Caenorhabditis elegans screen reveals role of PAR-5 in RAB-11 … · Caenorhabditis elegans screen...

ART I C L E S

Caenorhabditis elegans screen reveals role of PAR-5 inRAB-11-recycling endosome positioning andapicobasal cell polarityJulia Franziska Winter1,3, Sebastian Höpfner1,3,4, Kerstin Korn2,4, Benjamin O. Farnung1,4, Charles R. Bradshaw1,4,Giovanni Marsico1, Michael Volkmer1,4, Bianca Habermann1,4 and Marino Zerial1,5

Apically enriched Rab11-positive recycling endosomes (Rab11-REs) are important for establishing and maintaining epithelialpolarity. Yet, little is known about the molecules controlling trafficking of Rab11-REs in an epithelium in vivo. Here, we report agenome-wide, image-based RNA interference screen for regulators of Rab11-RE positioning and transport of an apical membraneprotein (PEPT-1) in C. elegans intestine. Among the 356 screen hits was the 14-3-3 and partitioning defective protein PAR-5,which we found to be specifically required for Rab11-RE positioning and apicobasal polarity maintenance. Depletion of PAR-5induced abnormal clustering of Rab11-REs to ectopic sites at the basolateral cortex containing F-actin and other apical domaincomponents. This phenotype required key regulators of F-actin dynamics and polarity, such as Rho GTPases (RHO-1 and the Rac1orthologue CED-10) and apical PAR proteins. Our data suggest that PAR-5 acts as a regulatory hub for a polarity-maintainingnetwork required for apicobasal asymmetry of F-actin and proper Rab11-RE positioning.

The polarity of simple columnar epithelial cells, such as those liningthe luminal surface of the intestine, is characterized by three maininter-dependent hallmarks: (1) apicobasally organized cytoskeletalcomponents1 and polarity regulators2; (2) two compositionally andfunctionally distinct apical and basolateral plasma membrane surfacesseparated by intercellular junctions3; (3) a sophisticated network ofmembrane organelles, some of which are asymmetrically distributed,such as apical Rab11-positive recycling endosomes (Rab11-REs; ref. 4).The epithelial membrane trafficking system is both complex androbust owing to multiple sorting mechanisms and parallel transportroutes to ensure the delivery of cargo to its correct destination5,6.Elucidating the regulatory networks that couple the organizationof the membrane traffic system to establishment and maintenanceof epithelial polarity remains an ongoing quest. In this respect,particular attention has been given to a set of evolutionarily conservedmaster regulators including PAR proteins and small GTPases of theRho and Rab families7.The PAR family comprises seven functionally diverse proteins

originally described in the roundworm Caenorhabditis elegans8. Theapical PAR proteins PAR3, PAR6, aPKC and PAR1 play specific roles in

1Max Planck Institute of Molecular Cell Biology and Genetics, MPI-CBG, Pfotenhauerstrasse 108, 01307 Dresden, Germany. 2High-Throughput TechnologyDevelopment Studio, MPI-CBG, 01307 Dresden, Germany. 3These authors contributed equally to this work. 4Present addresses: Bird & Bird LLP, Pacellistrasse 14,80333 Munich, Germany (S.H.); Cenix BioScience GmbH, Tatzberg 47, 01307 Dresden, Germany (K.K.); ETH Zürich, Institut für Biochemie, HPM E 17.1,Schafmattstrasse 18, 8093 Zürich, Switzerland (B.O.F.); Wellcome Trust and Cancer Research UK Gurdon Institute, University of Cambridge, Cambridge CB2 1QN, UK(C.R.B.); Max Planck Institute for Biology of Aging, Robert-Koch-Strasse 21, 50931 Cologne, Germany (B.H.).5Correspondence should be addressed to M.Z. (e-mail: [email protected])

