RESEARCH ARTICLE SUMMARY◥

PLANT BIOLOGY

Ubiquitin-dependentchloroplast-associatedprotein degradation in plantsQihua Ling*, William Broad*, Raphael Trösch, Mats Töpel, Tijen Demiral Sert,Panagiotis Lymperopoulos, Amy Baldwin, R. Paul Jarvis†

INTRODUCTION: Chloroplasts are plant or-ganelles responsible for the bulk of terrestrialphotosynthetic primary production. They evolvedvia endosymbiosis from a cyanobacterial orga-nismmore than a billion years ago. The biogen-esis and operation of chloroplasts depends onthe assembly and homeostasis of thousands ofnucleus-encoded proteins, which together con-stitute a large part of the organellar proteome.These proteins are imported by multiproteintranslocases in each of the chloroplast envelopemembranes after translation in the cytosol.Chloroplast proteins are subject to proteolyticregulation, which plays vital roles in main-taining normal organellar functions and indelivering responses to developmental andenvironmental cues. Turnover of internal chlo-

roplast proteins is controlled by proteases in-herited fromtheorganelle’s prokaryotic ancestor,but mechanisms underlying the degradationof chloroplast outer envelopemembrane (OEM)proteins are poorly defined.

RATIONALE:Wepreviously showed that com-ponents of the protein import translocases inthe OEM (so-called TOC proteins) are ubiq-uitinated by the OEM-localized ubiquitin E3ligase SP1 and subsequently degraded by thecytosolic 26S proteasome. Inherent in this pro-cess is a need to extricate the target proteinsfrom the chloroplast membrane (they are in-tegralmembrane proteins), and to achieve thisthere must exist mechanisms to overcome thephysical and thermodynamic barriers to ex-

traction. This implies that additional factorsare involved in OEM protein degradation, andwe sought to identify these by applying for-ward genetics and proteomic analysis in theplant Arabidopsis.

RESULTS: We identified two factors requiredfor the degradation of chloroplast OEM pro-teins: SP2 and CDC48. The former is anOmp85-type b-barrel channel of prokaryotic originlocated in the OEM, and the latter is a con-served eukaryotic AAA+ chaperone located inthe cytosol. We observed that inactivation ofeither component triggers the selective overac-

cumulation of target pro-teins, specifically at thechloroplast envelope. Weused genetic analyses todemonstrate that SP2 andCDC48 act together in thesame proteolytic pathway

as the SP1 E3 ligase and physical interactionstudies to show that the three components canform a complex at the surface of the chloro-plast. Furthermore, by applying complement-ary in vivo and in vitro assays, we demonstratedthat the SP2 and CDC48 proteins cooperate tobring about the extraction (“retrotranslocation”)of ubiquitinated proteins from the OEM. Over-all, the data are consistent with amodel (see thefigure) in which SP2 and CDC48 fulfil conduct-ance and motor functions, respectively, in theretrotranslocation of OEM proteins ubiquiti-nated by SP1 to enable their proteasomal de-gradation in the cytosol. These results extendthe range of known functions of Omp85 super-family proteins (which heretofore includedbacterial protein secretion, membrane pro-tein biogenesis, and organelle protein import)and of CDC48 (which has a well-characterizedrole in endoplasmic reticulum–associated pro-tein degradation). The broader importance ofthis proteolytic mechanismwas demonstratedby physiological analyses of plants with al-tered SP2 activity, which revealed defects inorganellar functions, plant development, andviability.

CONCLUSION: Collectively, our results de-scribe amulticomponent system for chloroplastenvelope protein removal, dependent on thecytosolic ubiquitin-proteasome system, which iscritically important for plant growth. A key partof the system is a protein retrotranslocationmechanism of chimeric prokaryotic-eukaryoticancestry that operates at the surface of the or-ganelle. We refer to this proteolytic system aschloroplast-associated protein degradation, orCHLORAD.▪

RESEARCH

Ling et al., Science 363, 836 (2019) 22 February 2019 1 of 1

The list of author affiliations is available in the full article online.*These authors contributed equally to this work.†Corresponding author. Email: paul.jarvis@plants.ox.ac.ukCite this article as Q. Ling et al., Science 363, eaav4467(2019). DOI: 10.1126/science.aav4467

Chloroplast-associated protein degradation. CHLORAD is a proteolytic system thatselectively removes chloroplast OEM proteins, including TOC components of the chloroplastprotein import machinery. The SP1 E3 ligase directs the ubiquitination (Ub) of targets; ithas a RING finger (RNF) domain for ubiquitin-conjugating enzyme (E2) recruitment and anintermembrane space (IMS) domain that binds to its targets. The SP2 and CDC48 proteinsmediate target retrotranslocation to the cytosol, respectively providing a conduit and drivingforce for the process. Upon release to the cytosol, targets are degraded by the 26S proteasome(26SP). Additional, as yet unknown factors are shown in gray.

ON OUR WEBSITE◥

Read the full articleat http://dx.doi.org/10.1126/science.aav4467..................................................

on May 21, 2020

http://science.sciencem

ag.org/D

ownloaded from

RESEARCH ARTICLE◥

PLANT BIOLOGY

Ubiquitin-dependentchloroplast-associatedprotein degradation in plantsQihua Ling1*, William Broad1*†, Raphael Trösch2‡, Mats Töpel2§, Tijen Demiral Sert2¶,Panagiotis Lymperopoulos2#, Amy Baldwin2**, R. Paul Jarvis1,2††

Chloroplasts contain thousands of nucleus-encoded proteins that are imported from thecytosol by translocases in the chloroplast envelope membranes. Proteolytic regulation ofthe translocases is critically important, but little is known about the underlyingmechanisms. We applied forward genetics and proteomics in Arabidopsis to identifyfactors required for chloroplast outer envelope membrane (OEM) protein degradation.We identified SP2, an Omp85-type b-barrel channel of the OEM, and CDC48, a cytosolicAAA+ (ATPase associated with diverse cellular activities) chaperone. Both proteins actedin the same pathway as the ubiquitin E3 ligase SP1, which regulates OEM translocasecomponents. SP2 and CDC48 cooperated to bring about retrotranslocation ofubiquitinated substrates from the OEM (fulfilling conductance and motor functions,respectively), enabling degradation of the substrates by the 26S proteasome in thecytosol. Such chloroplast-associated protein degradation (CHLORAD) is vital for organellarfunctions and plant development.

Chloroplasts are plant organelles responsi-ble for the bulk of terrestrial photosynthet-ic primary production, and they evolvedvia endosymbiosis from a cyanobacterialorganism more than 1 billion years ago

(1). The modern chloroplast proteome comprises~3000 proteins, most of which are nucleus en-coded and imported posttranslationally by multi-protein translocases located in the organelle’souter and inner envelope membranes; thesetranslocases are termed TOC and TIC (translo-cons at the outer and inner envelopemembranesof chloroplasts), respectively (2–5). The bio-genesis and operation of chloroplasts requires notonly the assembly of the constituent organellarproteins but also their coordinated homeostasis.Turnover of internal chloroplast proteins is gov-erned by several prokaryotic-type proteases in-herited from the endosymbiont (6). By contrast,outer envelope membrane (OEM) proteins are

degraded by the cytosolic ubiquitin-proteasomesystem (UPS) via poorly understood mecha-nisms (1).The RING (really interesting new gene)-type

ubiquitin E3 ligase SP1 is located in the chloro-plast OEM, where it mediates the ubiquitinationof OEM components of the chloroplast proteinimportmachinery (so-called TOC proteins), there-by promoting their degradation by the cytosolic26S proteasome (7). The TOC components af-fected by SP1 include the receptors Toc159 andToc33 and the channel protein Toc75. SuchSP1-mediated regulation of the TOC apparatuschanges the organellar proteome, which in turninfluences the developmental fate and functionsof the organelle (e.g., enabling plant adaptationto abiotic stress) (7, 8). Although the role of SP1 inmarking proteins for degradation is clear, otheraspects of this chloroplast protein degradationsystem have remained obscure. Because TOCproteins are integral membrane components,additional factors are most likely required toovercome the physical and energetic barriersto their extraction from the membrane beforedegradation in the cytosol, as is the case in othermembrane-associated proteolytic systems (9–11).

Identification and phenotypic andmolecular analysis of SP2

To improve understanding of the SP1-dependentproteolytic pathway, we revisited the forward-genetics screen (7) that originally identified SP1;this was a screen for extragenic suppressors ofthe Arabidopsis Toc33 mutation plastid proteinimport1 (ppi1), which causes chlorosis because of

defective chloroplast protein import (12). In ad-dition to sp1 mutants, mutants with lesions ata second, unlinked locus were identified in thescreen; these were termed suppressor of ppi1locus2 (sp2) mutants. Double mutant sp2 ppi1plants were larger and greener than the ppi1 pro-genitor (Fig. 1, A to C) and exhibited substan-tial improvements in chloroplast developmentand chloroplast protein import capacity (Fig. 1,D to F); in all of these respects, the sp2 mu-tants were phenotypically very similar to sp1mutants (7).The SP2 locus (At3g44160) (Fig. 1G) was iden-

tified by using a combination of genetic mappingand whole-genome sequencing of the three in-dependentmutant alleles identified in the screen(fig. S1). The original alleles were phenotypicallysimilar to one another and to a transferred-DNA(T-DNA) insertion mutant affecting the samegene (Fig. 1, H and I, and fig. S1C); unless spe-cifically stated otherwise, the latter (sp2-4, a nullallele) was used in all subsequent analyses. Theencoded protein is a member of the Omp85 su-perfamily of b-barrels, which are involved in pro-tein biogenesis and transport and are widelydistributed in the outer membranes of bacteria,mitochondria, and chloroplasts (13, 14). The SP2protein is of unknown function, but it is broadlyconserved in the angiosperms (flowering plants)and is closely related to the chloroplast outermembrane protein OEP80 (a protein that hasalso been termed Toc75-V) (fig. S2) (15). Thefunction of OEP80 is also uncertain (15), al-though it has been proposed to mediate outermembrane protein biogenesis (16, 17) by analogywith well-characterized homologs in bacteria(BamA and TamA) and mitochondria (Sam50/Tob55) (13, 14).The SP2 protein is located in the chloroplast

OEM (figs. S3 and S4) (15), and it has been shownto have the capacity to form a membrane chan-nel, like other members of the Omp85 super-family (18). UnlikeOEP80, SP2 lacks anN-terminalPOTRA (polypeptide transport–associated) do-main (such domains typically mediate protein-protein interactions), suggesting that the twoproteins have functionally diverged (Fig. 1G andfig. S2B), a notion that is also supported by phy-logenetic analyses of the proteins (fig. S2A) (17).Accordingly, OEP80 and SP2 have diametricallyopposing effects on TOC protein abundance, as isevident upon comparing published results [show-ing thatOEP80knockdowndepletes TOCproteins(16)] with those discussed below.

