Evidence for a Second, High Affinity G�� Binding Site onG�i1(GDP) Subunits*□S

Received for publication, December 29, 2008, and in revised form, April 9, 2009 Published, JBC Papers in Press, April 15, 2009, DOI 10.1074/jbc.M109.006585

Jingting Wang, Parijat Sengupta1, Yuanjian Guo, Urszula Golebiewska2, and Suzanne Scarlata3

From the Department of Physiology and Biophysics, Stony Brook University, Stony Brook, New York 11794-8661

It is well known that G�i1(GDP) binds strongly to G�� sub-units to form the G�i1(GDP)-G�� heterotrimer, and thatactivation to G�i1(GTP) results in conformational changesthat reduces its affinity for G�� subunits. Previous studies ofG protein subunit interactions have used stoichiometricamounts of the proteins. Here, we have found that G�i1(GDP)can bind a second G�� subunit with an affinity only 10-foldweaker than the primary site and close to the affinity betweenactivated G�i1 and G�� subunits. Also, we find that phospho-lipase C�2, an effector of G��, does not compete with thesecond binding site implying that effectors can be bound tothe G�i1(GDP)-(G��)2 complex. Biophysical measurementsand molecular docking studies suggest that this second site isdistant from the primary one. A synthetic peptide having asequence identical to the putative second binding site onG�i1competes with binding of the second G�� subunit. Injectionof this peptide into cultured cells expressing eYFP-G�i1(GDP) and eCFP-G�� reduces the overall association ofthe subunits suggesting this site is operative in cells. We pro-pose that this second binding site serves to promote and sta-bilize G protein subunit interactions in the presence of com-peting cellular proteins.

The plasma membranes of cells are organized as a series ofprotein-rich and lipid-rich domains (1–3).Many of the protein-rich domains, in particular those organized by caveolin pro-teins, are thought to be complexes of functionally related pro-teins that transduce extracellular signals (2). There is increasingevidence that heterotrimeric G proteins exist in pre-formedmembrane complexes with their receptors and their intracellu-lar effectors (4–8).The G protein signaling system is initiated when an extracel-

lular agonist binds to its specific G protein-coupled receptor(for review see Refs. 9–12). The ligand-bound receptor willthen catalyze the exchange of GTP for GDP on the G� subunitin the G protein heterotrimer. In the basal state, G�(GDP)binds strongly to G��, but in the GTP-bound state this affinity

is reduced, allowingG�(GTP) andG�� subunits to individuallybind to a host of specific intracellular enzymes and change theircatalytic activity.Although the interactions between G protein subunits have

been studied extensively in vitro, their behavior in cells maydiffer. For example, in pure or semi-pure systems, activation ofG�(GDP) sufficiently weakens its affinity for G�� resulting indissociation (13). However, in cells separation of the heterotri-mer is observed under some circumstances, but not others (7,14–17). The reason for these differences in behavior is notclear. There are four families of G� subunits that each containseveral members, and, additionally, there are many subtypes ofG�� subunits (18). It is possible that differences in dissociationbehavior reflect differences in affinity between G protein sub-unit subtypes (19), the presence of various protein partners,and/or differences in post-synthetic modifications of the sub-units (20).The mechanism that allows activated G proteins to remain

bound is not apparent from the crystal structure (21, 22). If Gprotein subunits do not dissociate in cells, then their interac-tion must change in such a manner as to expose the effectorinteraction site(s). We have found that phospholipase C�1(PLC�1),4 an important effector of G�q (23), is bound to G�qprior to activation and throughout the activation cycle (6)implying that G�q(GDP) interacts with PLC�1 in a non-func-tional manner.We have evidence that signaling complexes are stabilized by

a series of secondary interactions. Using purified proteins andmodel membranes, we have found that membranes of the G�q-G��/PLC�1/RGS4 signaling system have secondary, weakerbinding sites to members of this signaling system in addition totheir high affinity site(s) to their functional partner(s).We spec-ulate that secondary contacts allow for self-scaffolding of sig-naling proteins. To understand the nature of these secondarycontacts, we have studied the ability of the G�i1(GDP)-G��heterotrimer to remain complexed through the activation cycle(24). Here, we present evidence that G�i1(GDP) has two dis-tinct G�� binding sites that only differ in affinity by an order ofmagnitude and may allow for continued association betweenthe subunits upon activation.We also find that this site plays animportant role in stabilizing G protein associations in cells andprovides a mechanism of self-scaffolding.

* This work was supported, in whole or in part, by National Institutes of HealthGrant GM53132.

□S The on-line version of this article (available at http://www.jbc.org) containssupplemental text, Table S1, and Fig. S1.

1 Present address: Program in Neuroscience, Washington State University,Pullman, WA 99164.

2 Present address: Dept. of Biological Sciences and Geology, QueensboroughCommunity College, the City University of New York, Bayside, NY 11364.

3 To whom correspondence should be addressed: Dept. of Physiology & Bio-physics, Stony Brook University, Stony Brook, NY 11794-8661. Tel.: 631-444-3071; Fax: 631-444-3432; E-mail: Suzanne.Scarlata@sunysb.edu.

