© 2018 Nature America, Inc., part of Springer Nature. All rights reserved. systems membrane into machineries functional reconstituting recapitulated been have in Reactions membrane dynamics membrane synthesis of ATP verter, and an ATP synthase photocon a , membrane-bound two minimum a at requires a means of controlling coupled reactions. Converting light into provide a sustainable energy source for biomimetic vesicle systems and maintain, couple, and control cascade reactions. Photosynthesis could work has largely relied on spontaneous reactions, making it difficult to Massachusetts, USA. Massachusetts, K.K.P. ( for Nanobio Science and Technology, Harvard University, Cambridge, USA. Massachusetts, should Correspondence be addressed to K.S. ( trochemical gradient across the membrane, generating a proton motive Engineering, Engineering, John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, USA. Massachusetts, interactions –protein study to applied been have approaches 1 organelles of compartments modular the across distributed are reactions metabolic complex cells, Inside Keel Yong Lee A artificial organelles Photosynthetic sustain and control nature biotechnology nature and and Institute of Biological Interfaces, Sogang University, Seoul, Korea. Sungkyunkwan Sungkyunkwan University, Suwon, Korea. lipid vesicles in a simplifiedusing reactions and isolatedmetabolic intracellular specific of investigation ting Biomimetic membrane systems have been used as bioreactors, permit behaviors. cellular complex and homeostasis exhibit that networks regulatory with systems vesicle biomimetic of development the enable may organelles photosynthetic Switchable morphology. vesicle outer altering latter the with polymerization, actin and fixation carbon reactions, ATP-dependent two of control optical demonstrated and system protocellular a form to vesicle giant a in organelles photosynthetic the We encapsulated ATP synthesis. impedes light green and facilitates light red ATP synthesis: of control dynamic allows photoconverters two the of activation optical Independent ATP synthesis. enable proteorhodopsin) derived - and II photosystem (-derived photoconverters two and ATPsynthase (ATP) An reactions. intravesicular directing of means a and source energy sustainable a both provides that light-harvesting switchable, a tested and built, designed, we Here challenging. is reactions Received Received 5 April 2017; accepted 11 April 2018; published online 28 May 2018; L L Mahadevan Institute Institute of Biological Interfaces and Department of Chemistry, Sogang University, Seoul, Korea. T [email protected] P-dependent P-dependent reactions in a system protocellular 3– 5 . However, maintaining and controlling these these controlling and maintaining However, . 13 6 1 , , 1 7 7 – 4 Department Department of Organismic and Evolutionary Biology, Department of Physics, Wyss Institute for Biologically Inspired Engineering, Kavli Institute

4 , DNA

9 , 1

, , Kwang-Hwan Jung 0

, , Sung-Jin Park , and cytoskeleton formation cytoskeleton and , 1

5 5 , RNA . . The photoconverter establishes an elec advance online publication online advance ), ), T.K.A. ( 16 4 , 1 Sogang-Harvard Research Sogang-Harvard Center for Disease Biophysics, Sogang University, Seoul, Korea. 7 , and proteins [email protected] in vitro 2 , 4 , , Keon Ah Lee environment 5 , , Tae Kyu Ahn 18 , ), ), or K.-H.J. ( 1 1 4 9 , , in vitro in 11 2 . However, this . , 1 2 , as well as as well as , 2 , 6 6 , John John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge,

7 . These by by 5 [email protected] , , Se-Hwan Kim

3 , , Kevin Kit Parker 8 - - - , d o i photoconverters owing to their unique photophysical properties photophysical unique their to owing photoconverters ability, determined ability,by the determined p proton-pumping bidirectional Moreover, pH-dependent PR exhibits kDa) (350 complex protein large a is PSII whereas mainly green light initiates direct proton pumping in PR in pumping proton direct initiates light green mainly whereas membrane ( membrane the and growth filament actin between vesicle interaction lipid dynamic the and giant the inside assembly actin control to us enabling polymerization, actin with photosynthesis organelle switchable the ( respectively light, green PR or and red PSII with of stimulation by impeded or facilitated be dynamically can synthesis ATP Thus, PSII. of action the counteracting tion, direc pumping its changes PR pH, lower at but PMF, the increase phyll molecules inside chromophores, and PR is a single protein protein single a is kDa) (27 PR and chromophores, inside molecules phyll conversion of ADP to ATP to the ADP of catalyze conversion to synthase ATP the inducing membrane, organelle’s artificial the across PMF generates (PR), proteorhodopsin derived bacteria- and (PSII) II photosystem plant-derived photoconverters, form an encapsulated organelle system ( system form organelle an encapsulated (~10–100 vesicle lipid giant a within them encapsulated and modules energy as diameter) or couple with other reactions. intravesicular these proteoliposomes did not dynamically control chemical reactions (D97) light activates electron transport chains in the PSII reaction center in chains the PSII reaction transport electron activates light synthase into proteoliposomes enables ATP synthesis ATP enables proteoliposomes into synthase ATP an and photoconverter a reconstituting that demonstrated have combining photoconverters with specific properties. Pioneering and studiesselecting by ATPsynthesis and generation PMF control directly species have diverse biophysical properties biophysical havediverse species intravesicular reactions. metabolic As photoconverters from different ATP-dependentfor needed energy the provide would ATPsynthesis ATP to ADP conversionof ATPby drives that synthase. force(PMF) : 1 0 We engineered switchable photosynthetic organelles (~100 nm nm (~100 organelles photosynthetic switchable engineered We . 1 0 3 2 3 8 . At a higher pH, both PR and PSII work in conjunction to to conjunction in work PSII and PR both pH, higher a At . / ). ). n 2 b 2 2 5 holding only a single retinal chromophore. Blue and red red and Blue chromophore. retinal single a only holding t Disease Disease Biophysics Group, Wyss Institute for Biologically Inspired . , , Heeyeon Kim 4 Fig. 1 Fig. 1 4 0 2 c , 4 ). & Kwanwoo Shin µ m diameter) embedded with ionophores to to ionophores with embedded diameter) m K 5 a ( of proton the primary acceptor residue 1 Fig. 1 Fig. , , Yasmine Meroz a 3 Department Department of Energy Science, ). We selected PSII and PR as as PR and PSII selected We ). Fig. 1 Fig. Fig. Fig. 5 Department Department of Science 1 2 , 2 0 [email protected] 4 1 , it should be possible to possible be should it , 1

b containing 35 chloro 35 containing ). Activation ). of Activation the two ). Finally, we coupled coupled we Finally, ).

6 s r e t t e l , 13

, 1 4 . However, . ), ),

2 2 2 0 2 1  - - . . , © 2018 Nature America, Inc., part of Springer Nature. All rights reserved. proteoliposomes were in ranges similar to the estimated values as as values estimated the to similar co-reconstituted ranges in PSII–PR were the proteoliposomes of pH intravesicular the in change and to a change in the intravesicular pH of the proteoliposomes ( to pH a of change the proteoliposomes in the intravesicular patch clamp glass chips produced photocurrent ( that the two photoconverters in the lipid bilayers that formed on planar with ATP-dependent actin polymerization and morphological change of the vesicle. the of change morphological and polymerization actin ATP-dependent with ( protons. depleting by ATP synthesis Fig. 2 (Fig. peak) nm (520 PR and peaks) nm 670 and nm (440 PSII of bands absorbance individual of the summation the from resulting co- reconstituted PSII–PR proteoliposomes was demonstrated bythe their black color,in proteins two the of coexistence The Methods). proteins into vesicles ( terium pH measurement with an encapsulated pH indicator (SNARF-1) (SNARF-1) indicator pH ( encapsulated an Ratiometric with membrane. measurement their pH across PMF generated oliposomes prote co-reconstituted PSII–PR the illumination, light white Upon organelles into a giant lipid vesicle; see see vesicle; lipid giant a photosynthetic into the organelles photo (encapsulating the system finally, protocellular and, synthetic proteoliposomes), into synthase ATP of (reconstitution organelle photosynthetic the then vesicles), into ers of (reconstitution photoconvert the proteoliposome first, gressively: used during testing). during used solutions ionic and compositions lipid about information detailed ( membrane. organelle’s the into reconstituted ATPan and PR) synthase and (PSII photoconverters complementary two of activation coordinated the s r e t t e l  Figure 1 Figure b Supplementary Figs. 3 Figs. Supplementary

) Activation of PSII with red light facilitates ATP synthesis by generating protons inside the organelle, while activation of PR with green light impedes impedes light green with PR of activation while organelle, the inside protons generating by ATP synthesis facilitates light red with PSII of ) Activation We purified recombinantly expressed PR We assembled and tested the encapsulated organelle system pro system organelle encapsulated the tested and Weassembled c a Supplementary Fig. 3e Fig. Supplementary and PSII from

Design and applications of the photosynthetic artificial organelle. ( organelle. artificial photosynthetic the of applications and Design Proton photoconvertors (1) Activationof PR Photosynthetic organelle Spinacia oleracea Fig. Fig. 2 Giant vesicle and and Proton ADP H a Water + , , and ). The photosensitive current and the the and current photosensitive The ). 4 ) and patch clamp recording revealed revealed recording clamp patch and ) generation (2) PMF c Supplementary Fig. 2 Supplementary ∆pH ) After organelle energy modules are encapsulated in a giant vesicle, optical stimulation couples ATP synthesis ATP couples synthesis stimulation optical vesicle, a giant in encapsulated are modules energy organelle ) After PSII ATP H and reconstituted the purified + depletion H + Supplementary Fig. 1 Fig. Supplementary generation 2 from ADP Fig. Fig. 2c

Gamma proteobac ADP synthesis , d synthesis (3) ATP , , and Online ATP ), ), which led synthase ATP ATP impediment signal Green light: Fig. Fig. 2e

facilitation signal Red light: ATP for for b ). ). , - - - - - f

a b ATP synthesis facilitates Red light ATP synthesis impedes immediately Green light ) Upon optical stimulation, the artificial organelle synthesizes ATP by synthesizes organelle artificial the stimulation, optical ) Upon teoliposomes ( teoliposomes from calculated measurements of single PSII or PR reconstituted pro i, 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1 pid, phospholi charged negatively a removing by decreased sharply was in the the pH In lipid membrane. gradient addition, photoconverters 13.3–32.3% of the ATP turnover number of of number turnover ATP (13–33 s the of to up 13.3–32.3% was achieved synthase ATP per synthesis ATP the that cates The artificial organelle sustained ATP conversion for 3 days days 3 at month for 1 and ( °C 4 conversion temperature room ATP at efficiency) sustained (half-maximum organelle artificial ( The gradient potential electrochemical light-driven cyanide- the with (carbonyl (oligomycin), that demonstrating protonophore ATP was coupled synthesis directly a with and trifluoromethoxyphenylhydrazone) an inhibitor of ATP synthase treatment upon ( proteoliposomes into the integrated were ATP where at sites synthases the organelles artificial and PSII and PSII of activity and orientation the improved negatively phospholipid the charged that confirmed which PSII), for × 2.9 PR, for × (2.4 M die b otcl ciain a sfiin t idc ATP was 4.3 induce synthesis ATP to for number sufficient turnover maximum was The synthesis. activation optical by driven PMF The proteoliposomes. co-reconstituted PSII–PR into reconstituted Next, ATP synthase purified from from purified synthase ATP Next, ± 0.1 s 0.1 Fig. 2 Fig. advance online publication online advance −1 24 ) −1 , 2 26 h 5 , measured by the luciferin–luciferase assay. This indi This assay. luciferin–luciferase the by measured , ( (Supplementary Note 1 ). Supplementary Fig. 5 Fig. Supplementary