Analysis of the effects of SP2 on TOCproteins and plant development

To shed light on the role of SP2, further geneticanalyses were conducted. In addition to ppi1,two other TOC mutations (hypomorphic allelesof the genes encoding Toc159 and Toc75) (16, 19)were phenotypically suppressed by sp2 (fig. S5).By contrast, mutations that cause chlorosis be-cause of lesions affecting the TIC apparatus ofthe inner envelope membrane were not sup-pressed by sp2 (fig. S6). Together, these data im-plied a close functional relationship between SP2

RESEARCH

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 1 of 12

1Department of Plant Sciences, University of Oxford, OxfordOX1 3RB, UK. 2Department of Biology, University ofLeicester, Leicester LE1 7RH, UK.*These authors contributed equally to this work. †Present address:UK Research and Innovation, 58 Victoria Embankment, LondonEC4Y 0DS, UK. ‡Present address: Department of Biology,University of Kaiserslautern, Paul-Ehrlich-Strasse 23, DE-67663Kaiserslautern, Germany. §Present address: Department ofMarine Sciences, University of Gothenburg, Box 460, SE-40530Göteborg, Sweden. ¶Present address: Department of Biology,Faculty of Arts and Sciences, Harran University, 63300 Sanliurfa,Turkey. #Present address: Department of Plant and EnvironmentalSciences, University of Copenhagen, Thorvaldsensvej 40, 1871Frederiksberg C, Copenhagen, Denmark. **Present address:School of Biosciences, Cardiff University, Cardiff CF10 3AX, UK.††Corresponding author. Email: paul.jarvis@plants.ox.ac.uk

on May 21, 2020

http://science.sciencem

ag.org/D

ownloaded from

and the TOC apparatus, and this notion wassupported by the restored accumulation of Toc75protein in sp2 toc doublemutant plants (Fig. 2, Ato D, and fig. S7). In all of these respects, the sp2mutants were phenotypically very similar to sp1mutants (7).The close functional connection between SP2

and the TOC apparatus was further emphasized

by the observation that overexpression of SP2triggers the specific depletion of TOC proteins(Fig. 2, E and F), resembling closely the effect ofSP1 overexpression (7). Moreover, SP2 [like SP1previously (7)] was shown to interact physicallywith TOC components in coimmunoprecipita-tion (co-IP) experiments (fig. S4), suggestingthat its effect on TOC accumulation is mediated

through direct physical interaction with TOCproteins. The unlikely possibility that SP2 alsoinfluences peroxisomal protein import (20, 21)could be ruled out because of the absence of aneffect on the abundance of peroxisomal proteinimport machinery components (fig. S8).Further similarities between SP2 and SP1 were

observed when the expression profiles of the two

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 2 of 12

Fig. 1. The sp2 mutation suppresses the phenotype of the Toc33knockout mutation, ppi1. (A) Visible phenotypes of 30-day-old sp2-1 ppi1suppressor mutant and control plants grown on soil. (B) Chlorophyll contentsof 10-day-old sp2-1 ppi1 suppressor mutant and control seedlings grown invitro. (C to E) Transmission electron microscopy analysis of the ultrastructureof cotyledon chloroplasts in 10-day-old sp2-1 ppi1 suppressor mutant andcontrol plants grown in vitro. Typical organelles are shown (C). Scale bar,2 mm. These and other micrographs were used to estimate chloroplastcross-sectional area (D) and thylakoid development (E). (F) Analysis ofprotein import into chloroplasts isolated from sp2-1 ppi1 suppressor mutantand control plants by using 35S-labeled Rubisco small subunit precursorprotein as the import substrate, and corresponding quantification of thematuration (mat) of the radiolabeled precursor protein (pre). A representativephosphor screen image is shown (top); times (t) indicate minutes after thestart of each import reaction.Together with similar images from two additional

experiments, this was used to conduct the quantitative analysis shown(bottom); the amount of imported, mature protein in each sample wasexpressed as a percentage of that present at the final time point for theWT.(G) Domain map of the SP2 protein. Gray box, b-barrel domain; black boxes,predicted transmembrane spans.The sites of amino acid substitutions in twosp2mutant alleles are indicated with gray triangles. Numbers indicate aminoacid positions. R, arginine; H, histidine; G, glycine. (H and I) Independentmutantalleles of sp2 suppress ppi1 in similar fashion to sp2-1.The sp2-2 ppi1 andsp2-3 ppi1mutantswere identified in the EMSmutagenesis screen,whereas thesp2-4 T-DNA insertion mutant was obtained from a stock center and crossedto the ppi1 line.The plants were grown in vitro for 10 days before photography(H) and chlorophyll content analysis (I).The chlorophyll values for sp2-2 andsp2-3, which cause missense mutations, were statistically significantly differentfrom that for sp2-1 (Student’s t test, P ≤ 0.01), suggesting that they are weakalleles. All values are means ± SEM (n ≥ 3 experiments or samples).

RESEARCH | RESEARCH ARTICLEon M

ay 21, 2020

http://science.sciencemag.org/

Dow

nloaded from

genes were compared (fig. S9) and when sp2mutant and SP2 overexpressor (SP2-OX) plantswere analyzed physiologically in relation to leafsenescence (Fig. 2G) and salt stress tolerance(Fig. 2H and fig. S10). The activity of SP1 pro-motes both leaf senescence and abiotic stresstolerance (which it does by reconfiguring thechloroplast protein import machinery to producethe necessary organellar proteome changes) (7, 8),and a very similar pattern of phenotypes wasobserved for SP2 (Fig. 2, G and H).

Identification of CDC48 as a mediator ofOEM protein degradation

The identification of a channel-forming compo-nent (SP2) that putatively cooperates with SP1 inTOC protein degradation raised a parallel withendoplasmic reticulum (ER)–associated proteindegradation (ERAD) (9, 10, 22), where polytopicmembrane proteins form a retrotranslocon toenable substrate extraction from the membrane(23–25). In that system, the conserved eukary-otic, multifunctional AAA+ (ATPase associatedwith diverse cellular activities) chaperone CDC48(p97) (9, 10, 26) forms a cytosolic ATP-poweredmotor to drive such retrotranslocation.We soughtsimilar regulators of the TOC apparatus by usingco-IP followed by mass spectrometry and identi-fied CDC48 as a minor TOC-associated compo-nent (fig. S11). The major Arabidopsis isoform,CDC48A (At3g09840) (27) (Fig. 3A), is ubiquitousin the cytosol, but its association with chloro-plasts was evident upon analyzing lysed, cytosol-free cells containing a CDC48 mutant withstabilized substrate binding (fig. S12).Because CDC48 is an essential component in

Arabidopsis, we used plants that inducibly over-expressed a dominant-negative (DN) form of theprotein (CDC48-DN) (Fig. 3A) (27) to assess itsfunction. The expression of CDC48-DN causedchlorosis (Fig. 3B), indicating defective chloro-plast biogenesis, as well as a buildup of reactiveoxygen species (ROS), paralleling an effect of thesp1mutation (8) (fig. S13). In the case of sp1, suchROS overaccumulation was attributed to a fail-ure to properly regulate the TOC apparatus andchloroplast protein import, leading to a deregu-lation of photosynthetic activity (8). Similarly,the effects of CDC48-DN observed in this studywere linked to the specific overaccumulation ofTOC proteins (Fig. 3, C and D), revealing pheno-typic similarity to sp2mutants (Fig. 2, A to D) inaddition to sp1 mutants (7). The SP1 protein,which is subject to UPS-dependent autoregula-tion (fig. S14) (7), also accumulated in responseto CDC48-DN expression (Fig. 3E), indicatingthat SP1 is degraded via the same processes asTOC proteins.Because CDC48 is an abundant cellular con-

stituent distributed throughout the nucleo-cytosolic compartment, we wished to localize itsfunctional links to the chloroplast protein im-port machinery specifically to the OEM in intactcells. By using bimolecular fluorescence com-plementation (BiFC), a method that uses thereconstitution of yellow fluorescent protein(YFP) fluorescence to report on protein-protein

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 3 of 12

Fig. 2. Functional analysis of SP2 reveals roles in chloroplast proteostasis and development.(A to F) Immunoblot analyses of total protein extracts (two loading amounts per sample) fromthe indicated genotypes, including sp2 ppi1 suppressor mutants [(A) and (B)], sp2 and toc75-III-3single and double mutants [(C) and (D)], and two different SP2-OX lines [(E) and (F)]. Slp1,mitochondrial stomatin-like protein 1 (a nonchloroplastic membrane protein that served as aloading control); H3, nuclear histone H3. (G) Leaf senescence analysis of the indicated genotypesusing mature rosette leaves induced to senesce by covering with aluminum foil. Typical control(uncovered) and senescent (covered) leaves are shown (left). The maximum photochemicalefficiency of photosystem II (Fv/Fm) was measured to estimate the extent of senescence (right);the covered values for sp2-4 and SP2-OX were statistically significantly different from that for theWT (Student’s t test, P ≤ 0.0003). (H) Abiotic stress tolerance analysis of the indicated genotypesusing 14-day-old plants grown in vitro on NaCl medium. Typical plants (left) and chlorophyllcontents (right) are shown. The chlorophyll values for the mutant and overexpressor plantswere statistically significantly different from that for the WT (Student’s t test, P < 0.004). Allvalues are means ± SEM (n ≥ 3 experiments or samples).

RESEARCH | RESEARCH ARTICLEon M

ay 21, 2020

http://science.sciencemag.org/

Dow

nloaded from

interactions of interest, we demonstrated thephysical interaction of CDC48 with Toc159 spe-cifically at the chloroplast envelope (Fig. 3, F andG). Separate experiments further revealed thatthe CDC48-DN–triggered accumulation of OEM

proteins (shown in Fig. 3, C and D) occurredspecifically at the envelope in vivo (fig. S15).Together, these data supported a direct role forthe CDC48 chaperone in OEM protein degrada-tion, at the surface of the organelle.

Analysis of the functional and physicalrelationships among SP1, SP2, and CDC48Having identified two new components thatapparently mediate OEM protein degradation,like SP1, we next addressed whether the threecomponents act together in a common pathway.We began by considering the relationship be-tween SP1 and SP2. The absence of phenotypicadditivity in sp1 sp2 double mutants (in the ppi1background) in relation to plant greening andToc75 protein accumulation (Fig. 4, A to D) sup-ported the notion that SP1 and SP2 functiontogether. Accordingly, SP2 was essential for SP1action, because the sp2mutation abrogated theeffect of SP1 overexpression (fig. S16).Moreover, whereas the overexpression of

neither SP1 nor SP2 individually affected plantgreening in thewild-type (WT) background understandard conditions (fig. S10), the simultaneousoverexpression of both genes caused strong chlo-rosis linked to severe depletion of TOC proteins(Fig. 4, E to G), indicating functional interde-pendency between SP1 and SP2. Such interac-tions are often observed where the componentsinteract physically, andmay arise throughmutualstabilization (28, 29). SP1 and SP2 comigrated(with each other and with the TOC apparatus)on native gels, suggesting their coexistence inhigh–molecular weight complexes that eitherinclude or exclude the TOC apparatus (fig. S17A).In accordance with this interpretation, SP1 andSP2 interacted specifically in in vitro pull-downexperiments (fig. S17B).Next, the functional relationships between

CDC48 and the two SP components were as-sessed. This was done by determining the de-pendence of SP overexpression effects uponCDC48 function. We observed that the ability ofthe overexpression of either SP1 or SP2 to trig-ger TOC protein depletion was blocked whenCDC48-DN was coexpressed (Fig. 5). These dataindicated that the functions of the SP compo-nents in OEM protein degradation requireCDC48 and, thus, that all three factors act inthe same proteolytic pathway. In agreementwith this conclusion, reciprocal co-IP assays de-monstrated physical associations among CDC48,SP2, SP1, and TOC proteins (Fig. 6 and fig. S18).