4 The abbreviations used are: PLC�1, phospholipase C�1; GTP�S, guanosine5�-3-O-(thio)triphosphate; CPM, 7-diethylamino-3-(4�-maleimidylphenyl)-4-methylcoumarin; eYFP, enhanced yellow fluorescent protein; eCFP,enhanced cyan fluorescent protein; FRET, fluorescence resonance energytransfer; D-, DabcylSE; FCS, fluorescence correlation spectroscopy.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 284, NO. 25, pp. 16906 –16913, June 19, 2009© 2009 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

16906 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 284 • NUMBER 25 • JUNE 19, 2009

by guest on April 12, 2018

http://ww

w.jbc.org/

Dow

nloaded from

MATERIALS AND METHODS

Protein Expression and Purification—His6-G�i1 proteinswere expressed in Escherichia coli and purified as previouslydescribed. The plasmids of these proteins were kindly providedby Dr. Heidi Hamm (Dept. of Pharmacology, Vanderbilt Uni-versity). Her laboratory has shown that the single Cys mutantsbehave identically to wild type (see Refs. 25, 26 for full descrip-tion of their properties as well as their expression and purifica-tion). G�i1(GDP) was activated by incubation at 30 °C for 30min, with an activation buffer (50 mM Hepes, 100 mM

(NH4)2SO4, 150 mM MgSO4, 100 mM EDTA, and 100 �M

GTP�S) (25, 27).His6-G�1�2 was expressed in SF9 cells through baculovirus

infection (28). This method allows for post synthetic modifica-tions. The geranylgeranyl chain on the G�2 subunit wasassessed on LK5D linear-k silica gel TLC plates. His6-PLC�2was expressed using a baculovirus-Sf9 expression system.Protein Labeling—G�i1 proteins were labeled with the thiol-

reactive probe, 7-diethylamino-3-(4�-maleimidylphenyl)-4-methylcoumarin (CPM) at a 1:1 probe to protein ratio as deter-mined by absorption spectroscopy. Unreacted probe wasremoved by dialysis (3� for 30min) against a 100-fold excess ofbuffer containing dithiothreitol. G��was labeled at pH8.0withDabcylSE or Alexa488 carboxylic acid 2,3,5,6-tetrafluorophe-nyl ester (Invitrogen), which reacts with primary amines to givea labeling ratio of 1:5 probe:protein as determined by absorp-tion. This low level of labeling is due to acylation of a largeportion of the protein as determined by trypsin digestion fol-lowed bymass spectrometry. Unreacted probe was removed bygel chromatography. Labeling was also verified by measuringthe diffusion coefficient of Alexa-Gbg by fluorescence correla-tion spectroscopy.Fluorescence Measurements—Fluorescence experiments were

carried out on an ISS PC1 spectrofluorometer (ISS, Urbana, IL).10 nM CPM-labeled G�i was reconstituted on 200 �M largeunilamellar vesicles composed of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine, 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine, and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (1:1:1), and the solution was placed in a3-mm microcuvette. Spectra were recorded using a 384 nmexcitation wavelength and by scanning the emissions from 420to 520 nm. The area under the curve was calculated to give thetotal emission intensity.FCSMeasurements—FCSmeasurements were performed on

an Alba dual-channel confocal fluorescence correlation instru-ment (ISS), equippedwith an argon ion laser (Melles Griot) andinterfaced to an inverted microscope (TE300, Nikon). Excita-tion was at 488 nm, and the fluorescence was recorded using anavalanche photodiode through an emission filter (HQ535/50X). Alignment and calibration were performed using freshlyprepared 20 nM rhodamine 110 solutions. The experimentalautocorrelation function, G(�), was fit using the three-dimen-sional diffusion model, G(�) � 1/N (1 � �/�D)�1�(1 ��/a�D)�0.5, in which �D is the residence time of a molecule inFCS observation volume, N is the average number of particles,and a is the structural parameter describing the observed vol-ume. For photon counting histogram analysis, histograms were

generated from the raw data using software developed byEnrico Gratton (Laboratory of Fluorescence Dynamics at Uni-versity of California, Irvine) and were subsequently fit to anequation describing Gaussian beamwaist photon counting his-togram (PCH) distribution involving two components. Thebrightness ratio of the 1st and 2nd components was fixed at 2(see Ref. 29 for details), and the fraction of components werecalculated from the relative population ratio of the two species.Cell Culture and Transfection—HEK293 cells were cultured

in Dulbecco’s modified Eagle’s medium supplemented with10% fetal bovine serum, 50 units/ml penicillin, and 50 �g/mlstreptomycin sulfate at 37 °C in a 5% CO2 incubator. The cellswere transfected using calcium phosphate coprecipitation inwhich cells were grown on 60-mm dishes for 24–48 h toachieve 80–90% confluence, the media was then replaced. 5 �gof eYFP-G�i1 and eCFP-G�1 and 10 �g of HA-G�7 plasmidswere mixed with 120 mM CaCl2 and HBS buffer (21 mMHepes,123 mM NaCl, 5 mM KCl, and 0.9 mM Na2HPO4, pH 7.1), incu-bated on ice for 10 min, and added to cells dropwise. The cellswere then incubated at 37 °C, and themedia were replaced after8–14 h. The cells were allowed to recover for 8–14 h and splitinto 35-mm glass bottom Mattek dishes and imaged 48–72 hlater. Membrane preparations from cells transfected witheYFP-G�i1 and eCFP-G�1 activate PLC� effectors at a �50%higher level than membrane prepared from cells transfectedwith empty vector.Microinjection—Transfected cells were grown in Mattek