Fig. 2d Fig. polymerization

Actin Ionophore depletion H , + f Green PR ), indicating robust performance of the two two the of performance robust indicating ), Red generation Fig. 2 Fig. RPSII PR H + g ). ATP synthesis sharply decreased decreased sharply ). ATP synthesis ). ). ATP was produced outside the PSII

ails pseudofirmus Bacillus nature biotechnology nature ADP Actin filaments shrci coli Escherichia ′ synthase ATP -rac-glycerol) -rac-glycerol) synthase ATP ATP Fig. 2 Fig. was was

g p ). ). - - - -

© 2018 Nature America, Inc., part of Springer Nature. All rights reserved. decreased ATP synthesis after treatment with a protonophore (FCCP) and an inhibitor of ATP synthase (oligomycin). of ATP (oligomycin). an inhibitor and (FCCP) synthase a protonophore with ATPtreatment after decreased synthesis by significantly of ATP.as shown synthesis the membrane, the across catalyzed pH gradient synthase ATPby a light-induced driven was synthesis reconstituted proteoliposomes, showing independent activity of photoconverters in vesicles ( in vesicles of photoconverters activity independent showing proteoliposomes, reconstituted of single-photoconverter- measurements from as calculated measurements of these values estimated the and proteoliposomes co-reconstituted of PSII–PR generation pH gradient and current photosensitive measured the between no difference was There experiments). independent biologically stimulation, PSII and PR elicited photocurrents ( photocurrents PR elicited and PSII stimulation, of the chromophores ( chromophores of the coefficients absorption and spectrum absorption by the calculated was PR) and (PSII proteoliposomes in the of chromophores number The vesicles. into bars and the colored bands indicate mean mean indicate bands colored the and bars ( ratios peak GC/MS panel: Upper experiments, and representative images are shown. ( shown. are images representative and experiments, independent three from obtained were results Similar PR. and PSII photoconverters the containing proteoliposomes giant and proteoliposomes Lower panel: the mass spectroscopy confirmed the presence of OAA. Data shown represent two independent experiments with similar results. ( results. similar with experiments independent two represent shown Data of OAA. presence the confirmed spectroscopy mass the panel: Lower ( assay fluorometric with measured was concentration OAA it. suppressed inhibitor, carboxylase a pyruvate acid, phenylacetic whereas of OAA, concentration the increased g c a tography–mass spectrometry (GC/MS) showed that the four-carbon four-carbon that the showed (GC/MS) spectrometry tography–mass 6 Fig. Supplementary Figure 2 Figure the artificial organelle over 1 month ( 1 month over organelle artificial the ( assay luciferin–luciferase with measured ATP was at 20 min). concentration ore o te ope rato o cro fxto ( fixation carbon of reaction coupled the for source nature biotechnology nature

We ATP the synthesized tested whether could be as used an energy ATP / chromophores (n/n) 100 200 300 400 500 0

Light-driven ATP synthesis in the artificial organelle. ( organelle. artificial ATPthe in synthesis Light-driven 01 20 15 10 5 0 cp: 9.88×10 cp: 9.23×10 cp: 4.77×10

White light FCCP Oligomycin PR–PSII PR PSII PSII–PR PSII PR cp: chromophores ml eil Giantvesicle Small vesicle Time (min) ml eil Giantvesicle Small vesicle (number ofcp) n PR reconstituted = 7 biologically independent experiments). ( experiments). independent = 7 biologically Vesicle only Supplementary Fig. 2 Fig. Supplementary 4 4 4 ). Both fluorometric assay and gas chroma gas and assay fluorometric Both ).

100 pA 0.2 s advance online publication online advance Supplementary Fig. 6 Fig. Supplementary n d h = 3 biologically independent samples). ( samples). independent = 3 biologically Photocurrents / ATP / chromophores (n/n) chromophores (fA) 300 300 ± –4 –3 –2 –1 ). Data shown represent seven independent experiments with similar results. ( results. similar with experiments independent seven represent shown ). Data s.d. of independent experiments. ( experiments. of independent s.d. 0 0 0 501 5015 0 15 0 15 0 15 0 a 030 10 3 Day 0 SIPR PSII c P =0.006 , d , , b n ) Absorption spectrum of the proteoliposomes, demonstrating the reconstitution of PSII and PR and of PSII reconstitution the demonstrating proteoliposomes, of the spectrum ) Absorption = 7 biologically independent experiments), leading to changes in intravesicular pH ( in intravesicular to changes leading experiments), independent = 7 biologically ) of chromatograms between pyruvate and OAA were calculated as a function of illumination time. time. of illumination as a function calculated were OAA and pyruvate between ) of chromatograms Time (min) Measurement: i. 2i Fig. Estimate: a ) of small proteoliposomes and fluorescence images of small small of images fluorescence and proteoliposomes small of ) Suspension P =0.98 PSII–PR j

) Gas chromatography/mass spectrometry (GC/MS) analysis of carbon fixation products. products. fixation of carbon analysis (GC/MS) spectrometry ) chromatography/mass Gas , j PSII–PR co-reconstituted and and 4 °C RT PSII reconstituted ml eil Giantvesicle Small vesicle Giantvesicle Small vesicle - e n i = 7 biologically independent experiments). ( experiments). independent = 7 biologically was converted back into ADP and reused in the ATP–ADP cycle. cycle. ATP–ADP the in reused and ADP into back converted was A, coenzyme and bicarbonate, which means that the ATPsynthesized acetyl- carboxylase, pyruvate of presence the in pyruvate compound three-carbon the from formed was (OAA) oxaloacetate compound d

i OAA / chromophores (n/n) intra-vesicular pH ) Coupling of ATP synthesis with carbon fixation. Optical illumination illumination Optical fixation. of ATP carbon ) with Coupling synthesis , 7.4 7.6 7.8 8.0 100 200 300 400 500 f , g 0 ) ) P All scalebars=20µm 5 4 3 2 1 0 01 20 15 10 5 0 values were calculated by Student’s two-sided by Student’s two-sided calculated were values

PAA PR–PSII PR PSII d , f Time (min) ). ). ( Time (min) g ) Upon white light stimulation, reconstituted ATP reconstituted stimulation, light white ) Upon PSII–PR PR PSII b

Absorbance (A.U.) 0 0 0 700 600 500 400 P j f = × 3.5 10 GC peak ratio –9 Abundance (%) OAA / (PA + OAA) ∆ pH × 10 / chromophores / 5 min –10 –15 100 100 c –5 50 50 0 0 0 – h f ) Upon white light light white ) Upon 0102030450 350 250 150 50 ) Photosynthetic activity of activity ) Photosynthetic 01 20 15 10 5 0 Wavelength (nm) SIP PSII–PR PR PSII 73 P =0.01 133

−9 −9 PR–PSII 147 (PR vs. oligomycin oligomycin vs. (PR 171 t Si -test. Time (min) OAA 207 Measurement: O s r e t t e l O m/z e O Estimate: Si , O f , , O d n P =0.20 Si – PSII–PR PR PSII = 7 i) 389 Error Error 417

459  © 2018 Nature America, Inc., part of Springer Nature. All rights reserved. increased it at pH < 7.4 ( intravesicular pH of PR reconstituted proteoliposomes at pH > 7.4 but the of PR pumping decreased bidirectionality pH-sensitive the while slope, gradual a with decreased proteoliposomes reconstituted PSII ( pH intravesicular in change of direction the set photoconverters the of activity pH-dependent and PSII with red light and increased by activating PR with green light ( light green with PR activating by increased and light red with PSII activating independently by decreased was pH intravesicular range, pH bidirectional the at proteoliposomes; co-reconstituted PSII–PR of (<7.4) pH range intravesicular bidirectional the of PR determined actin polymerization actin ATP-dependent through formation cells—cytoskeleton in process ( mechanism for controlling coupling reactions, such as carbon fixation a provided PR and PSII of activation independent via ATP synthesis ( organelle ward photosensitive current of PR rapidly depleted protons inside the bled bled control of ATP synthesis ( ( 7.4 as pH intravesicular of limit upper mination led to the only pH, determining a in intravesicular decrease dimensional actin filaments ( filaments actin three- dimensional of growth to leading polymerization, actin ATP-dependent induced and synthesis ATP initiated stimulation Optical it. outside provided were ions magnesium and 7.8, at maintained was system light-off case ( case light-off the to compared ATP of rate synthesis quenching the improved tion ( range pH bidirectional the within was pH intravesicular when light green with PR activating by and impeded light red with PSII activating independently by facilitated was ( range pH to bidirectional pH the intravesicular the bring to enough was min 15 for mination (7.4), pH illu light White for ATPPMF intravesicular synthesis. a positive maintaining thus of limit upper the than higher organelles the outside level pH the keep to selected was (>7.8) level pH initial ( ATP synthesis maximize pH to and level intravesicular organelles’ the to similarly contribute would PR and PSII that to ensure chosen were respectively) 2,000:1, and 260:1, The ratios density of PSII, PR, and ATP over synthase lipid (19,000:1, ( pho ( currents pH-dependent toactive and wavelength- generated proteoliposomes ( toconverters of the activities pho and led the through wavelength- pH-dependent 7 Fig. Supplementary ( synthesis indicating ATP with activity, fixation carbon of carboxylase coupling direct pyruvate of inhibitor an acid, tic 6.5 ( 6.5 pH at current positive a but 8.0 pH at current negative activation a PR generated whereas 6.5, pH at magnitude reduced significantly a generated activation negative current at both pH 6.5 and pH 8, with ( cross-talk minimal with respectively, light, green and ( ciency, we optimized the artificial organelles’ size and electrical charge phenylace by abolished was OAA into pyruvate of conversion The s r e t t e l  and G-actin) within it to form the photosynthetic protocellular protocellular photosynthetic the ( form system to it within G-actin) and (ADP and reactants the organelles the and encapsulated vesicle giant Fig. 3 Fig. Fig. 3 Fig. Fig. 2 Fig. upeetr Fg 16 Fig. Supplementary

Finally, to organelles we simulate a applied ubiquitous the artificial Switchable PMF generation by the photosynthetic organelle ena organelle photosynthetic the by generation PMF Switchable PMF generation by the photosynthetic organelle could be control could organelle by the photosynthetic PMF generation Fig. 3 Fig. f c b and and and and ), PSII and PR could be independently activated with red red with activated independently be could PR and PSII ), upeetr Fg 15 Fig. Supplementary Fig. 3 Fig. b and and Supplementary Fig. 14 Fig. Supplementary Supplementary Video 1 Video Supplementary Supplementary Video 1 Video Supplementary Fig. 3 Fig. c Supplementary Figs. 8 Figs. Supplementary ). This wavelength- and pH-dependent control of of control pH-dependent and wavelength- This ). Fig. 3 Fig. e 2 and and 7 ). . We embedded magnesium ionophores in a in ionophores magnesium Weembedded . Supplementary Supplementary Fig. 10 a ). The initial pH inside the protocellular protocellular the inside pH initial The ). ). Consistent with their absorbance peaks peaks absorbance their with Consistent ). Supplementary Fig. 13 Fig. Supplementary Fig. 3d Fig. Fig. Fig. 3d Fig. 4 Fig. Fig. 3 Fig. . o mrv ecpuain effi encapsulation improve To ). ). , a e , ). In contrast, at pH > 7.4, illu 7.4, > pH at contrast, In ). , , e ). As expected, ATP synthesis synthesis ATP expected, As ). Supplementary Fig. 12 Fig. Supplementary and Supplementary Fig. 17 Fig. Supplementary c Fig. 3 Fig. ). PSII–PR co-reconstituted co-reconstituted PSII–PR ). ). The intravesicular pH of of pH intravesicular The ). Fig. 3 Fig. and and Supplementary Fig. Supplementary 11 c ). ). The pH dependence d 9 ). The wavelength- wavelength- The ). ). Thus, PR activa PR Thus, ). ) because the out the because ) Fig. 3 Fig. Fig. 2 Fig. b ). PSII PSII ). ). The The ). i , and and , and and ). ). ------

experiments). 0–5 min: * min: 0–5 experiments). impeded ATP synthesis, respectively ( respectively ATP synthesis, impeded and facilitated light) (green PR and light) (red PSII of activation optical min), ~15 (at range bidirectional the reached pH the after but synthesis, ATP facilitated PR and PSII both min), ~0–5 (after range bidirectional the in was pH intravesicular the Until ATP of synthesis. control enabled P ( PR with green light ( light green with PR activating by increased and light red with PSII activating by decreased was pH intravesicular range, bidirectional the In 7.4). < (pH organelle the of range bidirectional the determined PR of bidirectionality The proteoliposomes. co-reconstituted PSII–PR in pH intravesicular of direction the set PR and PRII of photoactivities pH-dependent indicate mean mean indicate ( independent experiments). ( experiments). independent proteoliposomes were influenced by pH ( pH by influenced were proteoliposomes sphere that occupied 75–100% of the interior of the protocellular protocellular the to the rupture enough large grew sphere actin the of Eventually system. interior the of 75–100% occupied that F-actin sphere an form to tangled and elongated filaments actin then and 2 Video Supplementary 15–20 min: * min: 15–20 ( ( 3 Figure d n f a c a e values were calculated by Student’s two-sided two-sided Student’s by calculated were values ) Control of the coupled reaction of carbon fixation ( fixation carbon of reaction coupled the of Control ) , , = 7 biologically independent experiments). ( experiments). independent biologically 7 = b e ATP / chromophores (n/n) Intra-vesicular pH ) The wavelength- ( wavelength- The ) ) The photocurrents of PSII and PR in PSII–PR co-reconstituted co-reconstituted PSII–PR in PR and PSII of photocurrents The ) pH 6.5 pH 8.0 100 150 200 250 6.2 6.5 7.0 7.3 7.8 8.1 50 White 0 advance online publication online advance