Investigating the roles of SP2 and CDC48in OEM protein retrotranslocation

In ERAD, CDC48 drives the retrotranslocationof substrate proteins from the ER to the cyto-sol. To address the hypothesis that CDC48 issimilarly involved in the extraction of substratesfrom the chloroplast OEM, we conducted retro-translocation assays by using two complemen-tary methodologies (23, 30, 31).First, we used an in vivo assay that evaluates

the distribution of polyubiquitinated substratebetween chloroplasts and the cytosol in intact,proteasome-inhibited cells. We observed that theextraction of polyubiquitinated Toc33 to the cyto-sol was inhibited in cells expressing CDC48-DN,providing strong evidence that CDC48 acts inOEM protein retrotranslocation (Fig. 7, A and B).Replication of this assay with sp2 mutant and

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 4 of 12

Fig. 3. Functional analysis of CDC48 reveals roles in chloroplast proteostasis and develop-ment. (A) Domain map of the CDC48A protein. Gray boxes, ATPase domains (D1 and D2). Thesites of the DN and Trap mutations are indicated with gray triangles. Numbers indicate aminoacid positions. K, lysine; A, alanine; E, glutamic acid; Q, glutamine. (B) Visible phenotypesof 9-day-old plants expressing the CDC48-DN protein or an equivalent nonmutated controlprotein (CDC48-WT), both induced with estradiol for 2 days. Two independent transgenic linesare shown for each construct. (C to E) Immunoblot analyses of total leaf [(C) and (D)] andprotoplast (E) protein extracts from the CDC48-WTand CDC48-DN transgenic plants after inductionwith estradiol or mock control treatment lacking the inducer. For (C) and (D), plants weresimultaneously treated with 200 mM mannitol for 2 days before protein extraction to triggerstress-dependent TOC protein degradation (8); this was essential to observe the effects ofCDC48-DN in this assay. (F and G) BiFC analysis of the CDC48-TOC interaction. Reconstitutionof YFP fluorescence was assessed after transient coexpression of the indicated pairs of fusionproteins, which carry complementary N- or C-terminal YFP fragments (nY and cY, respectively).OEP7 and CDKA1 acted as OEM and cytosolic controls, respectively. Representative imagesare shown (F). Scale bar, 10 mm. In addition, the frequency of protoplasts showing a BiFC signalwas quantified (G). All values are means ± SEM (n = 3 experiments).

RESEARCH | RESEARCH ARTICLEon M

ay 21, 2020

http://science.sciencemag.org/

Dow

nloaded from

SP2-OX cells revealed a similar requirement forSP2 in the extraction of polyubiquitinated Toc33(Fig. 7, C and D). Moreover, when both of theseassayswere repeatedwith SP1 as a second,modelsubstrate [selected because of its high turnoverrate (7)], essentially identical results were ob-tained (Fig. 8, A to D, and fig. S19).To provide corroboration of the in vivo assay

data, we used an in vitro retrotranslocationassay. This experiment tested the ability ofCDC48-DN in a cytosol extract to interfere withthe removal of polyubiquitinated SP1 from iso-lated chloroplasts (Fig. 8, E and F). The resultsclosely paralleled those from the in vivo assays,because the presence of CDC48-DNwas seen toinhibit OEM substrate extraction to the cytosol(in amanner that the control CDC48-WT protein

did not). Thus, we concluded that both CDC48and SP2 act in OEM protein retrotranslocation.

Discussion

This work identified SP2 and CDC48 as func-tional partners of the SP1 E3 ligase and, in sodoing, defined a multicomponent system forchloroplast protein degradation involving inte-gral OEM and cytosolic factors. Both compo-nents act in substrate protein retrotranslocation,and, on the basis of their structural character-istics, we conclude that they provide a retro-translocon for substrate conductance duringextraction and amolecular motor needed tomeetthe energetic threshold of the process (fig. S20).We designate this system chloroplast-associatedprotein degradation (CHLORAD), in recognition of

similaritieswith ERAD (including the involvementof the UPS, CDC48, and retrotranslocation), butemphasize that the central participation of a b-barrelchannel of prokaryotic ancestry and a number ofother key differences set it apart as a mechanis-tically and evolutionarily distinct process (9, 25).A major function of the analogous ERAD sys-

tem is the elimination of a broad range of mis-folded ER proteins (9, 10, 22), which it performsalongside another role in the removal of un-damaged proteins for regulatory reasons (32).The available evidence suggests that CHLORADhas a different focus, with its primary purposebeing regulatory and centered on the TOC appa-ratus of the chloroplast protein import machin-ery. Moreover, whereas ERAD processes lumenalproteins in addition to ER membrane proteins,there is presently no evidence to suggest thatCHLORAD similarly acts on internal chloroplastproteins, a fact which may be linked to the pres-ence of diverse proteases of prokaryotic originin the chloroplast interior (6). An early event inCHLORAD is substrate protein ubiquitination,mediated by SP1, but how substrates are identi-fied and transferred to SP1 is currently unclear.The SP1 intermembrane space domain is involvedin substrate binding (7), and the triggering factorappears to be related to developmental cues orenvironmental stress (7, 8); the stability of SP1may also be relevant, because the protein isautoubiquitinated (7) and is itself degraded byCHLORAD. Whether CHLORAD also acts inthe clearing of damaged proteins, in parallelwith its regulatory functions, remains to bedetermined.Whereas ERAD uses E3 ligases with multiple

transmembrane spans that may simultaneouslyserve as channels (9), CHLORAD requires a sepa-rate channel-forming component. This is pre-sumably because SP1 does not have a sufficientnumber of transmembrane spans to enable channelformation in isolation. Our data indicate thatCHLORAD is mediated by an SP1-SP2 complexinwhich SP2 provides a channel to enable a closelinkbetweenubiquitinationand retrotranslocation.It is possible that the operation of the SP2 channelinvolves lateral opening to enable substrate entryfrom the membrane (24, 33). Unlike any of theproposed ERAD channels [which are all basedon a-helical transmembrane spans (25)], SP2belongs to the Omp85 superfamily of b-barrelproteins originating from prokaryotic cells(13, 14). This indicates that CHLORAD recruiteda channel of prokaryotic origin and that thiscomponent underwent neofunctionalization dur-ing the evolution that followed endosymbiosis.Consistent with this notion, SP2 is unusual in thatit does not possess a partner-protein–interactingPOTRA domain, which is a typical feature ofOmp85-type proteins in other systems. The dataimply that SP2 instead collaborates with SP1 toprovide this recruitment function (7), and wehypothesize that the loss of the POTRA domainwas a key step in the protein’s acquisition of anew function. As has been proposed for ERAD(10), it is possible that more than one channeloperates in CHLORAD (e.g., under different

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 5 of 12

Fig. 4. SP2 acts in the same pathway of TOC protein degradation as SP1. (A to D) Analysis of ansp1 sp2 ppi1 triple mutant. Triple mutant plants were compared with both sp1 ppi1 mutants and sp2 ppi1mutants in relation to the extent of suppression of ppi1. No phenotypic additivity in the triple mutantswas apparent upon analysis of visible phenotypes (A), chlorophyll contents (B), or the abundance ofToc75 protein as determined by immunoblotting [(C) and (D)]; two loading amounts per sample wereanalyzed. (E to G) Analysis of plants simultaneously overexpressing both SP1 and SP2. Double-OXplants were compared with both single SP-OX genotypes, revealing a synergistic interaction in relationto the visible phenotype (E) (left), chlorophyll content (E) (right), and TOC protein depletion asanalyzed by immunoblotting [(F) and (G)]. Values are means ± SEM (n = 4 to 5 experiments).

RESEARCH | RESEARCH ARTICLEon M

ay 21, 2020

http://science.sciencemag.org/

Dow

nloaded from

circumstances); this may account for the ab-sence of a complete block in retrotranslocationin sp2mutants.Once CHORAD has been initiated, the con-

served eukaryotic chaperone CDC48 is recruitedto the chloroplast from the cytosol. This stepmayinvolve the binding of CDC48 to the polyubiquitinchain on a substrate protein with the assist-ance of unknown cofactors or with a putativechloroplast OEM tethering factor (9, 10, 22).Either way, the chaperone then drives substrateextraction to the cytosol for proteasomal degra-dation, which it does by overcoming the en-ergetic barrier to the removal of proteins fromthe membrane (fig. S20). The functional coop-eration of CDC48 with SP2 in this retrotrans-location process is notable, because it indicatesthat CHLORAD has a chimeric prokaryotic-eukaryotic origin, much like the chloroplastprotein import machinery, which also has anOmp85 superfamily protein (Toc75) at its core(2–5). Quite likely, other factors are involved inthe CHLORADprocess, and thesemay exist withinthe SP1-SP2 core complex or act peripherally in thebroader proteolytic pathway.Previously published data concerning SP1

function (7, 8), together with results reportedhere concerning SP2, show clearly that theCHLORAD system is critically important forplant growth and development, with roleslinked to the reconfiguration of the organellarproteome and functions and to organelle devel-opmental fate. These observations provide aclear indication of the physiological importanceof CHLORAD and suggest potential strategiesto improve crop performance (e.g., under envi-ronmental stress).

Materials and methodsPlant material and growth conditions

All Arabidopsis thaliana plants were of theColumbia-0 (Col-0) ecotype, except the ppi1 lineused for the genetic mapping of sp2, which wasintrogressed into Landsberg erecta (Ler) throughseven outcrosses. The sp1-1, sp1-3, ppi1, tic40-4,hsp93-V-1, ppi2-3 (fts1), and toc75-III-3 (mar1)mutants, as well as the 35S promoter–driven SP1overexpressor (SP1-OX) transgenic line, have allbeen described previously (7, 12, 16, 19, 34, 35).The sp2-4 (SALK_137135) mutant was obtainedfrom the Salk Institute Genomic Analysis Lab-oratory and confirmed by polymerase chainreaction (PCR) and reverse transcription PCR (RT-PCR) analyses, as described previously (36); thismutant was phenotypically similar to the threemutants with a chemically induced sp2 alleleidentified in this study and unlike a previouslydescribed T-DNAmutant (37). Formaking doubleand triple mutants, the sp1-3 and sp2-4 alleleswere used (because their mutations are easier todetect by PCR); double mutants were selectedand verified by phenotype analysis and by PCR-based genotyping (table S1). Unless specificallystated otherwise, the sp2-4 allele was used inthe experiments. For consistency with previouswork (7, 8), the sp1-1 allele was used in physiolog-ical experiments.

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 6 of 12

Fig. 5. CDC48 acts in the same pathway of TOC protein degradation as SP1 and SP2. Analysisof the effects of CDC48-DN on the ability of SP1 or SP2 overexpression to trigger TOC proteindepletion. Transgenic plants carrying the following constructs were analyzed by immunoblotting:CDC48-WT, CDC48-DN, SP1-OX, and the latter two in combination (A and B) and CDC48-WT,CDC48-DN, SP2-OX, and the latter two in combination (C and D). In each case, the analysis wasdone after treatment with [or without (mock)] estradiol (E2) to induce the expression of theCDC48 constructs. All values are means ± SEM (n = 3 to 4 experiments).

Fig. 6. CDC48 and SP2 interact physically with each other and with OEM proteins. Co-IP ofSP1, TOC components, and/or CDC48 with HA-tagged WT CDC48 (A) or Myc-tagged SP2(B) from protoplast extracts. Cells were transfected with the following constructs: CDC48-WT-HAor YFP-HA plus SP1-Myc (A) or SP2-Myc plus CDC48-WT-HA, SP1-HA, or YFP-HA (B). In both cases,YFP-HA acted as a negative control. TL, total lysate; poly-Ub, poly-ubiquitinated form; hc, heavychain; a-HA, anti-HA; a-Myc, anti-Myc. The asterisk indicates a nonspecific cross-reacting band.

