optodynamic simulation of β-adrenergic receptor...

ARTICLEReceived 9 Jun 2015 | Accepted 27 Aug 2015 | Published 28 Sep 2015

Optodynamic simulation of b-adrenergic receptorsignallingEdward R. Siuda1,2,3,4, Jordan G. McCall1,2,3,4, Ream Al-Hasani1,2,4, Gunchul Shin5, Sung Il Park5,

Martin J. Schmidt1,2,4, Sonya L. Anderson1,2, William J. Planer1,2, John A. Rogers5,6,7,8

& Michael R. Bruchas1,2,3,4,9

Optogenetics has provided a revolutionary approach to dissecting biological phenomena.

However, the generation and use of optically active GPCRs in these contexts is limited and it

is unclear how well an opsin-chimera GPCR might mimic endogenous receptor activity. Here

we show that a chimeric rhodopsin/b2 adrenergic receptor (opto-b2AR) is similar in

dynamics to endogenous b2AR in terms of: cAMP generation, MAP kinase activation and

receptor internalization. In addition, we develop and characterize a novel toolset of optically

active, functionally selective GPCRs that can bias intracellular signalling cascades towards

either G-protein or arrestin-mediated cAMP and MAP kinase pathways. Finally, we show how

photoactivation of opto-b2AR in vivo modulates neuronal activity and induces anxiety-like

behavioural states in both fiber-tethered and wireless, freely moving animals when expressed

in brain regions known to contain b2ARs. These new GPCR approaches enhance the utility of

optogenetics and allow for discrete spatiotemporal control of GPCR signalling in vitro and

in vivo.

DOI: 10.1038/ncomms9480 OPEN

1 Department of Anesthesiology, Division of Basic Research, Washington University School of Medicine, St Louis, Missouri 63110, USA. 2Washington UniversityPain Center, Washington University School of Medicine, St Louis, Missouri 63110, USA. 3Division of Biology and Biomedical Sciences, Washington UniversitySchool of Medicine, St Louis, Missouri 63110, USA. 4Anatomy and Neurobiology, Washington University School of Medicine, St Louis, Missouri 63110, USA.5Department of Materials Science and Engineering, Frederick Seitz Materials Research Laboratory, University of Illinois at Urbana-Champaign, Urbana,Illinois 61801, USA. 6Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801, USA.7Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801, USA. 8Department of Chemistry,University of Illinois at Urbana-Champaign, Urbana, Illinois 61801, USA. 9Department of Biomedical Engineering, Washington University in St Louis, St Louis,Missouri 63110, USA. Correspondence and requests for materials should be addressed to M.R.B. (email: [email protected]).

NATURE COMMUNICATIONS | 6:8480 | DOI: 10.1038/ncomms9480 |www.nature.com/naturecommunications 1

& 2015 Macmillan Publishers Limited. All rights reserved.

mailto:[email protected]

http://www.nature.com/naturecommunications

Over the past decade, optogenetics and chemogenetics havemade significant contributions to probing biologicalquestions. Traditional optogenetic approaches however,

utilize expression of depolarizing (channelrhodopsins)1–3 andhyperpolarizing (halorhodopsins and archaerhodopsins)4 ionchannels to selectively turn on/off neurons in the presence oflight. Chemogenetic approaches, such as designer receptorsexclusively activated by designer drugs5, have high utilityfor their modulation of GPCR signalling, but similar to otherligand-mediated responses, can sometimes be limited bypharmacokinetics and pharmacodynamics making themdifficult to use to examine real-time kinetics of receptoractivity, particularly in vivo. Thus, the development of opticallyactive G-protein coupled receptors (GPCRs) allows for more finetuned modulation of cellular activity. Through manipulation ofthe light stimulus, we can modulate the activity of optically activeGPCRs in an ‘optodynamic’ manner. This, in combination withthe spatiotemporal control offered through optogenetics, providesa more refined in vivo GPCR toolkit than is currently possiblewith chemogenetics and traditional pharmacology.

The chimeric rhodopsin/b2 adrenergic receptor (opto-b2AR)has been shown to activate cAMP, presumably through Gas-mediated signalling, and to modulate neuronal excitability6–8.However, it is unclear whether opto-b2AR behaves similarly toendogenous b2AR. Here we fully evaluated the in vitro activity ofopto-b2AR and b2AR by examining the temporal kinetics ofcAMP and MAP kinase activation in addition to receptorinternalization and desensitization, and demonstrate that opto-b2AR mimics the dynamic signalling profile of b2ARs.

Over the past few years the tenets of GPCR pharmacology havebeen challenged by the concept of ‘functional selectivity’ or‘biased agonism,’ demonstrating that ligands exert varying levelsof efficacies on intracellular signalling mechanisms and thatG-proteins may not be the sole determinants of intracellularactivity9–11. While arrestin was canonically thought to onlyterminate GPCR-mediated signalling through inactivation andinternalization of the receptor, it is now widely accepted thatarrestin acts to scaffold several intracellular signalling cascades,particularly MAP kinases. Several receptors, including opioidreceptors12–14, angiotensin II15, V2 vasopressin16 and bAR17display arrestin-dependent MAP kinase activation. In an effort tocombine spatiotemporal control of opto-b2AR with biased GPCRsignalling, we performed site-directed mutagenesis on opto-b2ARto generate optically active, functionally selective GPCRs.

Mutation of three key residues in b2AR (b2ART68F,Y132G,Y219A orb2ARTYY) generate an arrestin-biased mutant11, while modificationof C-terminal serines prevents arrestin binding by blockingG-protein-coupled receptor kinase phosphorylation resultingin a G-protein-biased mutant (b2ARS355A,S356G or b2ARSS)18–21.Analogous residues were altered in opto-b2AR to generatean arrestin-biased receptor, opto-b2ARL72F,Y136G,Y224A or opto-b2ARLYY and a G-protein-biased receptor, opto-b2ARS362A,S363Gor opto-b2ARSS. Here we determined the dynamic opticalproperties of these novel optically active, functionally selectivereceptors.

Heterologous expression systems are essential in characteriza-tion of receptor activity. They offer the ability to dissect receptorfunction not possible in more complex environments. However,to truly understand endogenous activity, it is essential toultimately look at in vivo function. To that end, we determinedwhether opto-b2AR could be used in vivo since it has yet to beutilized for inducing a significant behavioural phenotype6. Weexpressed opto-b2AR in a biologically relevant neural circuitknown to be under the influence of noradrenergic signalling andthat expresses b2ARs and its signalling moieties, the basolateralamygdala (BLA). The presence of all nine adrenergic receptor

subtypes within the amygdaloid complex22 has precludeddetermination of their roles and signalling pathways in vivo dueto the fact that pharmacological isolation of these receptors isdifficult within the amygdala. These caveats are true for manyGPCRs, due to the lack of spatiotemporal control of receptorfunction, isolation of specific cell types, and control of selectnoradrenergic or other modulatory inputs. To isolatenoradrenergic GPCR signalling in vivo, we used opto-b2AR anddemonstrated that in vivo photoactivation of b-adrenergicsignalling produced excitation of BLA neurons resulting inanxiety-like states in both fiber-tethered and wireless, freelymoving animals.

Here we have fully evaluated the utility of opto-b2AR inmimicking endogenous b2AR activity; we developed novel,optically active, functionally selective receptors to bias b2ARintracellular signalling mechanisms and we used opto-b2ARin vivo and define its ability to initiate a series of real-timebehavioural responses.

ResultsOptical control of b-adrenergic signalling. We first fullycharacterized a unique optical tool for activating b-adrenergicsignalling and compared its pharmacodynamic properties withb2-adrenergic receptors (b2AR). The opto-b2AR receptor is achimeric protein that includes transmembrane and extracellularcomponents of bovine rhodopsin, with intracellular domains andloops of the b2 adrenergic receptor (Fig. 1a)6. Photostimulationof HEK293 cells expressing opto-b2AR caused a real-time,light-power-dependent increase in cAMP (cyclic adenosinemonophosphate), a canonical product of the Gas signallingpathway (Fig. 1b,c), similar to isoproterenol-inducedconcentration-dependent cAMP generation (Fig. 1c)23. Thisreal-time cAMP increase in response to light was absent inuntransfected HEK293 cells (Supplementary Information,Supplementary Fig. 1a). Furthermore, the kinetics of cAMPactivation (ton) and inactivation (toff) are strikingly similarbetween opto-b2AR and endogenous b2AR in HEK293 cellssuggesting that although the extracellular regions of the receptorsdiffer greatly, the conformational change required to initiateintracellular signalling are maintained, making these receptorskinetically similar (Fig. 1d,e). In addition, cAMP triggersactivation of cyclic nucleotide-gated nonspecific cationchannels. Here we show opto-b2AR causes a robust increase inintracellular Ca! 2 in response to light stimulation, with similarresults obtained for b2AR following isoproterenol bathapplication, while untransfected HEK293 cells show noresponse to light (Supplementary Fig. 1b).

In addition to cAMP, extracellular-signal regulated kinase(ERK) 1/2 phosphorylation has been examined extensively inb2AR, showing a rapid, yet transient peak within 2–5min ofisoproterenol-induced activation11,17,24–26. To determine whetheropto-b2AR also activates ERK 1/2 kinases, we stimulated opto-b2AR with a 1min light pulse and generated a time course of ERKphosphorylation (pERK). Similar to isoproterenol-induced pERKin b2AR, opto-b2AR showed a rapid and transient increase inpERK that peaks within 2–5min and then rapidly declines(Fig. 1f,g, Supplementary Fig. 2a for full time course). The kineticeffects seen in b2AR were the same whether in the continuedpresence of isoproterenol (Fig. 1f,g) or following a 1min pulsewith isoproterenol (Supplementary Fig. 2b,c). We also show thatlevels of total ERK in b2AR remain constant over a two-hour trialperiod, suggesting the kinetics of pERK are not due todegradation of total ERK (Supplementary Fig. 2d,e). Inaddition, the extent of ERK activation in opto-b2AR displayed alight-power-dependent relationship that was absent in

ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms9480

2 NATURE COMMUNICATIONS | 6:8480 | DOI: 10.1038/ncomms9480 |www.nature.com/naturecommunications



untransfected HEK293 control cells (Supplementary Fig. 2f,g).The frequency of the light pulse showed the most effect on pERKat full (not pulsed) light or at 5 s on/5 s off (SupplementaryFig. 2h), while light pulse length had little effect (SupplementaryFig. 2i), suggesting that opto-b2AR activity can be modulated viamanipulation of the light stimulus (that is, optodynamic). Thesekinetically parallel data sets suggest that photoactivation of opto-b2AR induces rapid and transient increases in receptor signallingknown to be mediated through b-adrenergic, Gas-dependentpathways11.

Previous studies have shown that rhodopsin and somechimeric GPCRs display dark activity, or are constitutively activein the absence of light7,27–29. To test this, we quantified levels ofpERK in HEK293 cells stably expressing opto-b2AR anduntransfected controls. In the absence of light stimulation, bothcell types showed similar levels of pERK, suggesting that thepresence of opto-b2AR does not induce constitutive activity(Supplementary Fig. 2j)6,30,31.

Opto-b2AR internalization and desensitization. We alsodetermined if opto-b2ARs are regulated through their receptorinternalization and desensitization kinetics in a manner similar tob-adrenergic receptors32,33. Photostimulation (1min) of opto-b2AR resulted in rapid receptor internalization within 2–5minfollowing light exposure (ton" 2.8min) that peaked within15min and returned to baseline levels 90min later (Fig. 2a–c,Supplementary Fig. 3a). In the continued presence ofisoproterenol (1 mM), b2AR internalization was temporallymatched (ton" 2.8min) to opto-b2AR and yielded valuessimilar to those obtained by other groups (Fig. 2a–d,

Supplementary Fig. 3a,b)23,34. To better mimic theoptodynamic stimulation of opto-b2AR, b2AR cells were treatedwith a 1min isoproterenol pulse. b2ARs internalized with similarkinetics (ton" 2.2min) as opto-b2AR, yet in contrast tocontinuous agonist exposure, b2ARs return to baseline morerapidly following a 1-min pulse of agonist (SupplementaryFig. 3c,d). If we compare Fig. 2d and Supplementary Fig. 3d,there are significant differences at the 60, 90 and 120min timepoints suggesting that b2ARs return faster to the membrane in theabsence of agonist, than in its presence (Supplementary Fig. 3e).These kinetic differences in receptor internalization and recyclinghighlight a significant limitation of traditional pharmacologicalapproaches, as it can be difficult to rapidly remove ligand fromthe cell media/bath, or in particular following in vivo infusion.This dynamic function of the optically active GPCR, highlightsthe utility of these types of optical approaches that mimic GPCRactivity at time scales matched to endogenous neuromodulator(NE) uptake and degradation35,36.

We next determined the functional recovery from desensitiza-tion of opto-b2AR in a real-time cAMP assay. Following an initiallight pulse (P1), cells produced less cAMP in response to a secondlight pulse (P2) at short interstimulus intervals (ISI; Fig. 2e).Varying ISIs showed complete functional recovery of cAMP overtime (trec" 49min) (Fig. 2e–g). We also observed that thereduced cAMP responses seen at short ISIs are not due todegradation of the 9-cis chromophore, but rather internalizationand desensitization of the receptor (Supplementary Fig. 3f)7.These results suggest that opto-b2AR has optodynamicallymatched kinetics and signal transduction profiles to b2AR andis a useful tool for spatiotemporal control of b-adrenergicsignalling.

g

cb

d!2AR

Opto-!2AR

a

f0 1 2 5 15 30!2AR

Opto-!2AR

pERK

!-actin

4442

pERK

!-actin

4442

0 1 2 5 15 30 (min)

e

Opto-!2AR

–1 0 1 2 3 4 50

20406080

100120

0 10 20 300

25

50

75

100

pER

K (

% m

ax)

On Off0.0

0.2

0.4

0.6

0.8

1.0

30 W cm–2

3 W cm–2

0.3 W cm–2

0.03 W cm–2

0.003 W cm–2

–1 0 1 2 3 4 50123456

cAM

P (

Fol

d ba

selin

e) Opto-!2AR EP50

0.9 ± 0.1 W cm–2

!2AR EC5014 ± 6 nM

–4 –2 0 2

–12 –10 –8 –6

0

20

40

60

80

100

cAM

P (

% m

ax)

cAM

P (

% m

ax) !2AR

Opto-!2AR

!2AR Opto-!2AR

pERK

"s !#ARR

PP

cAMP pERK

isoiso iso

"s !#ARR

PP

pERKcAMP pERK

Time (min) Time (min)

Time (min) Power (log W cm-2)

Isoproterenol (log M)

43

43

(min)!

(min

)

Figure 1 | Opto-b2AR and b2AR exhibit similar G-protein signaling mechanisms. (a) Both b2AR (ligand) and opto-b2AR (light) activate intracellularcAMP and pERK pathways. (b) Representative traces show light-induced activation of cAMP in response to increasing powers of light (5 s pulse) inHEK293 cells expressing opto-b2AR. (c) Power response curve of cAMP of opto-b2AR (blue) (EP50"0.9±0.1Wcm# 2; n"4 experiments). Isoproterenolincrease cAMP in b2AR (black) expressing cells (EC50" 14±6 nM; n"6 experiments). (d) Opto-b2AR (blue, n" 14 experiments) and endogenous b2ARin HEK293 cells (black, n"4 experiments) display similar kinetics of cAMP activation and deactivation in response to photostimulation (5 s pulse) andisoproterenol (1 mM) respectively (mean" solid line, s.e.m." shaded area). (e) Time constants of cAMP activation (ton) and deactivation (toff) fit fromtraces in d for opto-b2AR (blue) and b2AR (black). Activation (opto-b2AR"0.86±0.12min; n" 18 experiments and b2AR"0.77±0.14min; n"4experiments) and deactivation (opto-b2AR"0.85±0.03min; n" 15 experiments and b2AR"0.88±0.09min; n"4 experiments) time constants are notstatistically different. (f) Representative pERK immunoblots in response to isoproterenol (1mM) in b2AR and photostimulation (1min) in opto-b2AR.(g) Quantification of immunoblots for both b2AR (black, n" 5 experiments) and opto-b2AR (blue, n"8 experiments) displayed over time. All data areexpressed as mean±s.e.m. All light pulses are 473 nm, 1Wcm# 2 unless otherwise noted.

NATURE COMMUNICATIONS | DOI: 10.1038/ncomms9480 ARTICLE




Functionally selective opto-b2ARs receptors. Rhodopsin andbARs are both Class A GPCRs and hence share similar sequencehomology (Supplementary Fig. 4). It is this high homology thatfacilitated the generation of chimeric opto-b2ARs to mimic bARsintracellular signalling. In an effort to combine the spatio-temporal control of opto-b2AR with biased intracellular signallingcascades, we performed site-directed mutagenesis on opto-b2ARto generate optically active, functionally selective, G-protein-coupled receptors.