PSII: 3.19×10 10 8 6 4 2 0 PR: 6.69×10 Optical control mechanism of the artificial organelle. organelle. artificial the of mechanism control Optical Red

Red * – i 15–20min 0–5 min

Green P =0.066 e Green Red P 0.5 fA ± = 0.0005; ** 0.0005; = FCCP 0.5 s s.d. of independent experiments. ( experiments. independent of s.d. ** 4 4 n chromophores chromophores 50 = 7 biologically independent experiments). experiments). independent biologically 7 = 0

d White ) and pH- ( pH- and ) Green P = 2.07 × 10 × 2.07 = Red ** ). The ATP triggered G-actin nucleation, nucleation, G-actin triggered ATP The ). b Green P –

Ligh OFF * = 0.0001; *** 0.0001; = f ) Error bars and the colored bands bands colored the and bars Error ) *** e ) dependent activities of PSII and PR PR and PSII of activities dependent ) n b d f = 10 biologically independent independent biologically 10 = −10 Photocurrents / chromophores (fA) a OAA / chromophores (n/n) ATP / chromophores (n/n) –2.0 –1.5 –1.0 –0.5 ) and light wavelength ( wavelength light and ) 100 100 200 300 400

0.0 0.5 ; ** ; 50 0 0 nature biotechnology nature P P P =0.08 c 5 4 3 2 1 0 02 040 30 20 10 0 = 6.76 × 10 × 6.76 = White = 1.59 × 10 × 1.59 = e Green Red ) The wavelength- and and wavelength- The ) t -test. pH 8.0 b , e Green Red P= 0.62 e n ) * ) = 10 biologically biologically 10 = Time (min) Time (min) P < 0.001. 0.001. < −6 P =0.28 P =0.19 −9 . .

SIPR PSII

White Estimate b

) Green Red

50 © 2018 Nature America, Inc., part of Springer Nature. All rights reserved. representative of three independent experiments with similar results. similar with experiments independent three of representative spherical membrane system into teardrop ( teardrop into system membrane spherical of local curvature. ( curvature. local of distribution a narrow with shape membrane a circular maintained filaments actin and membrane phase Lo2 the between interactions membrane–actin of inhibition the while curvature, local of distribution a broad to leading membrane, local the deformed filaments actin and membrane phase Ld attached to the Ld phase membrane ( membrane phase Ld the to attached the outer membrane. ( membrane. outer the to the amount of ATP produced over time, that is, is, that time, over produced ATP of amount the to proportional be to predicted was sphere F-actin the of radius the of process in actin filament growth. In the model, the analytical increase strated that ATP synthesis by the artificial organelles was the dominant 5 Video Supplementary ( case light-off the with of compared rate polymerization quenching actin the improved light green by activation PR tion, projected area of the F-actin sphere grows linearly in time, that is, is, that R time, in linearly grows sphere F-actin the of area projected experiments. experiments. ( results. similar with ( shown. are data and images representative and experiments, independent three from obtained were results Similar min). (260 ruptured membrane the until grew sphere F-actin The min). (~135 sphere F-actin a single of formation to leading min), (~60 elongation F-actin and min) (~15 nucleation G-actin initiated stimulation follows the relation relation the follows and PR activation ( activation PR and PSII independent via growth filament impeded light and green tated sion is the limiting process. These observations yield yield observations These process. limiting the is sion activation of the artificial organelles. Similar results were obtained from two independent experiments, and representative images are shown. ( shown. are images representative and experiments, independent two from obtained were results Similar organelles. artificial the of activation light red by triggered was lines) (white filaments actin of polymerization ATP-dependent The images. microscopy confocal from system complete the of membrane ( membrane 4 Figure nature biotechnology nature ( Figures 2g 2g Figures sphere. F-actin existing the to organelles the from ATP synthesized of with is process associated the when growth-rate-limiting Supplementary Fig. 18 Fig. Supplementary a 2 (t) (t) Both our analytical model and our experimental results demon results experimental our and model analytical our Both

∝

µ m) m) ( X 50 t Control of actin polymerization and membrane shape of the photosynthetic protocellular system by light. ( light. by system protocellular photosynthetic the of shape membrane and polymerization actin of Control wih s osset ih h agmn ta AP diffu ATP that argument the with consistent is which , and

Z (µm) P Fig. 4 Fig. values were calculated by Student’s two-sided two-sided Student’s by calculated were values 3 d

60 h b

show white-light-induced ATP production that that production ATP white-light-induced show ) The global deformation of phase-separated membranes composed of Ld and Lo mixtures. ( mixtures. Lo and Ld of composed membranes phase-separated of deformation global ) The

d and and 0 Fig. 4 Fig. e ) Optical control of actin growth ( growth actin of control ) Optical

h 0 ATP , f

) Local deformation of a liquid-disordered (Ld) or a liquid-ordered (Lo2) phase membrane. Polymerized actin was strongly strongly was actin Polymerized membrane. phase (Lo2) a liquid-ordered or (Ld) a liquid-disordered of deformation ) Local 0 Organelle ). and and Interaction Repulsive Lo2 phase Supplementary Video 3 Video Supplementary c

c ∝ Actin and and Interaction Attractive Ld phase: ) Actin polymerization was controlled by red and green light. The data are representative of four independent experiments experiments independent four of representative are data The light. green and red by controlled was polymerization ) Actin advance online publication online advance t Green: organelle Supplementary Note 1 Note Supplementary Red: membrane 1/2

Supplementary Video 4 Video Supplementary , while while , White: actin Y (µm) e ) but was separated from the Lo2 membrane by a gap ( a gap by membrane Lo2 the from separated was ) but Phase-separated

outer membrane 50 i ) or mushroom ( mushroom ) or iue 4 Figure b e Projected actin area (%) 20 40 60 80 0 organelle Photosynthetic Nucleation Strong attractiveinteraction(Ldphase) 010102020300 250 200 150 100 50 0 ). Red light facili light Red ). b Ld phase n shows that the the that shows 15 min = 4 biologically independent experiments). Error bars indicate mean mean indicate bars Error experiments). independent = 4 biologically actin 20 µm R(t) R(t) ). In addition, addition, In ). R(t) R(t) j i ) shapes via heterogeneous membrane–actin interactions. ( interactions. membrane–actin heterogeneous via ) shapes Fig. 4 Fig. t White: actin Green: organelle Red: membrane 20 µm -test. ( -test. ). In addi In ).

∝ elongation F-actin ∝ ATP(t)

ATP(t) 01520310 280 135 60 d e and and Time (min) – j ) Membrane deformation caused by impingement of polymerized actin on actin polymerized of impingement by caused deformation ) Membrane F-actin sphere growth - - - - , 01000.1 0.0 –0.1

Curvature (µm and and and eventually rupturing it ( it rupturing deforming eventually and membrane, outer the into protruded they grew, ments terol that produced a weak attractive interaction, and a liquid-ordered ordered phase mixture (Lo1) composed of sphingomyelin and choles a liquid- interaction, a that produced strong attractive phospholipids polyunsaturated of composed (Ld) mixture phase liquid-disordered a filaments: actin with interactions attraction–repulsion of range a exhibiting mixtures membrane different three used we interactions, ain f i-ope proteins for the tip-complex or of mation bundling additional without deformation, membrane of cause sole the was membrane the and filaments actin the between opooia cag o te ue mmrn o te protocel interactions the of membrane–actin via membrane system outer lular the of change morphological ( was on completely dependent ATP organelles by the synthesized artificial polymerization actin that ADP,indicating or organelles, ficial arti stimulation, of light absence in the not form did filaments actin Supplementary Fig. 19 Fig. Supplementary Green: Ldphase Red: Lo2phase Moreover, cytoskeleton formation and manipulation induced induced manipulation and formation cytoskeleton Moreover, 10 White: actin –1 20 µm – ) burst 12 f ). This behavior confirmed that the elastic interaction interaction elastic the that confirmed behavior This ). f ). ( ). Strong repulsiveinteraction(Lo2phase) c g Lo2 phase j

) Strong attractive interactions between the the between interactions attractive ) Strong Projected actin area (%) PEG2000 10 20 30 40 0 2 8 4 300 240 180 120 and and White: actin Green: organelle Red: membrane 20 µm e re Red Green Red Red: membrane(Lo1phase) Fig. 4 Fig. Supplementary Videos 6 Videos Supplementary i Time (min) , j 11 ) Actin polymerization deformed the the deformed polymerization ) Actin a , 01000.1 0.0 –0.1 ) Three-dimensional reconstruction reconstruction ) Three-dimensional 1 b 2 Curvature (µm and and Green: organelle . To control membrane–actin membrane–actin control To . e – j) White: actin The data and images are are images and data The Supplementary Videos 3 Videos Supplementary 20 µm ± s.d. of four independent independent four of s.d. –1 ) 27 g , d 2 8

Fraction Projected actin area s r e t t e l A atn fila actin As . 0.0 0.2 0.4 (%) / min 0.0 0.1 0.2 0.3 0.4 – Curvature (µm 01000.1 0.0 –0.1 9 Phase

). Red P =3 b Ld Lo2 Green ) Optical ) Optical P =0.02 × FWHM 10 0.065 0.008 OFF –4 –1 )  - - - - -