RESEARCH | RESEARCH ARTICLEon M

ay 21, 2020

http://science.sciencemag.org/

Dow

nloaded from

For in vitro growth, seeds were surface steri-lized, sown onMurashige-Skoog (MS) agar me-dium in petri plates, cold-treated at 4°C, andthereafter kept in a growth chamber, as de-scribed previously (38). All plants were grownunder a long-day cycle (16 hours of light and8 hours of darkness). For induction of CDC48-WTor CDC48-DN expression in the correspondingtransgenic lines, 7-day-old plants were transferredonto MS agar medium supplemented with 4 mMestradiol (Sigma).

Physiological studies

Chlorophyll measurements were performed byusing a Konica-Minolta SPAD-502 meter (39)for analysis of rosette-stage plants or by photo-metric quantification after extraction in N,N′-dimethylformamide (DMF) as describedpreviously(40) for analysis of seedlings.

Dark treatments for the induction of senes-cence were conducted as previously described(7, 41). Developmentally equivalent leaves of28-day-old plants were wrapped in aluminumfoil while still attached to the plant and then leftunder standard growth conditions for 5 days. Themaximum photochemical efficiency of photo-system II (Fv/Fm) was determined by measuringchlorophyll fluorescence with a CF Imager (Tech-nologica, UK) as described previously (42). Five ex-periments were performed, and approximatelyfive leaves (each one from a different plant) wereanalyzed per genotype in each experiment.Salt stress experiments were conducted as

described previously, with minor modifications(8). All seeds of the different genotypes used inthis workwere harvested at the same time. Seedswere germinated directly on MS agar medium(supplemented with 1% sucrose) containing 150

to 170mMNaCl. Stress tolerance was assessed bymeasuring chlorophyll accumulation after 14 days.Three experiments were performed, and ~25seedlings per genotype were analyzed in eachexperiment.Hydrogen peroxide was detected by staining

with 3,3′-diaminobenzidine (DAB) (Sigma) aspreviously described (43). Seven-day-old plantsgrown on MS agar medium were transferredonto similar medium containing 4 mM estradiol(for induction of CDC48-WT and CDC48-DN ex-pression). The plants were left to grow for afurther 2 days before the initiation of DAB stain-ing. Each experiment used approximately fiveseedlings per genotype. Three experiments wereperformedwith the same result, and typical imagesare presented. The area of stainingwas quantifiedby using ImageJ as described previously (8).

Identification of the sp2 mutants andgenetic mapping

The original sp2 mutants (sp2-1 [sp2-310], sp2-2[sp2-416], and sp2-3 [sp2-555]) were identified byscreening the M2 progeny of 7000M1 ppi1 seedsthat hadbeen treatedwith 100mMethylmethane-sulfonate (EMS) for 3 hours by using a publishedprocedure (7, 44). Allelism test crosses confirmedthat all of the mutations are allelic. All three sp2ppi1mutants were backcrossed to ppi1mutantsthree times before phenotypic analysis. Initialmapping of sp2was conducted by analyzing thegreenest plants in F2 populations from crossesbetween sp2-1 ppi1 (Col-0) and ppi1 mutantsintrogressed into the Ler ecotype by using PCRmarkers that detect Col-0/Ler polymorphisms.In a mapping population of 190 such F2 plants,six were heterozygous for the marker F21A17 atposition 12285000 on the upper arm of chromo-some 3 but homozygous for Col-0 downstreamof that, suggesting that the suppressor mutationwas in the downstream Col-0 region; F3 seedlingsfrom these six plants were grown and verifiedvisually to be nonsegregating, as expected for ahomozygous sp2-1 ppi1 double mutant. In a se-cond mapping population of 192 plants, the sp2mutation was further mapped to the south ofa more southerly marker, MJI6-2 at position12597802 on the upper arm of chromosome 3.However, it was not possible to determine theposition of the sp2 locus precisely because ofthe persistence of an “island” of Col-0 DNA inthe Ler-introgressed ppi1 line, near the sp2 locus(around the chromosome 3 centromere). Thus,final identification of the gene was achieved bywhole-genome sequencing.

Whole-genome sequencing and assembly

Approximately 100mg of plant inflorescence tissuefrom each of the original sp2 alleles (sp2-1 ppi1,sp2-2 ppi1, and sp2-3 ppi1) and from ppi1-1 washarvested and flash-frozen in liquid nitrogen. Totalgenomic DNAwas then extracted by using an E.Z.N.A. plant DNA kit (Omega Bio-tek) accordingto themanufacturer’s guidelines. TheDNAsampleswere quantified by comparison with standards.Library preparation and sequencing were con-

ducted at the Earlham Institute (Norwich, UK).

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 7 of 12

Fig. 7. CDC48 and SP2 are required for the retrotranslocation of polyubiquitinated Toc33.(A and B) Analysis of the role of CDC48 in the retrotranslocation of Toc33-HA substrate by anin vivo retrotranslocation assay. Protoplasts isolated from CDC48-WT and CDC48-DN transgenicplants that were transiently expressing Toc33-HA were treated, after estradiol induction, with 5 mMbortezomib proteasome inhibitor and then separated into chloroplast and cytosol fractions. In thisassay, retrotranslocation occurred in intact cells, and the retrotranslocated Toc33-HA was protectedby bortezomib inhibition, which initiated the experiment. After fractionation, Toc33-HA wasimmunoprecipitated from both fractions and detected by immunoblotting with anti-HA (a-HA) andanti-ubiquitin (a-Ub) (A). Retrotranslocation efficiency was assessed by quantifying the relativeamounts of ubiquitinated Toc33 in the chloroplasts and cytosol (B). (C and D) Analysis of the role ofSP2 in the retrotranslocation of Toc33-HA substrate by an in vivo retrotranslocation assay performedas described for (A) and (B) by using protoplasts from WT, sp2 mutant, and SP2-OX plants (andwithout the need for estradiol induction). Typical immunoblotting results are shown (C), along withquantification (D). All values are means ± SEM (n = 3 experiments).

RESEARCH | RESEARCH ARTICLEon M

ay 21, 2020

http://science.sciencemag.org/

Dow

nloaded from

Approximately 1 to 5 mg of genomic DNA persample at a minimum concentration of 20 ng/mlwas used in sequencing library preparation. Indi-vidual barcoded Illumina TruSeq DNA librarieswere generated for each genotype. The four sam-ples were then sequenced on one lane of IlluminaHiSeq2000,which generatedbetween32.2millionand 42.3 million 100–base pair (bp) paired-end reads for each sample. The first five bases ofthe 5′ ends of the reads were removed by usingfastx_trimmer v.0.0.14 (45), and any bases witha Phred quality score below 15 were removedfrom the 3′ end by using cutadapt (v.1.3) (46).Illumina TruSeq adaptors where also removedby using cutadapt, where a minimum overlap of10 bases with the adaptors, a maximum error rateof 0.1, and a minimum final read length of 50bases were set. Lastly, fastq_quality_filter v.0.0.14(45) was used to remove sequences with a Phredscore below20 inmore than 5%of the bases. Readpairs were identified by using pairSeq.py (47).The reads from the ppi1 single mutant were

mapped to the TAIR10 A. thaliana referencegenome (48) of the Phythozome v.9.0 release (49)by using clc_mapper v.4.0.13.86165 (50). A con-sensus sequence in FASTA format was then gen-erated by using clc_find_variations v.4.0.13.86165(50). Also, the transcript sequences from the TAIR10release were aligned to the reference genome tofacilitate manual examination of identified muta-tions and to visualize whether a particular muta-tion occurs in an exon or intron, and so on.The three datasets from the individual sp2

double mutants were independently aligned tothe ppi1 reference genome by using clc_mapperv.4.0.13.86165. The ppi1 reads were also alignedto the same reference in order to identify anyvariable sites resulting from allelic variation inthe ppi1 line.

In silico identification of mutations

The mutagen EMS used to generate the threesp2 mutants reacts with guanine in the DNAmolecule and is likely to (i) cause pointmutationsthat change guanine to adenine (or cytosine tothymine on the reverse strand). The respectivemutations affecting the three sp2mutants werefurthermore expected to (ii) occur in the samegene (or corresponding promotor region) in (iii)all three mutants but (iv) not necessarily in theexact same positions. These four search criteriawere implemented in the program “find_sp2.py”(51), which takes as input a gff3 file with genecoordinates and the single-nucleotide polymor-phism variant output from clc_mapper and out-puts a list of names of mutated genes from eachdataset. Genes found to bemutated in all the sp2datasets were thenmanually examined by visual-izing the alignment data with the genome viewerIGV (v.2.3) (52).This analysis showed that each sp2 allele con-

tains aG-to-Apointmutationwithin theAt3g44160gene, just to the south of the chromosome 3 cen-tromere. In sp2-1, a mutation was detected at thesplice junction preceding the final exon; this waslater shown to cause mis-splicing, frameshifts,and premature termination, implying that sp2-1 is

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 8 of 12

Fig. 8. CDC48 and SP2 are required for the retrotranslocation of polyubiquitinated SP1. (A toD) Analysis of the roles of CDC48 and SP2 in the retrotranslocation of SP1-HA substrate by an in vivoretrotranslocation assay. Protoplasts isolated from CDC48-WTand CDC48-DN transgenic plants [(A) and(B)] or from WT, sp2 mutant, and SP2-OX plants [(C) and (D)] that were transiently expressing SP1-HAwere treated [after estradiol induction in the case of (A) and (B)] with 5 mM bortezomib proteasomeinhibitor and then separated into chloroplast and cytosol fractions. In this assay, retrotranslocation occurredin intact cells (as in Fig. 7), and the retrotranslocated SP1-HA was protected by bortezomib inhibition, whichinitiated the experiment. After fractionation, SP1-HA was immunoprecipitated from both fractions anddetected by immunoblotting with anti-HA (a-HA) and anti-ubiquitin (a-Ub) [(A) and (C)]. Retrotranslocationefficiency was assessed by quantifying the relative amounts of ubiquitinated SP1 in the chloroplastsand cytosol [(B) and (D)]. (E and F) Analysis of the role of CDC48 in the retrotranslocation of SP1-HAsubstrate by an in vitro retrotranslocation assay. This assay [in contrast with the in vivo assay depictedin (A) to (D)] used a cell-free reaction. Chloroplasts were prepared from estradiol-induced CDC48-DNprotoplasts transiently expressing SP1-HA and FLAG-tagged ubiquitin. Cytosol was prepared from inducedCDC48-WTor CDC48-DN protoplasts (that were not expressing SP1-HA or FLAG-tagged ubiquitin).Reaction mixtures composed of chloroplasts and either cytosol fraction (as indicated above the gel images)were incubated for 1 hour and then refractionated into chloroplast (left) and cytosol (right) samples;any SP1-HA detected in the cytosol at the end of the reaction must have arisen from the chloroplasts.SP1-HA was enriched from 90% of the protein extract of each fraction by IP and detected byimmunoblotting with anti-HA (to detect unmodified SP1-HA) and anti-FLAG (to detect ubiquitinatedSP1-HA) (E), and the results were quantified (F). All values are means ± SEM (n = 3 to 4 experiments).