Opto-b2AR was altered to generate the putative arrestin-biased,opto-b2ARL72F,Y136G,Y224A or opto-b2ARLYY, and the putativeG-protein-biased, opto-b2ARS362A,S363G or opto-b2ARSS (Fig. 3a,Supplementary Fig. 4). It has been proposed and demonstrated by

several groups that G-protein-mediated signalling is rapid andtransient while arrestin-mediated signalling is slow and pro-longed10,11,17 (Fig. 3b). Here we show that activation ofendogenous bAR in HEK293 cells with isoproterenol shows arapid and transient increase in cAMP (Fig. 3c). In contrast,HEK293 cells overexpressing the arrestin-biased, b2ARTYY showsa marked reduction in cAMP. This reduction is suggestive of adecrease in G-protein interaction, and may also indicate apotential dominant negative effect of the mutant receptor onendogenous bAR. In contrast, HEK293 cells overexpressing theG-protein-biased b2ARSS had an exaggerated response in thecontinued presence of isoproterenol (Fig. 3c). Interestingly,HEK293 cells overexpressing b2ARWT show a longer and

a 0 min

!2AR

Opto-!2AR-YFP

15 min

1 min

30 min

5 min

60 min

Time (min)

% in

tern

aliz

atio

n

b

0 10 20 300

20

40

60

80

!2AR!on = 2.8 min

Opto-!2AR!on = 2.8 min

20

40

60

80%

inte

rnal

izat

ion

0

Time (min)0 1 906030155

****** ******

*

c

(122

)

(17)

(22)

(32)

(26)

(28)

(31)

Opto-!2AR

% in

tern

aliz

atio

n

0

20

40

60

80

0 1 120906030155Time (min)

*************** ***

d

(27)

(28)

(32)

(15)

(17)

(18)

(18)

(21)

!2AR

ISI (min)

cAM

P (

P2/

P1)

cAM

P (

P2/

P1)

2401200 60

1

0.5

018060 240 min1201552.5

Pulse 1 Pulse 2

Pulse 2

e f

2 min

1 fb

Opto-!2AR!rec = 49 min

ISI (min)24

018

012

060301552.50

0.2

0.4

0.6

0.8

1.0

1.2

**

g Opto-!2AR

***

*****(8

)

(8)

(7)

(10)

(9)

(8)

(8)

(8)

cAM

P

Figure 2 | Opto-b2AR and b2AR internalize and recover from desensitization. (a) Representative images show internalization of opto-b2AR–YFP inresponse to photostimulation (1-min pulse). Inset shows similar internalization of b2AR–YFP (colourized to black and white) in response to isoproterenol(1mM) at: 1, 5, 15, 30 and 60min post-stimulation. Scale bar,10mm. Arrowheads show examples of internalized punctate receptors. (b) Quantification ofinternalization in opto-b2AR (blue; ton" 2.8min) and b2AR (black; ton" 2.8min) with similar time constants of activation (ton). (c) Percent internalizationfor opto-b2AR–YFP (*Po0.05, ***Po0.001 via one-way ANOVA followed by Dunnett’s multiple comparison test to 0-min control; (n" number of cells pertime point). (d) Percent internalization in b2AR–YFP (***Po0.001 via one-way ANOVA followed by Dunnett’s multiple comparison test to 0min control;(n" number of cells per time point). (e) Representative traces of recovery from desensitization in opto-b2AR (P1 and P2; 5 s). (f) P2/P1 quantification ofFig. 1e, trec"49min (n"6–9 experiments per time point). (g) P2/P1 opto-b2AR functional recovery (*Po0.05, **Po0.01, ***Po0.001 via pairedStudent’s paired, two-tailed t-tests comparing P2 to P1 at each time point; (n" number of experiments). All data are expressed as mean±s.e.m. All lightpulses are 473 nm, 1Wcm# 2 unless otherwise noted.





sustained cAMP response in the presence of isoproterenol ascompared to endogenous bAR in HEK293 cells. These kineticdifferences are also clearly seen when repeated at 25 !C(Supplementary Fig. 5a). Further, all three receptor types showsimilar kinetic responses to forskolin, a general activator ofadenylate cyclase (Supplementary Fig. 5b), suggesting these

kinetic differences are mediated through Gas signalling and notdue to cAMP sensor expression.

In response to a 5-s pulse of blue light, opto-b2AR shows arapid and transient increase in cAMP, while the arrestin-biased,opto-b2ARLYY, shows an attenuated response to photostimula-tion (Fig. 3d). When we repeat this experiment at 25 !C,

a bModel

G-protein signalling

Arrestin signalling

Time

Res

pons

e

c dIsoproterenol

!2ARSS

!2ARWT

!2ARTYY

!2AR Opto-!2AR

Opto-!2AR

Opto-!2AR

Opto-!2AR

Opto-!2ARSS

Opto-!2ARSS

Opto-!2ARSS

Opto-!2ARLYY

Opto-!2ARLYY

Time (min)

On Off0

1

2

3

fOpto-!2AR

Opto-!2AR

!2AR

***

g

i

0 2 4 6 30 60 90 1200

25

50

75

100

Time (min) Time (min)0 2 4 6 30 60 90

0

25

50

75

100

pER

K (

% m

ax)

pER

K (

% m

ax)

***

***

*

!2AR

*

* *

****

******

h

Time (min)

HEK293 HEK293

eOpto-!2AR!2AR

****

***

***TYY LYY SSSS

cAM

P (

% W

T)

0

100

200

300

400

500

**HEK

0

2

4

6

8

! (m

in)

! (m

in)

On Off

**

***

**

!2ARSS

!2ARSS

!2ARWT

!2ARWT

!2ARTYY

!2ARTYY

HEK293

0 1 2 3 4 50

1

2

3

4

5

6

0 1 2 3 4 50

20

40

60

80

100

!2ARWT

(no iso)Opto-!2AR

(no light)

cAM

P (

Fol

d ba

selin

e)

cAM

P (

Fol

d ba

selin

e)

!2AR Opto-!2ARiso

iso iso

T Y Y

!#ARR

L Y Y

"s "s!#ARR

S S S S

Figure 3 | Mutations of b2AR and opto-b2AR alter intracellular signaling kinetics. (a) Schematic of point mutations. (b) Model of G-protein and arrestin/ MAP kinase signaling10. (c) Isoproterenol-induced (1 mM) cAMP kinetics of b2ARWT (black; n" 3), b2ARTYY (green; n" 3), b2ARSS (red; n" 3) andHEK293 cells (purple; n" 3). b2ARWTwith no isoproterenol (dashed brown; n" 3; mean" solid line, s.e.m." shaded area). (d) Light-induced (5 s) cAMPkinetics of opto-b2AR (dark blue; n"8), opto-b2ARLYY (dark green; n" 5) and opto-b2ARSS (dark red; n" 5). HEK293 controls and opto-b2AR with no light(dashed brown; n" 3) (mean" solid line, s.e.m." shaded area). (e) b2ARTYY (green; n" 3), opto-b2ARLYY (dark green; n" 5), HEK293 cells (purple;n" 3), b2ARSS (red; n" 3) and opto-b2ARSS (dark red; n" 5) cAMP compared with WT (*Po0.05, **Po0.01, ***Po0.001, ****Po0.0001 via one-wayANOVA followed by Dunnett’s multiple comparison test to WT). (f) Time constants of isoproterenol (1 mM)-induced cAMP of b2ARWT (black; n" 3),b2ARTYY (green; n" 3), b2ARSS (red; n" 3) and HEK293 cells (purple; n" 3) (*Po0.05, **Po0.01 via One Way ANOVA followed by Dunnett’s multiplecomparison test to WT). (g) Time constants of light (5 s pulse)-induced cAMP activation of opto-b2AR (dark blue; n"8) and opto-b2ARSS (dark red;n" 5) (***Po0.001 via Student’s unpaired, two-tailed t-tests to WT). (h) Time course of isoproterenol (1 mM)-induced pERK in b2ARWT (black; n" 5),b2ARTYY (green; n"4) and b2ARSS (red; n" 3) (*Po0.05, **Po0.01 via Student’s unpaired, two-tailed t-tests to WT). (i) Time course of light (1min)-induced pERK in opto-b2AR (dark blue; n" 9), opto-b2ARLYY (green; n" 8) and opto-b2ARSS (red; n" 9; *Po0.05, **Po0.01, ***Po0.001 via Student’sunpaired, two-tailed t-tests to WT). All data expressed as mean±s.e.m.





opto-b2ARLYY does show a small response (SupplementaryFig. 5c), yet is still significantly reduced from opto-b2AR, similarto the differences we and others have seen with b2ARTYYfollowing agonist treatment11. We next used an inhibitor of theGas subunit (NF 449). Using a concentration shown to effectivelyreduce isoproterenol-induced cAMP in b2ARWT cells(Supplementary Fig. 6a), we show that NF 449 (100 mM)reduces peak cAMP (Supplementary Fig. 6b,d). However, dueto the potential for off target activity at purinergic receptors, weused a nonselective P2 purinergic antagonist, suramin (100 mM),and showed no reduction in peak cAMP (SupplementaryFig. 6c,d), suggesting the effect is not due to purinergic receptorinteraction but likely to Gas inhibition. This reduction in real-time cAMP activity is similar to the profile of opto-b2ARLYY(Supplementary Fig. 5c), and suggests that the reduction in cAMPis likely due to reduced association between the Gas subunit andthe receptor.

In contrast, the G-protein-biased, opto-b2ARSS yields anexaggerated and prolonged cAMP response following photo-stimulation (Fig. 3d); an effect that is also reproduced at 25 !C(Supplementary Fig. 5c). Comparatively, b2ARSS and opto-b2ARSS show a significant robust enhancement of cAMPresponses when compared to WT, whereas both b2ARTYY andopto-b2ARLYY show a significant reduction (Fig. 3e).

The differences seen between receptors in activation (ton) anddeactivation (toff) time constants attest to the unique kinetics ofeach receptor type at both 37 !C (Fig. 3f) and 25 !C(Supplementary Fig. 5d). These kinetic differences may beamplified at cooler temperatures due to a reduction of cellularmetabolism, and/or variations in the temperature sensitivity ofthe ligand-induced conformation versus photoisomerization ofretinal. However, we did identify significant differences in thedeactivation time constants (toff) for both b2ARSS and opto-b2ARSS, which were significantly slower when compared withb2ARWT and opto-b2AR, respectively (Fig. 3f,g, SupplementaryFig. 5d). These slower rates suggest a lack of G-protein coupledreceptor kinase and arrestin recruitment to the membraneprolonging the activity of the receptor and yielding more cAMPoutput. In contrast, the reduced cAMP levels produced by bothb2ARTYY and opto-b2ARLYY is potentially due to inefficientcoupling to the G-proteins that are required to initiate cAMPproduction. It is also unlikely that these kinetic differences aredue to different light transduction properties of each receptortype as all three opto-b2ARs light power response curves yieldsimilar EP50 values (Supplementary Fig. 5e,f), although we didnote that the efficacy for generation of cAMP by opto-b2AR is notcompletely recapitulated comparedwith b2AR. This is most likelydue to the presence of endogenous b-adrenergic receptorsexpressed in HEK293 cells.

To confirm that the kinetic cAMP differences observedbetween opto-b2AR, opto-b2ARLYY and opto-b2ARSS are due tobiased intracellular signalling and not receptor expression levels,we quantified cell surface receptor expression. Using on-cellwesterns37,38 with a rhodopsin antibody, we show that rhodopsinexpression is not only significantly elevated in the three cell linesin comparison to untransfected HEK293, but are also equal(Supplementary Fig. 5g,h). We also calculated receptor surfaceexpression as a percent of total fluorescence and show thatboth opto-b2ARLYY and opto-b2ARSS have increased surfaceexpression compared to opto-b2AR (Supplementary Fig. 5i).The reduction in opto-b2AR surface fluorescence is most likelydue to the higher levels of diffuse internalized receptor at baseline(see 0-min time point in Fig. 2b,c).

Activation of MAP kinase cascades also shows dramatic kineticdifferences reminiscent of the model proposed by Luttrell andGetsey–Palmer (Fig. 3b)10. In the presence of isoproterenol,

b2ARWT and b2ARSS show a marked increase in ERKphosphorylation that peaks within 2–5min and is rapidlyattenuated, suggesting a G-protein phase of activation, whileb2ARTYY, shows a significantly reduced level of pERK that isprolonged and sustained (Fig. 3h, Supplementary Fig. 7a).Likewise, opto-b2AR and opto-b2ARSS show a comparabletemporal profile in the initial phase of ERK activation whileopto-b2ARLYY remains significantly slower and sustained (Fig. 3i,Supplementary Fig. 7b), in a similar manner to the kinetics ofb2ARTYY. Further, when examined as fold increase over baseline,the effects on pERK by b2ARWT and b2ARSS, show a markedincrease in ERK phosphorylation that peaks within 2–5min andis rapidly attenuated, while b2ARTYY shows a reduced andsustained level of pERK (Supplementary Fig. 7c). Likewise, opto-b2ARSS and opto-b2ARLYY differed significantly from opto-b2AR,particularly in the initial stages of ERK activation (SupplementaryFig. 7d). The kinetics of both cAMP and ERK activity areremarkably similar between b2AR, opto-b2AR, and theirrespective G-protein-biased and arrestin-biased mutantsstrongly supporting the current proposed model of rapid andtransient G-protein-mediated signalling, and slower sustainedarrestin-mediated signalling.

Functionally selective opto-b2AR mutant internalization.Activity of GPCRs is usually terminated following phosphoryla-tion via G protein-coupled receptor kinases (GRK) followed bysubsequent recruitment of arrestin, leading to receptor inter-nalization via clathrin coated pits39. To ascertain thecharacteristics of optically active, functionally selective mutantopto-b2ARs following desensitization we captured a time courseof receptor internalization following photostimulation. Opto-b2ARSS did not internalize at any time point tested following lightstimulation, suggesting that the ability of arrestin to initiateinternalization is significantly compromised in this receptor(Fig. 4a,b, Supplementary Fig. 8). Conversely, the arrestin-biasedmutant, opto-b2ARLYY was able to internalize rapidly followinglight stimulation, peaking at 15min (Fig. 4c,d). In comparison toopto-b2AR at the 15-min time point, we see that opto-b2ARLYY issignificantly slower in reaching maximal internalizationsuggesting less efficient coupling with G-protein subunitsthat facilitate GRK recruitment (Fig. 4e). In addition, previousstudies looking at b2ARTYY also demonstrated a decrease inrecruitment of arrestin which could also explain why opto-b2ARLYY does not reach the same levels of internalization asopto-b2AR11.

In addition to demonstrating a lack of internalization for opto-b2ARSS, we determined whether opto-b2ARSS receptors function-ally desensitize. Following an initial pulse of light (P1), opto-b2ARSS displayed a typical real-time cAMP response (Fig. 4f).Following a subsequent light pulse (P2), opto-b2ARSS showed amild reduction in cAMP, only at the earliest time point tested (P1versus P2 at 5min, P" 0.0308 via Student’s paired t-test).However, overall opto-b2ARSS did not show a significantreduction in P2-mediated cAMP (Fig. 4f–h). In comparison,opto-b2AR displayed a significant reduction in cAMP suggestingthat the receptor functionally desensitized, and eventuallyrecovered function within 120min (Figs 2e–g and 4g). Opto-b2ARSS also did not lose its ability to generate cAMP at most timepoints tested, unlike opto-b2AR, which had a dramatic loss incAMP signalling at shorter interstimulus intervals. Due to theabsence of a detectable cAMP signal at 37 !C, no functionalrecovery data were collected for opto-b2ARLYY. These datasuggest that the putative G-protein-biased opto-b2ARSS does notcouple to arrestin as it is not internalized and desensitizedfollowing photostimulation, yet the putative arrestin-biased





opto-b2ARLYY mutant interacts with arrestin since it is robustlyinternalized following photostimulation.

Photostimulation of opto-b2AR promotes neuronal firing. Thisseries of in vitro data allowed us to illustrate that b2AR and opto-b2AR share similar properties of cAMP generation, MAP kinaseactivation and receptor internalization. In addition, utilizingoptically active, functionally selective receptors, we gained addi-tional insight into how these tools bias receptor function towardsG-protein or arrestin-mediated signalling effects in cAMP, MAPkinase activation, and receptor internalization/desensitization.Therefore, we next examined how opto-b2AR functions in abiologically relevant neuronal context known to be modulatedthrough endogenous noradrenergic activation.