dishes for 48 h at 70–80% to achieve confluence. Prior tomicroinjecting, the media was changed to phenol-free Leibo-vitz-15. The sample microinjection solutions consisted of 130nM peptide with 0.2% deoxycholic acid in 20 nMHepes, 160mMNaCl, pH 7.2, with trace amounts of Cy5, and the control solu-tion was identical except that peptide was omitted.We used anInjectMan NI2 with a FemtoJet pump from Eppendorf tomicroinject the solutions into cytoplasm. We typically set theinjection pressure Pi � 30 hPa, and kept the compensationpressure Pc � 15 hPa. The injection time was t � 0.4 s. Typi-cally, we injected about 10–25 cells within a 10- to 20-minperiod. We examined the microinjected cells under the phasemicroscope (Axiovert 200M from Zeiss with 40� phase 2objective) to select viable cells. We then transferred the cells tothe Zeiss LSM 510 Confocor 2 apparatus (Jena, Germany) andcollected images. FRET analysis of the images were collectedand analyzed as previously described (24).

RESULTS

FRET Studies Suggests that G�i1 Binds Multiple G��Subunits—In a previous study, Hamm and coworkers devel-oped a series of single Cys mutants of G�i1 to monitor confor-mational changes that occur upon activation (25). We haveused these mutants to measure the affinity between G�� andG�i1(GDP) on 200 �M lipid bilayers to be Kd � 1 nM (24). Todetermine the changes in orientation between G�i1 and G��that occur upon activation, we labeled each of these single CysG�i1 mutants with an environmentally sensitive fluorescentprobe (CPM) and monitored the ability of these mutants totransfer excited state energy (i.e.FRET) toG�� subunits labeledon the N terminus with the non-fluorescent FRET acceptor,

A Second, High Affinity G�� Binding Site on G�i1

JUNE 19, 2009 • VOLUME 284 • NUMBER 25 JOURNAL OF BIOLOGICAL CHEMISTRY 16907

by guest on April 12, 2018

http://ww

w.jbc.org/

Dow

nloaded from

DabcylSE (D-), at a 0.3:1 probe to protein and note that this lowratio is due to significant fraction of acetylated protein as deter-mined by mass spectrometry. We also note that there is uncer-tainty of the labeled species due to the difficulty in assessing thesmall G�. Therefore, we are using FRET as a general indicatorof binding.Because Dabcyl is a non-fluorescent FRET acceptor,

transfer from CPM will be seen as a decrease in donor fluo-rescence. The Ro for these probes, which corresponds to thedistance at which 50% of donor fluorescence is lost to trans-fer, is 20 Š(30). Keeping in mind that the amount of FRETfollows the 6th power of the distance, then from the crystalstructure (31), only Cys-106 will be close enough to the Nterminus of G� to participate in FRET barring a largeamount of free rotation from the N terminus of G�. Unex-pectedly, the titration curves for all of the G�i1(GDP)mutants showed a biphasic behavior (Fig. 1). At low G��concentrations, an increase in C-G�i1 donor intensity wasobserved. Above 5 nM, the decreases in C-G�i1 intensities are

not seen when unlabeled G�� subunits are used. We havepreviously reported this increase in CPM-G�i1(GDP) inten-sity with the addition of unlabeled G�� and interpret it to bedue to protection of the CPM from solvent quenching asG�� subunits bind to G�i1 forming the heterotrimer. If thedata obtained for the unlabeled G�� is subtracted fromD-G��, then this increase is eliminated. Thus, the initialportions of the titration curves are interpreted to representthe primary G�� binding site of G�i1(GDP).Unexpectedly, at higher G�� concentrations, a decrease in

intensity, indicative of FRET, was seen. This decrease wasobserved for all mutants and saturated at �80 nM G��. Weinterpret the loss in C-G�i1 intensity to be due to FRET toD-G��. We find that the extent of this decrease varied for eachmutant (see Fig. 4), although the concentration dependence ofthe titration curves was similar. Fitting the titration curves to abimolecular association constant gives a similar apparent dis-sociation constant for all of the single G�i1 mutants rangingfrom 21 to 34 nM giving an average of Kd � 23 � 5 nM (see

FIGURE 1. A, raw data of a single set of titrations comparing the change in fluorescence intensity in arbitrary units of CPM-labeled Cys-217 with the addition ofunlabeled G�� (open circles) or D-G�� (closed circles), and then with the addition of PLC�2. For this plot, the intensity values of the control sample were offsetto better display the rise of the samples at low G�� concentrations. B, similar study showing the intensity changes of CPM-labeled Cys-3 upon the addition ofG�� or D-G��, and then with the addition of PLC�2, where the data are an average of five trials with � S.D. C, the results using activated CPM-labeled Cys-3.