© 2018 Nature America, Inc., part of Springer Nature. All rights reserved. Fig. 20 Fig. ( interaction repulsive the to owing shape spherical a maintained Lo2 of made system the but interaction, tive to attrac owing the curvature membrane local of altered and Ld Lo1 made system membrane single-phase the stimulation, optical Upon writing writing of the manuscript.final the theoretical description. All authors the resultsdiscussed and contributed to the performed reconstitution experiments. Y.M., K.Y.L., S.-J.P., and L.M. worked out the designed photosynthetic protocellular system organelles with artificial and K.Y.L. and S.-J.P. performed photocurrent measurements. K.Y.L., S.-J.P., and K.S. photoconverters and ATP synthase. K.Y.L. and H.K. calibrated intravesicular pH. experiments. K.Y.L.,supervised K.A.L., and outS.-H.K. carried purification of K.Y.L., S.-J.P., K.-H.J., T.K.A., K.K.P., and K.S. developed the concept and Korea (PJ009106022013). the Woo Jang Chun Project Special of the Rural Development Administration, andScience ICT, Korea. K.Y.L. and T.K.A. acknowledge the support provided by KRICT through the National Research Foundation, funded by ofthe Ministry and Program 2015R1D1A1A01058917, DRC-14-03- (2013K1A4A3055268), and (2016R1A2B3015239 Foreign2011-0017539), Research Institute Recruitment This work was supported by the Mid-Career Researcher Programs o Note: Any Supplementary Information and Source Data files are available in the the in available are references, and codes accession statements of including Methods, and data availability any associated Me a ments may contribute to the goal of long-standing achieving building environ cell-like artificial into organelles and proteins of networks Introducing replication. genetic for nucleus or protein for reticulum repre endoplasmic as the such organelles, modules additional senting isolated create to used be could approach tom-up Wechange). (morphological that envision our behavior bot cell-like exhibit as well as polymerization) actin fixation, (carbon reactions metabolic cascading control and induce to able was The system resulting system. protocellular the for module energy an as served here presented organelle artificial photosynthetic the nature, in chondria behaviors, cellular complex and homeostasis replicate that systems cell-like artificial developing of possibility the supports system protocellular the the of thus and morphology formation cytoskeleton ATP-dependent control can 13 Video Supplementary ( shapes mushroom or teardrop into cells spherical the deforming curvature, cell global the in changes induced interactions membrane–actin local between ( Ld-Lo2) and Ld-Lo1 as (such mixtures different two of made systems membrane that produced a strong repulsive interaction ( (polysaturated-PEG2000-PE) glycol)-2000] [methoxy(polyethylene 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine- polysaturated and cholesterol, sphingomyelin, of composed (Lo2) mixture phase s r e t t e l  t AUTHO Ac h n

l e Our demonstration that switchable photosynthetic organelles organelles photosynthetic switchable that demonstration Our i k n

th p e now

a v de novo de p e od , and and , r R R e s i r o CON . l n s edg

o Supplementary Videos 10 Videos Supplementary f

29 t TRIBUTI h Fig. 4 Fig. m , e 3

0 p en . a p e ts r Fig. 4i Fig. h . ON and and S ). , Supplementary Fig. 21 Fig. Supplementary j in vitro in , , Supplementary Figs. 21 Figs. Supplementary . Just like and mito and chloroplasts like Just . – 12 Fig. 4e Fig. ). In the phase-separated phase-separated the In ). Supplementary Fig. 20 – g , , on ), the difference difference the ), Supplementary Supplementary l i and and n e

v e r 22 s i on , and and , N

of ). ------11. 10. 9. 8. 7. 6. 5. 4. 3. 2. 1. claims jurisdictional in published maps and affiliations. institutional R The author declares no competing interests. 30. 29. 28. 27. 26. 25. 24. 23. 22. 21. 20. 19. 18. 17. 16. 15. 14. 12. 13 r CO e eprints and permissions information is available online at at online available is information permissions and eprints

p . r MP Choi, H.-J. & Montemagno, C.D. Artificial organelle: ATP synthesis from cellular synthesis ATP organelle: Artificial C.D. Montemagno, & H.-J. Choi, Carvalho, K. F.C.Keber, M2 protein principle. membrane-organizing a as rafts Lipid K. Simons, & D. Lingwood, Influenza R.A. Lamb, & G.P. Leser, X., Jing, J.S., Rossman, Elani, Y., Law, R.V. & Ces, O. Vesicle-based artificial cells as chemical microreactors model protocellular a in synthesis Sugar B.G. Davis, & K. Winzer, P.M., Gardner, supported in proteins Photosynthetic S.D. Evans, & X. Han, J.S., Roth, L., Wang, Rivas,Vogel,G.,Schwille, S.K.& of P. Reconstitution reconstitution cytoskeletalof vitro proteinassemblies in construction: under Biology D.A. E.S. Fletcher, & A.P. Boyden, Liu, & K.R. Guthrie-Honea, D.A., Martin-Alarcon, K.P., Adamala, modules from biology: synthetic of wave second The R. Weiss, P.E.M.& Purnick, Walde, P. Building artificial cells and models: experimental approaches experimental models: protocell and cells artificial Building P. cellular Walde, of principles design cell: the Building W.F. Marshall, & S.M. Rafelski, Mulli & D.A. Fletcher, disassembly and assembly by driven motility Cellular G.G. Borisy, & T.D. Pollard, (2004). kinetics. synthesis biphasic displays synthase Brusilow,W.S.A.The & O.B. Glagoleva, Tomashek,J.J., lipid-protein flip: to P.W. not Heacock, Dowhan, To or J., & flip Xie, Bogdanov,M., R. Tunuguntla, Proteorhodopsin. C. Glaubitz, & J. Wachtveitl, E., Bamberg, C., Bamann, O. Béjà, oxygen- of structure Crystal N. Kamiya, & J.R. Shen, K., Kawakami, Y., Umena, photosynthesis. of Evolution R.E. Blankenship, & M.F. Hohmann-Marriott, solar-powered for pumps efflux bacterial Engineering D. Wendell, & V. Kapoor, cell artificial an toward step a as bioreactor vesicle A A. Libchaber,V. & Noireaux, Controlled C.J. Kastrup, & C. Klein-Bosgoed, S., Law, S.K., V.,Novakowski, Chan, of proliferation Sustainable T. Yomo, & T. Sunami, S., Fujii, G., Tsuji, K. Kurihara, ATPaseand II photosystem of Coassembly J. Li, & P.,Y.,J. Cai, Jia, Fei, X., Feng, A.P. Liu, 1135–1139 (2014). 1135–1139 cell cycle. cell proteorhodopsin in artificial membranes. artificial in proteorhodopsin mediates ESCRT-independent membrane scission. ESCRT-independentmembrane mediates pathways. reaction segregated spatially with bacteria. in response signalling cell a to leads (2015). lipid bilayers: towards a biokleptic approach for energy capture. large-scalefor membrane transformation. function. cellular Chem. Nat. Engineering genetic circuit interactions within and between synthetic minimal cells. systems. to with lipid vesicles. lipid with architecture. filaments. actin of Biol. topology. protein membrane final of determinant a are interactions charge Å. 1.9 of resolution a at II photosystem evolving waters. surface of bioremediation assembly. Ed. . using platelets in mRNA exogenous of (2016). replication. RNA inner with compatible (2016). ATPSynthesis. light-driven for artificial as . mimetic shape change. shape 485–492 (2010). 485–492 Rev. Plant Biol. Plant Rev. , 46–50 (2010). 46–50 327, Biophys. Acta Biophys. sea. (2008). 789–793 i n t s E / , 13590–13593 (2015). 13590–13593 54, i Science n TI , 925–935 (2008). 925–935 182, advance online publication online advance d e NG Bacterial rhodopsin: evidence for a new type of phototrophy in the in phototrophy of type new a for evidence rhodopsin: Bacterial al. et x . ebaeidcd udig f ci filaments. actin of bundling Membrane-induced al. et Proc. Natl. Acad. Sci. USA Sci. Acad. Natl. Proc. h Nat. Commun. Nat. I Topology and dynamics of active nematic vesicles. nematic active of dynamics Topologyand al. et Nat. Rev. Mol. Cell Biol. Cell Mol. Rev. Nat. t , 431–439 (2017). 431–439 9, et al. Cell-sized liposomes reveal how actomyosin cortical tension drives A recursive vesicle-based model protocell with a primitive model primitive a with protocell model vesicle-based recursive A al. et m N Nat. Rev. Mol. Cell Biol. Cell Mol. Rev. Nat. , 1902–1906 (2000). 1902–1906 289, , 614–625 (2014). 614–625 1837, l T Proc. Natl. Acad. Sci. USA Sci. Acad. Natl. Proc. . . Publisher’s note:

Lipid bilayer composition can influence the orientation of orientation the influence can composition bilayer Lipid al. et , 515–548 (2011). 515–548 62, E Nat. Rev. Mol. Cell Biol. Cell Mol. Rev. Nat. R Cell BioEssays E STS ns, R.D. Cell mechanics and the cytoskeleton. the and mechanics Cell R.D. ns, , 453–465 (2003). 453–465 112, Nano Lett. Nano , 8352 (2015). 8352 6, , 296–303 (2010). 296–303 32, Nano Lett. Nano S , 2538–2542 (2005). 2538–2542 5, pringer Nature remains neutral with neutral Nature remains regard to pringer

, 410–422 (2009). 410–422 10,

, 17669–17674 (2004). 17669–17674 101, , 593–602 (2008). 593–602 9, Proc. Natl. Acad. Sci. USA Sci. Acad. Natl. Proc. Curr.Opin.Chem. Biol. Biophys. J. Biophys.

, 16456–16461 (2013). 16456–16461 110, , 644–650 (2009). 644–650 10, Nat. Commun. Nat. , 2189–2193 (2013). 2189–2193 13,

Nat. Chem. Nat. . il Chem. Biol. J. nature biotechnology nature Nature Cell , 1388–1396 (2013). 1388–1396 105, , 902–913 (2010). 902–913 142, F1F0 ATPF1F0 coli Escherichia C Nano ACS , 55–60 (2011). 55–60 473, h t , 377–383 (2009). 377–383 1, t , 5305 (2014). 5305 5, Small p :