We first determined whether photoactivation of opto-b2AR incells known to express wild-type bARs, the BLA, could promotetime-locked signalling effects in vivo and subsequent excitation ofBLA neurons. We virally targeted opto-b2AR-mCherry toexcitatory neurons under the control of the CaMKIIa promoterto the BLA (opto-b2ARBLA/CaMKIIa) (Fig. 5a–d). Utilizing a16-channel optrode array in the BLA for single-unit extracellularrecordings we delivered light stimulation (20-s pulse) in vivo and

demonstrated a significant increase in neuronal firing of BLAneurons (Fig. 5e–j). Repeated light pulses (both 5- and 20-sconstant light) also showed sustained activity over time(Supplementary Fig. 9a,b), whereas photostimulated cells incontrol virus (empty-vector lenti-virus) injected mice showed noeffect on neuronal firing (Fig. 5i,j). In some instances, cells did notrespond to light stimulation or showed inhibitory effects ofphotoactivation of opto-b2ARBLA/CaMKIIa (5% and 9%, respec-tively; Fig. 5g, Supplementary Fig. 9c). These differences inneuronal activity may potentially be due to lack of expression ofthe viral construct, or lateral inhibition from local inhibitoryneurons in this region40–42. In addition, we show that activationof opto-b2ARBLA/CaMKIIa in vivo showed a slow onset ofactivation (ton) and inactivation (toff), similar to previousreports (Supplementary Fig. 9d)6. We also confirmed that thepresence of opto-b2ARBLA/CaMKIIa in vivo does not induceconstitutive activity and alter neuronal activity as baseline firingrates between opto-b2ARBLA/CaMKIIa (n" 41 units) and viralcontrol (n" 11 units) neurons are not different (P" 0.2593 viaunpaired Student’s t-test) (Fig. 5j). These results demonstrate thatphotoactivation of opto-b2ARBLA/CaMKIIa signalling can robustlyincrease neuronal activity, in a time-locked and spatiallyrestricted manner in vivo.

0 1 5 15 600

20

40

60 Opto-!2ARSSb

(15)

(19)

(13)

(22)

(19)

Opto-!2ARLYY

0 1 5 15 600

20

40

60

******

***

***

(55)

(16)

(24)

(26)

(9)

d

Time (min)

% in

tern

aliz

atio

n%

inte

rnal

izat

ion

a

Time (min)

c

0 min 1 min 5 min 15 min 60 minOpto-!2ARSS

0 min 1 min 5 min 15 min 60 minOpto-!2ARTYY

0.0

0.2

0.4

0.6

0.8

1.0

Time (min)

P2/

P1

P2/

P1

*

15 30

*

50

20

40

60

80

OB LYY

Opto-!2AR@ 15 min

Opto-!2ARSSOpto-!2AR

****

5 15 30 6012018

024

00.00.20.40.60.81.01.2

Time (min)

30 120 min155

Pulse 1 Pulse 2

f

2 min

1 fb

60

Opto-!2ARSS

Opto-!2ARSSe g h

(5)

(5)

(4)

(4)

(4)

(4)

(4)

(5)

(5)

(5)

(8)

(8)

(9)

*

% in

tern

aliz

atio

n

Figure 4 | Opto-b2ARSS and opto-b2ARLYY internalization and desensitization. (a) Representative images show opto-b2ARSS–YFP (pseudocoloured red)in response to photostimulation (1min). Scale bar, 10mm. (b) Quantification of internalized opto-b2ARSS–YFP (red; (n" number of cells per time point).(c) Representative images show internalization of opto-b2ARLYY–YFP (pseudocoloured green) following light exposure (1min). Arrowheads denoteinternalized receptor. Scale bar, 10mm. (d) Quantification of internalized opto-b2ARLYY–YFP (green; (n" number of cells per time point; ***Po0.001 viaOne-Way ANOVA followed by Dunnett’s multiple comparison test to 0-min control). (e) Comparison of internalization at 15min post photostimulation foropto-b2AR (dark blue; n" 32 cells), opto-b2ARLYY (dark green; n" 26 cells) (****Po0.0001 via Student’s unpaired, two-tailed t-tests). (f) Representativetraces show recovery from desensitization in opto-b2ARSS expressing cells. P1 is cAMP response to initial light. P2 is cAMP response to second light pulsefollowing different interstimulus interval. (g) Comparison of recovery from desensitization between opto-b2AR (dark blue) and opto-b2ARSS (dark red) atlow interstimulus intervals (n" number of of experiments; *Po0.05 via Student’s unpaired, two-tailed t-tests). (h) Quantification of opto-b2ARSS recoveryfrom desensitization in (f) (*Po0.05, paired two-tailed t-tests comparing P2 with P1 at each time point; (n" number of experiments). All data expressed asmean±s.e.m. All light pulses are 473 nm, 1Wcm# 2.





0 80 160 240Time (s)

% in

ligh

t

40

20

0

Tim

e in

ope

n (s

)

60

40

20

80

100

01

1

2 3 4 5

2 3 4 5

6Bins (1 min)

** ** *** ** ** **

High

Low

Control

Elevated zero maze

Dis

tanc

e (c

m)

RF antenna RF poweramplifier

LV

RF signalgenerator 60

100

80

Tim

e in

dar

k (s

) 250

200

150

100

50

0

Bins (1 min)

**

**

Light dark box140

70

0Late

ncy

(s)

*

Dar

k

Control

Control

Control

0

200

400

600

800

a

e

h

k

n o p q

l m

i

j

f

g

b c d

lenti CaMKII" opto-!2AR-mCherry

Rec

150 mV50 ms

ON

ON ON

7 s50 mV1

1

2

2

3

3

500 ms50 mV

Time (s)

Firi

ng r

ate

(Hz)

00

10

10

20

20

30

30

40

40 50Time (s)

0 10 20 30 40 50

Increase (86%)Decrease (9%)Unresponsive (5%)

Firi

ng r

ate

(Hz)

Off On Off

Control

0

5

10

15

20***

Firi

ng r

ate

(Hz)

0

10

20

30

40Opto-!2AR

Opto-!2AR

Opto-!2AR

Opto-!2AR

Opto-!2AR

Opto-!2AR

Control

NisslOpto-!2AR-mCherry

CeA

BLANissl NisslOpto-!2AR-mCherry Opto-!2AR-mCherry

Figure 5 | BLA neurons expressing Opto-b2AR promote anxiety-like behaviours. (a) Bilateral viral injection sites. (b–d) Lenti-CaMKIIa-opto-b2AR-mCherry expression in BLA (Scale bar, 50mm (b), 25mm (c) and 5mm (d). (e) Representative single-unit recording (with expanded views at 1, 2 and 3) ofBLA neuron reversibly increases in firing rate during light stimulation (20 s). (f) Representative waveforms. (g) Distribution of neurons that increase(green), decrease (grey) or show no change (black) in response to photostimulation. (h) Representative histogram shows increase in firing rate in responseto photostimulation of opto-b2AR expressing BLA neuron. (i) Representative histogram shows no change in firing rate in response to photostimulation incontrol animals. (j) Opto-b2AR BLA expressing neurons (blue; n"41 units) shows reversible increase in neuronal firing rate in response to 20 sphotostimulation (***Po0.001; one-way repeated measures ANOVA). Virally injected controls (black; n" 11 units) are not light responsive.(k) Representative heat maps show behaviour in EZM, lighter colours indicate more time spent in a position. (l) Open arm cumulative time course ofphotostimulated opto-b2AR (blue, n" 7) and control animals (black; n" 10) (5 s off/on; **Po0.01, ***Po0.001; multiple Student’s unpaired, two-tailedt-tests). (m) Viral control (black, n" 10) and opto-b2AR (blue, n" 7) expressing animals do not show differences in total distance traveled during the EZMtrial (Student’s unpaired, two-tailed t-test, P"0.1657). (n) Viral injection site, unilateral m-ILED implant and brief overview of wireless transmission system.(o) Representative traces of control (black) and opto-b2AR (blue) animal in LDB. (p) Wireless photostimulation of mice expressing opto-b2AR (blue; n" 7)in the BLA enter the dark box faster than viral controls (black; n" 11) (*Po0.05; log-rank (Mantel-Cox) Test; inset *Po0.05 via Student’s unpaired,two-tailed t-test). (q) Opto-b2AR animals (blue) spend more cumulative time in the dark box than viral controls (black; *Po0.05; multiple Student’sunpaired, two-tailed t-tests). All data expressed as mean±s.e.m. All light pulses are 473 nm, 1Wcm# 2.





Photostimulation of opto-b2AR promotes anxiety-like behaviour.We next explored whether spatiotemporal activation ofb-adrenergic signalling in vivo was sufficient to promote robustbehavioural effects. We bilaterally expressed opto-b2ARBLA/CaMKIIa

in the BLA (Fig. 5a–d, Supplementary Figs 10a,b and 11) andphotoactivated (5 s on, 5 s off) animals during two differentyet widely used rodent anxiety-like behavioural models, theelevated zero maze (EZM) and the light–dark box (LDB)43,44.Photoactivation of opto-b2ARBLA/CaMKIIa signalling in the BLAproduced rapid and sustained anxiogenic-like behaviour with micespending significantly more time in the closed arm of the EZM(Fig. 5k,l), with no changes in locomotor activity (Fig. 5m,Supplementary Fig. 10c,d). We next examined the effects of opto-b2ARBLA/CaMKIIa activation on rapid acute anxiety-like behaviourusing the LDB45. In this 10min assay, rapid entry into a dark boxfrom a light chamber and increased time spent in the dark box aremeasures of anxiogenesis46. This model consists of an enclosed darkenvironment with a small doorway inaccessible to fiber opticimplants, we therefore utilized our recently developed wirelessoptogenetic approach to drive microscale inorganic light emittingdiodes (m-ILEDs)47,48 (Fig. 5n, Supplementary Fig. 12a–d). Here wedeveloped and utilized a ‘wireless 2.0’ version, that is much smallerallowing for even more unrestricted animal activity, to remotelyphotoactivate opto-b2ARBLA/CaMKIIa injected mice. In theseexperiments, photoactivation of opto-b2ARBLA/CaMKIIa causedmice to rapidly enter the dark chamber as demonstrated by botha significant decrease in latency to enter the dark, and significantlymore time spent in the dark during the trial (Fig. 5o–q) with noeffect on locomotor behaviour (Supplementary Fig. 12e,f).Altogether, these results demonstrate that activation of opto-b2ARsignalling in vivo can robustly modulate behavioural responses inbAR expressing regions. Furthermore, this can be done in awireless manner, providing a unique method for spatiotemporalengagement of GPCR signalling in vivo in an unrestricted manner,such as the home cage, or other more diverse behaviouralenvironments.

DiscussionManipulation of endogenous intracellular GPCR signallingin vivo has historically required pharmacological techniques. Tosome degree, optogenetics has filled this niche by allowing cell-type specificity in addition to spatiotemporal control49. However,a vast majority of optogenetic studies utilize light sensitive ionchannels or pumps providing only binary control of neuralactivity. Here we show for the first time, that a chimericrhodopsin/b2-adrenergic receptor (opto-b2AR) behaves in akinetically similar manner, in a host of signalling readouts, tothe human b2AR. We show that opto-b2AR activates GPCRsignalling in a power dependent manner, mimicking theconcentration dependence of isoproterenol on b2AR.Importantly, we also report that the kinetic responses of thesetwo receptors are similar and that opto-b2AR not onlyinternalizes following stimulation, but also functionallydesensitizes, sharing the same kinetics as b2AR. Taken togetherthese data strongly support using opto-b2AR as a tool to mimicb2AR activity (see summary data in Supplementary Table 1). Inaddition to modelling b2AR at the receptor level, incorporation ofopto-b2AR as a tool allows for modelling the kinetics ofendogenous NE release as the activation and deactivation ofthis receptor is controlled instantly through optogenetictechniques. In contrast, small molecules may confound analysisat the circuit level, as drug clearance becomes an issue andsystemic half-life makes conclusions regarding kinetics ofbehavioural onset/offset difficult to interpret. Conversely, itmust be taken into account that opto-b2AR is not identical to

b2AR, and it is not currently known whether the trafficking andrecycling pathways are the same in vivo. Future studies will needto further characterize the intracellular dynamics of opto-b2ARin vivo using real-time imaging approaches.

In addition to showing the utility of optically activated GPCRs,we show here for the first time, optically active, functionallyselective GPCRs. The concept of functional selectivity or biasedagonism is becoming increasingly important to understandGPCR biology and in the development of novel therapeu-tics9,50,51. Here we show two functionally distinct receptors: thearrestin-biased, opto-b2ARLYY, and the G-protein-biased, opto-b2ARSS. Through an array of biochemical analyses we show thatopto-b2ARLYY: mobilizes less cAMP, shows reduced andprolonged activation of ERK and internalizes in response tolight stimulation in comparison to opto-b2AR. Conversely, opto-b2ARSS: shows enhanced and transient cAMP signalling, showselevated transient activation of ERK, lacks internalization, andshows little desensitization as compared with opto-b2AR. Theseoptically sensitive, functionally selective GPCRs provide theadvantages of not requiring a biased ligand and have directspatiotemporal control over receptor activation and deactivation(see summary data in Supplementary Table 2).

The information gathered from in vitro studies regarding theintracellular characteristics of opto-b2AR in a heterologousexpression system provided confidence that opto-b2AR closelymimics the pharmacological properties of b2AR. To expand onthis potential utility, we further validated this approach in vivo, ina biologically relevant anxiety circuit known to be modulatedthrough noradrenergic activation, the BLA. Given that the BLA iscomposed mostly of excitatory neurons52,53, we packagedopto-b2AR under the control of the CaMKIIa promoter todrive robust expression in the BLA (opto-b2ARBLA/CaMKIIa).Photostimulation of opto-b2ARBLA/CaMKIIa in vivo altered thebaseline firing properties of BLA neurons and revealed aheterogeneous population of cells. While the majority of cellsincreased firing rate, some exhibited no change and some showeda reduction in firing rate. Utilizing traditional pharmacologicalapproaches would not allow for the isolation of a cell type withina given anatomical region and hence the roles of individual celltypes, and adrenergic receptor subtypes were not previouslypossible.

When opto-b2ARBLA/CaMKIIa was activated in vivo, miceexhibited an anxiety-like phenotype. Anxiogenesis was demon-strated in two commonly used models of anxiety-like behaviourin rodents, the EZM43 and the LDB46. Utilization of the LDB wasonly possible due to new wireless optogenetic technology recentlydeveloped47,48. Here we also demonstrate for the first time, real-time wireless control of GPCR signalling. Opto-b2ARBLA/CaMKIIa

expressing mice, when stimulated wirelessly, displayed ananxiety-like phenotype in the LDB assay. This technologyallowed us to use a common model of anxiogenic behaviour,but also sets the stage for important future work utilizing wirelessmanipulation of GPCR signalling in vivo allowing for a largeexpansion of GPCR-mediated behaviours that can be paired withcurrent optogenetic techniques. That being said, optogeneticscomes with certain caveats. It is possible that photostimulation inthe BLA also activates axon collaterals, and hence confoundsubsequent behavioural output, or that more complex cell typesand expression patterns are needed to truly hone in on b2ARsignalling in real-time, in vivo. However, this is a unique firstapproach, and as additional mouse genetic tools and targetingschemes become available54, further isolation of cell types andGPCR signalling in neural circuits will become increasinglypossible.

Taken together our data demonstrate that chimeric, opticallyactive GPCRs can behave in a similar manner to their





endogenous counterparts, making them particularly useful forboth in vitro and in vivo applications. Future studies will utilizethese tools to engage the diversity of GPCR signalling in vivo anddetermine if spatiotemporal control of biased signalling promotesa series of pluridimensional behavioural phenotypes. In thisreport, we were able to show that the integration of the excitatorynoradrenergic influence in the BLA is mediated via activation ofb-adrenergic pathways that ultimately promotes anxiogenic-likebehavioural states. Some exciting additional extensions includeusing them for examining the role of signalling inside cells, sincethe light used to activate these receptors can penetrate the cellmembrane. Recent work from von Zastrow and colleagues hasshown that Gas-coupled receptors have multi-phasic signallingproperties, one that is membrane bound, and another that occursfrom endosomes55,56. Other uses of these receptors include ameans to better understand the process of G-protein activationwithout confounds of ligand binding. These findings have broadimplications for our understanding of the mechanisms of GPCRsignalling in vivo and in the development of novel therapeuticsthat depend on interactions with GPCRs.

MethodsChemicals. Isoproterenol (10mM dimethyl sulfoxide (DMSO) stock), forskolin(10mM DMSO stock) and NF 449 (10mM water stock) were obtained from TocrisBiosciences. 9-cis-retinal (10mM DMSO stock and shielded from light) wasobtained from Sigma. Vehicle controls were used in all cases.

Light stimulation. All light stimulation was constant at 473 nm, at powers andpulse lengths indicated in figure legends. Light was delivered via a 100-mW,473-nm diode-pumped solid-state laser (OEM Laser Systems)4,57.