A Second, High Affinity G�� Binding Site on G�i1

16908 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 284 • NUMBER 25 • JUNE 19, 2009

by guest on April 12, 2018

http://ww

w.jbc.org/

Dow

nloaded from

supplementalmaterial). These apparentKd values are�20-foldweaker than those measured for G�i1(GDP)-G�� by othermethods. Addition of excess PLC�2, which binds strongly toG�� subunits, caused a partial recovery of the CPM donorintensity (Fig. 1) suggesting that the binding of the second G��subunit is reversible.The results of the FRET titrations are surprising because all

mutants display a significant amount of FRET that begins abovethe concentration at which the G�i1-G�� heterotrimer shouldhave formed (i.e. apparent Kd � 1 nM (24)), and also because,with one exception, all of the probe positions should be associ-ated with little or no FRET. We do not believe that this weakerapparent affinity is due to the presence of fluorescent tags,because the location of the probes are not interfacial andbecause the apparent Kd values obtained for the mutants areclose. The simplest interpretation of these data is that a secondG�� molecule is binding to the heterotrimer, which is in closeenough proximity to participate in FRETwith the different sin-gle Cys sites (i.e. within 20 Å). Association of a second G��subunit to the heterotrimer at higher concentrations was con-firmed by chemical cross-linking (see supplemental material).We repeated the FRET titrations using the activated forms of

the Cys-29 and Cys-106 mutants. We could not detect bindingat low G�� concentrations in accord with the reduced affinitythat accompanies activation (e.g. Fig. 1C). We did, however,clearly observe aweaker association that is on the same order asthe apparent second site observed using FRET (Kd(app)� 45�

5 nM for Cys-29 and 62 � 11 nM forCys-106). These results suggestthat, in the activated state, thehigh affinity site is replaced by asecond, weaker site. Cross-linkingstudies also suggest that only oneG�� binding site is operative whenG� is activated (see supplementalmaterial).Brightness Analysis Shows That

G�i1 Can Bind to Multiple G��Subunits—Toverify the existence ofa second site, we labeled G�� withAlexa488 and measured the molec-ular brightness of the molecules insolutions containing different stoi-chiometric amounts of G�i1 (Fig. 2)(32) on an FCS instrument. Meas-urements were performed withoutlipid vesicles, which would interferewith diffusionmeasurements. In theabsence of G�i1, the brightness ofA-G��, expressed as counts per sec-ond per molecule, was �40% lowerthan free Alexa 488 most likely dueto the presence of local quenchinggroups.At low amounts of G�i1, the

brightness of A-G�� does not sig-nificantly change. However, as theamount of addedG�i1 increases, the

brightness A-G�� also increases. The onlymechanism that canincrease the molecular brightness is the presence of oligomersthat contain more than one A-G�� molecules. As the stoichio-metric amount of G�i1 rose 5- to 10-fold above A-G��, thebrightness decreased as the equilibrium shifted to a singleA-G�� molecule per G�i1 subunit. The corresponding diffu-sion coefficients measured by FCS were 6-fold slower than freeAlexa488 suggesting that the complexes are small. Importantly,the diffusion coefficientswere relatively unchanged throughoutthe titration curves showing that the proteins were notaggregating.We fit the data from the 80 nMA-G�� :40 nMG�i1 in Fig. 2 to

a model in which one A-G�� binds to the A-G��-G�i1i com-plex (see “Materials andMethods” and supplemental material).The results show that at this stoichiometry, 30% of theG�� is ina 2 G��-G�i1 complex. The labeling ratio of A-G�� is 1:5probe:protein, and so a significant population of unlabeledmaterial is present that is optically silent. Thus, the amount ofG��-G�i1-G�� is expected to be far greater than 30%.We notealso that, in these studies, the presence of lipids wouldbe expected to increase the amount of higher order complexesdue to the effective reduction of dimensionality and effectiveincrease in concentration (see Ref. 28).Our FRET titration curves suggest that, in the activated state,

the primary high affinity binding site of G�� to G�i1 is replacedby a second, weaker site. To determine whether activated G�i1is capable of binding one or two A-G�� subunits, we repeated

FIGURE 2. Dependence of molecular brightness of 80 nM A-G�� at different G�i1(GDP) concentrations(closed circles) and G�i1(GTP�S) (open circles) where each point is an average of 18 –23 measurementsand �S.D. is shown. Inset, plot of the ln of the G�i1(GDP) versus concentration shown to better distinguish thebehavior of the molecular brightness at low concentrations of G�i1(GDP).