RESEARCH | RESEARCH ARTICLEon M

ay 21, 2020

http://science.sciencemag.org/

Dow

nloaded from

a knockout allele. In sp2-2 and sp2-3, the detectedmutations were both predicted to cause an aminoacid substitution. Further details are provided inthe fig. S1 legend. The transmembrane b-strandsand the three-dimensional structure of the SP2protein were predicted with Phyre2 by using theIntensivemodelingmode,whichproduced amodelbased on five structures for bacterial TamA andBamA proteins (c4c00a, c5ekqA, c4k3bA, c4n75A,and c4k3cA) to increase confidence (53).

Phylogenetic analysis

Sequences were obtained by BLAST searches ofthe Phytozome 12 database (54) (table S2). Se-quenceswere aligned bymultiple alignment usingfast Fourier transform (MAFFT) (55), and manualalignment adjustmentsweremade by usingMes-quite 1.12 (Tangient). Phylogeny was inferred byusing MrBayes 3.2 software (56). Two runs wereperformed in parallel, with each using eightMarkov chain Monte Carlo chains for 8 milliongenerations and the temperature set to 0.2. Thestandard deviation of split frequencies (StdDev)was 0.001228 at the end of the analysis, and it wastherefore assumed to have converged. Trees weresampled every 1000 generations, reaching a totalof 8000 trees. Burn-in was set to 25%, and so thefirst 2000 trees were discarded. The resultingphylogenywas aminimum 50% consensus of theremaining 6000 sampled trees. Parameters notmentioned were retained at the default setting.

Gene identifiers

Gene sequences for the following proteins fromA. thaliana were used experimentally in thisstudy: SP2 (At3g44160), AtCDC48A (At3g09840),Toc33 (At1g02280), CDKA1 (At3g48750), Toc159(At4g02510),OEP7 (At3g52420),OEP80(At5g19620),and Rubisco small subunit (SSU) (At1g67090).

Plasmid constructs

All primers used are listed in table S1. The SP1-HA, YFP-HA, FLAG-tagged ubiquitin, SP1-YFP,GST-SP1flex, and YFP-Toc33 constructs have allbeen described previously (7, 21). The codingsequence (CDS) for the CDC48-DN mutant(AtCDC48AK254A,K527A) was amplified from apre-existing plasmid (H6T7-DN-B, which has anethanol-inducible promoter) (27) by using pri-mers that add a C-terminal FLAG tag. All otherArabidopsisCDSs (including that forWTAtCDC48A,amplified with and without a FLAG tag) werePCR amplified from Col-0 cDNA, and the CDS en-coding theCDC48 “Trap”mutant (AtCDC48AE581Q)was generated by overlap-extension PCR (57).The Gateway cloning system (Invitrogen) wasused tomakemost of the constructs, and all entryclones were verified by DNA sequencing. To gen-erate C-terminal 6×Myc tag fusion proteins, theSP1 and SP2 CDSs were cloned into the pE3cvector (58) and then subcloned into the p2GW735S-driven expression vector (59) for protoplasttransfection (generating the SP1-Myc and SP2-Myca constructs). The SP2CDS,with andwithoutthe Myc tag, was cloned into the pB2GW7 binary35S-driven overexpression vector (59) for stableplant transformation (generating the SP2-OX and

SP2-Myc constructs). The C-terminally FLAG-tagged WT and DN mutant CDC48 CDSs werecloned into thepMDC7binary vector (60) for stableplant transformation, enabling estradiol-inducibletransgene expression [providing more stableexpression than the original ethanol-inducible sys-tem (27)] (generating the CDC48-WT and CDC48-DN constructs). UntaggedWT and Trap mutantCDC48 CDSs were cloned into both a modifiedp2GW7 plant expression vector providing aC-terminal hemagglutinin (HA) tag (generating theCDC48-WT-HA and CDC48-Trap-HA constructs)and the p2GWC7 (for WT) or p2GWY7 (for Trap)plant expression vector (59), which provides aC-terminal cyan fluorescent protein (CFP) or YFPtag, respectively (generating the CDC48-WT-CFPand CDC48-Trap-YFP constructs). To generateC-terminally HA-tagged Toc33, the correspond-ing CDS was similarly cloned into the modifiedp2GW7vector (generating theToc33-HA construct).To generate N-terminally TAP (tandem affinitypurification)–tagged Toc33, the correspondingCDSwas cloned into the NTAPi (N-terminal TAPtag) binary vector (61) (generating the TAP-Toc33construct). To generate BiFC constructs, selectedgene sequences were cloned into pBlueScript IISK− by using a single SmaI restriction site, se-quenced, and then subcloned into 5′ KpnI and 3′XmaI sites of the pSAT4A-cEYFP-N1 vector (62)for CDC48-cYFP; 5′XhoI and 3′ EcoRI sites of thepSAT4A-cEYFP-N1 vector (62) for CDKA1-cYFP;5′ EcoRI and 3′ SalI sites of the pSAT4-nEYFP-C1vector (62) for nYFP-Toc159; and 5′ EcoRI and 3′SalI sites of the pSAT4A-nEYFP-N1 vector (62) forOEP7-nYFP.

Transient assays and stableplant transformation

Protoplast isolation and transient assays werecarried out as described previously (7, 63). Whenrequired, MG132 (Sigma), epoxomicin (Merck),or bortezomib (Selleckchem) (all three chemicalsprepared as a 10 mM stock solution in dimethylsulfoxide) or E-64 (Melford) (prepared as a 10mMstock solution in water) was added to the proto-plast culture medium at 15 hours after transfec-tion, to a final concentration of 1 to 10, 1 to 10, 5,or 10 mM, respectively; subsequently, the cul-ture was incubated for a further 2 to 3 hoursbefore analysis. When protoplasts isolated fromthe CDC48-WT and CDC48-DN transgenic lineswere used, 10 mMestradiol (prepared as a 10mMstock solution in ethanol) was included in the cul-ture medium throughout the incubation of proto-plasts (either for 15 hours in the case of transfectedprotoplasts or for 2 days when transfection wasnot needed). For XFP fluorescence and IP assays,protoplast aliquots of 0.1 ml (105 protoplasts) or1 ml (106 protoplasts) were transfected with 5or 100 mg of DNA, respectively, and the sampleswere analyzed after 15 to 18 hours.Transgenic lines carrying the SP2-OX, SP2-

Myc, CDC48-WT, and CDC48-DN constructs weregenerated by Agrobacterium-mediated transfor-mation (16, 36). Transformants were selected byusing MS medium containing either phosphino-thricin (for the SP2 constructs) or hygromycin B

(for the CDC48 constructs). At least 12 T2 lines foreach transformationwere analyzed, and at least twolines with a single T-DNA insertion (which showeda 3:1 segregation on selective MS medium in theT2 generation) were chosen for further analysis.

Microscopy

Transmission electronmicroscopywas performedas described previously (16). Measurements wererecorded by using at least 30 different plastids pergenotype and were representative of three indi-viduals per genotype. Chloroplast cross-sectionalarea was estimated as described previously (16, 34)by using the equation p × 0.25 × length × width.Numbers of thylakoid lamellae per granal stackand of interconnections between granal stackswere determined as previously described (7, 16)in at least 96 resolvable grana across three indi-viduals per genotype.All fluorescence microscopy and BiFC experi-

ments were conducted at least twice with thesame results, and typical images are presented.For the imaging of CFP, YFP, and chlorophyllfluorescence signals, in most cases (except forfig. S14) protoplasts were examined by using aZeissLSM510META laser-scanning confocalmicro-scope (Carl Zeiss) as described previously (8). Tovisualize signals associatedwith chloroplastswith-out interference from cytosolic signals, protoplastswere ruptured by gently tapping the cover glass;this enabled the release of the cytosol and of intactchloroplasts. For fig. S14, fluorescence imageswere captured by using aNikonEclipseTE-2000Einvertedmicroscope as described previously (36).For BiFC assays, plasmid DNA for two con-

structs [onenYFP (N-terminal YFP fragment) fusionand one cYFP (C-terminal YFP fragment) fusion]was cotransfected into WT protoplasts and thenYFP signals were analyzed by confocal imaging.All images were captured by using the samesettings to enable comparisons. The frequencyof protoplasts that successfully expressed a BiFCYFP fluorescence signalwas determined by count-ing the number of positive cells and the totalnumber of cells per microscope field (each fieldtypically contained ~40 protoplasts) in approx-imately five microscope fields per transfection.Each combination of constructs was analyzedthree times to evaluate the frequency.

In vitro translation and in vitropull-down analysis

The SP2 and OEP80 CDSs were cloned intopBlueScript II SK− by using a single SmaI re-striction site and verified by DNA sequencing.The SSU precursor construct was described pre-viously, aswas the in vitro transcription-translationprocedure (16, 64).The purification of glutathione S-transferase

(GST)–SP1flex and GST proteins from bacteriaand the procedure used for in vitro pull-downanalysis were described previously (7).

Chloroplast isolation, protein import,and topology analysis

Chloroplastswere isolated from 14-day-old in vitro–grown plants (or, when stated, from protoplasts).

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 9 of 12

RESEARCH | RESEARCH ARTICLEon M

ay 21, 2020

http://science.sciencemag.org/

Dow

nloaded from

Isolations, protein import, and protease treat-ments were performed as described previously(16, 38, 65–68). The presented chloroplast pro-tein import data are representative of three inde-pendent experiments.

Immunoblotting, immunoprecipitation,and blue native PAGE

Immunoblotting was performed as previouslydescribed (34, 69) withminormodifications. Totalprotein samples of 10 to 20 mg, prepared fromseedlings, were typically analyzed. Primary anti-bodies were as follows. To identify TOC andTIC proteins, we used anti–atToc75-III antibody(36), anti-atToc159 antibody (70), anti-atToc33(G-domain) antibody (36), anti-atTic110 antibody(71, 72), and anti-atTic40 antibody (36). To iden-tify non-TOCOEMproteins, we used anti-OEP80antibody (73) and anti-SFR2 antibody (74). Toidentify chloroplast stromal proteins, we usedanti-cpHsc70 (AgriSera, AS08 348) (75), anti-Hsp93 (heat shock protein, 93 kDa) antibody(16, 76), and anti-PRPL35 antibody (7). To iden-tify proteins of other cellular compartments, weused anti-Slp1 (mitochondria) (77), anti-calreticulin(ER) (78, 79), anti-H3 histone (Abcam) (nucleus)(36), and anti-PEX13 and anti-PEX14 (the latterfrom Agrisera, AS08 372) (peroxisome) (20, 21).Other primary antibodies we used were anti–HAtag (Sigma), anti–c-Myc tag (Sigma), anti–greenfluorescent protein (GFP) (which detects bothGFPandYFP) (Sigma), and anti–FLAG tag (Sigma).As most of the proteins analyzed in this studywere membrane proteins, we used Slp1 (a non-chloroplastic membrane protein) and Tic110 [aninternal chloroplast membrane protein unaffectedby CHLORAD (7, 8)] as loading controls.Secondary antibodieswere anti–rabbit immuno-

globulin G (IgG) conjugated with horseradishperoxidase (Santa Cruz Biotechnology) or, in thecase of anti–c-Myc and anti-FLAG, anti–mouseIgG conjugated with horseradish peroxidase (GEHealthcare). Chemiluminescence was detected byusing ECL Plus Western blotting detection re-agents (GEHealthcare) and an LAS-4000 imager(Fujifilm). Band intensities were quantified byusing Aida software (Raytest). Quantificationdata were obtained from the results of at leastthree experiments all showing a similar trend.Typical images are shown in all figures.For the IP of HA-tagged proteins, total protein

(~500 mg) was extracted from protoplasts in IPbuffer (25 mM tris-HCl, pH 7.5; 150 mM NaCl;1 mM EDTA; 1% Triton X-100) containing 0.5%plant protease inhibitor cocktail (PPIC) (Sigma)and centrifuged at 20,000 × g for 10 min at 4°C.The clear lysate was then incubated with 50 mlof EZview Red anti-HA affinity gel (Sigma) for2 hours to overnight at 4°C with slow rotation.After six washes with 500 ml of IP washing buf-fer (25 mM tris-HCl, pH 7.5; 150 mMNaCl; 1 mMEDTA; 0.5% Triton X-100), bound proteins wereeluted by boiling in 2× SDS–polyacrylamide gelelectrophoresis (PAGE) loading buffer [50 mMtris-HCl, pH 6.8; 20% glycerol; 1% sodiumdodecylsulfate (SDS); and 0.1 M dithiothreitol (DTT)] for5 min and analyzed by SDS-PAGE and immuno-

blotting. A similar procedurewas adopted for theIP of Myc-tagged proteins, except that 50 ml ofEZview Red anti–c-Myc affinity gel (Sigma) wasused instead of the anti-HA gel. For detection ofubiquitinated proteins, the IP buffer also con-tained 10mMN-ethylmaleimide (NEM) (Sigma).Two-dimensional blue native (BN)–PAGE was

performed by using a procedure described pre-viously (80).