Cell culture. HEK293 cells (ATCC, CRL-1573) were grown in DMEM supple-mented with 10% fetal bovine serum containing 1$ pen/strep (Invitrogen) andmaintained at 37 !C in a humidified incubator with 5% CO2. Stable HEK293 celllines expressing pcDNA3.1 containing opto-b2AR, b2AR and their respectivemutants were generated by transfecting HEK293 cells with identical amounts ofcDNA (10 mg in 100-mm dishes) for 4 h using JetPrime (Polyplus) reagent permanufacturer’s instructions. Cells were placed under selective pressure with G418(400 mgml# 1) and FACS (Washington University FACS Sorting Facility) sortedfor equal yellow fluorescent protein (YFP) fluorescence to ensure equivalentreceptor expression. All experiments utilizing opto-b2AR and its mutants wereperformed in the dark in the presence of 1 mM 9-cisretinal (Sigma).

Site-directed mutagenesis. Human adrenergic receptor beta 2 (ADRB2), GeneBank Accession Number: NM_000024.3 was purchased from cDNA.org.

b2AR–YFP fusion protein was created using human ADBR2 as a template in ahigh fidelity PCR using the following primers: with primers EcoRI-b2AR-forward(50-AGT GTG GTG GAA TTC GAT TAT CCA CC-30) and XhoI-b2AR-reverse(50-CCT CTA GAC TCG AGT aAC AGC AGT GA-30) with the stop codonmutated to leucine. The PCR product was then digested and cloned into the50 EcoRI and 30 XhoI sites of pcDNA3–YFP (Addgene Plasmid 13033).

b2ARSS–the ‘G-protein biased’ mutation (S355A/S356G) was created usinghuman ADBR2 as a template in a high fidelity PCR using the following primers:internal primers containing point mutations in lower case, forward (50-TAT GGGAAT GGC TAC gCC gcC AAC GGC AAC ACA GG-30) and reverse (50-CCT GTGTTG CCG TTG gcG GcG TAG CCA TTC CCA TA-30) with external primersEcoRI-b2AR-forward (50-AGT GTG GTG GAA TTC GAT TAT CCA CC-30) andXhoI-b2AR-reverse (50-CCT CTA GAC TCG AGT aAC AGC AGT GA-30) withthe stop codon mutated to leucine. The PCR product was then digested and clonedinto the 50 EcoRI and 30 XhoI sites of pcDNA3–YFP (Addgene Plasmid 13033).

b2ARTYY–the ‘arrestin biased’ mutation was obtained from Robert Lefkowitz(Duke University). We created a fusion protein between b2ARTYY and YFP.b2ARTYY was amplified via high fidelity Taq with the following primers: EcoR1-b2ARTYY-forward (50-TAC AAG GAC GAT GAa ttC atg GGG CAA CCC GGGAAC GGC A-30) and XhoI-b2ARTYY-reverse (50-GCG GCC GTT ctc gag tgc CAGCAG TGA GTC ATT TGT ACT-30) with stop codon changed to an alanine. ThePCR product was then digested and cloned into the 50 EcoRI and 30 XhoI sites ofpcDNA3–YFP (Addgene Plasmid 13033).

Opto-b2ARSS–the ‘G-protein biased’ mutation (S362A/S363G) was createdusing a COBALT alignment against human b2AR (S355A/S356G). Opto-b2AR wasobtained from Karl Deisseroth (Stanford University) and used as the template in ahigh-fidelity PCR using the following primers: internal primers containing pointmutations in lower case, forward (50-TCC AAA GCG TAC GGA AAT GGC TATgCA gga AAC AGC AAC GGA AAG ACT GAT TAT-30) and reverse (50-ATA

ATC AGT CTT TCC GTT GCT GTT tcc TGc ATA GCC ATT TCC GTA CGCTTT GGA-30) with external primers HindIII- opto-b2ARWT-forward (50-CCAAGC TGG CTA GTT AAG CTT GCC ACC-30) and NotI–opto-b2ARWT–rev(50-GCT CAC GGC GGC CGC GGC CGG AGC GAC-30). PCR product was thendigested and cloned into the 50 HindIII and 30 NotI sites of pcDNA3.1–YFP(generously provided by Deisseroth Lab).

Opto-b2ARTYY, the ‘arrestin biased’ mutation was created using a COBALTalignment between opto-b2AR and b2ARTYY point mutations: L72F, Y136G,Y224A were generated using the opto-b2ARWT as a template in a high-fidelity PCRusing the following primers: opto-b2ARL72F-forward (50-CTC CAA ACC GTG TTtAAC TAC ATA CTC CTT-30), opto-b2ARL72F-reverse (50-AAG GAG TAT GTAGTT aAA CAC GGT TTG GAG-30); opto-b2ARY136G-forward (50-TTG GCC ATAGAG AGG ggC GTG GTG GTC ACA-30), opto-b2ARY136G-reverse (50-TGT GACCAC CAC Gcc CCT CTC TAT GGC CAA-30); opto-b2ARY224A-forward (50-ATCTTT TTC TGT gcC GGC AGG GTG TTC CAG-30), opto-b2ARY224A-reverse(50-CTG GAA CAC CCT GCC Ggc ACA GAA AAA GAT-30) with the externalprimers HindIII- opto-b2ARWT-forward (50-CCA AGC TGG CTA GTT AAGCTT GCC ACC-30) and NotI- opto-b2ARWT-rev (50-GCT CAC GGC GGC CGCGGC CGG AGC GAC-30). PCR product was then digested and cloned into the50 HindIII and 30 NotI sites of pcDNA3.1–YFP (generously provided by DeisserothLab).

All mutations were confirmed by DNA sequencing (AGCT Inc., Wheeling, IL).

Real time cAMP assay. Stable HEK cell lines containing opto-b2AR, b2AR andtheir respective mutants were transfected with the pGloSensor-22F cAMP plasmid(Promega E2301) using JetPrime (Polyplus) transfection reagent per manu-facturer’s instructions. Stable co-transfected cells were maintained under bothG418 (400 mgml# 1) and hygromycin (200 mgml# 1) selective pressure. The daybefore an experiment, cells were plated on 96-well tissue culture treated plates(Costar) and allowed to recover overnight at 37 !C, 5% CO2. Optimal results wereobtained when b2AR (and respective mutants) were plated at 20 K cells per welland when opto-b2AR (and respective mutants) were plated at 100 K cells/well. Thenext day, media was replaced with 2% GloSensor reagent (Promega) suspended inCO2-independent growth medium (Gibco) and incubated for 2 h at 37 !C or 25 !Cdepending on experiment. For real time cAMP, a baseline was first obtained withno treatment by recording relative luminescent units (RLUs) every 6 s for 1minusing a SynergyMx microplate reader (BioTek; Winooski VT; USA). Drug or lightwould then be used to stimulate the cells, and subsequent RLUs recorded every 6 sfor 5–10min depending on experiment. For data expressed as cAMP (fold base-line), RLUs for 1min of baseline were averaged and all subsequent RLUs were thendivided by this average. For data expressed as cAMP (% max), raw RLUs wereentered into GraphPad Prism (v5.0d, GraphPad Software, San Diego CaliforniaUSA) and the normalization function used to assign the lowest RLU a value of 0%and the highest RLU a value of 100%. Time constants were calculated in GraphPadPrism using one-phase association (Y"Y0! (Plateau#Y0)$ (1# exp(#K$ x)))and one-phase decay (Y" (Y0# Plateau)$ exp(#K$X)! Plateau) nonlinearregression analyses yielding a time constant value (t).

Concentration/power response curves. For b2AR experiments, baseline relativeluminescence recordings were taken and cells exposed to varying concentrations ofisoproterenol in serial half log dilutions diluted from 10mM stock in DMSO. RawRLUs were normalized to the peak response evoked by isoproterenol and repre-sented as cAMP (% max). Subsequent concentration response curves were fit usingstandard nonlinear regression to obtain EC50 values using GraphPad Prism andexpressed as mean±s.e.m., with triplicate data points averaged per experimentwith a total of six individual experimental replicates. For opto-b2AR experiments,individual wells were exposed to a 5-s blue light pulse (473 nm) at varying powersto generate a power response curve with data normalized to maximal cAMPresponse. Subsequent power response curves were fit with standard nonlinearregression to obtain EP50 values using GraphPad Prism. Data are expressed asmean±s.e.m.

Recovery from desensitization. Opto-b2ARWT and opto-b2ARSS cells grown in96-well plates were individually exposed to a single 5-s blue light pulse (473 nm,1Wcm# 2) called P1 (pulse 1) with the subsequent cAMP response recorded.After varying interstimulus intervals (0, 2.5, 5, 15, 30, 60, 120, 180 and 240min)each well was then re-exposed to a second light pulse (P2), and subsequent cAMPresponse recorded. Peak RLU for P2 were then divided by peak RLU for P1.These points were then fit with a one-phase association (Y"Y0! (Plateau#Y0)$(1# exp(#K$ x))) curve in GraphPad Prism to obtain a time constant of recoveryfrom desensitization (trec). Data are expressed as mean±s.e.m.

Immunoblots. Western blots for phospho-MAPKs were performed as describedpreviously58. Cells were grown overnight in 6- or 12-well plates, then serum-starved a minimum of 4 h before treatment to avoid serum growth factor-inducedMAPK activation. Cells were treated at various time points at 37 !C and thencollected in lysis buffer (50mM Tris-HCl, 300mM NaCl, 1mM EDTA, 1mMNa3VO4, 1mM NaF, 10% glycerol, 1% Nonidet P-40, 1:100 of phosphataseinhibitor mixture set 1 (Calbiochem), and 1:100 of protease inhibitor mixture set 1





(Calbiochem) on ice. Lysates were sonicated for 15 s, centrifuged for 20min(14,000 rcf at 4 !C), then stored at # 20 !C. Protein concentration was determinedby Pierce BCA (Thermo Scientific) with bovine serum albumin as the standard.Each gel contained the same amount of total protein and varied between 20–40 mg,depending on experiment. Nondenaturing 10% bisacrylamide precast gels(Invitrogen) were run at 180V for 1 h. For determination of molecular weights,pre-stained molecular weight ladders (Life Technologies; Novex Sharp ProteinStandard; LC5800) were loaded along with protein samples. Blots were transferredto nitrocellulose (Whatman, Middlesex, UK) for 1.5 h at 30mV, blocked in 5%bovine serum albumin in tris-buffered saline (TBS) for 1 h, incubated overnight at4 !C with goat anti-rabbit phospho-ERK 1/2 (Thr-202/Tyr- 204) antibody (1:1,000,Cell Signaling) and mouse b-actin (1:20,000, Abcam). Membranes were thenwashed 4x for 10min in TBST (Tris-buffered saline, 1% Tween 20) and thenincubated with the IRDyeTM 800 (1:5,000, donkey anti-rabbit) and 700 (1:20,000,donkey anti mouse) conjugated affinity purified IgG in a 1:1 mixture of 5% milk/TBS and Li-Cor blocking buffer (Li-Cor Biosciences, Lincoln, NE) for 1 h at roomtemperature in the dark. Membranes were then washed three times for 10min inTBST then once for 10min in TBS to remove Tween. Immunoblots were scannedusing the Odyssey infrared imaging system (Li-Cor Biosciences). Band intensitywas measured using Odyssey software following background subtraction andintegrated intensity measured for each band in high-resolution pixels. All pERKbands were normalized to b-actin, as an equal protein loading control. For dataexpressed as pERK (fold baseline), raw pERK/actin values were normalized to the0-min time point. For data expressed as pERK (% max), raw pERK/actin valueswere entered into GraphPad Prism (v5.0d, GraphPad Software, San Diego CA,USA) and the normalization function used to assign the lowest pERK/actin a valueof 0% and the highest raw pERK/actin a value of 100%. Data are expressed asmean±s.e.m. Concentration-response data were fit using nonlinear regression inGraphPad Prism. Positive controls are cell lysates obtained from HEK293 cellsstably expressing the Nociceptin/Orphanin FQ Opioid Peptide Receptor (NOPR)harvested following a 5-min incubation in nociceptin (1 mM)58. These independentpositive controls are to ensure successful execution of western blots and in no wayaffect the data presented. Full unaltered scans of all western blot images withcorresponding molecular markers can be found in Supplementary Figs 13–16.

On cell western. On cell westerns were performed following previously publishedprotocols37,38. HEK293 cells stably expressing opto-b2AR, opto-b2ARSS, opto-b2ARLYY and untransfected control HEK293 cells were plated on 24-well tissueculture treated plates at 200K cells/well and grown in DMEM containing 10% fetalbovine serum and penicillin/streptomycin at 37 !C in 5% CO2. Plate was placed onice, media removed and cells immediately fixed with 4% paraformaldehyde for30min at room temperature. Cells were washed five times for 30min in PBS,blocked for 90min in Li-COR Odyssey Blocking Buffer at room temp with gentlerocking. Cells were incubated overnight at 4 !C in mouse rhodopsin antibody(4D2) (Novus NBP1–48334) diluted 1:1,000 in Odyssey Blocking Buffer. Wash fivetimes in Tris-buffered saline containing 0.1% Tween-20 (TBST) for 30min.Incubate in Li-Cor IRDye 680RD (donkey a mouse), diluted 1:1,000 in Odysseybuffer! 0.1% TWEEN-20 at room temperature for 1 h. Wash five times in TBSTfor 30min. After final wash, remove solution from wells, tap or blot gently onpaper towels to remove traces of wash buffer and scanned using the Odysseyinfrared imaging system (Li-Cor Biosciences). Well intensity was measured usingOdyssey software following background subtraction and integrated intensitymeasured for each well in high-resolution pixels. Data were analysed in GraphPadPrism and are expressed as mean±s.e.m.

Receptor internalization. b2AR–YFP, opto-b2ARWT–YFP, opto-b2ARSS–YFPand opto-b2ARLYY–YFP were plated on collagen/poly-D-lysine coverslips in24-well plates at 50K cells per well and placed in 37 !C, 5% CO2 humidifiedincubator overnight. Following treatment the following day, cells were washedthree times with PBS and then fixed in 4% paraformaldehyde for 20min, washedthree times in PBS, washed twice in PB and then mounted with VECTASHIELD(Vector Laboratories, Burlington, CA). All imaging was performed within theWashington University Pain Center Confocal Imaging Center. Images, cells, andtreatment groups were chosen and analysed in a blinded fashion. Semi-quantitativeanalysis of internalization was calculated as previously described using Metamorph(Molecular Devices, CA, USA) analysis algorithm for pixel intensity measurementsof internalized fluorescence measures58. To determine internalized percentages,equal cell shapes and sizes were always chosen; concentric circles around thefluorescence, background internal fluorescence (untreated controls) or internalized(treated) portions of the entire cell were drawn in Metamorph, integrated pixelintensities were recorded for each using the Metamorph algorithm for integratingintensity and internalized receptors were calculated using: Inside F/Total F toproduce the internalization ratio. Data are expressed as mean±s.e.m.

In vitro calcium imaging. Cells were plated on collagen/poly-D-lysine glass cov-erslips, loaded with Fura-2 acetoxymethyl ester (2.5–5mM), and incubated for60min at room temperature in 1.5mM of pluronic acid (Molecular Probes,Eugene, OR) in a HEPES-buffered saline (2mM Ca2! ). Coverslips were placed ina laminar flow perfusion chamber (Warner Instrument Corp.) and constantly

perfused with HEPES-buffered saline (2mM Ca2! ). Images of Fura-2-loaded cellswith the excitation wavelength alternating between 340 and 380 nm were captured.Following subtraction of background fluorescence, the ratio of fluorescenceintensity at the two wavelengths was calculated. Ratio levels were analyzed usingMetaFluor (Universal Imaging Corporation).

Animals. Adult (25–35 g or 2 to 3 months old) male C57BL/6J mice were used inall in vivo experiments. Mice were group-housed, given access to food and water adlibitum and maintained on a 12 h:12 h light:dark cycle. All animals were held in afacility in the lab 1 week before surgery, post-surgery and throughout the durationof the behavioural assays to minimize stress from transportation and disruptionfrom foot traffic. All procedures were approved by the Animal Care and UseCommittee of Washington University and conformed to US National Institutes ofHealth guidelines.

Viral preparation. Plasmid encoding pLenti-CaMKIIa-opto-b2AR-mCherry (finaltiter 4.8$ 108 IUml# 1) was obtained from Deisseroth Laboratory at StanfordUniversity and packaged at the WUSTL Hope Center Viral Vector Core. Lenti-PGK-GFP (viral control; final titer 1.3$ 108 IUml# 1) was provided by theWUSTL viral core facility. AAV5-CaMKIIa-HA-GSD-IRES-mCitrine (final titer3$ 1012 virus molecules per ml) and AAV5-CaMKIIa-eGFP (final titer 5$ 1012

virus molecules per ml) were obtained from University of North Carolina GeneTherapy Center Vector Core and Virus Vector Core.