Received 6 March 2012; accepted 19 April 2012; published online 27 May 2012; DOI: 10.1038/ncb2508

epithelial polarity, such as cytoskeletal organization9,10, apical junctionformation11 and maintenance12, and basal membrane positioning13.Rho GTPases (ref. 14) such as RhoA, Cdc42 and Rac1 co-ordinatelyregulate actin polymerization, actomyosin contractility or both throughboth distinct and shared effector proteins2. PAR proteins are botheffectors and regulators of Rho GTPases, modulating their activity andcrosstalk, as well as the localization of other cytoskeletal regulators andcomponents2. Rab GTPases are key determinants of functional identityof membrane organelles15,16.Among the membrane organelles, Rab11-REs play a prominent role

both in establishing and maintaining epithelial polarity17,18. Rab11-REsmediate recycling and sorting of both apically19,20 and basolaterally21

endocytosed membrane components, and can function as secretorytransport intermediates between the Golgi complex and plasmamembrane22. Notably, Rab11-REs enrich apically in polarized epitheliaacross metazoans4,23, including the intestine of C. elegans (ref. 24, thisstudy). The C. elegans intestine has become an attractive organ modelsystem for studying epithelial polarity11,25, trafficking26 and cytoskeletalregulation27,28. It comprises 20 enterocytes that line up bilaterally withtwo or four cells per segment29.

666 NATURE CELL BIOLOGY VOLUME 14 | NUMBER 7 | JULY 2012

© 2012 M acmillan Publishers Limited. A ll rights reserved.

mailto:[email protected]

ART I C L E S

In this study, we aimed to gain insights into the coordination betweenepithelial membrane traffic, cytoskeleton and polarity machineriesin vivo. In particular, we were interested in identifying regulators ofRab11-REs and of the trafficking routes sustaining apical membranetransport. To achieve this in a systematic and unbiased manner,we carried out a genome-wide RNA interference (RNAi) screen inthe C. elegans intestine.

RESULTSPrimary RNAi high-content screenTo establish a high-content, image-based assay, we generated adouble transgenic screening strain (MZE1) that enabled monitoringof the intensity, subcellular position and morphology of threeintestine-specific fluorescent markers (Fig. 1a): (1) green fluorescentprotein (GFP) fusion to the Rab11a orthologue, RAB-11.1 (henceforthRAB-11), which labels apically-enriched Rab11-REs, (2) DsRed fusionto the apical membrane protein PEPT-1, a conserved H(+)-coupledoligopeptide transporter30, and (3) autofluorescent lysosome relatedorganelles (LROs; ref. 26), as indicators of general versus recycling-endosome-specific effects on organelle positioning and integrity. RNAi-treatedwormswere imaged by automated epifluorescencemicroscopy.In pilot experiments, we verified that the distribution of PEPT-

1::DsRed and GFP::RAB-11 change in a predictable fashion on RNAi ofestablished transport regulators. We confirmed that PEPT-1 is suitablefor reporting perturbations of apical membrane transport, and thatGFP::RAB-11 recycling endosomes in C. elegans enterocytes shareproperties with Rab11-REs studied in mammalian and Drosophilaepithelia23,31, as they are distinct from Golgi and early endosomes(Supplementary Fig. S1g–l), traversed by apical membrane proteins(Fig. 1a and Supplementary Figs S1c,d, S5a,d), and their apicalenrichment31 changes on depletion of RAB-11 effectors such asSEC-15 (ref. 23; Supplementary Fig. S1e,f). For the genome-wideRNAi analysis, we developed a semi-automated workflow (Fig. 1b)that uses the Ahringer RNAi feeding library32, whose 16,757 bacterialclones cover 87% of the C. elegans protein-coding genome. Eachgene was analysed in at least two independent runs. The primaryRNAi high-content screen (HCS) was split into two steps (primaryscreen methods are described in Supplementary Text S1). First, in theemb screen, we screened 1,037 embryonic lethal (emb) genes (!65% ofgenes essential for embryonic development and viability; WormBaserelease WS172) in the young-adult stage of the first generation. Wevalidated the emb-screen images manually for altered phenotypes inmarker distributions and intensities. Second, in the genome-widescreen (GWS), we screened the complete library in the young-adultstage of the second generation owing to increased RNAi phenotypefrequency (percentage of worms with phenotypes) and increasedstrength (phenotype intensity in individual worms). To efficientlyand objectively score RNAi phenotypes in the !2.3 million imagesof the GWS, we developed an image analysis software and trainedit on the emb screen images. This custom-built software is basedon a noise-tolerant medial axis transform33, enabling the analysis ofmultiple biometric parameters, for example worm shape, size (length,width, area), marker intensity, subcellular distributions and organellemorphology (Fig. 1d–g and Supplementary Fig. S2c).In total, 356 candidate genes (emb screen, 209; GWS, 147;

Supplementary Table S1) of sequence-authenticated feeding clones

were reproducibly identified. Overall, 88% of the 356 genes haveorthologues in higher metazoans (82% in human, 3% in fly and 3%in mouse), underscoring the conservation of the epithelial membranetrafficking and polarity machineries.