/ 22 / Angew. Chem. Int. Chem. Angew. 4465–4470 279, w

, 18–26(2014)., w 11

, 590–595 113, w a. Phys. Nat. 556–561 10, Science . , 3306–3318 n Nature a t u Biochim. r Science e J. Cell J. . Annu. c 345, 463, o m 4, / © 2018 Nature America, Inc., part of Springer Nature. All rights reserved. spinach using the sucrose gradient method gradient sucrose the using spinach II. photosystem of Purification MO). Louis, (St. Co. Sigma–Aldrich and isopropyl isopropyl and TBDMCS) (1% 1% Tert-Butyldimetheylchlorosilane with (MTBSTFA), mide (FCCP), nylhydrazone 4(trifluoromethoxy)phe cyanide Carbonyl CO). (Denver, Inc. Cytoskeleton, (Mountain Inc. BioVision, from purchased were from phalloidin 670 Acti-stain and View,protein Actin CA). purchased were reagents other and and Kit Assay chemicals Oxaloacetate An NJ). Plains, (Morris Organics Acros Molecular Probes (Eugene, OR). Phenyl- from purchased were Kit Determination ATP an and (SNARF-1) SNARF-1 5-(and-6)-carboxy and fluo (alexa-488) ester The TFP 488 Fluor Alexa CA). probes rescent (Hercule, Inc. Bio-Rad, from purchased was mesh) (25–50 SM-2 Bio-Beads Polystyrene OH). (Maumee, Inc. Anatrace, from obtained Polar Avanti AL). from (Alabaster, Inc. purchased , were cholesterol and (SM), sphingomyelin (DOTAP), salt) (chloride 1,2-dioleoyl-3-trimethylammonium-propane PE), lcl-00 (E20-E, 1,2-dipalmitoyl-sn-glycero-3-phosphoeth anolamine- (PEG2000-PE), glycol)-2000] 5 mM 5 MgCl mM NaCl, M 0.4 7.8), (pH Tricine-KOH mM 20 containing solution buffer a in dopsin (MBP) (MBP) dopsin proteorho Bay (Monterey (PR) proteorhodopsin the carrying pKJ900 1- 1,2-dioleoyl-sn-glycero-3-phospho- (POPE), palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine Reagents. O doi:10.1038/nbt.4140 1,2-dioleoyl-sn-glycero-3-phospho-ethanolamine- (DOPE), 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPC), choline ride)-24-norcholesterol (TopFluor-Chol), 1,2-dioleoyl-sn-glycero-3-phospho difluo 23-(dipyrrometheneboron (Rhod-PE), salt) (ammonium sulfonyl) B 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine- 3-phosphoethanolamine- m MgCl mM 5 NaCl, M 0.15 7.8), (pH Tricine-KOH mM 20 containing solution buffer a in for 10 min at 1,000 was centrifuged solution aproic acid (1 mM), and phenylmethylsulfonyl fluoride (PMSF) (0.2 mM). The containing PSII core ( core PSII containing Brea, CA) for 36 h at 4 °C. We were Coulter, able bands to (B0–B7) eight distinct detect (Beckman rotor 28 SW Coulter Beckman a in gradients sucrose on 80,000 at centrifuged were samples solubilized The complete. was thawing until refrigerator °C 4 a in kept then were tubes The min. 1 for temperature room at water in tubes the immersing by initiated was mixture frozen the of Thawing h. 4 for °C −80 at CA) Brea, Coulter, (Beckman tubes Ultra-Clear at 30 °C. Cells were induced with 1 mM IPTG and 10 10 and IPTG mM OD600) 1 with (0.4 induced nm were 600 Cells at °C. 30 at units OD 0.4 of absorbance an produce and to 1:100 grown diluted were UT5600 in cells transformed of cultures Overnight and and 5 mM MgCl 6,000 at stacked min thylakoids, the 10 mixture was for incubated in again 3% DDM (w/v), 15 mM centrifuged NaCl, was solution the and ing were made with 0.7 M sucrose, 10 mM HEPES (pH 7.5), 10 mM NaCl, NaCl, mM 5 10 mM MgCl 7.5), (pH HEPES mM 10 sucrose, M 0.7 with made were ing thaw and freezing by prepared gradients density sucrose The thawing. and 10 mM NaCl, and 5 and MgCl mM NaCl, mM 10 were resuspended in a small volume of 10 mM HEPES (pH 7.5), 0.4 M sorbitol, 2,000 at centrifugation 5-min a material, lized was resuspended in 20 15 mM (pH 7.5), mM HEPES was NaCl, resuspended and 5 mM MgCl 4,000 at min 10 for centrifuged was PMSF.solution mM This 0.2 duced to the the to duced Heterologous expression and purification of proteorhodopsin. ( PSII of condition the confirm to obtained was spectrum absorption an and ( (SDS-PAGE) by determined was band each of Supplementary Fig. 23c Fig. Supplementary NLIN E M Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), (POPC), Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine N 2 2 , , and DDM. (w/v) 0.03% were These mixed and kept in thin-wall Escherichia , and the protease inhibitors benzamidine (0.2 mM), h-aminoc mM), (0.2 benzamidine inhibitors protease , the and -(methoxy(polyethylene glycol)-2000] (polysaturated PEG2000- glycol)-2000] -(methoxy(polyethylene E 2 , 0.2 mM benzamidine, 1 mM h-aminocaproic acid, and and acid, h-aminocaproic mM 1 benzamidine, mM 0.2 , Gamma proteobacterium Gamma TH β 2 - for 30 min on ice with soft agitation. To agitation. soft for on nonsolubi remove 30 with min ice d OD -1-thiogalactopyranoside (IPTG) were purchased from from purchased were (IPTG) -1-thiogalactopyranoside S Supplementary Fig. 23a Fig. Supplementary N -Methyl- ). N -(7-nitro-2-1,3-benzoxadiazol-4-yl) (NBD-PE), (NBD-PE), -(7-nitro-2-1,3-benzoxadiazol-4-yl) UT5600 strain for heterologous expression heterologous for strain UT5600 n 2 -Dodecyl- . Sucrose gradients were prepared by freezing by freezing prepared were gradients . Sucrose Photosystem II (PSII) was purified from from purified was (PSII) II Photosystem N l -(tert-butyldimethylsilyl) trifluoroaceta -(tert-butyldimethylsilyl) -serine -serine (DOPS), 1,2-dioleoyl-sn-glycero- , database ID: AAG10475) was intro was AAG10475) ID: , database p -benzoquinone was -benzoquinone purchased from β - g d 31 , and the pellet was resuspended was resuspended , and pellet the -maltopyranoside (DDM) was was (DDM) -maltopyranoside , 3 g 2 was performed. Membranes Membranes performed. was . Leaves were gently ground ground gently were Leaves . ′ ). The protein composition composition protein The ). -rac-glycerol) (POPG), (POPG), -rac-glycerol) Supplementary Fig. 23b Fig. Supplementary N N -(methoxy(polyethylene -(methoxy(polyethylene -(lissamine rhodamine rhodamine -(lissamine µ M all-trans-Retinal all-trans-Retinal M g . To solubilize solubilize To . The g , the pellet pellet the , 33 , 3 4 2 ), ), g ------. ,

Inc., Barker, TX) and incubated at 37 °C for 30 min with gentle rocking. rocking. gentle with min 30 for MgCl mM 10 °C 37 at incubated and TX) (GenDEPOT Barker, cocktail Inc., inhibitor protease 1× and cell), of mg/g (1 lysozyme ugation. Cells were resuspended in 50 mM Tricine-KOH (pH 8.0), 200 mM KCl, NaCl at pH 9.7. Cells were grown to late log 0.5% phase with and medium then nutrient harvested alkaline byan in VA) centrifgrown was Manassas, ATTC, 30 °C. Induced cells were harvested by centrifugation at 2,500 at centrifugation by harvested were cells at Induced h °C. 4 30 for incubated and OD600 0.4 at MO) Louis, St. Co., (Sigma–Aldrich 10 10 mM MgCl KCl, mM 15 8.0), (pH Tricine-KOH mM 20 with washed were Cells h. 1 for 3,200 at centrifugation 20-min a by removed were cells Unbroken sonication. with disrupted were cells and tail, 50 mM Tricine-KOH 200 mM KCl, (pH 8.0), - and 1 cock × inhibitor protease centrifugation at 125,000 gentle with rocking. All subsequent steps were min atperformed 4 15 °C. Pellets were byfor collected temperature room at incubation before added were The pTrc99a vector containing pyruvate carboxylase was used to transform transform to used was carboxylase pyruvate containing vector pTrc99a The This Argentina). construct was Co., kindly provided by Biotech P. (AP Kim (Catholic Universityvector of pTrc99a Korea, Korea). a into cloned and DNA from amplified was gene carboxylase pyruvate fixation. carbon for an carboxylase, pyruvate of Purification ( gels SDS-polyacrylamide on profile polypeptide and activity, synthase ATP content, protein for analyzed were fractions gradient The fractions. 0.35-ml at at 125,000 lets lets were dissolved in 5% glycerol, 50 mM Tricine-KOH (pH 8), 5 mM MgSO ATP The pellet. synthase–containing supernatants were removed, and the pel This was centrifuged at 75,000 at 4 °C. stirring with mg/ml) (198 to brought saturation was 60% supernatant ATPwhich contained negligible pellets, activity,synthase were The discarded. at 75,000 centrifuged was 4 suspension The °C. at stirring with mg/ml) (176 to 30% saturation extract to the crude was added addition of cholate sodium to 1% (100 mg/10 ml). Ammonium sulfate powder the after fractionation sulfate ammonium to subjected was ATPthe synthase of most contained that supernatant orange dark and the discarded, was pellet 235,000 at centrifuged and shaking occasional on for ice 30 with min incubated was of 50 mM The suspension and 3 mg/ml. concentrations final to added were dithiothreitol) 1 mM and glucoside, octyl tion in 50 mM Tricine-KOH (pH 8.0), 25 mM K mM 25 8.0), (pH Tricine-KOH mM 50 in tion solu stock mg/ml (50 of M 0.5 asolectin DDM and solutions 50 mg/ml Stock were centrifuged for 17 h in a Beckman-Coulter SW 50.1 rotor (Beckman (Beckman rotor 200,000 at CA) 50.1 Brea, SW Coulter, Beckman-Coulter a in h 17 for tubes the centrifuged and ml, were 4.75 was volume gradient The gradients. sucrose 1–40% onto loaded were ml) (0.2 Portions mg/ml. to 20–30 and 1 mM dithiothreitol previously Purification of ATP synthase. ( purity approximate its SDS-PAGE to check to subjected was PR Purified DDM. 0.02% and NaCl, purified proteins were finally solubilized in 50 mM Tris-HCl (pH 7.0), 150 mM MA). The Bedford, Corp., (Millipore filter MWCO 10,000 Ultra-4 centrifugal Amicon an with concentrated and collected was fraction colored The zole). imida mM 250 and DDM, 0.02% NaCl, mM 150 7.0), Tris-HCl mM (pH (50 buffer with eluted was protein Finally, MO)). Louis, St. Co., (Sigma–Aldrich (50 mM Tris-HCl (pH 7.0), 150 mM NaCl, 0.02% DDM, and 20 mM imidazole buffer with washed Venlo, Netherlands), Ni-nitrilotriacetic (Qiagen, beads with (NTA)-agarose acid incubated was fraction solubilized The °C. 4 at h 1 for DDM 1% with treated was pellet suspended The fraction. membrane the Coulter, 70Ti, IN) Beckman at Indianapolis, 35,000 type rotor ultracentrifuge, XL-90 (Beckman ultracentrifuged and taken was 2,500 at sonicated were cells collected again Danbury, for CT) 250, 10 on min and ice centrifuged Sonifier (Branson The NaCl. mM 150 and 7.0) (pH mM 50 Tris-HCl in suspended and Germany) 5810R, centrifuge (Eppendorf min 20 20% 20% glycerol, 50 mM Tricine-KOH (pH 8.0), 25 mM K Supplementary Fig. 24b Fig. Supplementary The washed membrane vesicles were suspended to about 10–15 mg/ml in in mg/ml 10–15 about to suspended were vesicles membrane washed The g and 4 °C for 20 min to discard unbroken cells. The soluble fraction fraction soluble The cells. unbroken discard to min 20 for °C 4 and g 3 for 1 h. 1 for 5 and suitably modified. modified. suitably and 2 2 , , and 1× protease inhibitor They cocktail. were then centrifuged , DNase I, and RNase A (0.1 mg/g of cell; Sigma–Aldrich Co.) Co.) Sigma–Aldrich cell; of mg/g (0.1 A RNase and I, DNase , Supplementary Fig. 24a Fig. Supplementary g for 1 h. The pellets were resuspended with 160 ml of ). g for 15 min to separate the supernatant from the g Overall purification procedures were described at 4 °C. The gradients were fractionated into into fractionated were gradients The °C. 4 at g Bacillus pseudofirmus Bacillus n frhr etiue a 125,000 at centrifuged further and ). g nature biotechnology nature g Bacillus subtilis Bacillus for 15 min, and the resulting for and resulting the 15 min, 2 for 2 h. The grayish brown brown grayish The h. 2 for SO g and 4 °C for 1 h to isolate 4 , 5 mM MgSO mM 5 , 2 SO 4 , , and 5 mM MgCl (ATCC 700159, 700159, (ATCC g 168 genomic genomic 168 and 4 °C for for °C 4 and 4 , 50 mM mM 50 , The The 4 2 g - - - - . ,