Stereotaxic surgery. Mice were anaesthetized in an induction chamber(5% isoflurane) and placed in a stereotaxic frame (Kopf Instruments, Model 1900)where they were maintained at 1–2% isoflurane throughout the procedure.Following craniotomy mice were injected bilaterally with 1.2 ml of either lenti-EF1a-GFP or lenti-CaMKIIa-optob2AR-mCherry in the BLA at stereotaxiccoordinates: # 1.3mm posterior to bregma; ±2.9mm lateral to bregma and# 4.9mm ventral to bregma. For wireless m-ILED BLA studies, animals wereinjected unilaterally, not bilaterally. Mice were then implanted with chronic fiberoptic implants or m-ILED wireless devices with coordinates adjusted from viralinjection to: # 1.3mm posterior to bregma; ±2.9mm lateral to bregma and# 3.9mm ventral to bregma. For bio-dissolvable samples, the device wasimplanted at the desired target, ACSF was applied to the portion of the device thatremained outside of the skull to facilitate dissolution of the adhesive, and then theepoxy needle was removed after a delay of 15min47,48. The fiber optic implants andwireless m-ILED devices were secured using two bone screws (CMA, 743102) andaffixed with TitanBond (Horizon Dental Products) and dental cement (LangDental)48. Mice were allowed to recover for at least 3–6 weeks before behaviouraltesting; this interval also permitted optimal viral expression.

In vivo electrophysiology. Spontaneous single unit activity was recordedfollowing previous published protocols47,57. Briefly, mice were lightly anesthetized(1% isoflurane), placed in a stereotactic frame and two skull screws were placedon either side of the midline to ground the electrode array. The recordingapparatus consisted of a 16-channel (35-mm tungsten wires, 150-mm spacingbetween wires, 150-mm spacing between rows, Innovative Physiology) electrodearray. This array was epoxied to a fiber optic and lowered into the BLA (stereotaxiccoordinates from bregma: # 1.3mm (AP),±2.9mm (ML) and # 4.9mm (DV).Spontaneous and photostimulated neuronal activity was recorded from eachelectrode, bandpass-filtered with activity between 250 and 8,000Hz, and analysedas spikes. Voltage signals were amplified and digitally converted using Omniplexand PlexControl (Plexon). For opto-b2ARWT, 5 s constant light, followed by 5-s nolight was repeated for 12 cycles or 20-s constant light (on) followed by 1-minrecovery with no light (off) was repeated for 12 cycles. Principle componentanalysis and/or evaluation of t-distribution with expectation maximization wereused to sort spikes using Offline Sorter (Plexon). Cells were considered excited ifthere was than a 10% increase in baseline firing frequency, and inhibited if therewas 4 10% decrease in baseline firing frequency in the presence of constantphotostimulation.

Wireless powering and RF scavenger for wireless optogenetics. Wirelesspowering of the m-ILED devices was performed following previously publishedprotocols47,48. The wireless power transmitter includes an RF signal generator(Agilent N5181A), a power supply (Agilent U8031A), a RF power amplifier(Empower RF Systems 1119-BBM3K5KHM), an RF signal splitter (RF LambdaRFLT2W0727GN), and two panel antennas (ARC Wireless ARC-PA2419B01). TheRF signal generator is internally modulated to delivery sufficient power to light them-iLEDs at the given stimulation protocol (10Hz, 50-ms pulse widths). The RFpower amplifier that is powered by the power supply enlarges the modulated RFsignal from the RF signal generator. The RF power is then transmitted from thepanel antenna to the headstage power harvesters. The RF signal generator has apower output from # 10 to 0 dBm at 1.5 GHz, optimized daily to ensure equivalentlight power throughout the space of the LDB assay. Mice with chronicallyimplanted m-ILED devices were acutely connected to the headstage powerharvesters immediately before any wireless photostimulation.





Behaviour. Behavioural assays were performed in a special sound attenuated roommaintained at 23 !C. Lighting was measured and stabilized at B4 lux for anxietytests and B200 lux for place testing. All behavioural apparatuses were cleaned with70% ethanol in between animals. In each assay, animals received constant pho-tostimulation throughout the entire trial. For all behavioural experiments, lenti-EF1a-GFP and lenti-CaMKIIa-optob2AR-mCherry received 5 s of constant pho-tostimulation (473 nm) followed by 5 s of no light throughout the entire trial.Movements were video recorded and analyzed using Ethovision Software.

Elevated zero maze. The EZM (Harvard Apparatus) was made of grey plastic,200 cm in circumference, comprised of four 50-cm sections (two opened and twoclosed). The maze was elevated 50 cm above the floor and had a path width of 4 cmwith a 0.5 cm lip on each open section. Animals were connected to cables coupledto a function generator, positioned head first into a closed arm, and allowed toroam freely for 6min. Mean open arm time was the primary measure of anxiety-like behaviour.

Light/dark box. The LDB was a 50$ 50 cm square plexiglass enclosure with a16.5 cm$ 49 cm dark insert. For testing, animals were connected to wirelessharvester and placed into the corner of the open enclosure and allowed to roamfreely for 10min.

Immunohistochemistry. At the conclusion of behavioural testing, mice wereanaesthetized with sodium pentobarbital and transcardially perfused with ice coldPBS, followed by 4% phosphate-buffered paraformaldehyde following previouslypublished protocols47. Brains were removed, post-fixed overnight in para-formaldehyde, and saturated in 30% phosphate-buffered sucrose. Sections of 30mmwere cut, washed in 0.3% Triton X100/5% normal goat serum in 0.1M PBS, stainedwith fluorescent Nissl stain (1:400 Neurotrace, Invitrogen, Carlsbad, CA) for 1 h, andmounted onto glass slides with Vectashield (Vector Laboratories, Burlingame, CA).opto-b2AR expression was verified using fluorescence (Olympus, Center Valley, PA)and confocal microscopy (Leica Microsystems, Bannockburn, IL). Images wereproduced with Leica Application Suite Advanced Fluorescence software. Animalsthat did not show targeted expression were excluded from analyses.

Statistics/data analysis. All data are expressed as mean±s.e.m. Data werenormally distributed, and differences between two groups were determined usingindependent Students’ two-tailed, unpaired or paired t-tests as appropriate. Dif-ferences between multiple groups were determined via one-way or two-way ana-lysis of variances (ANOVAs) followed by post hoc Bonferroni or Dunnett’s multiplecomparisons if the main effect was significant at Po0.05. Statistical significancewas taken as *Po0.05, **Po0.01, ***Po0.001, ****Po0.0001 and all analyseswere conducted using Prism 5.0 (GraphPad). Grubbs’ test was used to remove anystatistical outliers.

References1. Boyden, E. S., Zhang, F., Bamberg, E., Nagel, G. & Deisseroth, K. Millisecond-

timescale, genetically targeted optical control of neural activity. Nat. Neurosci.8, 1263–1268 (2005).

2. Li, X. et al. Fast noninvasive activation and inhibition of neural and networkactivity by vertebrate rhodopsin and green algae channelrhodopsin. Proc. NatlAcad. Sci. USA 102, 17816–17821 (2005).

3. Ishizuka, T., Kakuda, M., Araki, R. & Yawo, H. Kinetic evaluation ofphotosensitivity in genetically engineered neurons expressing green algae light-gated channels. Neurosci. Res. 54, 85–94 (2006).

4. Zhang, F., Aravanis, A. M., Adamantidis, A., de Lecea, L. & Deisseroth, K.Circuit-breakers: optical technologies for probing neural signals and systems.Nat. Rev. Neurosci. 8, 577–581 (2007).

5. Armbruster, B. N., Li, X., Pausch, M. H., Herlitze, S. & Roth, B. L. Evolving thelock to fit the key to create a family of G protein-coupled receptors potentlyactivated by an inert ligand. Proc. Natl Acad. Sci USA 104, 5163–5168 (2007).

6. Airan, R. D., Thompson, K. R., Fenno, L. E., Bernstein, H. & Deisseroth, K.Temporally precise in vivo control of intracellular signalling. Nature 458,1025–1029 (2009).

7. Bailes, H. J., Zhuang, L.-Y. & Lucas, R. J. Reproducible and sustained regulationof Gas signalling using a metazoan opsin as an optogenetic tool. PLoS ONE 7,1–11 (2012).

8. Kim, J.-M. et al. Light-driven activation of b2-adrenergic receptor signaling bya chimeric rhodopsin containing the b2-adrenergic receptor cytoplasmicloopsw. Biochemistry 44, 2284–2292 (2005).

9. Kenakin, T. & Christopoulos, A. Signalling bias in new drug discovery:detection, quantification and therapeutic impact. Nat. Rev. Drug Discov. 12,205–216 (2013).

10. Luttrell, L. M. & Gesty-Palmer, D. Beyond desensitization: physiologicalrelevance of arrestin-dependent signaling. Pharmacol. Rev. 62, 305–330 (2010).

11. Shenoy, S. K. et al. beta-arrestin-dependent, G protein-independent ERK1/2activation by the beta2 adrenergic receptor. J. Biol. Chem. 281, 1261–1273 (2006).

12. Bruchas, M. R., Macey, T. A., Lowe, J. D. & Chavkin, C. Kappa Opioid ReceptorActivation of p38 MAPK Is GRK3- and Arrestin-dependent in Neurons andAstrocytes. J. Biol. Chem. 281, 18081–18089 (2006).

13. Bruchas, M. R. & Chavkin, C. Kinase cascades and ligand-directed signaling atthe kappa opioid receptor. Psychopharmacology (Berl.) 210, 137–147 (2010).

14. Bruchas, M. R. et al. Selective p38a MAPK deletion in serotonergic neuronsproduces stress resilience in models of depression and addiction. Neuron 71,498–511 (2011).

15. Wei, H. et al. Independent b-arrestin 2 and G protein-mediated pathways forangiotensin II activation of extracellular signal-regulated kinases 1 and 2. Proc.Natl Acad. Sci USA 100, 10782–10787 (2003).

16. Ren, X.-R. et al. Different G protein-coupled receptor kinases govern G proteinand b-arrestin-mediated signaling of V2 vasopressin receptor. Proc. Natl Acad.Sci. USA 102, 1448–1453 (2005).

17. Ahn, S., Shenoy, S. K., Wei, H. & Lefkowitz, R. J. Differential Kinetic andSpatial Patterns of b-Arrestin and G Protein-mediated ERK Activation by theAngiotensin II Receptor. J. Biol. Chem. 279, 35518–35525 (2004).

18. Krasel, C. et al. Dual role of the beta2-adrenergic receptor C terminus forthe binding of beta-arrestin and receptor internalization. J. Biol. Chem. 283,31840–31848 (2008).

19. Hausdorff, W. P. et al. A small region of the beta-adrenergic receptor isselectively involved in its rapid regulation. Proc. Natl Acad. Sci. USA 88,2979–2983 (1991).

20. Seibold, A., January, B. G., Friedman, J., Hipkin, R. W. & Clark, R. B.Desensitization of b2-adrenergic receptors with mutations of the proposed gprotein-coupled receptor kinase phosphorylation sites. J. Biol. Chem. 273,7637–7642 (1998).

21. Vaughan, D. J. et al. Role of the G protein-coupled receptor kinase site serinecluster in b2-adrenergic receptor internalization, desensitization, and b-arrestintranslocation. J. Biol. Chem. 281, 7684–7692 (2006).

22. Hieble, J. P. et al. International Union of Pharmacology. X. Recommendationfor nomenclature of alpha 1-adrenoceptors: consensus update. Pharmacol. Rev.47, 267–270 (1995).

23. Drake, M. T. et al. b-arrestin-biased agonism at the b2-adrenergic receptor.J. Biol. Chem. 283, 5669–5676 (2008).

24. Friedman, J., Babu, B. & Clark, R. B. b2-adrenergic receptor lacking the cyclicamp-dependent protein kinase consensus sites fully activates extracellularsignal-regulated kinase 1/2 in human embryonic kidney 293 cells: lack ofevidence for Gs/Gi switching. Mol. Pharmacol. 62, 1094–1102 (2002).

25. Schmitt, J. M. & Stork, P. J. S. b2-adrenergic receptor activates extracellularsignal-regulated kinases (ERKs) via the small G protein Rap1 and the serine/threonine kinase B-Raf. J. Biol. Chem. 275, 25342–25350 (2000).

26. Nobles, K. N. et al. Distinct phosphorylation sites on the {beta}2-adrenergicreceptor establish a barcode that encodes differential functions of {beta}-arrestin. Sci. Signal. 4, ra51 (2011).

27. Cornwall, M. C., Matthews, H. R., Crouch, R. K. & Fain, G. L. Bleached pigmentactivates transduction in salamander cones. J. Gen. Physiol. 106, 543–557(1995).

28. Fain, G. L., Matthews, H. R. & Cornwall, M. C. Dark adaptation in vertebratephotoreceptors. Trends Neurosci. 19, 502–507 (1996).

29. Oh, E., Maejima, T., Liu, C., Deneris, E. & Herlitze, S. Substitution of 5-HT1Areceptor signaling by a light-activated G protein-coupled receptor. J. Biol.Chem. 285, 30825–30836 (2010).

30. Cohen, G. B., Yang, T., Robinson, P. R. & Oprian, D. D. Constitutive activationof opsin: Influence of charge at position 134 and size at position 296.Biochemistry 32, 6111–6115 (1993).

31. Pierce, K. L., Premont, R. T. & Lefkowitz, R. J. Seven-transmembrane receptors.Nat. Rev. Mol. Cell Biol. 3, 639–650 (2002).

32. Zastrow, M. von & Kobilka, B. K. Ligand-regulated internalization andrecycling of human beta 2-adrenergic receptors between the plasma membraneand endosomes containing transferrin receptors. J. Biol. Chem. 267, 3530–3538(1992).

33. Zastrow, M. von & Kobilka, B. K. Antagonist-dependent and -independentsteps in the mechanism of adrenergic receptor internalization. J. Biol. Chem.269, 18448–18452 (1994).

34. Morrison, K. J. et al. Repetitive endocytosis and recycling of the beta 2-adrenergic receptor during agonist-induced steady state redistribution. Mol.Pharmacol. 50, 692–699 (1996).

35. Iversen, L. L. Role of Transmitter Uptake Mechanisms in SynapticNeurotransmission. Br. J. Pharmacol. 41, 571–591 (1971).

36. Pelton, II E. W., Kimelberg, H. K., Shipherd, S. V. & Bourke, R. S. Dopamineand norepinephrine uptake and metabolism by astroglial cells in culture. LifeSci. 28, 1655–1663 (1981).

37. Daigle, T. L., Kearn, C. S. & Mackie, K. Rapid CB1 cannabinoid receptordesensitization defines the time course of ERK1/2 MAP kinase signaling.Neuropharmacology 54, 36–44 (2008).

38. Straiker, A., Wager-Miller, J., Hutchens, J. & Mackie, K. Differential signallingin human cannabinoid CB1 receptors and their splice variants in autaptichippocampal neurones. Br. J. Pharmacol. 165, 2660–2671 (2012).

39. Sorkin, A. & von Zastrow, M. Endocytosis and signalling: intertwiningmolecular networks. Nat. Rev. Mol. Cell Biol. 10, 609–622 (2009).





40. Wolff, S. B. E. et al. Amygdala interneuron subtypes control fear learningthrough disinhibition. Nature 509, 453–458 (2014).

41. Spampanato, J., Polepalli, J. & Sah, P. Interneurons in the basolateral amygdala.Neuropharmacology 60, 765–773 (2011).

42. Ehrlich, I. et al. Amygdala inhibitory circuits and the control of fear memory.Neuron 62, 757–771 (2009).

43. Bailey, K. R. & Crawley, J. N. in Methods of Behavior Analysis in Neuroscience.(ed. Buccafusco, J. J.) (CRC Press, 2009).

44. Silverman, J. L., Yang, M., Lord, C. & Crawley, J. N. Behavioural phenotypingassays for mouse models of autism. Nat. Rev. Neurosci. 11, 490–502 (2010).

45. Crawley, J. & Goodwin, F. K. Preliminary report of a simple animal behaviormodel for the anxiolytic effects of benzodiazepines. Pharmacol. Biochem. Behav.13, 167–170 (1980).

46. Lepicard, E. M., Joubert, C., Hagneau, I., Perez-Diaz, F. & Chapouthier, G.Differences in anxiety-related behavior and response to diazepam in BALB/cByJ and C57BL/6J strains of mice. Pharmacol. Biochem. Behav. 67, 739–748(2000).

47. Kim, T. et al. Injectable, cellular-scale optoelectronics with applications forwireless optogenetics. Science 340, 211–216 (2013).

48. McCall, J. G. et al. Fabrication and application of flexible, multimodal light-emitting devices for wireless optogenetics. Nat. Protoc. 8, 2413–2428 (2013).

49. Tye, K. M. & Deisseroth, K. Optogenetic investigation of neural circuitsunderlying brain disease in animal models. Nat. Rev. Neurosci. 13, 251–266(2012).

50. Luttrell, L. M. Minireview: more than just a hammer: ligand ‘bias’ andpharmaceutical discovery. Mol. Endocrinol. 28, 281–294 (2014).

51. Whalen, E. J., Rajagopal, S. & Lefkowitz, R. J. Therapeutic potential ofb-arrestin- and G protein-biased agonists. Trends Mol. Med. 17, 126–139 (2011).