A Second, High Affinity G�� Binding Site on G�i1

JUNE 19, 2009 • VOLUME 284 • NUMBER 25 JOURNAL OF BIOLOGICAL CHEMISTRY 16909

by guest on April 12, 2018

http://ww

w.jbc.org/

Dow

nloaded from

the above studies using activated G�i1. We find that the bright-ness is constant through a wide range of G�i1 concentrations(Fig. 2) suggesting that, in the activated state, only one G��binding site is available.FRET Studies between G�� Subunits Support a G��-G�i1-

G�� Complex—To support the idea that G�i1 is capable ofbinding two G�� subunits, we carried out a FRET study inwhich we added CPM-labeled G�� to a stoichiometric amountof D-G�� andmeasured the ability of the two proteins to FRETwhen unlabeled G�i1 is added. We find that G�i1 producesFRET between theG�� subunits. In Fig. 3 we show results froma similar study in which D-G�� was added to the preformedG�i1(GDP)-G�� heterotrimer where G�� was labeled withCPM. A decrease in fluorescence intensity is seen suggestingthat FRET between the G�� subunits is occurring.The Second G�� Site Does Not Occlude PLC�2 Effector

Binding—PLC�2 is strongly activated byG�q but not G�i1 sub-units, and it is also activated by G�� subunits (for review seeRefs. 33, 34).Wedeterminedwhether PLC�2 could displace thesecondG�� from theG�i1-G��heterotrimer by adding PLC�2to the C-G�i1-D-G�� complex and measuring the change inthe amount of FRET. Addition of 80 nM PLC�2 to solutionscontaining 10nMG�i1 and 80nMG��did not affect the amountof FRET most likely due to binding of the enzyme to unboundG�� subunits (Figs. 1 and 4). However, addition of 236 nMPLC�2 resulted in a partial loss in FRET for Cys-106, Cys-217,and Cys-305, which had a high degree of FRET, and a completeloss in FRET for Cys-3, Cys-29, and Cys-273, which had a lowerdegree of FRET. Thus, either the PLC�2 directly displaced thesecond G�� site or addition of excess PLC�2 caused dissocia-tion of G�� by a shift in equilibrium. To distinguish betweenthese mechanisms, we added D-G�� to C-G�i1 in the presenceof PLC�2where the PLC�2was at a concentration high enoughto be completely bound to G�� (i.e. 20 nM). We found thatPLC�2 did not affect binding implying that the site of interac-tion between the second G�� and G�i1(GDP) does not overlapwith the effector binding site (data not shown).Identification of the Second Interaction Site—To identify the

nature of the second binding site, we used GRAMM (Dr. I.Vasker, University of Kansas) to dock G�� to the G�i1G��using the crystal structure of Sprang and coworkers (31). Of the20 lowest energy conformations, approximately half involvedirect G��-G�� interactions, which is inconsistent with theFRET and FCS data. Of the remaining models, we eliminatedseveral on the basis of the FRET results, the partial occlusionof G� 86–105, which comprises the PLC�2 activation site(35–37), and the possible membrane orientation of the pro-teins (38, 39).We then carried out a study to distinguish between the

remaining models based on differences in the solvent accessi-bilities of the single Cys side chains in the G�i1 mutants whenthe first and second G�� subunits are bound using the thiol-reactive probe, CPM. In its unreacted state, CPM is not fluores-cent but becomes highly fluorescent upon covalent linkage to athiol group (see Invitrogen product literature) and thus, Cysaccessibility can be judged by the amount of fluorescence. Wethen added CPM to a solution containing membrane-boundG�i1(GDP) andG�� subunits whoseCys residues were blocked

by pretreatment with iodoacetamide. We compared theamount of fluorescence at 0 and 30 min for the Cys-106, Cys-217, and Cys-273 of G�i1 at G�� concentrations where the firstsite should be primarily occupied (10 nM G��) and at concen-trations the second site would be occupied (80 nM G��). Cys-3and Cys-29 were not tested due to their close proximity to thefirst site and themembrane interface.We find that Cys-217 andCys-273 were more (�38 � 5%) shielded and less solvent-ac-cessible (i.e. exhibited less CPM fluorescence) at 80 nM G�� ascompared with 10 nM. However, Cys-106 shows the sameamount of CPM fluorescence at both concentrations suggest-ing that it is exposed when the second G�� is bound. Thisfinding narrows down the potential models to a set of threesimilar models (the lowest energy one is shown in Fig. 5). Thismodel also correlates well with the FRET measurements.

FIGURE 3. Titration showing the onset of FRET, as seen by a decrease inthe normalized fluorescence intensity (FI), as D-G�� is added to CPM-G�� - G�i1(GDP) showing that the two G�� subunits are in FRET dis-tance, where n � 3 and �S.D. is shown. Control studies, which substitutedunlabeled G�� or omitted G�i1(GDP) from the cuvette solution, did not affectthe fluorescence intensity.

FIGURE 4. Compiled results of the decrease in fluorescence intensities ofthe single Cys CPM- G�i1(GDP) mutants with the addition of 80 nM D-G��and 80 and 236 nM of unlabeled PLC�2 is shown, where n � 3– 6 and�S.D. is shown.