© 2018 Nature America, Inc., part of Springer Nature. All rights reserved. tion, unlysed tion, cells and unlysed cell debris were removed by at centrifugation 4,000 by collected were Cells centrifugation °C. with at 2,750 37 at induced h 4 were another Cells for grown °C. and 37 IPTG at mM 0.1 nm 600 at units an OD produce 0.4 to of grown and absorbance 1:100 diluted were cultures Overnight °C. 37 at of phospholipids in a single giant proteoliposome (15- proteoliposome giant single a in phospholipids of phores in the photoconverters, their concentrations were divided. The number oliposomes were suspended in a solution containing 0.2 M sucrose, 2 mM mM 2 sucrose, HEPES (pH 8.0), 2 prote mM M MgSO 0.2 The containing solution a method. in suspended were detergent-mediated oliposomes a by prepared proteoliposomes chromophores. of number by the divided 15 mg Bio-Beads per milligram of DDM for 1 h each, followed by a fifth fifth a by followed each, DDM of h milligram per 1 Bio-Beads mg 30 of addition for DDM of milligram per Bio-Beads mg 15 of additions sequential four of consisting procedure slow a by removed was DDM SM-2. Bio-Beads prewashed adding by °C 4 at removed was DDM the mixture. detergent–lipid After a for incubation 30-min equilibration, micellar the to added were PR) and (PSII photoconverters solubilized the and DDM, for (w/w) 1 of ratios detergent/lipid at mixed into solubilized were 2.5 mM NaH mM 2.5 molecules 30 mg Bio-Beads SM-2 per milligram of DDM of of addition milligram fifth per a SM-2 by Bio-Beads followed mg 30 each, h 1 for DDM of milligram per SM-2 Bio-Beads mg 15 of additions sequential four of consisting procedure slow a removed at 4 °C by SM-2. adding DDMprewashed Bio-Beads was removed by was DDM the equilibration, micellar for incubation 30-min a After mixture. detergent–lipid the to added were PR) and (PSII photoconverters solubilized the and DDM, with (w/w) 1.5 of ratios at detergent/lipid micelles mixed into tion spectrum and absorption coefficients of the chromophores. The The chromophores. the ( of coefficients absorption by of in concentration PSII applying chlorophyll-a was an calculated equation and spectrum - tion absorp the by calculated was PR) and (PSII photoconverters the in phores converter residues and then centrifuged at 2,000 at centrifuged then and residues converter photo remove to MA) Billerica, Inc., (Millipore device ultra-filter Amicon = 8.70 × 10 × 8.70 = Escherichia coli Escherichia nature biotechnology nature 2 mM MgSO mM 2 2 M (0.2 and solution mM sucrose, (pH volume 8.0), of buffer HEPES desired more for vacuum high in a defined hydrated were under Lipids residue. solvent 1 h than to organic remove then and gas nitrogen in kept was obtained film thin the and tube, test glass a in chloroform visualiza in for mixed were Lipids added tion. was Rhod-PE of mol% 0.25 or NBD-PE and of 2:1:1:1, mol% of 0.25 ratio molar a at cholesterol and POPG, POPE, POPC, from organelles. artificial thetic and photosyn proteoliposomes, vesicles, of Preparation unilamellar small ( purity approximate its SDS-PAGE check to (Qiagen, beads to subjected was (NTA)-agarose carboxylase pyruvate Finally, purified Venlo, Netherlands)). acid Ni-nitrilotriacetic using purified was carboxylase pyruvate and obtained, was fraction soluble The °C. 4 at min 20 (50 of the presence ampicillin in medium broth lysogeny in grown was transformant The purification. and = 1.96 × 10 × 1.96 = tuted into SUVs by a detergent-mediated method detergent-mediated a by SUVs into tuted scattering. light by dynamic measured 100-nm with extruded through the membrane 11 filter and was times, the of size distribution was vesicles passed and MA) solution Newton, Inc., (Whatman, lipid filters membrane The polycarbonate AL). Alabaster, Lipids, 40 at bath water a in sample the thawing and nitrogen liquid in vesicles the ing freez times: 5 cycle following the repeating by obtained were SUVs thawed and Frozen h. 1 for transition phase of that above temperature a at vortexed culated from the retinal absorption using a molar absorption coefficient of of M 50,000 coefficient absorption molar a using absorption retinal the from culated diameter vesicle = vesicle 8.70 × diameter 10 µ g/ml = 12.25·A = g/ml Artificial organelles containing ATP synthases were reconstituted into into reconstituted were synthases ATP containing organelles Artificial Proteoliposomes Proteoliposomes and organelles artificial were transferred to a 100,000 MW Proteoliposomes containing photoconverters (PSII and (PSII PR) were reconsti photoconverters containing Proteoliposomes ± 1 °C. The extrusion was performed by a mini-extruder (Avanti Polar Polar (Avanti mini-extruder a by performed was extrusion The °C. 1 −1 3 9 cm 9 . . 6 lipid molecules), a lipid bilayers patch (2- patch bilayers lipid a molecules), lipid lipid molecules), and a single small proteoliposome (0.1- proteoliposome small single a and molecules), lipid 4 Supplementary Figure 2 Figure Supplementary 2 ) to yield a total lipid concentration of 5 mg/ml. This was then then was This mg/ml. 5 of concentration lipid total a yield to ) PO −1 strain BL21 (Novagen, Darmstadt, Germany) for expression expression for Germany) Darmstadt, (Novagen, BL21 strain (ref. 4 663.6 , 2.5 mM KH mM 2.5 , g for 15 min at 4 °C. After cells were disrupted by sonica 3 –2.55·A 8 ). ). To to of the ratios chromo estimate phospholipids 4 Small unilamellar vesicles (SUVs) were prepared were prepared (SUVs) vesicles unilamellar Small lipid molecules) were calculated based on based POPC were calculated lipid molecules) µ 4 646.6 , , 0.1 mM phenyl- g/ml) (Duchefa Biochemie B.V., Biochemie (Duchefa g/ml) Netherlands) 2 PO ) 3 7 4 , and the concentration of PR was cal was PR of concentration the and , , and 5 mM NaHCO mM 5 and , shows the number of lipid molecules molecules lipid of number the shows Supplementary Fig. 24c Fig. Supplementary p 3 -benzoquinone, 2 -benzoquinone, mM ADP, 6 . The number of chromo of number The . g 3 6 for 30 min. The pellets pellets The min. 30 for 3 . SUVs were solubilized solubilized were SUVs . 6 . µ µ m diameter vesicle vesicle diameter m 3 m diameter patch patch diameter m . Proteoliposomes Proteoliposomes . ). g µ for m ------

of two-sided indium-tin oxide–coated glass slides (9 cm (9 slides glass oxide–coated indium-tin of two-sided 200 visualization. at mol% of for 0.25 lesterol NBD-PE and a an of ratio molar additional 2:1:1:1 cho and POPG, POPE, POPC, and chloroform in lipids of solution mg/ml 1 with ultraviolet and visible laser lines. GUVs, giant proteoliposomes, and and proteoliposomes, giant GUVs, lines. laser visible and ultraviolet with equipped Germany) Mannheim, (Leica-Microsystems, microscope confocal microscopy. Confocal above. as electroformation performed were and hydration Lipid GUVs. phase-separated to Lo2 and added Ld was of PEG2000-PE mol% polysaturated 0.25 mol% 3 or additional An TopFluor-Chol of Rhod-PE. mol% 0.25 DOPC:DOPS:SM: of incorporating ratio molar cholesterol 2:1:3:1.5 a with prepared were Lo1) and ductive ductive side of indium-tin oxide-coated two-sided slides (9 cm and diluted to 1 mg/ml. 200 using a vacuum desiccator at 4 °C for 3 h. The lipids were rehydrated with with rehydrated 0.2 were M sucrose, 2 lipids The h. 3 for °C 4 at desiccator vacuum a using added. Lipid hydration and electroformation were performed as above. as performed were electroformation and hydration Lipid added. was of mol% 0.25 Rhod-PE To interactions. visualization, enable brane–actin - mem prevent to PEG2000-PE mol% 3 and 1:1 of ratio molar at a cholesterol SM with and were prepared GUVs at Lo2 of a 1:1. ratio molar and cholesterol SM with prepared were GUVs Lo1 2:1:1. of ratio molar a at cholesterol and DOPS, DOPC, with prepared were GUVs Ld Lo2). and (Lo1 mixtures phase method proteoliposomes. giant electroformation the and by synthesized were (GUVs) vesicles vesicles unilamellar Giant unilamellar giant of Preparation temperature. at room thawing min by 3 followed nitrogen liquid in freezing of 5 min consisting cycles freeze-thaw three using thawed and frozen then and solution buffer desired the in resuspended were (0.2 M glucose, 2 mM MgSO mM 2 glucose, M (0.2 solution buffer resuspension in 1:100 diluted tip, pipette wide-bore a using harvested then were GUVs min. 30 for slides lipid-coated the to applied was generator. To wave a of sine 4-Hz function 2 GUVs, the V an detach arbitrary 8 for h using chamber the across applied V was of 1.2 wave sine a 10-Hz then ADP, 3 mM NaH organelles, 0.2 M sucrose, 20 mM Tris-HCl photosynthetic (pH 7.8), of 0.1 mg/ml G-actin, 5 mg/ml mM 2 with rehydrated were lipids The MO). Louis, St. Co., St. Louis, MO), and 1 mol% magnesium ionophore I Co., (Sigma–Aldrich Rhod-PE for visualization, 1 mol% calcium ionophore A23187 (Sigma–Aldrich 2:1:2:1, 1 mol% PEG2000-PE to prevent nonspecific interactions, 0.25 mol% of prepared as above with DOPC, POPE, DOPS, and cholesterol at a molar ratio of Preparation of system. protocellular electroformation. of h 4 within imaged were proteoliposomes 1% Pluronic with F-127 (Sigma–Aldrich Co., blocked St. Louis, MO) been for 1 had h before that use. All giant dishes confocal in prote imaged Giant were use. oliposomes before min 30 for settle to allowed and 8.0)) (pH HEPES biological molecules ( molecules biological swelling, which allows encapsulation of artificial organelles (100 nm) and other 670 for phalloidin causes filament actin Electroformation staining. membrane drated with 0.2 M sucrose, 2 2 sucrose, M 0.2 with drated rehy were lipids The slides. glass oxide–coated indium-tin lipid-coated two tion chambers by were 3-mm between spacers silicon constructed positioning by placing the slides in a vacuum desiccator for 1 h. After drying, electroforma nitrogen gas. Organic solvents on the glass slides were removed by evaporation liquid-disordered phase mixture (Ld) and from two different liquid-ordered liquid-ordered different two from and (Ld) mixture phase liquid-disordered oliposomes in suspension oliposomes dye. SNARF-1 external the dilute to 6–9 pH 2 M (0.2 mM atsolution HEPES glucose, buffer in 1:100 resuspension diluted slides for 30 min. GUVs were using then harvested a pipette wide-bore tip and lipid-coated the to applied was V 2 of wave sine 4-Hz a GUVs, To the detach OR). Inc., (Tektronix generator function arbitrary an using h 4 for chamber the across applied was V 1.2 of wave sine 10-Hz a and 6–9, of at pH a HEPES For phase-separated protocellular system protocellular phase-separated For To investigate membrane–actin interactions, GUVs were prepared from a from prepared were GUVs interactions, membrane–actin Toinvestigate Giant proteoliposomes were prepared through dehydration of small prote of small dehydration through were prepared Giant proteoliposomes 4 0 . For calibration of intravesicular pH, GUVs were prepared from a from prepared were GUVs pH, intravesicular of calibration For . µ 2 M M SNARF-1, 2 mM HEPES (pH 8.0), and 2 mM MgSO µ PO l l of the lipid solution on was deposited the side conductive 4 Supplementary Fig. 15a Fig. Supplementary , , 3 mM KH Confocal microscopy was Confocal conducted using a SP5 Leica 41 , µ 4 2 l l of the lipid solution was dehydrated on the con . Small proteoliposomes were . proteoliposomes prepared Small as above 4 µ , 0.1 mM phenyl- mM 0.1 , M SNARF-1 (a pH-sensitive dye), and 2 mM mM 2 and dye), pH-sensitive (a SNARF-1 M 2 PO Giant vesicles unilamellar (GUVs) were 4 , 0.2 , mM 0.2 CaCl 4 3 , , phase-separated GUVs (Ld (Ld GUVs phase-separated , b p ). -benzoquinone, and 2 mM mM 2 and -benzoquinone, 2 , , and 0.1 2 doi:10.1038/nbt.4140 area, 30–50 area, 2 area, 30–50 µ M Acti-stain M Acti-stain Ω ) with ) with 4 , , and Ω ------)