Phenotypic profiles for functional inferenceTo classify alterations in intensity, position and morphology of thethree fluorescent markers, we defined 14 phenotypic defect categories(Fig. 2a and Supplementary Fig. S2a and Table S1). To each gene weassigned a phenotype profile consisting of scores between 0 (wild type)and 3 (strong phenotypic defect) for each category. Genes sharingsimilar functions, such as those encoding subunits of multiproteincomplexes (for example endosomal sorting complex required fortransport (ESCRT), vacuolar H+-ATPases (V-H+-ATPases)), yieldedsimilar phenotype profiles (Supplementary Table S2).In the hit set, ‘membrane trafficking’ genes constituted the most

abundant group (Fig. 2b and Supplementary Fig. S2b and Table S1)and ‘intracellular protein trafficking’ was the most enriched term(Protein Analysis Through Evolutionary Relationships (PANTHER)ontology34, Fig. 2c), confirming that the screen indeed identified thegenes it was designed for.We next carried out hierarchical clustering35 on all 356 phenotypic

profiles (Supplementary Fig. S3e), capitalizing on the guilt-by-association principle36. Within clusters, functionally known genes canbe used to infer functional predictions for as-yet uncharacterized genesand genes previously not linked to epithelial polarity or membranetrafficking (illustrated with the compact clustering dendrogram inSupplementary Fig. S3a–d). Examples of functionally known andunknown candidates where we assigned functional roles are presentedbelow and in Supplementary Table S3.

Comparative secondary assay analysisFor further comparative and functional analysis in secondary assays,we selected 60 genes mainly with phenotypic defects in GFP::RAB-11positioning, PEPT-1::DsRed trafficking, or both, with known orunknown functions (Supplementary Table S3). The RNAi phenotypeof each candidate was examined in transgenic worm strains expressingthe apical membrane cargo PEPT-1::DsRed in combination withGFP-tagged markers for specific organelles, such as GFP::RAB-5 (earlyendosomes) or LMP-1::GFP (late endosomes) (Fig. 3b–f).Among the 60 genes, 38 served as ‘reference candidates’, whose

functions in membrane trafficking were mainly inferred from theirorthologues, and on the basis of these we defined six phenotypically andfunctionally distinct ‘secondary assay’ groups (Supplementary TableS3 and Fig. 3a). For several secretory regulators and for componentsof the dynein and dynactin, ESCRT I and III and exocyst complexeswe provided evidence for their role in membrane trafficking, organellepositioning or both in C. elegans enterocytes. The remaining 22 ‘testcandidates’ comprised both functionally known and unknown genesand were assigned to the six secondary assay groups (SupplementaryTable S3 and Fig. 3a). This confirmed numerous functional predictionsinferred from the hierarchical clustering (Supplementary Fig. S3a,c). Togain mechanistic insights into RAB-11-RE positioning and function,we focused on candidates with either a general effect on endosomepositioning or a more specific effect on RAB-11-REs (SupplementaryTable S3, groups 5 and 6).

NATURE CELL BIOLOGY VOLUME 14 | NUMBER 7 | JULY 2012 667


ART I C L E S

Anterior Posterior

Pharynx Intestine

Apical

Basal

Grow feeding bacteria overnight

Bacteria seeding, induction of dsRNA synthesis

Addition of L1-stage worms per well (six-well plates, solid culture)