52. Carlsen, J. Immunocytochemical localization of glutamate decarboxylase in therat basolateral amygdaloid nucleus, with special reference to GABAergicinnervation of amygdalostriatal projection neurons. J. Comp. Neurol. 273,513–526 (1988).

53. Smith, Y. & Pare, D. Intra-amygdaloid projections of the lateral nucleus in thecat: PHA-L anterograde labeling combined with postembedding GABA andglutamate immunocytochemistry. J. Comp. Neurol. 342, 232–248 (1994).

54. Fenno, L. E. et al. Targeting cells with single vectors using multiple-featureBoolean logic. Nat. Methods 11, 763–772 (2014).

55. Irannejad, R. et al. Conformational biosensors reveal GPCR signalling fromendosomes. Nature 495, 534–538 (2013).

56. Tsvetanova, N. G. & von Zastrow, M. Spatial encoding of cyclic AMP signalingspecificity by GPCR endocytosis. Nat. Chem. Biol. 10, 1061–1065 (2014).

57. Sparta, D. R. et al. Construction of implantable optical fibers for long-termoptogenetic manipulation of neural circuits. Nat. Protoc. 7, 12–23 (2012).

58. Zhang, N. R. et al. Serine 363 is required for nociceptin/orphanin FQ opioidreceptor (NOPR) desensitization, internalization, and arrestin signaling. J. Biol.Chem. 287, 42019–42030 (2012).

AcknowledgementsWe thank Drs Robert Lefkowitz for generously providing the b2ARTYY cDNA and KarlDeisseroth for the opto-b2AR cDNA. We particularly thank Drs Robert Gereau 4th,Thomas Baranski, Joe Henry Steinbach and N. Gautam for their helpful and criticalinsight into the preparation of this manuscript. We also thank the members of theBruchas laboratory for all their help and technical support, The HOPE Center viralvector core (NINDS, P30NS057105) and Bakewell Imaging Center. This work is sup-ported by NIDA R01DA037152 (M.R.B), R21DA035144 (M.R.B.), R00DA025182(M.R.B.), NIMH F31MH101956 (J.G.M.), TR01NS081707 and the McDonnell Center forSystems Neuroscience (M.R.B.).

Author contributionsE.R.S. designed and performed the experiments, collected and analysed the data andwrote the manuscript. J.G.M., R.A., M.J.S., S.L.A. designed and performed the experi-ments, collected and analysed the data. G.S., S.I.P. designed and fabricated the wirelessm-iLED devices and RF power harvesters. W.J.P. provided technical support.J.A.R. helped the design and oversee the wireless device operation. M.R.B. designed andoversaw experiments, and wrote the manuscript.

Additional informationSupplementary Information accompanies this paper at http://www.nature.com/naturecommunications

Competing financial interests: The authors declare no competing financial interests.

Reprints and permission information is available online at http://npg.nature.com/reprintsandpermissions/

How to cite this article: Siuda, E. R. et al. Optodynamic simulation of b-adrenergicreceptor signalling. Nat. Commun. 6:8480 doi: 10.1038/ncomms9480 (2015).

This work is licensed under a Creative Commons Attribution 4.0International License. The images or other third party material in this

article are included in the article’s Creative Commons license, unless indicated otherwisein the credit line; if the material is not included under the Creative Commons license,users will need to obtain permission from the license holder to reproduce the material.To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/






http://npg.nature.com/reprintsandpermissions/

http://npg.nature.com/reprintsandpermissions/

http://creativecommons.org/licenses/by/4.0/


!"##$%&%'()*+,-./"*%,01,2#(34!567,)'8,!

567,9:)*%,9.&.$)*,;6<=,)'8,#>7?,9./')$.'/,&%;:)'.9&9@,A)B,

!"#$%&'%(!)&*()+,%&-%./).0!1"2%3456+7%-%8%9:%+;/+<&=+-)'>%&-%<+'/.-'+%).%5&?@)%3A%'+!7%%,.))+,%5&-+>B%$6</5+%)<(!+%

'@.C%DEF0/G5.%!+55'% &-% <+'/.-'+% ).% )@+%'(=+% 5&?@)%')&=656'%3-%8%:%+;/+<&=+-)'>% 3=+(-%8%'.5&,% 5&-+7%HE#%8%'@(,+,%(<+(>B,ACB,I-)<(!+5565(<%J(K1%<+'/.-'+%.L%./).0!

1"2%3456+M%9%=&-%5&?@)M%-%8%9N%!+55'>7%!

1"2%345(!OM%9% #%

&'./<.)+<+-.5M%-%8%P%!+55'>%(-,%DEF1QN%!+55'%3/6</5+M%9%=&-%5&?@)M%-%8%Q%!+55'>%!.0)<(-'L+!)+,%C&)@%JRG"%3=+(-%8%'.5&,%5&-+7%HE#%8%'@(,+,%(<+(>B,"55%,()(%(<+%+;/<+''+,%('%=+(-%S%HE#B%"55%5&?@)%/65'+'%(<+%:PN%-=%()%9%TU!=1%6-5+''%.)@+<C&'+%-.)+,B

!"##$%&%'()*+,-./"*%,0

)V/).0

1"2

DEF1QN

W&=+%3=&->!"#$%3L.5,%4('+5&-+>

09 X 9 1 N : AX

9

1

N

:

A

YC

1"2

V/).01"2

DEF1QN

J(5!&6=%I=(?&-?

X 1X :X YX

9BX

9B1

9B:

9BY

Z%3-.<=>

W&=+%3'+!>

!

!"#$%&'()*+

,-)./%'0/1-/234

55

5

6 7 8 9 : ; <

'-))

;%.%(=(

6>7%?@

6>9%?@

7%?@

9%?@

76%?@ &9+

&9+

&:+

&:+

&;+

&:+

&A+

B!C(D8E#

" #

$%&&'()(*+,-./0#!%-(/12/3&+45!167/,*8/!

167/9",-(/9#)#',-/:6;</,*8/&=7>/9#!*,'#*!/)(:",*#9)9?/@,A/

F-))%!"#$%CGH/%3(-0./%I)(C.%'0(H%0#!%-(/BCD!>%@EA%!"#$%CGH/%3(-0./%'())(JG2K%7%HG2%G.(!0(C/0/2()%&7% L+%JM.N%G2%!

8E#%&2%O%8%/P!/0GH/2C.+>%@8A%!"#$%CGH/%3(-0./%G2%!

8E#%3(H!M0/*%C(%C(CM)%"#$>%@(A%Q-M2CG'G3MCG(2%('%C(CM)%

"#$%G2C/K0MC/*%G2C/2.GC4%2(0HM)G@/*%C(%6%HG2%CGH/%!(G2C%&2%O%:%/P!/0GH/2C.+>%@CA%#/!0/./2CMCGR/%(!C(D!8E#%M2*%

?"$8S9%3(2C0()%!"#$%GHH-2(I)(C.%G2%0/.!(2./%C(%G230/M.G2K%)GKNC%!(J/0>%@!A/!"#$%!(J/0%0/.!(2./%3-0R/%('%(!C(D!

8E#%G2%0/.!(2./%C(%G230/M.G2K%)GKNC%!(J/0%&",

;6%O%:>9%T%7>:%U=3H8V%2%O%:%/P!/0GH/2C.+>%@"A%W(2CG2-(-.%

)GKNC%&'-))+%(0%;%./3%)GKNC%!-)./.%(2=(''%X(=(Y%'(0%7%HG2%.N(J%.GK2G'G3M2C)4%/)/RMC/*%)/R/).%('%!"#$%'0(H%IM./)G2/%G2%

(!C(D!8E#% &5!%Z%6>6;[% 55!%Z%6>67%RGM%B2/DUM4%E\B]E% '())(J/*%I4%^-22/CC_.%H-)CG!)/%3(H!M0G.(2% C/.C% C(%2(%

C0/MCH/2C%3(2C0()V%2%O%&`+%/P!/0GH/2C.+>%@#A/^G''/0/2C%)/2KCN.%('%3(2CG2-(-.%)GKNC%.N(J%.GHG)M0%G230/M./.%G2%!"#$%G2%(!C(D!

8E#%&I)-/+%&2%O%&`+%/P!/0GH/2C.+>%%@FA%#MJ%!"#$%G2C/K0MC/*%G2C/2.GC4%RM)-/.%'0(H%.MH/%K/)%('%-2.CGH-D

)MC/*%?"$8S9% 3/)).% JGCN% (!C(D!8E#% &I)-/+% M2*%JGCN(-C% &!-0!)/+% &2% O% :% /P!/0GH/2CM)% 0/!)G3MC/.+>%E))% *MCM% M0/%

/P!0/../*%M.%H/M2%T%a"L>%E))%)GKNC%!-)./.%M0/%:A9%2H%MC%7%U=3H8%-2)/..%(CN/0JG./%2(C/*>

B!C(D8E#%",

;6%

:>9%T%7>:%U=3H8

D: D9 D8 D7 6 7 8 9

6

;6

766

,(J/0%&)(K%U=3H8+

B!C(D8E#

6 7

;%./3

7;%./3

96%./3

7%HG2

8%HG2

9%HG2

:%HG2

;%HG2

8 9 : ; <

&;+

&;+

&;+

&;+

&;+

&;+

&<+

&:+

!"#$%&'()*+

,-)./%)/2KCN

B!C(D8E#

$%&&'()(*+,-./0#!%-(/1

!"#$

b(CM)%"#$

bGH/%&HG2+

b(CM)%"#$%&2(0H+

7 8 ; 7; 96 <6 S6 786

6>6

6>;

7>6

7>;

8>6

6 7 8 ; 7; 96 &HG2+

!"#$

DM3CG2

:::8

:9

<6 S6 786

8

6 96 <6 S6 7866

86

:6

<6

c6

766

bGH/%&HG2+%

:

!"#$%&d%HMP+

6 7 8 ; 7; 968E#

<6 S6 786 &HG2+

:::8

( C

!"#$%&d%HMP+

B!C(D8E#

!"#$

DM3CG2

6 9/D97/D9 9/D87/D89/D: 7/D7 9/D7 7/6 9/6 7/7 9/7

:::8

&U=3H8+

:9

?"$8S9

!"#$

DM3CG2

:::8

:9

!"#$

&e2C/K0MC/*%G2C/2.GC4+

B!C(D

8E#

?"$

8S9

B!C(D8E#?"$8S9

!"#$

DM3CG2

:::8

:9

F

6

76

86

96

:6

;6

!"#$

DM3CG2

!"#$

,

6 7 8 ; 7; 96

8E#

<6 S6 786 &HG2+

:::8

:9

7;6

DM3CG2

B!C(D8E#

f%3C0)

&HG2+

:::8

:9

E8E#%&7%HG2%G.(+

8E#

&7%HG2%G.(+

8E#

:::8

! "#$%&'

()*'+,-

()*

./0123 ()*

45$$67173%!890 ,2:5870 ;<0 #$%&'!()*=0 >53?%2&3!6690 23%783!62@70 !3A0 87?&B780 >8&10 A7=73=2%2@!%2&3C0 D!E0

!"#$"%"&'(')*"+),(-"+%./0%+)&'"$&(1)2(')/&+/3+/#'/4!56!4789+$":"#'/$%+)&+$"%#/&%"+'/+1)-.'+%'),;1(')/&+<=+,)&+

#;1%">+3)?"@+('+AB+,)&+'),"+#/)&'+#/%'+%'),;1(')/&C+D/'):"+1(:E+/3+)&'"$&(1)2"@+$":"#'/$+<%:(1"+F($+G+=B+H,>C+I&%"'+%./0%+ )&'"$&(1)2(')/&+/3+!

56!4789+<:/1/$)2"@+ '/+F1(:E+(&@+0.)'">+ )&+ $"%#/&%"+ '/+=+HJ+ )%/#$/'"$"&/1+('+%(,"+

'),"+ #/)&'C+6$$/0."(@+ %./0%+ #;&:'('"+ )&'"$&(1)2"@+ $":"#'/$+ <%:(1"+ F($+ G+ =B+ H,>C+ D"E+!"#$"%"&'(')*"+ ),(-"+%./0%+)&'"$&(1)2(')/&+/3+!

56!4789+<:/1/$)2"@+'/+F1(:E+(&@+0.)'">+$":"#'/$%+)&+$"%#/&%"+'/+)%/#$/'"$"&/1+<=+HJ>+

3)?"@+=5B+,)&+'),"+#/)&'+#/%'+%'),;1(')/&+<%:(1"+F($+G+K+H,>C+D?E+L;(&')3):(')/&+/3+)&'"$&(1)2(')/&+)&+!56!4789+

<-$"MN+/&+G+5C5+,)&>C+ +DAE+9"$:"&'+ )&'"$&(1)2(')/&+3/$+!

56!4789+<O#+P+BCBKQ+OOO#+P+BCBB=+*)(+R&"4S(M+6DRT6+

3/11/0"@+FM+U;&&"''V%+,;1')#1"+:/,#($)%/&+'"%'+'/+B+,)&+:/&'$/1N+<&+G+<W>+:"11%+#"$+'),"+#/)&'>C+D7E+X/,#($)%/&+/3+8)-;$"+5@+<:/&%'(&'+)%/N+F1(:E>+(&@+Y;##1","&'($M+8)-;$"+5@+<=+,)&+)%/N+-$"M>+<OOO+#+P+BCBB=Q+OOOO+#+P+BCBBB=+*)(+Z0/4S(M+6DRT6+3/11/0"@+FM+[/&3"$$/&)V%+,;1')#1"+:/,#($)%/&+'"%'>C+D>E+6@@)')/&+/3+A4:)%4$"')&(1+<=+ J>+'/+0"11%+F"3/$"+"(:.+1)-.'+#;1%"C+611+1)-.'+#;1%"%+($"+\]^+&,+('+=S_:,5C

F(/0123

45$$67173%!890,2:5870;

B = K =K ^B `B AB=5B

O

OOOOOO OOO

OOOOOO

5B

\B

`B

aB

b+I&'"$&(1)2(')/&

B

56!

<=+,)&+)%/>

A

Z),"+<,)&>

<=^>

<a>

<=]>

<=5>

<A>

<A>

<=5>

<=`>

B =B 5B ^BB

5B

\B

`B

aB

56!+<=+,)&+)%/>

/&+G+5C5+,)&+

?56!

<=+,)&+)%/>

Z),"+<,)&>

B K =B =KB

^

`

A

=5

> R#'/456!

cA4:)% cA4:)%

Z),"+<,)&>:6J9+<8/1@+F(%"1)&">

B = K =K ^B `B AB =5BB

5B

\B

`B

aB

b+I&'"$&(1)2(')/&

Z),"+<,)&>

56!

<X/&%'(&'+)%/>56!

<=+,)&+)%/>OOO

OOOO

OOOO

7

b+I&'"$&(1)2(')/&

!

"

#

$% &'

()

*

+ ,

-

./

0

1

2 34 2,.+,&2/ /%.+$.

44

)

$.

(

!

$

%

'

)

+

,

./

2

3

4

2

,.

+

,&

2

/

/

%

.+

$

.

*

) 3+ **

3 *

0+ *

* 4

4

/+

'!$

.

#"

,

2

&% )(

-

*

4

/

+'

!

$

.

#

"

,

2

&

%

) (

-

0

*4

/ +'

!

$

.#

",

2

&

%

)

3

(

-

0

*

*40***))

*3

/+2+& &

*4& &*

24+

+2

&

2 ,

* +

+ &)* 2 20 ) *

1 % * // * )

0 $ 4/ / /

*

'0 3

2/)

+&1/

3)*

)-) )

.*/

2 1% !

402

$

3(

- %

-20# 4 4 & .

"$

$

&

,

, $ % +

/ 0 4

3 + / /" + 0

00 . *

/ 0 +

+ - 4 2

/ '

( ('

'

!+

% . , * ((

!%2

2*

!!

%%'+*'

$')*!3

/0 0

30 /)

+*0

-1*.

2 ) 4

/)+4

0+&

"(2 % # + 2

.0

+ 3 & 0

. ) +

+ ) &

% ) /

4 , . /0 4 0

3 3 , '( + !

0)

+

($

**

*

!!

%%

%')22 ,%%,, %2'&42$)%2(*2($'$%$!*-$#

.2& $ % + / - ( & /%*% *# % (2 ! - % &, #% .*& & $ % ( )/ ). 2

%

%

!567689:;<=!56>

?@A7BC;CBD:E<=

?)!>

!56<F<?)!

,6<56G6H6DI

.6:;J<GKJAJ:6;

LM?M

NM

OL

PP

Q?