© 2018 Nature America, Inc., part of Springer Nature. All rights reserved. Calibration of intravesicular pH. intravesicular of Calibration ( imaging during pH that firmed the microscope’sconfocal laser scanning sources did not influence at and were it imaged fps, cells the 1.97 was con source, from laser the effects and GmbH, 2012 Zen (Carl Zeiss Jena, ToZeiss Germany). any minimize side and using processed software (ImageJ,image-processing MD) NIH, Bethesda, analyzed were images of reconstructions three-dimensional and images All 2. track on nm 638–755 at imaged and splitter beam 488/543/633 a with line Actin-stain 670 phalloidin samples were excited at 633 nm using an HeNe 2. laser track on nm laser 563–628 at imaged HeNe and splitter an beam using 488/543/633 a with nm line 543 at excited were samples Rhod-PE 1. track on nm at 493–563 imaged and splitter beam 488/543/633 a with line laser argon an using nm at 488 excited were TopFluor-Choland samples NBD-PE mode. multitrack in used was objective immersion water 40×/1.2 C-Apochromat a with equipped Germany) Jena, GmbH, Zeiss (Carl microscope fluorescence autofluorescence. and chlorophyll of spectra tion of absorp the outside nm lies as 532 autofluorescence, and other chlorophyll The use of 532 nm GUV minimized observation. as wavelength the excitation proteoliposomes small of solution a of dehydration through prepared were proteoliposomes Giant residues. and vesicles unincorporated the reducing of effects side the (438 (438 PR, or both photoconverters) to form a bilayer a form to photoconverters) both or PR, Germany). We used giant proteoliposomes (i.e., GUVs reconstituted with PSII, Munich, GmbH, Technologies (Nanion Patch system clamp patch automated Port-a- the with measured were bilayer the in photoconverters both and PR, PR. and PSII of photocurrents the Measuring pH. intravesicular calibrating in 15–30 of diameter a with vesicles Giant lipid solutions. the to NBD-PE of mol% 0.25 added we microscopy, light by GUVs of visualization To assist chloroform. in lipids of solution mg/ml 1 a from ated gener was 2:1:1:1 of ratio molar a at cholesterol and POPG, POPE, at POPC, containing film detected lipid A 6–9. was of pH a at SNARF-1 calibrated of was this and nm, intensity 570/630 fluorescence of ratio The nm. 543 to tuned source laser HeNe a SNARF-1 with microscope containing confocal a with GUVs imaged were dye. SNARF-1 external the dilute to HEPES) mM 2 glucose, M (0.2 solution buffer resuspension in 1:200 diluted and tip 2 mM HEPES) at pH 6–9. GUVs were then harvested using a wide-bore pipette formation with 2 temperature 630 520 at channels three in captured were emissions fluorescence SNARF-1 lens. immersion oil 63× a using units 3 or at 1 set was pinhole a and power, laser 10% at nm 532 and 488 at excited were SNARF-1 doi:10.1038/nbt.4140 mately 2 engine (Lumencor, Beaverton, OR) at 1.14 mW/cm light by a SPECTRA by illumination induced were a Photocurrents necessity. was rapid required, volumes small the to Owing femtofarads. few metric analysis of SNARF-1 fluorescence SNARF-1 of analysis metric fluorescence SNARF-1 of analysis ratiometric by measured was proteoliposomes giant of pH. intravesicular light-induced of Measurement temperature. at out room carried were experiments All Canada). Quebec, Inc., Lens (Doric fiber optical an via bilayer lipid the onto directed was Light least 1 G 1 least at of resistance seal a with bilayers planar formed and burst they chip, the of surface glass the touched proteoliposomes giant the When addition. posome was sufficient for reliable positioning within a few seconds after giant proteoli micro-constructed chip by applying the negative of aperture the pressure. on positioned were Generally,proteoliposomes Giant chip. −5 tostructed −20 mbar was ontosolution pipetted a patch clamp teoliposome-containing micro-con Germany). Munich, GmbH, For imaging protocellular systems, a Zeiss LSM 710 confocal laser-scanning For the formation of a planar containing PR, 10 10 PR, containing bilayer lipid planar a of formation For the ± ± 20 nm. As SNARF-1 has been shown previously to be responsive to to responsive be to previously shown been has SNARF-1 As nm. 20 24 nm), green (540 (540 green nm), 24 µ Ω m m in diameter) of a patch clamp chip Nanion(NPC-1, Technologies 4 6 4 4 . Membrane capacitance could be estimated on the order of a of order the on estimated be could capacitance Membrane . 4 5 , temperature was maintained between 22 °C and 24 °C during during °C 24 and °C 22 between maintained was temperature , . Choosing . to Choosing measure giant single minimized proteoliposomes , 41 µ M M SNARF-1 encapsulated in buffer solution (0.2 M sucrose, 4 2 . . Small proteoliposomes were prepared as described above Supplementary Fig. 4a Fig. Supplementary ± 27 nm), or red (660 (660 red or nm), 27 Intravesicular pH was measured by ratio by measured was pH Intravesicular 4 5 . GUVs were prepared by electro by prepared were GUVs . ). 46 The photocurrents of PSII, PSII, of photocurrents The , ± ± 2 4 , which was filtered by blue 7 µ 52 nm) band-pass filters. filters. band-pass nm) 52 20 nm, 570 570 nm, 20 in an aperture (approxi aperture an in m were collected for use use for collected were m The intravesicular pH pH intravesicular The µ ± 20 nm, and and nm, 20 l of the pro the of l ------

and diluted to 200 1 and diluted mg/ml. and OAA, dried residues were resuspended in 25 a vacuum concentrator at room temperature. For the derivatization of pyruvate at 4,000 and centrifuged then reactants collect to MA) Billerica, Inc., (Millipore device filter ultra Amicon MW 3,000 a to transferred were min 20 and 15, 10, 5, 1, for illuminated were organelles GC/MS analysis, containing artificial PSII-PR photoconverters that Tokyo,(Hitachi, an FL using Japan). F-7000 Spectrophotometer For emission cuvette durations quartz various (1–20 a min, for 1-min intervals). Samplesand were in analyzed by wavelengths measuring fluorescence light samples various with Multiple illuminated meter. were power a with NJ) Newton, tives of pyruvate and OAA were separated on a DB-5 capillary column column capillary DB-5 a on separated (30 m were × mm, 0.25 0.250 OAA and pyruvate of tives and 25 artificial organelles was determined with the luciferin–luciferase system luciferin–luciferase the with determined was organelles artificial by ATPsynthesis organelles. artificial of ATPsynthesis of Quantification The NJ). imaging. during off Newton, Thorlabs, (PM160, meter 2 min, and every the were light was turned illuminated giant proteoliposomes power a with NJ) Newton, 1.14 mW/cm 1.14 20 20 mM 3 Tris-HCl mM 8.0), (pH KH containing solution a in suspended were organelles artificial The used. were method (GC/MS) spectrometry chromatography/mass gas the and CA) Inc., (BioVision kit assay fluorometric OAA an (OAA), oxaloacetate of organelles. level the artificial of fixation carbon of Quantification Tokyo,(Hitachi, Japan). Spectrophotometer FL F-7000 an using nm 560 at measur emission by fluorescence ing analyzed were Samples intervals). 1-min min, (1–33 durations quartz cuvette were illuminated with various light wavelengths and for various (NDL-25C-4, Thorlabs, Newton, NJ) with a power meter. Multiple samples in a MA) that emitted broadband white light or light that was filtered by blue blue by filtered was that (438 light or light white broadband emitted that MA) Lawrence, Industries, Dolan-Jenner Lite PL-800, (Fiber filters heat-absorbing with equipped bulb a halogen with illuminator a fiber-optic with illuminated were proteoliposomes giant The use. before min 30 for settle to allowed and 1.14 mW/cm 1.14 was filtered by blue (438 filtered was that light or light white broadband emitted that MA) Lawrence, Industries, gen bulb equipped with heat-absorbing filters (Fiber Lite PL-800; Dolan-Jenner a with halo illuminator a with fiber-optic was illuminated suspension stirred gomycin (an inhibitor of ATP synthase) were added as the negative control. The 2 2 mM MgSO glucose, M (0.2 solution buffer resuspension in 1:100 diluted and tip pipette generator. an function arbitrary using GUVs a were then wide-bore harvested 8 for h using chamber the across applied V was of 1.2 wave sine a 10-Hz then 5 mM ADP, 2 mM MgCl mM 2 ADP, mM 5 using a vacuum desiccator at 4 °C for 3 h. The lipids were rehydrated with with rehydrated 0.2 were M sucrose, 2 lipids The h. 3 for °C 4 at desiccator vacuum a using adapted from a paper by Andrea Hazard Hazard Andrea by paper a from adapted was procedure The protocol. manufacturer’s the Probes, following and OR) Eugene, (Molecular Kit Determination ATP an using solution bulk-vesicle 2 2 mM MgCl tive side of two-sided indium-tin oxide–coated slides (9 cm (9 slides oxide–coated indium-tin two-sided of side tive (pH 8.0), 3 mM KH mM 3 8.0), (pH in 20 a mM containing were Tris-HClsolution suspended organelles The artificial above. described as prepared were synthase ATP and PSII, PR, taining MA) that emitted broadband white light or light that was filtered by blue blue by filtered was that (438 light or light white broadband emitted that MA) Lawrence, Industries, Dolan-Jenner Lite PL-800; (Fiber filters heat-absorbing with equipped bulb was a halogen with illuminator suspension a fiber-optic with illuminated stirred The control. negative a as added were carboxylase) of pyruvate (inhibitor acid and developer, 2 probe. mM enzyme, phenylacetic kit’s assay the containing solution mix buffer with mixed was suspension ing 0.1 adps fles t .4 mW/cm 1.14 at filters band-pass µ ± ± M acetyl-coenzyme A (acetyl-CoA), and reaction buffer. The result The buffer. reaction and (acetyl-CoA), A acetyl-coenzyme M 24 nm), green (540 (540 green 24 nm), 24 nm), green (540 (540 green 24 nm), µ L of acetonitrile and then incubated for 1 h and incubated then deriva at L 60 of °C. Silylated acetonitrile 2 , , and reaction buffer. 1 4 2 2 , 0.1 , mM phenyl- 0.1 by controlling a neutral density filter (NDL-25C-4, Thorlabs, Thorlabs, (NDL-25C-4, filter density neutral a controlling by by controlling a neutral density filter (NDL-25C-4, Thorlabs, Thorlabs, (NDL-25C-4, filter density neutral a controlling by µ M M SNARF-1, 2 mM HEPES (pH 8.0), and 2 mM MgSO 2 PO µ ± m; Agilent m; Technologies, Agilent Foster City, Analyses CA). 4 24 nm), green (540 (540 green 24 nm), , 3 mM NaH mM 3 , ± ± 2 , 5 5 , 27 nm), or red (660 or (660 red 27 nm), 27 nm), or red (660 or (660 red 27 nm), µ g l of the solution was dehydrated on conduc l the was dehydrated of solution the for were 30 and in taken min. The dried pellets µ p -benzoquinone, and 2 mM HEPES (pH 8.0)) 8.0)) (pH and 2 mM HEPES -benzoquinone, M pyruvate, 0.5 0.5 pyruvate, M 2 µ by controlling a neutral density filter filter density neutral a controlling by M M FCCP (a protonophore) and 1 2 PO 2 PO 4 , , 3 mM NaH et al. et 4 , 5 mM NaHCO mM 5 , ± nature biotechnology nature 27 nm), or 27 nm), red (660 4 µ ± ± 9 l l MTBSTFA + 1% TBDMCS 52 nm) band-pass filters at filters band-pass 52 nm) 52 nm) band-pass filters at filters band-pass 52 nm) . Artificial organelles con organelles Artificial . µ M pyruvate carboxylase, carboxylase, pyruvate M 2 PO 4 , , 5 mM NaHCO 2 area, 30–50 area, To determine determine To 3 , 5 mM ADP, mM 5 , ± 52 nm) µ M M oli 4 , , and 4 8 Ω in in 3 ------) ,