Hit rescreening, clone sequencing, and hit list assembly with manually

confirmed phenotype profiles

Automated epifluorescence microscopy, 96-well plate, ArrayScan VTI HCS reader

~4 worms COPAS BIOSORT~200 worms

RNAi until young-adult stage of worms in

first generation

(P0)

second generation

(F1)

a c

Image analysis software

Manual phenotype scoring

PEPT-1::DsRedGFP::RAB-11 LROs

F07F

6.4

RN

Ai

F07F

6.4

RN

Ai

d

f

b

e

g

dhc-1 RNAiControl RNAi dhc-1 RNAiControl RNAi

GWS: ~16,750 clones

emb screen: ~1000 clones

Figure 1 In vivo readout assay, semi-automated workflow and image analysisof the primary high-content screen. (a) Schematic depiction of the doubletransgenic nematode strain MZE1, and confocal and differential interferencecontrast microscopy images of the anterior intestinal region of an L4-stageanimal. (b) Enlarged images corresponding to the boxed area in a showingthe three markers. Small GFP::RAB-11-labelled recycling endosomes(arrow) align along the PEPT-1::DsRed-labelled apical membrane. Largerrecycling endosomes (LROs; arrowhead) that can contain small amountsof PEPT-1::DsRed (arrowhead, brightness of enlarged image increased)concentrate subapically. Autofluorescent LROs in blue. (c) Scheme for

the semi-automated workflow of the primary screen. Black arrows indicaterobotic steps, grey arrows manual steps. (d–g) Morphometric parametersdetermined by the custom-built image analysis include identification ofthe worm midline (d,e, cyan), intracellular punctate intensity maxima ofPEPT-1::DsRed (d,e, magenta circles, RNAi of the C. elegans ARFGAP3orthologue) and detection of subcellular recycling endosome positioning(f,g). (g) Different intensity profiles of GFP::RAB-11 along apicobasal andmedial axes of the intestines shown in f with REs enriched in the apicaldomain (yellow) in control RNAi (left) or in the basal domain (grey) on dhc-1RNAi (right). Scale bars, 10 µm in b and 20 µm in a,d and f.

Microtubule-based recycling-endosome-positioning factorsStrikingly, all known candidates required for apical RAB-11-REpositioning were functionally associated with the microtubulecytoskeleton, including cytoplasmic dynein and dynactin complexes,the kinesin-1 orthologue UNC-116 and dynein-associated factors37

such as LIS-1, NUD-1 and NUD-2, as well as F55C12.1, the orthologueof the Rab11 effector NUF (ref. 38, Supplementary Table S3, group5, and Fig. 3a,d). This group included as-yet uncharacterized genes,such as C32D5.11, a predicted E3 ubiquitin ligase, which may regulatemotor or cytoskeleton components. RNAi of these components causedsimilar phenotypes on all endosomal compartments tested, includingthe scattering or basal accumulation of recycling endosomes, lateendosomes (Fig. 3d) and early endosomes (not shown). Furthermore,whereas only 20% of intestines treated with control RNAi (n= 54)had PEPT-1::DsRed present in GFP::RAB-11-REs, over 50% of

intestines treated with arp-1, dhc-1 or nud-1 RNAi (n= 30) exhibitedincreased amounts of PEPT-1::DsRed in scattered and basallymisplacedGFP::RAB-11-REs (data not shown).UNC-116 (kinesin-1 heavy chain) was the only identified

microtubule plus-end kinesin. Its depletion similarly altered the apicaldistribution of both recycling endosomes and late endosomes (Fig. 3d),suggesting that kinesin-1 transports different endosomes in C. elegansenterocytes. It may also be required for dynein-mediated transport,as shown in C. elegans neurons39. However, the complementingphenotypes observed on depletion of dynein and dynactin componentsversus UNC-116 (kinesin-1) rather suggest that apical enrichment ofRAB-11-REs in C. elegans enterocytes requires coordinated minus- andplus-end microtubule-based transport.Overall, our results highlight the importance of the dynein–dynactin

machinery and identify further components for (1) minus-end



ART I C L E S

PEPT-1::DsRed

GFP::RAB-11 (REs)

1. Accumulated in punctate structures

4. Basally enriched REs

2. Accumulated in fuzzy structures

vha-9 (degradation) pept-1

par-5

syn-3

Control RNAi

Control RNAi

Library-covered genomeHit set

Membrane traffic18%

Translation11%

Protein metabolism9%

Cytoskeleton associated

7%

Unknown7%

RNA processing7%

Signalling7%

Lipid metabolism

6%

Mitochondrial function 6%

Carbohydrate metabolism 5%

DNA binding5%

Transport3%

Other9%

!!!

!!!!!

!!!