A Second, High Affinity G�� Binding Site on G�i1

16910 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 284 • NUMBER 25 • JUNE 19, 2009

by guest on April 12, 2018

http://ww

w.jbc.org/

Dow

nloaded from

According to this model, the most prominent contact madebetween G�i1(GDP) and the second G�� subunit involves res-idues 267–271 of G��. We prepared a peptide that has thissequence and should compete with binding of the second G��.We note that because this contact region is small, the test pep-tide may not be as effective for competition with the secondbinding site as comparedwith larger peptides (e.g.Ref. 40). Twotypes of assays were carried out. In the first, we addedD-G�� toC-G�i1 until FRET was clearly observed indicating the forma-tion of D-G��-(C-G�i1)-G�� complex.We then added peptideto the complex and found a complete reversal of FRET suggest-ing dissociation of the D-G�� subunit (Fig. 6, top). In a secondseries of studies, we titrated D-G�� into a solution containingC-G�i1 in the absence andpresence of 10�Mpeptide.We foundthe peptide shifts the binding of the second D-G�� to higherconcentrations from Kd(app) � 15.4 � 2.7 nM to 32 � 9.6 nMand reduced the extent of binding as judged by the amount ofFRET (Fig. 6, bottom). These results suggest that the peptidecompetitively binds to the second site.Our FRET and FCS studies show that, in the activated

state, G�i1 only binds one G�� subunit, and our titrationcurves show this binding has an affinity on par with the sec-

ond G�� site seen for G�i1 in the deactivated state. To deter-mine whether the second G�� binding site in the deactivatedstate overlaps with the single binding site in the activatedstate, we carried out a titration study of D-G�� toC-G�i1(GTP�S) in the presence of 10 �M peptide. We foundthat the presence of peptide had little affect on the binding ofD-G�� to C-G�i1 suggesting that the loop encompassing thesecond binding sites has little or no overlapwith theG�i1(GTP)site. However, it is still possible that, even though there is nodirect overlap, binding of the full-length proteins could becompetitive.Addition of Peptide Reduces G�i-G�� in Cells—To deter-

mine if the second site has functional significance, we deter-minedwhether the peptide could disruptG��-G�i1 associationin HEK293 cells. These studies were carried out by transfectingHEK293 cells with the FRET pair, eCFP-G�1�7 and eYFP-G�i1,and determining the change in FRET after microinjecting thepeptide. We note that we added a tracer, the red dye (Cy5),whose emission is out of range of the CFP/YFP channels, to

FIGURE 5. Two views of the lowest energy molecular model of the G��-G�i1(GDP)-G�� complex as predicted by GRAMM that best fits the fluo-rescence measurements where G�267–271 is shown in orange, G�86 –105 is in ball-and-stick format in green, and G�273 is in CPK.

FIGURE 6. Top, recovery of CPM fluorescence, expressed as normalized fluo-rescence intensity, due to the reversal of FRET when G�267–271 is added tothe D-G��-C-G�i1(GDP) complex. Bottom, plot of the degree of association, ascalculated by the decrease in the fluorescence intensity of CPM-labeledG�i1(GDP) as D-G�� is added in the absence and presence of G�267–271.

A Second, High Affinity G�� Binding Site on G�i1

JUNE 19, 2009 • VOLUME 284 • NUMBER 25 JOURNAL OF BIOLOGICAL CHEMISTRY 16911

by guest on April 12, 2018

http://ww

w.jbc.org/

Dow

nloaded from

allow us to identify the microinjected cells. The microinjectedcells were viewed within 1 h after injection and showed no vis-ible changes in morphology (Fig. 7A). We compared theamount of FRET between eYFP-G�i1(GDP) and eCFP-G�� inmicroinjected cells that contained peptide versus cells microin-jected with the carrier solution. The results from three inde-pendent studies show that peptide causes a reduction in FRETbetween eYFP-G�i1(GDP) and eCFP-G�� (Fig. 7B). We note

that the large variability in the FRETvalues of the peptide samples aremost likely due to the variation inthe amount of peptide deliveredinto the cells. On the whole, thesestudies show that peptide disruptsG protein subunits in cells.

DISCUSSION

In this study, we have shown thatG�i1 subunits contain a secondbinding site for G�� subunits.Molecular modeling coupled withfluorescence and accessibility stud-ies suggests that this site encom-passes residues 267–271 of G�.Activation of G�i1 eliminates thehigh affinity site but promotes theformation of a low affinity site thatdoes not share the identical inter-facewith the second site of the deac-tivated protein. The G�� affinity ofthe second binding site is only10-fold lower than its functionalone, and injection of a competitivepeptide into cells reduces thisassociation.The structural changes that occur

uponGprotein activation have beenintensely studied (see Ref. 41).In the deactivated heterotrimer,G�i1(GDP) contacts G�� subunitsat several points, including one thatencompasses the first �30 residuesof G�i1(GDP), resulting in a highaffinity interaction (31). Exchangeof GDP for GTP produces confor-mational changes in three keyregions that result in a loss in affin-ity of�10-fold (13, 21). This drop inaffinity may be too small to causesubunit dissociation in cells espe-cially considering G proteins areassociated with receptors and mostlikely other partners that may stabi-lize the heterotrimer (e.g. Refs. 4, 7,15). Additionally, we have foundthat activation of G�i1 changes itsorientation with respect to G��subunits allowing exposure of