TAP and mass spectrometry

Chloroplasts were isolated from a complementedppi1 mutant line carrying the TAP:Toc33 con-struct and then used as startingmaterial for TAP.The TAP procedure was performed as describedpreviously (81), with the omission of the secondaryaffinity purification step, which was not essentialfor our analysis. The tobacco etch virus nuclear-inclusion-a endopeptidase eluates were concen-trated 1:10 by using Vivaspin 500 ultrafiltrationspin columns (Sartorius Stedim Biotech), boiledwith 1 volume of 2× SDS-PAGE loading buffer,and loaded onto SDS-PAGE gels for analysis. Sil-ver staining was used to visualize proteins andestimate their sizes and migration positions. Forthe identification of CDC48, the 75- to 100-kDaregion of a Coomassie brilliant blue–stained SDS-PAGE gel slice was subjected to in-gel trypsindigestion and liquid chromatography–tandemmass spectrometry analysis. Scaffold (ProteomeSoftware) and Mascot database searches wereused to interpret the results.

In vivo retrotranslocation assays

The method used for in vivo retrotranslocationassays was adapted and modified from similarapproaches commonly applied in ERAD studies(23). Transfected protoplasts were used for theseassays to facilitate the detection of the substrateand enable the efficient and uniform applicationof proteasome inhibitor; in addition, protoplas-tation applies a stress that triggers the TOC de-gradationprocess (8). First, the substrate (Toc33-HAor SP1-HA) and ubiquitin were transiently overex-pressed in 106 protoplasts for each genotype to in-crease the detection sensitivity for higher–molecularweight (ubiquitinated) forms of the substrate (82).The transformed protoplasts were incubated for15 hours, and then bortezomib was applied to afinal concentration of 5 mM before an additional3 hours of incubation. Subsequent fractionationsteps to produce separate chloroplast and cytosolsamples were all carried out on ice or at 4°C andused previously described procedures with mod-ifications (65, 83). Protoplasts were pelleted bycentrifugation at 100 × g for 2 min and gently re-suspendedwithprotoplast-washingbuffer (500mMmannitol; 4 mM 4-morpholineethanesulfonic acid–KOH, pH 5.6). Then the protoplasts were pelletedagain and resuspended by gentle agitation in 500 mlof HS buffer [50 mM 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid (HEPES)–NaOH, pH 8.0;0.3 M sorbitol] containing 0.5% PPIC and 5 mMbortezomib and gently forced twice through10-mm nylon mesh to release chloroplasts. Thecollected flow-throughwas centrifuged at 1000 × gfor 5 min to produce a chloroplast-containing

pellet and a cytosol-containing supernatant (S1).The pellet was gently resuspended in 500 ml ofHS buffer, and the chloroplasts were purified bya two-step Percoll (Fisher Scientific) gradient(38). Intact chloroplasts were washedwith 500 mlof HS buffer and then pelleted by centrifuga-tion at 1000 × g for 5 min. The S1 sample wascentrifuged at 10,000 × g for 15 min. The result-ing supernatant (S10) was recovered and ultra-centrifuged at 100,000 × g for 1 hour, producing afurther supernatant (S100) that was concentratedto 50 ml by using Vivaspin 500 ultrafiltration spincolumns; this was the cytosolic fraction. The pel-leted chloroplasts were lysed in 100 ml of denatur-ing buffer (25 mM tris-HCl, pH 7.5; 150mMNaCl;5 mM EDTA; 10 mMNEM; 1% SDS; 2% Sarcosyl;5 mMDTT) containing PPIC, whereas the cyto-solic fractionwasmixedwith 50 ml of 2× denatur-ing buffer containing PPIC. Lastly, the substrateprotein was purified by IP by using a previouslydescribed procedure to improve the sensitivity ofdetection of the ubiquitinatedprotein (7). Experi-ments were repeated three times, and similarresults were obtained.

In vitro retrotranslocation assays

Themethod for in vitro retrotranslocation assayswas adapted andmodified from previous reports(84, 85). Approximately 4 × 106 protoplasts fromCDC48-WT or CDC48-DN plants were incubatedwith 10 mM estradiol for 2 days to induce abun-dant expression of CDC48 protein in the cytosol.After incubation, the protoplasts were washedonce with 2 ml of protoplast-washing buffer, andthen a cytosolic fraction was prepared from eachgenotype, essentially as described above except thatthe protoplasts were resuspended in import buffer[50 mM HEPES-NaOH, pH 8.0; 3 mM MgSO4;0.3 M sorbitol; 5 mM Mg–adenosine triphosphate(ATP); 20mMgluconic acid (potassiumsalt); 10mMNaHCO3; 0.2% bovine serum albumin] contain-ing 5 mM bortezomib and complete EDTA-freeprotease inhibitor cocktail (PIC) (Roche) (the lat-ter according to the manufacturer’s instructions),instead of HS buffer. Cytosol was concentratedto ~10 mg/ml by using Vivaspin 500 ultrafiltra-tion spin columns and was used immediatelyor aliquoted and stored at −80°C. Chloroplastswere prepared from 106 CDC48-DN protoplaststransiently overexpressing SP1-HA and FLAG-taggedubiquitin (and inducedwith 10mMestradiol)by using the procedure described above. Chloro-plast pellets were gently resuspended in reactionbuffer [import buffer containing 30mMMgATP,ubiquitin (Sigma) at 0.1 mg/ml, 5 mM bortezomib,and PIC] and then immediately used in in vitroretrotranslocation reactions. Each reaction mix-ture contained reaction buffer (to a final reactionvolume of 50 ml), 10 ml of cytosol, and 2 × 107

chloroplasts. Reaction mixtures were incubatedat 25°C for 1 hour with occasional agitation toresuspend the chloroplasts, and reactions werethen stopped by the addition of 1 ml of 250 mMNEM to a final concentration of 5 mM. Subse-quently, at 4°C, the chloroplasts were pelleted bycentrifugation at 10,000 × g for 1 min, and theresulting supernatant was further centrifuged at

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 10 of 12

RESEARCH | RESEARCH ARTICLEon M

ay 21, 2020

http://science.sciencemag.org/

Dow

nloaded from

20,000 × g for 10 min to produce another super-natant corresponding to the cytosolic fraction.The SP1-HA protein was enriched from the result-ing chloroplast and cytosol fractions by IP, as de-scribed above. Ubiquitinated SP1 protein (eitherresident in the chloroplasts or retrotranslocatedinto the cytosol) was detected by anti-FLAG im-munoblotting; use of transiently expressed FLAG-tagged ubiquitin in this way increased the detectionsensitivity. Experimentswere repeated three times,and similar results were obtained.

Statistical analysis

Statistical calculations (mean, SEM, and t test)were performedbyusingMicrosoft Excel software.The statistical significance of differences be-tween two experimental groups was assessedby using a two-tailed Student’s t test. Differencesbetween two datasets were considered significantat P < 0.05.

REFERENCES AND NOTES

1. P. Jarvis, E. López-Juez, Biogenesis and homeostasis ofchloroplasts and other plastids. Nat. Rev. Mol. Cell Biol. 14,787–802 (2013). doi: 10.1038/nrm3702; pmid: 24263360

2. L. G. L. Richardson, R. Singhal, D. J. Schnell, The integration ofchloroplast protein targeting with plant developmental andstress responses. BMC Biol. 15, 118 (2017). doi: 10.1186/s12915-017-0458-3; pmid: 29216893

3. E. Demarsy, A. M. Lakshmanan, F. Kessler, Border control:Selectivity of chloroplast protein import and regulation at theTOC-complex. Front. Plant Sci. 5, 483 (2014). doi: 10.3389/fpls.2014.00483; pmid: 25278954

4. L. X. Shi, S. M. Theg, The chloroplast protein import system:From algae to trees. Biochim. Biophys. Acta 1833, 314–331(2013). doi: 10.1016/j.bbamcr.2012.10.002; pmid: 23063942

5. H. M. Li, C. C. Chiu, Protein transport into chloroplasts.Annu. Rev. Plant Biol. 61, 157–180 (2010). doi: 10.1146/annurev-arplant-042809-112222; pmid: 20192748

6. K. Nishimura, Y. Kato, W. Sakamoto, Essentials of proteolyticmachineries in chloroplasts. Mol. Plant 10, 4–19 (2017).doi: 10.1016/j.molp.2016.08.005; pmid: 27585878

7. Q. Ling, W. Huang, A. Baldwin, P. Jarvis, Chloroplast biogenesisis regulated by direct action of the ubiquitin-proteasomesystem. Science 338, 655–659 (2012). doi: 10.1126/science.1225053; pmid: 23118188

8. Q. Ling, P. Jarvis, Regulation of chloroplast protein importby the ubiquitin E3 ligase SP1 is important for stress tolerancein plants. Curr. Biol. 25, 2527–2534 (2015). doi: 10.1016/j.cub.2015.08.015; pmid: 26387714

9. X. Wu, T. A. Rapoport, Mechanistic insights into ER-associatedprotein degradation. Curr. Opin. Cell Biol. 53, 22–28 (2018).doi: 10.1016/j.ceb.2018.04.004; pmid: 29719269

10. D. Zattas, M. Hochstrasser, Ubiquitin-dependent proteindegradation at the yeast endoplasmic reticulum and nuclearenvelope. Crit. Rev. Biochem. Mol. Biol. 50, 1–17 (2015).doi: 10.3109/10409238.2014.959889; pmid: 25231236

11. R. J. Braun, B. Westermann, With the help of MOM:Mitochondrial contributions to cellular quality control. TrendsCell Biol. 27, 441–452 (2017). doi: 10.1016/j.tcb.2017.02.007;pmid: 28291566

12. P. Jarvis et al., An Arabidopsis mutant defective in the plastidgeneral protein import apparatus. Science 282, 100–103(1998). doi: 10.1126/science.282.5386.100; pmid: 9756470

13. T. Ulrich, D. Rapaport, Biogenesis of beta-barrel proteins inevolutionary context. Int. J. Med. Microbiol. 305, 259–264(2015). doi: 10.1016/j.ijmm.2014.12.009; pmid: 25596888

14. I. E. Gentle, L. Burri, T. Lithgow, Molecular architecture andfunction of the Omp85 family of proteins. Mol. Microbiol. 58,1216–1225 (2005). doi: 10.1111/j.1365-2958.2005.04906.x;pmid: 16313611

15. P. M. Day, D. Potter, K. Inoue, Evolution and targeting ofOmp85 homologs in the chloroplast outer envelope membrane.Front. Plant Sci. 5, 535 (2014). doi: 10.3389/fpls.2014.00535;pmid: 25352854

16. W. Huang, Q. Ling, J. Bédard, K. Lilley, P. Jarvis, In vivoanalyses of the roles of essential Omp85-related proteinsin the chloroplast outer envelope membrane. Plant Physiol.