RM

LMP

LLR

LNS

LPM

LQL

LTR

?MM

?LT

??P

?OM

?SM

?RM

NMP

N?M

NNT

NS?NTM

O?M ONO

!"##$%&%'()*+,-./"*%,0

!"##$%&%'()*+, -./"*%, 01, !%*#%'(.'%, 2#(23456, )'7, &"()'(, 8%9"%':%;, -ABJ66;< 9CUKC;EC< 6V< 68J6@

?)!<

7C8:EJ:;D<B567689:;<=WHKC><A;7<?)!<=WB6X;><BC9:7KC9Y<)G:;6<AE:79<5:D5H:D5JC7<:;<BC7<956X<BC9:7KC9<GKJAJC7<J6<

ABBC9J:;@W:A9C7<=68J6@?)!*Q?+Z4LNS2Z4??O)<6B<68J6@

?)!*44><A;7<2@8B6JC:;@W:A9C7<=68J6@

?)!%NS?)[%NSN2<6B<68J6@

?)!%%><

GKJA;J9Y<%:G:HAB<GKJAJ:6;9<:;<ABBC9J:;@W:A9C7<=?)!*Q?+Z4LNS2Z4??O)<6B<

?)!&44><A;7<2@8B6JC:;@W:A9C7<=

?)!%NPP)Z%NPS2<

6B<?)!%%><

?)!<GKJA;J9Y

\8J6@?)!

!

!"#$%&'())

!"#$%&'(

!"#$%&'(*++

,-.&/0

!"#$%&'(

"

#$%%&!'!()*+,- ./0$+!- 12- 3%)/4*&- 45()+5&- 5"- 6*7+!(!+0/4- 8/0(*&/(09- :*;- - !&'(123 456789:3 ;3 <3 0=>3 !

&'(2++3

4?@AA;:3;3<30=>3!&'())34@AB:3;3<30=37;B3,-.&/034"C@"6A:3;3<3D=3766378#EF7#A38'GH3E;3@AI"$;IA3#$3EI$"@$#A@A;$634J3

G=3KE#L3BEMMA@A;#E7639E;A#E8IN34<=3!&'(123456789:3;3<30=>3!

&'(2++34?@AA;:3;3<30=>3!

&'())34@AB:3;3<30=337;B3,-.&/03

4"C@"6A:3;3<3&=3766378#EF7#A38'GH3E;3@AI"$;IA3#$3M$@I9$6E;34JO3 G=3KE#L3IEPE67@39E;A#E83"@$ME6AIN3:4;3!"#$%&'(3456CA:3

;3<3JD=>3$"#$%&'(*++34?@AA;:3;3<3Q=37;B3$"#$%

&'())34@AB:3;3<3JR=3766378#EF7#A38'GH3E;3@AI"$;IA3#$373D3IA8356CA3

6E?L#3"C6IAN3,-.&/03#@7;IMA8#AB3KE#L3SCI#3#LA3T6$)A;I$@3"67IPEB3IL$KAB3;$3AMMA8#3#$3D3IA836E?L#3"C6IAN347=32@78AI3M@$P344=3KA@A3ME#3KE#L3;$;6E;A7@3$;A3"L7IA37II$8E7#E$;34

$;=37;B3BA87U34

$MM=38C@FAI3#$3$5#7E;38'GH378#EF7#E$;37;B3

BA78#EF7#E$;3#EPA38$;I#7;#I34VV"W3ONOJ>3VVV"3W3ONOOJ3FE73!;A317U3'X!Y'3M$66$KAB35U3ZC;;A##[I3PC6#E"6A38$P"7@E%I$;3#AI#3#$312=N34!=3!"#$%

&'(3456CA:3;3<3Q=>3$"#$%

&'(*++34?@AA;:3;3<3\=37;B3$"#$%

&'())34@AB:3;3<3/=3766378#EF7#A3

8'GH3 E;3 @AI"$;IA3 #$3 56CA3 6E?L#3 E;3 73 "$KA@3 BA"A;BA;#3 P7;;A@N3 ];#@7;IMA8#AB3 ,-.&/03 8A66I3 4"C@"6A:3 ;3 <3 \=38$;#7E;E;?3#LA3T6$)A;I$@3"67IPEB3IL$K3;$3@AI"$;IA3#$37;U3"$KA@3$M356CA36E?L#N3:";3H$KA@3@AI"$;IA38C@FAI3E;3:!;3KA@A3ME#3KE#L3;$;%6E;A7@3@A?@AIIE$;3#$3"@$BC8A3-H

DO3F76CAI3M$@3$"#$%

&'(34B7@9356CA:3;3<3R=>3$"#$%

&'(*++34B7@93

?@AA;:3;3<30=37;B3$"#$%&'())3 4B7@93 @AB:3;3<3Q=N3 :0;3(A"@AIA;#7#EFA3$;38A663KAI#A@;356$#N3 :=;3!;38A663KAI#A@;3

IL$KI3 E;8@A7IAB3 @L$B$"IE;3 675A63 E;3$"#$%&'(34B7@9356CA:3;3<3\=>3$"#$%

&'(*++3 4B7@93?@AA;:3;3<3\=37;B3$"#$%

&'())34B7@93@AB:3;3<3\=3$FA@3C;#@7;IMA8#AB3,-.&/038A66I376$;A3456789:3;3<3\=34VVV3"3W3ONOOJ3FE73!;A317U3'X!Y'3

M$66$KAB35U3ZC;;A##[I3PC6#E"6A38$P"7@EI$;3 #AI#3 #$3,-.&/038A66I376$;A=N3 :/;3HA@8A;#3$M3IC@M78A3 M6C$@AI8A;8A3 #$3#$#763M6C$@AI8A;8A3M$@3$"#$%

&'(34!^3B7@9356CA:3;3<3J&&=>3$"#$%

&'(*++34*++3B7@93?@AA;:3;3<3DD=37;B3$"#$%

&'())3

4))3B7@93@AB:3;3<3JD=34VVV3"3W3ONOOJ3FE73!;A317U3'X!Y'3M$66$KAB35U3^$;MA@@$;E[I3PC6#E"6A38$P"7@EI$;3#AI#N3'663B7#737@A3A_"@AIIAB37I3PA7;3`3)-GN3'6636E?L#3"C6IAI37@A3\R03;P37#3J1a8P&3C;6AII3$#LA@KEIA3;$#ABN

!"#$%&'(

!^ ))*++

*&'(

b$@I9$6E;&'(<

&'())

&'(12

&'(2++

2EPA34PE;=

H$KA@346$?31a8P&=

2EPA34PE;=

O J & 0 \ DO

0

c

/

J&

4 !"#$%&'(

!"#$%&'())

!"#$%&'(

!"#$%&'(*++

,-.&/0

2EPA34PE;=

#$%%&!'!()*+,-./0$+!-1

!"#$%&'())

!"#$%&'(

!"#$%&'(*++

8'GH34b$6B357IA6E;A=

dI$"@$#A@A;$6

&'())

&'(12

&'(2++

O J & 0 \ DO

DO

JOO

JDO

,-.&/0

%J O J & 0 \ DO

&D

DO

RD

,-.&/0

O

DO

JOO

JDO

!^ ))*++,-.

d;#A;IE#U34'N]N=

>>>>>>>>>

!"#$%&'())

!"#$%&'(

!"#$%&'(*++

,-.&/0

%J3'5ae&3'5

eJ3'5a%&3'50 =

O

&O

\O

cO

QO

JOO

f3GAP5@7;A3FI

#$#763E;#A;IE#U

!^ ))*++

/

VVVVVV

!"#$%&'(

7

!; !MM

>>

>>

>>>

O

&

\

c

Q

JO

ONO

ONJ

ON&

ON0

ON\

OND

-HDO341a8P&=

%0 %& %J O JO

&

\

c

Q

JO

J&




34GE;=

!!"#$% &'()*

!"#"

#$%%&'(')*!+,-./0$+'-12-3%*/4!&-45)*+5&-56-4789-:/0)!&/)0;-<!=--+,--./01)20/3/4)5465(73(8)5/96'65965(/769:4(8)5/85/80)'7)(6765);*859:469/<=/ >?/4">@/85/!

!"#$%/<A;34BC/5/D/-?/6E'760085F/46;;0/<GH

IJ/D/-IJ/KL*/!JM/ >C/

5/D/-?N//<"=/&'()*!"#/<5/D/O/6341/(763(P65(/F7):'?/6E'760085F/46;;0/01)2/3/769:4(8)5/85/;8F1(/<I/064?*859:469/

4">@/85/(16/'7606546/)Q/+,/--./<;8;34C/=JJ/ >?N/<4=/&'()*!"#/<5/D/O/6341/(763(P65(/F7):'?/6E'760085F/46;;0/

01)2/3/769:4(8)5/85/;8F1(/<I/064?*859:469/4">@/85/(16/'7606546/)Q/0:73P85/<)735F6C/=JJ/ >?N/<>=/+,/--.R/A:(/5)(/0:73P85R/01)20/3/08F58Q8435(/769:4(8)5/85/'63B/4">@/<SS/'/D/JNJJMT/U83/V(:965(W0/:5'38769/(2)*(38;69/(*(60(C/5/D/O/6341/F7):'?N/";;/93(3/376/3(/7))P/(6P'673(:76/:5;600/)(1672806/5)(69N/";;/93(3/376/6E'760069/30/P635/X/VY>N/";;/;8F1(/':;060/376/-ZT/5P/3(/=$L4P!/:5;600/)(1672806/5)(69N/

#$%%&'(')*!+,-./0$+'-1

*= J = ! T - IJ=!T-IOZ

J = ! T - IJ=!T-IOZ

J

!

-

O

M

**

*M *Z *O *I *- *T *!J

!J

-J

OJ

MJ

=JJ

+,/--./[;)F/>\

]/>3E/80)'7)(6765);

H)5(7);

+,/--.<=JJ/ >?

V:73P85<=JJ/ >?

H)5(7);

H)5(7); H)5(7);+,/--.<=JJ/ >?

V:73P85<=JJ/ >?

&'()*!"#4 >

%8P6/<P85? %8P6/<P85?

4">@/<,);9/A306;856?

4">@/<,);9/A306;856?

4">@/<@63B?

!"##$%&%'()*+,-./"*%,01,2#(.3)$,34'(*4$,45,#678,9./')$.'/:,;)<,,!"#$"%"&'(')*"+),,-&./0.'%+%1.2+#3!4+'),"5.-$%"+6.$+!

78!9:+;/0(5<=>+!

78!:??+;@$""&=+(&A+!

78!BB+;$"A=C+;==+!"#$"%"&'(')*"+),,-&./0.'%+%1.2+#3!4+

'),"5.-$%"+6.$+.#'.D!78!+;A($<+/0-"=>+.#'.D!

78!E??+;A($<+@$""&=+(&A+.#'.D!

78!BB+;A($<+$"A=C+;3=+:),"+5.-$%"+.6+

)%.#$.'"$"&.0+;F+ G=D)&A-5"A+#3!4+"H#$"%%"A+(%+6.0A+/(%"0)&"+6.$+!78!9:+;/0(5<I+&+J+K=>+!

78!:??+;@$""&I+&+J+L=+

(&A+!78!BB+;$"AI+&+J+M=+;N#+O+PCPK>+NN#+O+PCPF+*)(+B'-A"&'%Q+-&#()$"A+'D'"%'+'.+9:=C++;>=+:),"+5.-$%"+.6+0)@1'+;F+

,)&=D)&A-5"A+#3!4+"H#$"%%"A+(%+6.0A+/(%"0)&"+6.$+.#'.D!78!+;A($<+/0-"I+&+J+FP=>+.#'.D!

78!E??+;A($<+@$""&I+&+J+

L=+(&A+.#'.D!78!BB+;A($<+$"AI+&+J+R=+;N#+O+PCPK+*)(+B'-A"&'%Q+-&#()$"A+'D'"%'+'.+9:=C+800+A('(+($"+"H#$"%%"A+(%+

,"(&+S+B3GC+800+0)@1'+#-0%"%+($"+LRM+&,+('+F+9T5,7+-&0"%%+.'1"$2)%"+&.'"AC

!"##$%&%'()*+,-./"*%,0

P 7 L U MP UP VPP

FK

MP

LK

N

NN

N N

3 >78! W#'.D

78!

:),"+;,)&= :),"+;,)&=

W#'.D78!BB

W#'.D78!

W#'.D78!E??

78!BB

78!9:

78!:??

P 7 L U MP UP VP F7PP

7

L

U

X

FP

N

N

N N

P F 7 K FK MP78!9:

#3!4

D(5')&

UP VP P F 7 K FK MPW#'.D78!

#3!4

D(5')&

UP VP F7P

P F 7 K FK MP78!:?? ;,)&=

#3!4

D(5')&

LL

L7

LM

UP VP P F 7 K FK MPW#'.D78!E??

#3!4

D(5')&

UP VP F7P

P F 7 K FK MP78!BB ;,)&=

#3!4

D(5')&

LL

L7

LM

UP VP P F 7 K FK MPW#'.D78!BB ;,)&=

#3!4

D(5')&

LL

L7

LM

UP VP F7P

) =

#3!4+;Y.0A+/(%"0)&"=

#3!4+;Y.0A+/(%"0)&"=

LL

L7

LM

;,)&=

LL

L7

LM

;,)&=

LL

L7

LM

;,)&=

!"#$%&'()*+

,-.//

0"#$%& 1,"#$%&

!"##$%&%'()*+,-./"*%,0

!"##$%&%'()*+,-./"*%,01,2#(34!567!!,.'(%*')$.8)(.3'9#-22%)%*&34#)%$5#(*%&)6#67*8#439:#*;#*()*+!

,-.//#

<(65=2*9*4*>52#>52?#;*44*8%&@#(7*)*6)%$=43)%*&A#/9345#B3>#C#1"# $A#

!"##$%&%'()*+,-./"*%,01,2#(34567,%8#*%99.'/,:)&;<<!,=>6,'%"*3'9,)*%,3#(.?)$$+,9%'9.(.@%,!"#$!$%

)'A, A*.@%, %8?.()(3*+, )?(.@.(+B,!"#$"%"&'(')*"+ ,)%'-.$(/%+ 0122+/%+ 3)&4+ -5+ )%-6('"7+ %)&.6"+ 8&)'+ %,-9)&.+!"#$%&'%+ )&+&"8$-&(6+5)$)&.+)&+-#'-:

5;!+"<#$"%%)&.+(&)/(6%+)&+$"%#-&%"+'-=+C)D+>2+%"?+-5+6).,'+%')/86(')-&+

5-66-9"7+3@+A2+%"?+-5+&-+6).,'B++CED+C+%"?+-5+6).,'+-&B+C+%"?+-5+6).,'+-55D+C?D+!"#$"%"&'(')*"+,)%'-.$(/+0122+/%+3)&4+-5+ )%-6('"7+%)&.6"+8&)'+%,-9%+(%#$%&'%+ )&+&"8$-&(6+ 5)$)&.+ )&+ $"%#-&%"+ '-+>2+%"?+ 6).,'+%')/86(')-&D+ CAD+E"8$-&(6+$"%#-&%"%+'-+6).,'+0&+F+1>+8&)'%4+)&+-#'-:

>;!+"<#$"%%)&.+(&)/(6%+5)'+9)',+&-&:6)&"($+$".$"%%)-&+'-+

-3'()&+-&+0"-&4+(&7+-55+0"

-554+')/"+?-&%'(&'%+0GG#+H+2D221+*)(+I'87"&'%J+8&#()$"7+':'"%'4D+;66+6).,'+#86%"%+($"+KLM+

&/B+1+NO?/>D

!"##$%&%'()*+,-./"*%,0

P#'-:

>;!+QR;+

')/"+?-&%'(&'%+0%4 >2

1C

12

C

2-& -55

A P#'-:>;!

GG

EP#'-:

>;!

C%+O+C%

2 12 >2 M2 K2S)/"+0%"?4

2

L

1K

>1

T)$)&.+$('"+0UV4

:>2 :1C :12 :C 2 C 12 1C >2S)/"+0%"?4

2

12

>2

IWX213+*%D+YZS21

W"$)"*"&'+!(%'"$%B+$"5"$"&?"+F+YZS21B+3)&+F+122+/%

T$"[8"

&?@+0)/

#O%"?4

2 >2 K2 A2 \2 122 1>2 1K2 1A2 1\2 >22S)/"+0%"?4

2

C

12

IWX2M(+*%D+YZS21

W"$)"*"&'+!(%'"$%B+$"5"$"&?"+F+YZS21B+3)&+F+122+/%

T$"[8"

&?@+0)/

#O%"?4

2

K

\

1>

T)$)&.+$('"+0UV4

>2%+O+A2%

P#'-:>;!?

)

>2%+O+A2%

P#'-:>;!