PLC�2 interaction sites without dissociation of the heterotri-mer (24).Previous studies of subunit interactions in the G protein het-

erotrimer have typically utilized stoichiometric amounts ofsubunits or used methods that masked this interaction. Thus,the presence of a second G�� association site has not yet beenobserved despite the strong affinity. In the studies carried outhere, association between the subunits was promoted by the

FIGURE 7. A, images of HEK293 cells expressing eYFP-G�i1 and eCFP-G�� with microinjection of Cy5 alone andwith G�267–271 peptide. The bright cytosolic spot is the point of injection. This region was not consideredwhen the FRET was calculated. Control studies used the identical microinjection solution without peptide.B, effect of G�267–271 on the normalized FRET between eYFP-G�i1 and eCFP-G��.

A Second, High Affinity G�� Binding Site on G�i1

16912 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 284 • NUMBER 25 • JUNE 19, 2009

by guest on April 12, 2018

http://ww

w.jbc.org/

Dow

nloaded from

effective reduction in dimensionality by carrying out the meas-urements on membrane-bound proteins (see Ref. 28). It isthus important to consider the affinities reported here asapparent affinities, and the true affinities are expected to beweaker. A lower affinity was qualitatively observed in thebrightness measurements, which were carried out in theabsence of membranes.Elementary docking studies of the second site suggest that

there are several contact points along the surface of G�� thatare responsible for binding to the second site and in particular,a unique contact encompassing the loop residues 267–271.This interaction is not found in the crystal structures of G��bound to G protein receptor kinase or the peptide bindingregion of adenylyl cyclase (42, 43). This region is not involved inPLC�2 binding (36, 44).We note, however, that the second siteinvolves multiple interactions over four G�� blades, two ofwhich are close to the PLC�2 binding site.The observation that the affinity of the secondG�� for deac-

tivated G�i1 is in range of the affinity seen for the G�� bindingsite in activated G�i1 leads to the speculation that the sitesoverlap. If this were the case, then G�i1 activation may causeG�� to move to this site stabilizing the heterotrimeric state.However, our data indicate that the interface of the secondbinding site encompassing residues 267–271 is not operative inthe activated state as indicated by the inability of the peptide todisplace G�� from deactivated G�i1 but not from activatedG�i1. Because G�i1 subunits undergo large conformationalchanges upon activation, it is likely that the G�� 267–271 toG�i1 interaction is weakened or replaced by other contacts.

While it is unclear whether some, if any, of the contactsbetween the second G�� site and G�i1(GDP) are utilized byactivated G�i1, our studies using cultured cells show thatmicroinjecting a peptide whose sequence corresponds to the267–271 disrupts G�i1(GDP)-G�� association. This observa-tion suggests that this second site may serve to organize higherorder G protein complexes on the plasma membrane serving ascaffolding function. It is also possible that the second bindingsite serves to increase the density of G�� effector binding. Thecomparative affinities of the primary and secondary bindingsites set a window of the concentration range of G�� that willallow the subunits to be released in cells and subsequently con-tact distant effectors.

Acknowledgments—We thank Stuart McLaughlin (Department ofPhysiology & Biophysics, Stony Brook University) for the use of theZeiss microscope. We also thank Stephen Sprang (Department ofStructural Biology, University of Montana) for reading this manu-script and Dr. Heidi Hamm for sending us the single Cys G�constructs.

REFERENCES1. Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J. (1994)

Molecular Biology of the Cell, pp. 734–758, Garland, New York2. Edidin, M. (2003) Annu. Rev. Biophys. Biomol. Struct. 32, 257–2833. Anderson, R. G. (1998) Annu. Rev. Biochem. 67, 199–2254. Bunemann, M., Frank, M., and Lohse, M. J. (2003) Proc. Natl. Acad. Sci.

U. S. A. 100, 16077–160825. Nobles,M., Benians, A., andTinker, A. (2005)Proc. Natl. Acad. Sci. U. S. A.

102, 18706–187116. Dowal, L., Provitera, P., and Scarlata, S. (2006) J. Biol. Chem. 281,

23999–240147. Philip, F., Sengupta, P., and Scarlata, S. (2007) J. Biol. Chem. 282,

19203–192168. Golebiewska, U., and Scarlata, S. (2008) Biophys. J. 95, 2575–25829. Neer, E. J. (1995) Cell 80, 249–25710. Neer, E. J. (1994) Protein Science 3, 3–1411. Birnbaumer, L. (2007) Biochim. Biophys. Acta 1768, 772–79312. Exton, J. H. (1997) Eur. J. Biochem. 243, 10–2013. Runnels, L. W., and Scarlata, S. F. (1999) Biochemistry 38, 1488–149614. Janetopoulos, C., Jin, T., and Devreotes, P. (2001) Science 291, 2408–241115. Hughes, T. E., Zhang, H., Logothetis, D. E., and Berlot, C. H. (2001) J. Biol.