157, 147–159 (2011). doi: 10.1104/pp.111.181891;pmid: 21757633

17. M. Töpel, Q. Ling, P. Jarvis, Neofunctionalization within theOmp85 protein superfamily during chloroplast evolution.Plant Signal. Behav. 7, 161–164 (2012). doi: 10.4161/psb.18677;pmid: 22307047

18. Y. C. Hsueh et al., Chloroplast outer envelope protein P39 inArabidopsis thaliana belongs to the Omp85 protein family.Proteins 85, 1391–1401 (2017). doi: 10.1002/prot.24725;pmid: 25401771

19. J. D. Woodson et al., Ubiquitin facilitates a quality-controlpathway that removes damaged chloroplasts. Science 350,450–454 (2015). doi: 10.1126/science.aac7444;pmid: 26494759

20. R. Pan, J. Satkovich, J. Hu, E3 ubiquitin ligase SP1 regulatesperoxisome biogenesis in Arabidopsis. Proc. Natl. Acad. Sci. U.S.A.113, E7307–E7316 (2016). doi: 10.1073/pnas.1613530113;pmid: 27799549

21. Q. Ling, N. Li, P. Jarvis, Chloroplast ubiquitin E3 ligase SP1:Does it really function in peroxisomes? Plant Physiol. 175,586–588 (2017). pmid: 28956781

22. Y. Liu, J. Li, Endoplasmic reticulum-mediated protein qualitycontrol in Arabidopsis. Front. Plant Sci. 5, 162 (2014).doi: 10.3389/fpls.2014.00162; pmid: 24817869

23. S. Neal et al., The Dfm1 Derlin is required for ERADretrotranslocation of integral membrane proteins. Mol. Cell 69,306–320.e4 (2018). doi: 10.1016/j.molcel.2017.12.012;pmid: 29351849

24. S. Schoebel et al., Cryo-EM structure of the protein-conductingERAD channel Hrd1 in complex with Hrd3. Nature 548,352–355 (2017). doi: 10.1038/nature23314; pmid: 28682307

25. K. Römisch, A case for Sec61 channel involvement in ERAD.Trends Biochem. Sci. 42, 171–179 (2017). doi: 10.1016/j.tibs.2016.10.005; pmid: 27932072

26. Y. Ye, W. K. Tang, T. Zhang, D. Xia, A mighty “proteinextractor” of the cell: Structure and function of the p97/CDC48 ATPase. Front. Mol. Biosci. 4, 39 (2017). doi: 10.3389/fmolb.2017.00039; pmid: 28660197

27. S. Park, D. M. Rancour, S. Y. Bednarek, In planta analysis ofthe cell cycle-dependent localization of AtCDC48A and itscritical roles in cell division, expansion, and differentiation.Plant Physiol. 148, 246–258 (2008). doi: 10.1104/pp.108.121897; pmid: 18660433

28. E. Lécuyer et al., Protein stability and transcription factorcomplex assembly determined by the SCL-LMO2 interaction.J. Biol. Chem. 282, 33649–33658 (2007). doi: 10.1074/jbc.M703939200; pmid: 17878155

29. P. R. Johnson, R. Swanson, L. Rakhilina, M. Hochstrasser,Degradation signal masking by heterodimerization ofMATa2 and MATa1 blocks their mutual destruction by theubiquitin-proteasome pathway. Cell 94, 217–227 (1998).doi: 10.1016/S0092-8674(00)81421-X; pmid: 9695950

30. S. Xu, G. Peng, Y. Wang, S. Fang, M. Karbowski, TheAAA-ATPase p97 is essential for outer mitochondrialmembrane protein turnover. Mol. Biol. Cell 22, 291–300 (2011).doi: 10.1091/mbc.e10-09-0748; pmid: 21118995

31. K. Nakatsukasa, G. Huyer, S. Michaelis, J. L. Brodsky,Dissecting the ER-associated degradation of a misfoldedpolytopic membrane protein. Cell 132, 101–112 (2008).doi: 10.1016/j.cell.2007.11.023; pmid: 18191224

32. J. Stevenson, E. Y. Huang, J. A. Olzmann, Endoplasmicreticulum-associated degradation and lipid homeostasis.Annu. Rev. Nutr. 36, 511–542 (2016). doi: 10.1146/annurev-nutr-071715-051030; pmid: 27296502

33. M. G. Iadanza et al., Lateral opening in the intact b-barrelassembly machinery captured by cryo-EM. Nat. Commun. 7,12865 (2016). doi: 10.1038/ncomms12865; pmid: 27686148

34. S. Kovacheva et al., In vivo studies on the roles of Tic110, Tic40and Hsp93 during chloroplast protein import. Plant J. 41,412–428 (2005). doi: 10.1111/j.1365-313X.2004.02307.x;pmid: 15659100

35. J. P. Stanga, K. Boonsirichai, J. C. Sedbrook, M. S. Otegui,P. H. Masson, A role for the TOC complex in Arabidopsis rootgravitropism. Plant Physiol. 149, 1896–1905 (2009).doi: 10.1104/pp.109.135301; pmid: 19211693

36. A. R. Kasmati, M. Töpel, R. Patel, G. Murtaza, P. Jarvis,Molecular and genetic analyses of Tic20 homologues inArabidopsis thaliana chloroplasts. Plant J. 66, 877–889 (2011).doi: 10.1111/j.1365-313X.2011.04551.x; pmid: 21395885

37. Y. C. Hsueh et al., The outer membrane Omp85-like proteinP39 influences metabolic homeostasis in mature Arabidopsisthaliana. Plant Biol. (Stuttgart) 20, 825–833 (2018).doi: 10.1111/plb.12839; pmid: 29758131

38. H. Aronsson, P. Jarvis, A simple method for isolatingimport-competent Arabidopsis chloroplasts. FEBS Lett. 529,215–220 (2002). doi: 10.1016/S0014-5793(02)03342-2;pmid: 12372603

39. Q. Ling, W. Huang, P. Jarvis, Use of a SPAD-502 meter tomeasure leaf chlorophyll concentration in Arabidopsis thaliana.Photosynth. Res. 107, 209–214 (2011). doi: 10.1007/s11120-010-9606-0; pmid: 21188527

40. D. Constan, R. Patel, K. Keegstra, P. Jarvis, An outer envelopemembrane component of the plastid protein import apparatusplays an essential role in Arabidopsis. Plant J. 38, 93–106(2004). doi: 10.1111/j.1365-313X.2004.02024.x; pmid: 15053763

41. S. Schelbert et al., Pheophytin pheophorbide hydrolase(pheophytinase) is involved in chlorophyll breakdown duringleaf senescence in Arabidopsis. Plant Cell 21, 767–785 (2009).doi: 10.1105/tpc.108.064089; pmid: 19304936

42. G. Gálvez-Valdivieso et al., The high light response inArabidopsis involves ABA signaling between vascular andbundle sheath cells. Plant Cell 21, 2143–2162 (2009).doi: 10.1105/tpc.108.061507; pmid: 19638476

43. A. Daudi et al., The apoplastic oxidative burst peroxidase inArabidopsis is a major component of pattern-triggeredimmunity. Plant Cell 24, 275–287 (2012). doi: 10.1105/tpc.111.093039; pmid: 22247251

44. O. H. M. Leyser, I. J. Furner, Characterisation of three shootapical meristem mutants of Arabidopsis thaliana. Development116, 397–403 (1992).

45. FASTX-Toolkit, http://hannonlab.cshl.edu/fastx_toolkit/.46. M. Martin, Cutadapt removes adapter sequences from

high-throughput sequencing reads. EMBnet J. 17, 10–12 (2011).doi: 10.14806/ej.17.1.200

47. pairSeq.py, GitHub; https://github.com/topel-research-group/pairSeq.

48. TAIR10 Arabidopsis thaliana reference genome; ftp://ftp.jgi-psf.org/pub/compgen/phytozome/v9.0/Athaliana/assembly/Athaliana_167.fa.gz.

49. Joint Genome Institute, Phythozome; www.phytozome.net/.50. Qiagen, www.qiagenbioinformatics.com/.51. find_sp2.py, GitHub; https://github.com/topel-research-

group/sp2.52. J. T. Robinson et al., Integrative genomics viewer.

Nat. Biotechnol. 29, 24–26 (2011). doi: 10.1038/nbt.1754;pmid: 21221095

53. L. A. Kelley, S. Mezulis, C. M. Yates, M. N. Wass,M. J. Sternberg, The Phyre2 web portal for protein modeling,prediction and analysis. Nat. Protoc. 10, 845–858 (2015).doi: 10.1038/nprot.2015.053; pmid: 25950237

54. D. M. Goodstein et al., Phytozome: A comparative platform forgreen plant genomics. Nucleic Acids Res. 40 (D1),D1178–D1186 (2012). doi: 10.1093/nar/gkr944;pmid: 22110026

55. K. Katoh, D. M. Standley, MAFFT multiple sequence alignmentsoftware version 7: Improvements in performance andusability. Mol. Biol. Evol. 30, 772–780 (2013). doi: 10.1093/molbev/mst010; pmid: 23329690

56. F. Ronquist et al., MrBayes 3.2: Efficient Bayesian phylogeneticinference and model choice across a large model space.Syst. Biol. 61, 539–542 (2012). doi: 10.1093/sysbio/sys029;pmid: 22357727

57. R. Higuchi, B. Krummel, R. K. Saiki, A general method of invitro preparation and specific mutagenesis of DNA fragments:Study of protein and DNA interactions. Nucleic Acids Res. 16,7351–7367 (1988). doi: 10.1093/nar/16.15.7351;pmid: 3045756

58. M. J. Dubin, C. Bowler, G. Benvenuto, A modified Gatewaycloning strategy for overexpressing tagged proteins in plants.Plant Methods 4, 3 (2008). doi: 10.1186/1746-4811-4-3;pmid: 18211686

59. M. Karimi, B. De Meyer, P. Hilson, Modular cloning in plantcells. Trends Plant Sci. 10, 103–105 (2005). doi: 10.1016/j.tplants.2005.01.008; pmid: 15749466

60. M. D. Curtis, U. Grossniklaus, A Gateway cloning vector set forhigh-throughput functional analysis of genes in planta. PlantPhysiol. 133, 462–469 (2003). doi: 10.1104/pp.103.027979;pmid: 14555774

61. J. S. Rohila, M. Chen, R. Cerny, M. E. Fromm, Improved tandemaffinity purification tag and methods for isolation of proteinheterocomplexes from plants. Plant J. 38, 172–181 (2004).doi: 10.1111/j.1365-313X.2004.02031.x; pmid: 15053770

