, Shenoy A, Fukata Y, Nicoll RA, Bredt DS (2007) Stargazin interacts func- 627–633. tionally with the AMPA receptor glutamate-binding module. Neuropharmacology – 9. Derkach VA, Oh MC, Guire ES, Soderling TR (2007) Regulatory mechanisms of AMPA 52(1):87 91. receptors in synaptic plasticity. Nat Rev Neurosci 8(2):101–113. 29. Kim KS, Yan D, Tomita S (2010) Assembly and stoichiometry of the AMPA receptor 10. Nakagawa T, Cheng Y, Sheng M, Walz T (2006) Three-dimensional structure of an and transmembrane AMPA receptor regulatory protein complex. J Neurosci 30(3): AMPA receptor without associated stargazin/TARP proteins. Biol Chem 387(2): 1064–1072. 179–187. 30. Shi Y, Lu W, Milstein AD, Nicoll RA (2009) The stoichiometry of AMPA receptors and 11. Vandenberghe W, Nicoll RA, Bredt DS (2005) Stargazin is an AMPA receptor auxiliary TARPs varies by neuronal cell type. Neuron 62(5):633–640. subunit. Proc Natl Acad Sci USA 102(2):485–490. 31. Ulbrich MH, Isacoff EY (2007) Subunit counting in membrane-bound proteins. Nat 12. Fukata Y, et al. (2005) Molecular constituents of neuronal AMPA receptors. J Cell Biol Methods 4(4):319–321. 169(3):399–404. 32. Ulbrich MH, Isacoff EY (2008) Rules of engagement for NMDA receptor subunits. Proc 13. Tomita S, et al. (2005) Stargazin modulates AMPA receptor gating and trafficking by Natl Acad Sci USA 105(37):14163–14168. distinct domains. Nature 435(7045):1052–1058. 33. Tomita S, et al. (2003) Functional studies and distribution define a family of trans- 14. Priel A, et al. (2005) Stargazin reduces desensitization and slows deactivation of the membrane AMPA receptor regulatory proteins. J Cell Biol 161(4):805–816. – 2+ AMPA-type glutamate receptors. J Neurosci 25(10):2682 2686. 34. Letts VA, et al. (1998) The mouse stargazer gene encodes a neuronal Ca -channel 15. Yamazaki M, et al. (2004) A novel action of stargazin as an enhancer of AMPA re- gamma subunit. Nat Genet 19(4):340–347. – ceptor activity. Neurosci Res 50(4):369 374. 35. Yu Y, et al. (2009) Structural and molecular basis of the assembly of the TRPP2/PKD1 16. Chen L, et al. (2000) Stargazin regulates synaptic targeting of AMPA receptors by two complex. Proc Natl Acad Sci USA 106(28):11558–11563. – distinct mechanisms. Nature 408(6815):936 943. 36. Schnell E, et al. (2002) Direct interactions between PSD-95 and stargazin control 17. Kott S, Werner M, Körber C, Hollmann M (2007) Electrophysiological properties of synaptic AMPA receptor number. Proc Natl Acad Sci USA 99(21):13902–13907. AMPA receptors are differentially modulated depending on the associated member 37. Borgdorff AJ, Choquet D (2002) Regulation of AMPA receptor lateral movements. of the TARP family. J Neurosci 27(14):3780–3789. Nature 417(6889):649–653. 18. Körber C, Werner M, Kott S, Ma ZL, Hollmann M (2007) The transmembrane AMPA 38. Groc L, et al. (2004) Differential activity-dependent regulation of the lateral mobilities receptor regulatory protein gamma 4 is a more effective modulator of AMPA re- of AMPA and NMDA receptors. Nat Neurosci 7(7):695–696. ceptor function than stargazin (gamma 2). J Neurosci 27(31):8442–8447. 39. Tardin C, Cognet L, Bats C, Lounis B, Choquet D (2003) Direct imaging of lateral 19. Cho CH, St-Gelais F, Zhang W, Tomita S, Howe JR (2007) Two families of TARP iso- – forms that have distinct effects on the kinetic properties of AMPA receptors and movements of AMPA receptors inside synapses. EMBO J 22(18):4656 4665. synaptic currents. Neuron 55(6):890–904. 40. Gill MB, et al. (2011) Cornichon-2 modulates AMPA receptor-transmembrane AMPA 20. Milstein AD, Zhou W, Karimzadegan S, Bredt DS, Nicoll RA (2007) TARP subtypes receptor regulatory protein assembly to dictate gating and pharmacology. J Neurosci – differentially and dose-dependently control synaptic AMPA receptor gating. Neuron 31(18):6928 6938. 55(6):905–918. 41. Schwenk J, et al. (2009) Functional proteomics identify cornichon proteins as auxiliary 21. Menuz K, Stroud RM, Nicoll RA, Hays FA (2007) TARP auxiliary subunits switch AMPA subunits of AMPA receptors. Science 323(5919):1313–1319. receptor antagonists into partial agonists. Science 318(5851):815–817. 42. von Engelhardt J, et al. (2010) CKAMP44: A brain-specific protein attenuating short- 22. Mi R, et al. (2004) AMPA receptor-dependent clustering of synaptic NMDA receptors term synaptic plasticity in the dentate gyrus. Science 327(5972):1518–1522. is mediated by Stargazin and NR2A/B in spinal neurons and hippocampal interneur- 43. Sobolevsky AI, Rosconi MP, Gouaux E (2009) X-ray structure, symmetry and mecha- ons. Neuron 44(2):335–349. nism of an AMPA-subtype glutamate receptor. Nature 462(7274):745–756.

5168 | www.pnas.org/cgi/doi/10.1073/pnas.1218765110 Hastie et al. Downloaded by guest on September 25, 2021