© 2018 Nature America, Inc., part of Springer Nature. All rights reserved. illuminator that emitted either broadband white light at 1.14 mW/cm 1.14 at light white broadband either emitted that illuminator fiber-optic a with them illuminating by systems protocellular the in induced was 30 min to polymerization ions into allow actin to the cells, be transported ImageJ program and the following equation: equation: following of the and program ImageJ curvature contour the and sphere systems. protocellular actin the of area the Calculating measurements. the during off turned was light the and fps 1.97 at imaged were cells the source, Toand from laser the durations. for lengths various any effects side minimize wave at various light using illuminated were dishes confocal in samples The NJ). Newton, Thorlabs, (PM160, meter power a with NJ) Newton, Thorlabs, bin positions for each time point ( point time each for positions bin as bin a with fixed and widths of was histogram calculated contour curvatures distribution probability The effects. end avoid to ignored were curve digital where all derivatives are with respect to t. respect are with derivatives all where estimation. of curvature method spline–based cubic a Rhod- with were reanalyzed systems of outer the membranes protocellular PE–labeled systems, protocellular the of curvature contour the calculate To of mode scan A eV. 70 at mode EI 15–650 in operated MSD The ml/min). (1.0 were gas 280 temperature °C. Helium and as was source used interface carrier temperature was inlet 250 °C with a operating split ratio of The 1:10. MSD ion front The min. 20 for at 8 °C °C/min 300 to finally and 5 min over °C/min 20 chromato following graphic conditions: temperature initial an 50 °C for the 2 min, increasing to 150 to °C with at (MSD) coupled detector chromatograph mass-selective gas 5973 Agilent 6890 Agilent an with performed were nature biotechnology nature filters at 1.14 mW/cm 1.14 at filters (540 by green filtered was that light sidered significant when when significant sidered was measured using two-tailed unpaired Student’s Statistics. than than 22 pixels were ignored in the analysis. Cubic splines (( fewer knots with curves Digital between pixels. 22 was splines) the distance in sites join The (polynomial image. gray-scale original the in pixels to ing - correspond MATLAB routines custom-written with performed was analysis in resuspension buffer solution (0.2 M glucose, 20 mM Tris-HCl (pH 7.8), 7.8), (pH 0.5 mM CaCl Tris-HCl mM 20 glucose, 1:10 M (0.2 diluted solution was buffer systems resuspension in protocellular of suspension The actin. and glass the of bottom the between interactions nonspecific prevent to use before h 1 Co., St. MO) for Louis, 1% with F-127 Pluronic (Sigma–Aldrich blocked been had that dishes confocal to transferred was systems protocellular of pension systems. protocellular photosynthetic in actin of Polymerization WorkStation MassHunter software. Agilent using processed filament curve by applying the following definition of curvature: curvature: of definition following the by applying curve filament the along apart pixel one points at measured was Curvature polyfit. the to fit rjce raAtnpeeae rao rtclua system protocellular of area area Actinsphere area Projected m Data are presented as mean / z was used for the analyses. GC/MS data were acquired and and acquired were data GC/MS analyses. the for used was 2 , , and 0.5 mM MgCl (%) = 2 The area of the actin sphere was calculated using the the using calculated was sphere actin the of area The by controlling a neutral density filter (NDL-25C-4, (NDL-25C-4, filter density neutral a controlling by P > 0.05 and significant when when and significant > 0.05 k = 2 ) ( ). ). After incubation at room temperature for Supplementary Note 2 Supplementary y x x y y x ± ′ ′′ ′ 27 nm) or red (660 or 27 nm) (660 red ± + s.d. The significance between data sets / −

′ The first and last two pixels of each pixels two and last first The ′′ ′ /2 2 3 2 t-test.

All automated curvature automated All curvature P P values were not con < 0.05 and < 0.05 ± x ). 52 nm) band-pass band-pass 52 nm) ( t ), ), y ( t )) )) were then P ss < 0.01. A sus A 2 or or - - - -

37. 36. 35. 34. 33. 32. 31. 49. 48. 47. 46. 45. 44. 43. 42. 41. 40. 39. 38. upon reasonable request. reasonable upon author from data are the corresponding source available Additional files. tion informa supplementary its and paper the within available are study this of availability. Data is available in the Nature Research Summary linked Reporting to this article. Life Sciences Reporting Summary. lar system organelle membrane systems is supplied in Code Code availability.

(2010). giant unilamellar vesicles. unilamellar giant ik, .. Kuwc, .. uiiain n rcnttto o te F1F0-ATP the of reconstitution and Purification T.A. Krulwich, & D.B. Hicks, or, .. h ceurd itr o te eeomn ad s o simultaneous of use and development the of history chequered liposomes. The into R.J. Porra, proteins membrane of Reconstitution D. Lévy, & J.L. Rigaud, A.K. Dioumaev, active highly of isolation S.Y.,Kim, characterization and Screening K.H. Jung, & Waschuk,L.S. Brown, S.A., Fast C.H. Yang, & C. Liu, T.H., Xu, Z.G., Functional Wang, R. Croce, & E.J. Boekema, S., Kereïche, R., Kouril, S., Caffarri, Hazard, A. & Montemagno, C. Improved purification for thermophilic F1F0 ATP F1F0 thermophilic for purification Improved C. Montemagno, & A. Hazard, synthesis ATP J.L. Rigaud, & G. Girault, M., Duñach, P., Richard, B., Pitard, Gutsmann, T., Heimburg, T., of Keyser, U., Mahendran, K.R. screening & Winterhalter, M. Protein Rapid N. Fertig, & M. George, M., Beckler, C., Farre, M., Kreir, the of analysis microscopy Spectroscopic J.M. Harris, & G.A. Myers, E.C., Heider, intracellular an as SNARF-1 J. F.Bolard, Sureau, Henry-Toulmé,& N., O., Seksek, H.J. Kaiser, on proteins of Reconstitution D.A. Fletcher, & D.L. Richmond, E.M., Schmid, P. Girard, Angelova, M.I. & Dimitrov, D.S. electroformation. and preparations vesicles: Giant P. Stano, & H. Engel, K., Cosentino, P.,Walde, E.F.DeLong, & M. Leclerc, J.L., Spudich, E.N., Spudich, O., Béjà, Proteorhodopsin Soc. Natl. Acad. Sci. USA Sci. Acad. Natl. synthase from alkaliphilic Bacillus firmus OF4. Evidence that the enzyme enzyme the Enzymol. Methods that Evidence OF4. firmus Na+. not Bacillus but H+ translocates alkaliphilic from synthase endogenous with coli Escherichia synthesis. retinal in color-tuning proteorhodopsin of spinach. from complexes core II photosystem synthase using n-dodecyl beta- electrophysiological for membranes characterization. lipid planar freestanding into reconstitution proteoliposomes. in reconstituted OmpF of analysis electrophysiological activity: protein membrane vesicles. phospholipid (2011). individual of pH interior (1991). 49–54 193, pH indicator in laser microspectrofluorometry: a critical assessment. vesicles. (2015). unilamellar giant electroformed applications. ocean. the in phototrophy (2002). 149–156 73, b. and a chlorophylls of determination accurate the for equations architecture of higher plant photosystem II supercomplexes. supercomplexes. II photosystem plant higher of architecture proteorhodopsin. f T snhss ih bacteriorhodopsin. with (1996). synthases into ATP reconstitution reconstituted optimal of the PS3 defining Bacillus Factors 1. thermophilic bacteriorhodopsin. from with liposomes synthase ATP F0F1 the by (2002). 3052–3063 (2009). 3052–3063 , 303–311 (1986). 303–311 81, nw ehd o te eosiuin f ebae rtis into proteins membrane of reconstitution the for method new A al. et re o lpd hss n oe ad lsa membranes. plasma and model in phases lipid of Order al. et ChemBioChem Custom code for calculating contour curvature of protocellu The authors confirm that the data supporting the findings findings the supporting data the that confirm authors The rtn rnfr i te htceia rato cce of cycle reaction photochemical the in transfers Proton al. et Nat. Protoc. Nat. Biochemistry Lab Chip Lab Bba-Bioenergetics , 65–86 (2003). 65–86 372,

, 16645–16650 (2009). 16645–16650 106, Nature Biophys. J. Biophys. , 587–595 (2008). 587–595 8, , 848–865 (2010). 848–865 11, d , 188–198 (2015). 188–198 10, J. Biol. Chem. Biol. J. , 5348–5358 (2002). 5348–5358 41, -maltoside. , 786–789 (2001). 786–789 411, Further information on experimental design , 504–513 (2008). 504–513 1777, , 419–429 (2004). 419–429 87, Arch. Biochem. Biophys. ehd Cl Biol. Cell Methods

265 u. . Biochem. J. Eur. J. Integr. Plant Biol. Plant Integr. J. , 20547–20554 (1990). 20547–20554 , nl Chem. Anal. Supplementary NoteSupplementary 2. Faraday Discuss. Chem. doi:10.1038/nbt.4140

8230–8238 83, Photosynth. Res. Photosynth. 769–778 235, 319–338 128, Anal. Biochem. 407, 117–124

MO J. EMBO 52 , 793–800 793–800 , Proc.

28 - - ,

nature research | life sciences reporting summary

Corresponding author(s): K-H. Jung, T. Ahn, K. Parker, K. Shin Initial submission Revised version Final submission Life Sciences Reporting Summary Nature Research wishes to improve the reproducibility of the work that we publish. This form is intended for publication with all accepted life science papers and provides structure for consistency and transparency in reporting. Every life science submission will use this form; some list items might not apply to an individual manuscript, but all fields must be completed for clarity. For further information on the points included in this form, see Reporting Life Sciences Research. For further information on Nature Research policies, including our data availability policy, see Authors & Referees and the Editorial Policy Checklist.

` Experimental design 1. Sample size Describe how sample size was determined. No statistical methods were used to predetermine sample size 2. Data exclusions Describe any data exclusions. No data were excluded for proteoliposome and artificial organelles experiments. For encapsulated organelle membrane system, It is excluded that the encapsulation does not proceed stochastically (See Supplementary Figure 16b). 3. Replication Describe whether the experimental findings were No attempts at replication failed. reliably reproduced. 4. Randomization Describe how samples//participants were For purification of membrane proteins, a random selection of spinach and protein allocated into experimental groups. expressed e.coli was chosen. For proteolipsomes and artificial orangelles, there was no biased use of particular sample tubes or chemical reagents. Though across batches, individuals were randomly processed. No other randomization is applicable for this study. 5. Blinding Describe whether the investigators were blinded to Not applicable. Group allocation was not a feature for experiments and associated group allocation during data collection and/or analysis. data for in-vitro experiments. Note: all studies involving and/or human research participants must disclose whether blinding and randomization were used.

6. Statistical parameters For all figures and tables that use statistical methods, confirm that the following items are present in relevant figure legends (or in the Methods section if additional space is needed).

n/a Confirmed

The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement (animals, litters, cultures, etc.) A description of how samples were collected, noting whether measurements were taken from distinct samples or whether the same sample was measured repeatedly A statement indicating how many times each experiment was replicated The statistical test(s) used and whether they are one- or two-sided (note: only common tests should be described solely by name; more complex techniques should be described in the Methods section) A description of any assumptions or corrections, such as an adjustment for multiple comparisons June 2017 The test results (e.g. P values) given as exact values whenever possible and with confidence intervals noted A clear description of statistics including central tendency (e.g. median, mean) and variation (e.g. standard deviation, interquartile range) Clearly defined error bars

See the web collection on statistics for biologists for further resources and guidance.

1 Nature Biotechnology: doi:10.1038/nbt.4140 ` Software nature research | life sciences reporting summary Policy information about availability of computer code 7. Software Describe the software used to analyze the data in this Student’s unpaired t-tests (Microsoft Excel 2016 MSO Version 16.0.9029.2167 study. 64bit), Confocal image (Carl zeiss ZEN 2.3 SP1 black 64bit 14.0.0.201), 3D reconstruction of confocal images, (Image.J 1.51n), Curvature calculation (MATLAB 9.1.0.441655 R2016b), Curve fitting (OriginPro 8 SR0 v8.0724 B724)

For manuscripts utilizing custom algorithms or software that are central to the paper but not yet described in the published literature, software must be made available to editors and reviewers upon request. We strongly encourage code deposition in a community repository (e.g. GitHub). Nature Methods guidance for providing algorithms and software for publication provides further information on this topic.

` Materials and reagents Policy information about availability of materials 8. Materials availability Indicate whether there are restrictions on availability of No restrictions. unique materials or if these materials are only available for distribution by a for-profit company. 9. Antibodies Describe the antibodies used and how they were validated No Antibodies were used. for use in the system under study (i.e. assay and species). 10. Eukaryotic cell lines a. State the source of each eukaryotic cell line used. No Eukaryotic cell lines were used in this study.

b. Describe the method of cell line authentication used. No Eukaryotic cell lines were used in this study.

c. Report whether the cell lines were tested for No Eukaryotic cell lines were used in this study. mycoplasma contamination.

d. If any of the cell lines used are listed in the database No Eukaryotic cell lines were used in this study. of commonly misidentified cell lines maintained by ICLAC, provide a scientific rationale for their use.

` Animals and human research participants Policy information about studies involving animals; when reporting research, follow the ARRIVE guidelines 11. Description of research animals Provide details on animals and/or animal-derived No animals were used in this study materials used in the study.

Policy information about studies involving human research participants 12. Description of human research participants Describe the covariate-relevant population No human research participants were used in this study characteristics of the human research participants. June 2017

2 Nature Biotechnology: doi:10.1038/nbt.4140