Chem. 276, 4227–423516. Hynes, T. R.,Mervine, S.M., Yost, E. A., Sabo, J. L., and Berlot, C. H. (2004)

J. Biol. Chem. 279, 44101–4411217. Digby, G. J., Lober, R.M., Sethi, P. R., and Lambert, N. A. (2006) Proc. Nat.

Acad. Sci. U. S. A. 103, 17789–1779418. Hildebrandt, J. D. (1997) Biochem. Pharmacol. 54, 325–33919. Sarvazyan, N. A., Lim, W. K., and Neubig, R. R. (2002) Biochemistry 41,

12858–1286720. Wedegaertner, P. B., Wilson, P. T., and Bourne, H. R. (1995) J. Biol. Chem.

270, 503–50621. Tesmer, J. J., Berman, D. M., Gilman, A. G., and Sprang, S. R. (1997) Cell

89, 251–26122. Tesmer, J. J., Sunahara, R. K., Gilman, A. G., and Sprang, S. R. (1997)

Science 278, 1907–191623. Rebecchi, M. J., and Pentyala, S. N. (2000) Physiol. Rev. 80, 1291–133524. Wang, J., Golebiewska, U., and Scarlata, S. (2009) J. Molec. Biol. 387,

92–10325. Medkova,M., Preininger, A.M., Yu,N. J., Hubbell,W. L., andHamm,H. E.

(2002) Biochemistry 41, 9962–997226. Skiba, N. P., Bae, H., and Hamm, H. E. (1996) J. Biol. Chem. 271, 413–42427. Chidiac, P., Markin, V. S., and Ross, E. M. (1999) Biochem. Pharm. 58,

39–4828. Runnels, L. W., and Scarlata, S. F. (1998) Biochemistry 37, 15563–1557429. Chen, Y., Muller, J. D., So, P. T. C., and Gratton, E. (1999) Biophys. J. 77,

553–56730. van der Meer, W., Coker G., and Chen, S. S. Y. (1994) Resonance Energy

Transfer, Theory andData, pp. 148–164, VCHPublishers, Inc., New York31. Wall, M. A., Coleman, D. E., Lee, E., Iniguez-Lluhi, J. A., Posner, B. A.,

Gilman, A. G., and Sprang, S. R. (1995) Cell 83, 1047–105832. Schwille, P. (2001) Cell. Biochem. Biophys. 34, 383–40833. Exton, J. H. (1994) Annu. Rev. physiol. 56, 349–36934. Suh, P. G., Park, J. I., Manzoli, L., Cocco, L., Peak, J. C., Katan, M., Fukami,

K., Kataoka, T., Yun, S., and Ryu, S. H. (2008) BMB reports 41, 415–43435. Li, Y., Sternweis, P. M., Charnecki, S., Smith, T. F., Gilman, A. G., Neer,

E. J., and Kozasa, T. (1998) J. Biol. Chem. 273, 16265–1627236. Panchenko,M. P., Saxena, K., Li, Y., Charnecki, S., Sternweis, P.M., Smith,

T. F., Gilman, A. G., Kozasa, T., and Neer, E. J. (1998) J. Biol. Chem. 273,28298–28304

37. Buck, E., Li, J., Chen, Y., Weng, G., Scarlata, S., and Iyengar, R. (1999)Science 283, 1332–1335

38. Murray, D., McLaughlin, S., and Honig, B. (2001) J. Biol. Chem. 276,45153–45159

39. Drin, G., and Scarlata, S. (2007) Cell. Signal. 19, 1383–139240. Drin, G., Douguet, D., and Scarlata, S. (2006) Biochemistry 45,

5712–572441. Cabrera-Vera, T. M., Vanhauwe, J., Thomas, T. O., Medkova, M., Prein-

inger, A., Mazzoni, M. R., and Hamm, H. E. (2003) Endocr. Rev. 24,765–781

42. Lodowski, D. T., Pitcher, J. A., Capel, W. D., Lefkowitz, R. J., and Tesmer,J. J. (2003) Science 300, 1256–1262

43. Diel, S., Klass, K., Wittig, B., and Kleuss, C. (2006) J. Biol. Chem. 281,288–294

44. Buck, E., Schatz, P., Scarlata, S., and Iyengar, R. (2002) J. Biol. Chem. 277,49707–49715

A Second, High Affinity G�� Binding Site on G�i1

JUNE 19, 2009 • VOLUME 284 • NUMBER 25 JOURNAL OF BIOLOGICAL CHEMISTRY 16913

by guest on April 12, 2018

http://ww

w.jbc.org/

Dow

nloaded from

ScarlataJingting Wang, Parijat Sengupta, Yuanjian Guo, Urszula Golebiewska and Suzanne

(GDP) Subunitsi1α Binding Site on GγβEvidence for a Second, High Affinity G

doi: 10.1074/jbc.M109.006585 originally published online April 15, 20092009, 284:16906-16913.J. Biol. Chem. 

  10.1074/jbc.M109.006585Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

  http://www.jbc.org/content/suppl/2009/04/15/M109.006585.DC1

  http://www.jbc.org/content/284/25/16906.full.html#ref-list-1

This article cites 42 references, 18 of which can be accessed free at

by guest on April 12, 2018

http://ww

w.jbc.org/

Dow

nloaded from