62. T. Tzfira et al., pSAT vectors: A modular series of plasmids forautofluorescent protein tagging and expression of multiplegenes in plants. Plant Mol. Biol. 57, 503–516 (2005).doi: 10.1007/s11103-005-0340-5; pmid: 15821977

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 11 of 12

RESEARCH | RESEARCH ARTICLEon M

ay 21, 2020

http://science.sciencemag.org/

Dow

nloaded from

63. F. H. Wu et al., Tape-Arabidopsis Sandwich—a simplerArabidopsis protoplast isolation method. Plant Methods 5, 16(2009). doi: 10.1186/1746-4811-5-16; pmid: 19930690

64. A. Baldwin et al., A molecular-genetic study of the ArabidopsisToc75 gene family. Plant Physiol. 138, 715–733 (2005).doi: 10.1104/pp.105.063289; pmid: 15908591

65. L. M. Fitzpatrick, K. Keegstra, A method for isolating a highyield of Arabidopsis chloroplasts capable of efficient import ofprecursor proteins. Plant J. 27, 59–65 (2001). doi: 10.1046/j.0960-7412.2001.01061.x; pmid: 11489183

66. H. Aronsson, R. P. Jarvis, Rapid isolation of Arabidopsischloroplasts and their use for in vitro protein importassays. Methods Mol. Biol. 774, 281–305 (2011). doi: 10.1007/978-1-61779-234-2_17; pmid: 21822845

67. J. Froehlich, Studying Arabidopsis envelope protein localizationand topology using thermolysin and trypsin proteases.Methods Mol. Biol. 774, 351–367 (2011). doi: 10.1007/978-1-61779-234-2_21; pmid: 21822849

68. C. C. Chu, H. M. Li, Determining the location of an Arabidopsischloroplast protein using in vitro import followed byfractionation and alkaline extraction. Methods Mol. Biol. 774,339–350 (2011). doi: 10.1007/978-1-61779-234-2_20;pmid: 21822848

69. S. Kovacheva, J. Bédard, A. Wardle, R. Patel, P. Jarvis, Furtherin vivo studies on the role of the molecular chaperone, Hsp93,in plastid protein import. Plant J. 50, 364–379 (2007).doi: 10.1111/j.1365-313X.2007.03060.x; pmid: 17376159

70. J. Bauer et al., The major protein import receptor of plastids isessential for chloroplast biogenesis. Nature 403, 203–207(2000). doi: 10.1038/35003214; pmid: 10646606

71. T. Inaba et al., Arabidopsis Tic110 is essential for the assemblyand function of the protein import machinery of plastids.Plant Cell 17, 1482–1496 (2005). doi: 10.1105/tpc.105.030700;pmid: 15829604

72. H. Aronsson et al., Nucleotide binding and dimerization at thechloroplast pre-protein import receptor, atToc33, are notessential in vivo but do increase import efficiency. Plant J. 63,297–311 (2010). doi: 10.1111/j.1365-313X.2010.04242.x;pmid: 20444229

73. S. C. Hsu, M. Nafati, K. Inoue, OEP80, an essential proteinparalogous to the chloroplast protein translocation channelToc75, exists as a 70-kD protein in the Arabidopsis thalianachloroplast outer envelope. Plant Mol. Biol. 78, 147–158 (2012).doi: 10.1007/s11103-011-9853-2; pmid: 22094888

74. N. Fourrier et al., A role for SENSITIVE TO FREEZING2 inprotecting chloroplasts against freeze-induced damage inArabidopsis. Plant J. 55, 734–745 (2008). doi: 10.1111/j.1365-313X.2008.03549.x; pmid: 18466306

75. P. H. Su, H. M. Li, Stromal Hsp70 is important for proteintranslocation into pea and Arabidopsis chloroplasts. Plant Cell

22, 1516–1531 (2010). doi: 10.1105/tpc.109.071415;pmid: 20484004

76. E. Nielsen, M. Akita, J. Davila-Aponte, K. Keegstra, Stableassociation of chloroplastic precursors with proteintranslocation complexes that contain proteins from bothenvelope membranes and a stromal Hsp100 molecularchaperone. EMBO J. 16, 935–946 (1997). doi: 10.1093/emboj/16.5.935; pmid: 9118955

77. B. Gehl, C. P. Lee, P. Bota, M. R. Blatt, L. J. Sweetlove, AnArabidopsis stomatin-like protein affects mitochondrialrespiratory supercomplex organization. Plant Physiol. 164,1389–1400 (2014). doi: 10.1104/pp.113.230383;pmid: 24424325

78. J. Denecke et al., The tobacco homolog of mammaliancalreticulin is present in protein complexes in vivo. Plant Cell 7,391–406 (1995). doi: 10.1105/tpc.7.4.391; pmid: 7773014

79. O. K. Teh, I. Moore, An ARF-GEF acting at the Golgi and inselective endocytosis in polarized plant cells. Nature 448,493–496 (2007). doi: 10.1038/nature06023; pmid: 17653191

80. S. Kikuchi, J. Bédard, M. Nakai, One- and two-dimensional bluenative-PAGE and immunodetection of low-abundancechloroplast membrane protein complexes. Methods Mol. Biol.775, 3–17 (2011). doi: 10.1007/978-1-61779-237-3_1;pmid: 21863435

81. C. Andrès, B. Agne, F. Kessler, Preparation of multiproteincomplexes from Arabidopsis chloroplasts using tandem affinitypurification. Methods Mol. Biol. 775, 31–49 (2011).doi: 10.1007/978-1-61779-237-3_3; pmid: 21863437

82. J. Bloom, M. Pagano, Experimental tests to definitivelydetermine ubiquitylation of a substrate. Methods Enzymol. 399,249–266 (2005). doi: 10.1016/S0076-6879(05)99017-4;pmid: 16338361

83. J. Ito et al., Analysis of the Arabidopsis cytosolic proteomehighlights subcellular partitioning of central plant metabolism.J. Proteome Res. 10, 1571–1582 (2011). doi: 10.1021/pr1009433; pmid: 21166475

84. R. M. Garza, B. K. Sato, R. Y. Hampton, In vitro analysis ofHrd1p-mediated retrotranslocation of its multispanningmembrane substrate 3-hydroxy-3-methylglutaryl (HMG)-CoAreductase. J. Biol. Chem. 284, 14710–14722 (2009).doi: 10.1074/jbc.M809607200; pmid: 19324879

85. K. Nakatsukasa, J. L. Brodsky, in vitro reconstitution of theselection, ubiquitination, and membrane extraction of apolytopic ERAD substrate. Methods Mol. Biol. 619, 365–376(2010). doi: 10.1007/978-1-60327-412-8_21; pmid: 20419421

ACKNOWLEDGMENTS

We thank M. Rashbrooke for assistance with initial analyses andrough mapping of sp2; Y. Zeng, J. Bédard, and N. Li for technical

assistance; N. Allcock and S. Hyman (University of Leicester EMLaboratory) for electron microscopy; A. R. Bottrill (University ofLeicester Protein Nucleic Acid Chemistry Laboratory) for massspectrometry; L. Dolan for comments on the manuscript; L. J.Sweetlove (Slp1), J. Denecke (calreticulin), and B. Bartel (PEX13)for antibodies; F. Wu for the Toc159 BiFC construct; S. Y. Bednarekfor the H6T7-DN-B construct; G. Benvenuto (and Addgene) for thepE3c construct; J. Chory and J. Woodson for ppi2-3 (fts1) seeds;and SIGnAL and NASC for the sp2-4 allele. Funding: This workwas supported by grants from the BBSRC (BB/D016541/1, BB/K018442/1, BB/R009333/1, and BB/R016984/1) to R.P.J.; theRoyal Society Rosenheim Research Fellowship to R.P.J.; aDepartment of Plant Sciences DPhil studentship to R.P.J and W.B.;a Gatsby Sainsbury Ph.D. studentship to R.T.; and a Carl TryggersStiftelse för Vetenskaplig Forskning fellowship (CTS 11:479) toM.T. Author contributions: Q.L. and W.B. formulated the researchplan, designed and performed experiments, and interpretedresults. Q.L. carried out the functional analysis of CDC48 andcontributed to the preparation of the manuscript. W.B. completedgenetic, molecular, physiological, and detailed functional studies ofSP2, with assistance from Q.L. R.T. performed the genetic mappingof sp2, prepared samples for whole-genome sequencing, andconducted protein import assays with assistance from T.D.S.M.T. analyzed the whole-genome sequencing data and identified acandidate sp2 locus. P.L. performed the TAP experiment. A.B.conducted the mutant screen and identified the original sp2 alleles.R.P.J. conceived of the study, supervised the work, analyzed thedata, and prepared the manuscript. All authors discussed theresults and commented on the manuscript. Competing interests:This work is the subject of pending UK patent applicationsGB 1803833.1, GB 1803834.9, and GB 1815206.6 (the inventoris R.P.J. in each case), which cover the use of CHLORAD tomanipulate plastid development in crops. Data and materialsavailability: Sequencing data used in this study, as well as thereference genome for ppi1, are available at the National Center forBiotechnology Information under Bioproject PRJNA488548. Themass spectrometry proteomics data have been deposited tothe ProteomeXchange Consortium via the PRoteomics IDEntifications(PRIDE) repository with the identifier PXD010954. All other data areavailable in the manuscript or the supplementary materials.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/363/6429/eaav4467/suppl/DC1Figs. S1 to S20Tables S1 and S2References (86–92)

17 September 2018; accepted 15 January 201910.1126/science.aav4467

Ling et al., Science 363, eaav4467 (2019) 22 February 2019 12 of 12

RESEARCH | RESEARCH ARTICLEon M

ay 21, 2020

http://science.sciencemag.org/

Dow

nloaded from

Ubiquitin-dependent chloroplast-associated protein degradation in plants

Paul JarvisQihua Ling, William Broad, Raphael Trösch, Mats Töpel, Tijen Demiral Sert, Panagiotis Lymperopoulos, Amy Baldwin and R.

DOI: 10.1126/science.aav4467 (6429), eaav4467.363Science 

, this issue p. eaav4467Scienceubiquitination of the targets by the chloroplast-localized E3 ubiquitin ligase SP1.protein processing by the cytosolic proteasome. Such chloroplast-associated protein degradation was initiated bymotor functions in the retrotranslocation of target proteins from chloroplasts. This process thus enabled outer membrane

chaperone that fulfill conductance and+-barrel outer membrane channel and a cytosolic AAAβidentified an Omp85-type degradation of proteins in the outer membrane of chloroplasts (the organelles responsible for photosynthesis). They

identified mechanisms underlying the UPS-mediatedet al., Ling Arabidopsisdegradation. Working in the model plant network in the eukaryotic cytosol. In such cases, the organelle membrane presents a substantial barrier to protein

a major proteolytic−−organelles. Some organellar proteins are targeted by the ubiquitin-proteasome system (UPS) Protein degradation is vital for cellular functions, and it operates selectively with distinct mechanisms in different

Chloroplast-associated protein degradation

ARTICLE TOOLS http://science.sciencemag.org/content/363/6429/eaav4467

MATERIALSSUPPLEMENTARY http://science.sciencemag.org/content/suppl/2019/02/20/363.6429.eaav4467.DC1

REFERENCES

http://science.sciencemag.org/content/363/6429/eaav4467#BIBLThis article cites 86 articles, 24 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceScience, 1200 New York Avenue NW, Washington, DC 20005. The title (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

Science. No claim to original U.S. Government WorksCopyright © 2019 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on May 21, 2020

http://science.sciencem

ag.org/D

ownloaded from