2 >2 K2 A2 \2 122 1>2 1K2 1A2 1\2 >22S)/"+0%"?4

2

12

>2

M2

IWX213+*%D+YZS21

W"$)"*"&'+!(%'"$%B+$"5"$"&?"+F+YZS21B+3)&+F+122+/%

T$"[8"

&?@+0)/

#O%"?4

2

12

>2

M2

K2

T)$)&.+$('"+0UV4

2

S)/"+0%"?4

>2 K2 A2 \2 122 1>2 1K2 1A2 1\2 >22

2

S)/"+0%"?4

>2 K2 A2 \2 122 1>2 1K2 1A2 1\2 >22

!"##$%&%'()*+,-./"*%,012,34(.5)(.6',67,8#(69:3;,.',(<%,=>3,#*6&6(%?,)'@.%(+9$.A%,B%<)5.6*C,,!),9,B"#

$%&'())*+,#+-# .(,/*0$12 0314#!5'((,"# *,#/6(#78)+.8/('8.#89:5;8.8<#=*)).#&)(>;+?+.+'(;#'(;@#)?8.(#78'#A#2BB#C9#!)"#8,;#DE#C9#!B"<#!4"#FG('85(#G(.+?*/:#8,;#!D"#8G('85(#(,/'*()#*,#$HI#8'(#,+/#)/8/*)/*?8..:#;*--('(,/#7(/J((,#+&/+0!

DFK#!'(;@#,#A#L"#8,;#?+,/'+.#8,*98.)#!7.8?MN#,#A#2B"<#F..#;8/8#(%&'())(;#8)#9(8,#O#P$I<#

!"##$%&%'()*+,-./"*%,01

Q(F

RSF

DE

DB

2E

2B

E

BT&/+0

DFKQ+,/'+.

U(.+?*/:#!?9V)"

4

T&/+0DFKQ+,/'+.

$,/'*()

LE

EB

B

DE

D$.(G8/(;#W('+#98W(

E.??$>%'(.9F-G

E.??$>%'(.9F-G

H%3

H%3

=>3 =>3

) B

!"#$%&'()*+,'%%

!"#$%&'()*--'%%

!"#$%&'()*./'%%

!"#$%&'()*/,'%%

!"#$%&'()*01'%%

2#345(6&7899 (:;4:(-<=(%6>#""?

@;45A'B5C#"2#345(DEF@;45A'B5C#"

,

/

,

0

/

.

)

-

G

1)+

)+

0

-,

G

1

H

.)

/

) .

1

H

-

0

G

G

H

,

1

)+

0

/

.

-

)

H

/

,

0

/

.

)-

G1

)+

)+

0

-

,

G

1

H

. )

/

)

.

1

H

-

0G

,

G

H

,

1)+

0

/

.

-)

H

!"##$%&%'()*+, -./"*%, 001,2')(3&.4)$, 43'5.*&)(.3', 35, 6.*)$, %7#*%88.3', )'9, 5%**"$%, .&#$)'(8:,!5I&4#"&I'I#345(6&7899 (:;4:(!

-<=(%6>#""?' J"#KL' #M;"#NN5:3' :B' )+' 53K5O5KP&I' &35%&IN*' !I&AQ' 5N' A:""#N;:3K53$' I#345(

6&7899 (:;4:(!-<=(%6>#""?, B5C#"':;45A' 5%;I&34*'!5I&4#"&I' I#345(RE) (DEF''J$"##3L'#M;"#NN5:3':B')+'53K5O5KP&I'

&35%&IN*'D"#?'5N'A:""#N;:3K53$'I#345(RE) (DEF'B5C#"':;45A'5%;I&34*'SP%C#"N'53K5A&4#'53K5O5KP&I'&35%&IN'&3K'O5"&I'I&C#I'T&N'$#3#"&II?';"#N#34'4>":P$>:P4'4>#'534#3K#K'4&"$#4*,,

!"##$%&%'()*+,-./"*%,00

!"##$%&%'()*+,-./"*%,01

2

3 4

!"#"$%&'(

))#*

!"#"$%&'(

))#*

+

,-./01

0%'23 04 05 06 07 08 09+

!"#"$%&'(+:9,*

9

)

;<=/<<;

0%'23

!"#"$%&'(

9#*

>?2@$&'(

):6?A

!"#"$%&'(

))#*

!"#"$%&'(

))#*

0%'23

!"#"$%&'(

))#*

!"#"$%&'(

))#*

!"#"$%&'(

))#*

!"#"$%&'(

))#*

.%BC&D2"(EDF'G

0%1&"?$3DH$,I

J+++

K+++

)+++

9+++

+L#&'-

);M!'?&('N

%

/?&(%(31D%?&'D2"(E

+

9+

)+

K+

J+

L#&'-);M!'?&('N

5

)

!"##$%&%'()*+,-./"*%,016,7.*%$%88,#9:(:;)4(.<)(.:',:5,=#(:;1>?,.',(9%,@A>,#*:&:(%8,)'B.%(+;

$.C%,3%9)<.:*D,,H)ID!%($@%&D2%"B(",D'OD"?D3?3(BPDC"(Q31&3(D'#3("&%?BD"&D9:4DRCS:D>&D$'?1%1&1D'OD"?&3??"TD%,#32"?$3D ,"&$C%?BD $%($@%&1TD !'$E$('O&-U"N&'?D ,@N&%#N%3(TD "?2D ,%$('-./01:D H3ID /?N"(B32D Q%3VD 'ODV%(3N311D '#3("&%'?D %?D &C3D "%(:D W$"N3D F"(D XD 9D ,,:D H4ID Y'@13D V%&CD "&&"$C32D ("2%'O(3Z@3?$PD #'V3(DC"(Q31&3(:DH2IDM"2%'O(3Z@3?$PD#'V3(DC"(Q31&3(DV%&CD[-./01:D[-./01D"(3D23&"$C"FN3DO(',D&C3DC"(Q31&3(D"1D1C'V?D%?D&C3D%?13&:DW$"N3DF"(1DXD4D,,:DH%IDL#&'-!

);MDH(32TD?DXD6ID"?2D$'?&('ND"?%,"N1DHFN"$E\D?DXD99ID

&("Q3ND1%,%N"(D2%1&"?$31D"?2D,"E3D1%,%N"(D3?&(%31DH5ID %?&'D&C3D2"(ED %?D &C3D N%BC&D2"(EDF'GD"11"P:D;NND2"&"D3G#(31132D"1D,3"?D]DW/Y:DD

!"##$%&%'()*+,-./"*%,01

!"#$%&'()*(+(!*"!,(!*-./-0()12&30(&4.5555

.'()*(+(6$2178(9$7:"$(;<=>?@A/550(&4.5555((

BC<D&'()*(+(1?EE(F$G%!E$%G(H&5&60(&4&555

.'()*(+(6$2178(9$7:"$(;<=>?@AI550(&43555((

2,3(*$2,3(*$

J$GK8?(&L(!%M(FKBBE?,?%#!8>(J$GK8?(.!((

.)<

!"##$%&%'()*+,-./"*%,014,5)6,6%7(%*',8$9(,/%$7:

2,3(*$ 2,3(*$

J$GK8?(&L(!%M(FKBBE?,?%#!8>(J$GK8?(.!NB#72

.)<

!"#$%&'()*(+(!*"!,(!*-./-0()12&30(&4.5555

.'()*(+(6$2178(9$7:"$(;<=>?@A/550(&4.5555((

BC<D&'()*(+(1?EE(F$G%!E$%G(H&5&60(&4&555

.'()*(+(6$2178(9$7:"$(;<=>?@AI550(&43555((

FKBBE?,?%#!8>(J$GK8?(.*

.)<(O&,$%($:7(P!:QR

ST(U=)

2,3(*$ 2,3(*$

!"#$%&'()*(+(!*"!,(!*-./-0()12&30(&4.5555

.'()*(+(6$2178(9$7:"$(;<=>?@A/550(&4.5555((

BC<D&'()*(+(1?EE(F$G%!E$%G(H&5&60(&4&555

.'()*(+(6$2178(9$7:"$(;<=>?@AI550(&43555((

SS(U=)S.(U=)

ST(U=) SS(U=)S.(U=)

ST(U=) SS(U=)S.(U=)

!"##$%&%'()*+,-./"*%,01

!"#$%&'()'*'+,--'./012-/10'34%5%678'%9%555

:&'()'*'6/;+<='>/<?@/'''1)))))))))'A#BC,DEF558'%9G555''

2,3(*$ 2,3(*$

.H!!-,I,1J2=C'K/0H=,':L

:(#'3J<J2-'"#$7

!"##$%&%'()*+,-./"*%,014,5)6,6%7(%*',8$9(,/%$7:

2,3(*$ 2,3(*$

.H!!-,I,1J2=C'K/0H=,':MN!J<;

:(#'3!<O,=7

D<J2-'"#$%&'()'*'+,--'./012-/10'3PQ4Q.78'%9RG55

:&'()'*'6/;+<='>/<?@/'A#BC,DER558'%9%5555''

!"#$%&'()'*'+,--'./012-/10'4%5%68'%9%555

:&'()'*'6/;+<='>/<?@/'A#BC,DEF558'%9G555''

.H!!-,I,1J2=C'K/0H=,':ST"$:4U'21L'N!J<;

:(#

3L2=V'2@J/W/JC72@J/1

%&'()'*'2)@2I'2)Q:RQ8'(+;%G8'%9:5555:&'()'*'6/;+<='>/<?@/'A#BC,DER558'%9:5555''

!"#$%&'()'*'+,--'./012-/10'4%5%68'%9%555

:&'()'*'6/;+<='>/<?@/'A#BC,DEF558'%9G555''

N!J<;:(#T"$:4U N!J<;

:(#T"$:4U

2@J/1%&'()'*'2)@2I'2)Q:RQ8'(+;%G8'%9:5555

:&'()'*'6/;+<='>/<?@/'A#BC,DER558'%9:5555''

2,3(*$ 2,3(*$

.H!!-,I,1J2=C'K/0H=,':MT"$:4U'3!<O,=7

!"#$%&'()'*'+,--'./012-/10'4%5%68'%9%555

:&'()'*'6/;+<='>/<?@/'A#BC,DEF558'%9G555''

2@J/1%&'()'*'2)@2I'2)Q:RQ8'(+;%G8'%9:5555

:&'()'*'6/;+<='>/<?@/'A#BC,DER558'%9:5555''

PU'VB( PP'VB(P:'VB(

PU'VB(PP'VB(P:'VB(

PU'VB(

PP'VB(P:'VB(

PP'VB(P:'VB(

PP'VB(P:'VB(

!"##$%&%'()*+,-./"*%,01

!"#$%&'()'*'+,--'./012-/10'3%4%56'%7%444

8&'()'*'5/9+:;'</:=>/'?#@A,BCD446'%7E444''

2,3(*$ 2,3(*$

.F!!-,G,1H2;A'I/0F;,'J2

8(#KB

!"##$%&%'()*+,-./"*%,014,5)6,6%7(%*',8$9(,/%$7:

2,3(*$ 2,3(*$

.F!!-,G,1H2;A'I/0F;,'J2

8(#BLL

!"#$%&'()'*'+,--'./012-/10'3%4%56'%7%444

8&'()'*'5/9+:;'</:=>/'?#@A,BCD446'%7E444''

2>H/1%&'()'*'2)>2G'2)M8JM6'(+9%E6'%784444

8&'()'*'5/9+:;'</:=>/'?#@A,BCJ446'%784444''

.F!!-,G,1H2;A'I/0F;,'J2

8(#..

!"#$%&'()'*'+,--'./012-/10'3%4%56'%7%444

8&'()'*'5/9+:;'</:=>/'?#@A,BCD446'%7E444''

2>H/1%&'()'*'2)>2G'2)M8JM6'(+9%E6'%784444

8&'()'*'5/9+:;'</:=>/'?#@A,BCJ446'%784444''

2>H/1%&'()'*'2)>2G'2)M8JM6'(+9%E6'%784444

8&'()'*'5/9+:;'</:=>/'?#@A,BCJ446'%784444''

NN'O@(N8'O@(

NP'O@(

NN'O@(N8'O@(

NP'O@(

NN'O@(N8'O@(

NP'O@(

!"##$%&%'()*+,-./"*%,01

!"#$%&'()'*'+,--'./012-/10'3%4%56'%7%444

8&'()'*'5/9+:;'</:=>/'?#@A,BCD446'%7E444''

.F!!-,G,1H2;A'I/0F;,'J2K!H:9

8(#LB

!"##$%&%'()*+,-./"*%,012,3)4,4%5(%*',6$7(,/%$58

9,:(*$ 9,:(*$

.F!!-,G,1H2;A'I/0F;,'J2'K!H:9

8(#BMM

!"#$%&'()'*'+,--'./012-/10'3%4%56'%7%444

8&'()'*'5/9+:;'</:=>/'?#@A,BCD446'%7E444''

2>H/1%&'()'*'2)>2G'2)N8JN6'(+9%E6'%784444

8&'()'*'5/9+:;'</:=>/'?#@A,BCJ446'%784444''

.F!!-,G,1H2;A'I/0F;,'J2'K!H:9

8(#..

!"#$%&'()'*'+,--'./012-/10'3%4%56'%7%444

8&'()'*'5/9+:;'</:=>/'?#@A,BCD446'%7E444''

2>H/1%&'()'*'2)>2G'2)N8JN6'(+9%E6'%784444

8&'()'*'5/9+:;'</:=>/'?#@A,BCJ446'%784444''

2>H/1%&'()'*'2)>2G'2)N8JN6'(+9%E6'%784444

8&'()'*'5/9+:;'</:=>/'?#@A,BCJ446'%784444''

OO'P@(O8'P@(

OQ'P@(

OO'P@(O8'P@(

OQ'P@(

OO'P@(O8'P@(OQ'P@(

!"##$%&%'()*+,-).$%,/

!"##$%&%'()*+,-).$%,/!"!#$%%&'(!)*!+&,$-#!)./&01-2!3)%4&'015!678!&12!)4/)"

6789!:-,-3/!'-*-'"

-13-2!+&,$-#!)./&01-2!*')%!,0/-'&/$'-9

37;< 678 )4/)" 678

=>?@A=<?@ BC!D!E!1; @9F!D!@9B!GA3%6

)1!H%01I @9JE!D!@9BC @9K?!D!@9@L

)**!H%01I @9KF!D!@9@F @9KE!D!@9B

4=8M

=<?@ B"L!N;6? C9L!D!B9C!GA3%6

01/-'1&,0O&/0)1

)1!H%01I 69K 69K

'-3)+-'(!*')%!2-#-1#0/0O&/0)1

'-3!H%01I BB96"BE9JLCP?FPE@ CF

1-$')1&,!&3/0+0/(

)1!H#-3I " K9B!D!B

)**!H#-3I " C9K!D!@9C

!"##$%&%'()*+,-).$%,/

!"##$%&%'()*+,-).$%,/!"!#$%%&'(!)*!+&,$-#!)./&01-2!3)%4&'015!6789:;!

678<<;!

678:==;&12!)4/)"

678;!)4/)"

678<<;!)4/)"

678>==?

37@A )4/)" 678 )4/)" 678<< )4/)" 678

>==678

9:678

<<678

:==

BCDEFBADE 6DGC E?6!H!E?E6 E?I!H!E?EJ E?K6!H!E?EL I?M!H!E?L K?L!H!E?K E?N!H!E?6

)1!O%01P INGC E?LQ!H!E?EJ E?QL!H!E?EJ " E?J!H!E?L K?L!H!E?K E?N!H!E?6

)**!O%01P INGC E?QL!H!E?EJ 6?6!H!E?I " K?6!H!E?EK D?D!H!E?N K?Q!H!E?K

)1!O%01P 6DGC K?K!H!E?K 6?D!H!E?6 K?Q!H!E?6 I?Q!H!E?D L?K!H!E?L 6?6!H!E?6

)**!O%01P 6DGC I?I!H!E?I N!H!K N?J!H!K?N I?I!H!E?EL D!H!E?M L?N!H!E?D

01/-'1&,0R&/0)1

)1!O%01P INGC 6?M " 6?M " " "

'-3)+-'(!*')%!2-#-1#0/0R&/0)1

'-3!O%01P INGC LJ 1F& " " " "

!"##$%&%'()*+,-%.%*%'/%0

!"#$ %&''&()$ *#)$+,-./&,(.0)$ *#$ 1$ 2345.)$6#$ 7#$ *.89.5:0;:&3,$ ;,-$ 0.<=<>&,($ 3?$ @.:;$ AB;-0.,.0(&<$ 0.<.':305$

'.0C&:$0.<.':30$0.5.,5&:&D;:&3,#$63>#$%4;0C;<3>#$EF)$GGGHGFG$IJ""!K#

GL#$ M;0;N)$2#$*#)$6.,;0-)$2#)$O.0(953,)$*#$*#$P#)$Q3>;'&.:03)$+#B6#$1$Q;03,)$6#$P#$R4.$Q3,5.0S.-$*.S.,B

R0;,5C.C@0;,.$*.89.,<.$T%IUKA)VW$3?$:4.$PB%03:.&,BQ39'>.-$X.<.':30$*9'.0?;C&>=$X.(9>;:.5$69>:&'>.$%03'.0B

:&.5$3?$:4.$#@.:;#AB+-0.,.0(&<$X.<.':30#$M&3<4.C&5:0=$I635<#K$VE)$J!ELFHJ!EJE$IJ""!K#

optodynamic simulation of β-adrenergic receptor...

Documents