5.9 2.87.4 2.9Enrichment factor:

Rel

ativ

e fr

eque

ncy

(%)

dyci-1 (organelle transport) tba-4 (microtubule integrity)

pgp-1 (lysosomal integrity)

sec-24.2 (secretion)Control RNAi

8. Accumulated REs 5. Scattered REs

Autofluorescent LROs 11. Enlarged LROs 12. Smooth AF / LROs 13. Luminal AF

vps-33 (lysosomal integrity)

0

5

10

15

20

Protei

n tar

getin

g

and lo

caliz

ation

P = 5

" 10–4

Intra

cellu

lar

protei

n tra

ffic

P = 2

" 10–2

6

Cell st

ructu

re

and m

otilit

y

P = 8

" 10–5

Protei

n meta

bolism

and m

odific

ation

P = 8

" 10–1

9Biological process (PANTHER ontology):

3. Less PEPT-1::DsRed

a

b

Control phenotype Phenotype defect categories (9 of 14)

c

Figure 2 Phenotypic profiling and biological functions of identified genes.(a) Representative epifluorescence microscopy images from the primaryscreen showing the control phenotypes (control RNAi, left-hand column)and nine of the 14 phenotypic defect categories with specific changesin marker distribution, intensity or organelle morphology. Scale bar,50 µm, and 20 µm for the enlarged view. Perturbations of specificbiological functions in membrane trafficking are coupled to specificphenotypic defect categories, as indicated by the gene name and therespective biological function in parentheses. RE, recycling endosome.

(b) Distribution of functional classes in the complete hit set (356 hits).(c) Relative frequencies of biological process terms in the complete hitset and the library-covered genome based on the PANTHER ClassificationSystem. The x axis shows biological process terms that are over 2.5-foldenriched. The enrichment factor is the ratio of relative frequencies ofa term in the hit set over the library-covered genome. ‘Intracellularprotein traffic’ is the most enriched or overrepresented biological processterm. Significance levels, ""P <0.01, """P <0.001; cumulative P -values,hypergeometric distribution.

microtubule transport and apical enrichment of RAB-11-REs,(2) steady state positioning of endosomes and (3) apical mem-brane transport.

par-1 and par-5 RNAi misplace RAB-11-REsIn unpolarized cells, depletion of PAR-6 or the Rho GTPase CDC-42specifically affects the positioning of RAB-11 and RME-1-positiverecycling endosomes40. Surprisingly, in C. elegans enterocytes, RNAiof par-3 or par-6 scattered all types of endosome tested (recycling

endosomes, late endosomes, early endosomes; Fig. 3e, group 5).In addition, the GFP-fluorescence intensities of organelle markerswere reduced, whereas the apical localization of PEPT-1::DsRedremained unaffected (Fig. 3e). These results suggest that PAR-3 andPAR-6 are of general importance for endosome positioning, possiblythrough cytoskeletal roles9.Apart from the ‘reference gene’ SEC-15, group 6 comprised only

three candidates causing different Rab11-RE-specific positioningphenotypes, with only weak or no effect on late endosomes, early



ART I C L E S

GFP::RAB-5 (EEs) AMANS-2::GFP (Golgi) GFP::SP12 (ER)

gbf-1 F40G9.1

Group 6: RAB-11 RE-positioning

FC

GFP:RAB-11 (REs) LMP-1::GFP (LEs)

LMP-1::GFP (LEs)

Biological function of examined genes (sum = 60)

Cytoskeleton associated (15)Protein metabolism (2)Transport (1)Unknown (11)Signalling (8)Lipid metabolism (1)Membrane traffic (22)

4/7

4/16

4/5

vps-33.1

par-3

par-6

GFP:RAB-11 (REs)

2/19

3/4

5/9

GFP:RAB-11 (REs) LMP-1::GFP (LEs)Group 5: endosome positioning

F55C12.1 (nuf)

GFP:RAB-11 (REs) LMP-1::GFP (LEs)

GFP::RAB-5 (EEs)

FC par-5

AMANS-2::GFP (Golgi)

FC

F55C12.1 (nuf)

FC

Predicted E3 ubiquitin ligasePredicted E3 ubiquitin ligase

FC

C45G9.5

vps-33.1

FC

C45G9.5

vps-33.1

par-3

par-6

par-1

par-5

par-1

par-5

FC par-5

GFP::RAB-11par-5 par-5 Anti-CASP

unc-116

FC

dyrb-1

dnc-1

unc-116

FC

dyrb-1

dnc-1

1 2 3 4 5 6(b) (d,e) (f)(c)‘Core’

functionsLROs

Biological functions

of genes in secondary

assay groups

df

g

Group 4: secretionGroup 3: endocytosis and degradation

a

b c

Group 5 (continued)e

C32

D5.

11D

ynac

tinD

ynei

n

Figure 3 Comparative secondary assays reveal functional roles for candidategenes. (a) Compositions of the six functional secondary assay groups.Above each bar, the fraction of genes that was previously not linked to theassigned function is indicated. In total, we proposed functional roles for22 of the 60 examined genes (for details see Supplementary Table S3).(b–f) Confocal microscopy images of RNAi phenotypes representative forfour secondary assay groups showing intestinal regions of live secondaryassay worms expressing PEPT-1::DsRed (red) and an organelle-specificGFP-marker (green) as indicated above the images. (b) RNAi of group 3genes perturbs endocytosis and early degradation, causing PEPT-1::DsRedaccumulations predominantly in early endosomes (EEs) and late endosomes(LEs). vps-33.1 RNAi also affects recycling endosome morphology andpositioning. FC, feeding control. (c) RNAi of group 4 genes perturbs secretionand causes PEPT-1::DsRed to accumulate in round, filled structures positivefor Golgi and endoplasmic reticulum (ER) markers. (d) RNAi of group 5genes affects different organelles to a similar extent. On perturbation ofminus-end-directed microtubule transport, recycling and late endosomes(and early endosomes, not shown) scatter (dyrb-1 RNAi), or enrich basally

(dnc-1 RNAi). On depletion of the microtubule plus-end-directed motorUNC-116 (kinesin-1), recycling endosomes and late endosomes concentrateapically in a discontinuous fashion. Both depletion of F55C12.1, theorthologue of the RAB-11 effector NUF, and depletion of the as-yetuncharacterized gene C32D5.11, a predicted E3 ubiquitin ligase, causescattering of both recycling endosomes and late endosomes. (e) Depletionof apical PAR proteins PAR-3 or PAR-6 causes diffuse distribution andreduced GFP-fluorescence intensities of both recycling endosome andlate endosome markers. (f) RNAi of group 6 genes such as par-1 andpar-5 affects RAB-11-RE positioning differently, more strongly, or both,than other organelles. Furthermore, and similarly to RNAi of par-3 andpar-6 (e) and par-1 (f), par-5 RNAi (f) results in a diffuse distribution of lateendosomes with reduced fluorescence. Early endosomes and Golgi bodiesare affected similarly. However, early endosomes were occasionally observedto cluster weakly at the periphery of these cells (arrowheads). (g) par-5RNAi-treated, dissected, methanol-fixed intestines immunostained for theendogenous Golgi marker CASP show that Golgi bodies do not enrich at sitesof GFP::RAB-11-RE clusters. Scale bar, 20 µm.



ART I C L E S

endosomes, or Golgi bodies (Supplementary Table S3, group 6, andFig. 3f). These were the as-yet uncharacterized WD40 repeat proteinH24G06.1 and, unexpectedly, non-apical PAR proteins PAR-1 andPAR-5. Depletion of PAR-1 (ref. 41) resulted in enrichment ofRAB-11-REs along cell–cell contact sites (Fig. 3f). Other organellessuch as late endosomes were more scattered, and GFP-fluorescenceintensities were reduced as for par-3 and par-6 RNAi. In the PAR family,PAR-5 (ref. 42) (the orthologue of mammalian 14-3-3!) is functionallyits most complex and poorest-understoodmember. 14-3-3 proteins aredimeric adapters of up to 200 interaction partners, thereby regulatingdiverse biological functions simultaneously43–45. Depletion of PAR-5(ref. 42) induced prominent, peripheral clusters of GFP::RAB-11-REsat the basolateral domain of enterocytes (Fig. 3f). In contrast, organellemarkers for late endosomes, early endosomes and Golgi bodies(Fig. 3f,g) became diffuse on par-5 RNAi, with similarly reducedfluorescence intensities as on RNAi of the other par genes. Weakerclusters of early endosomes were occasionally observed at the cellperiphery (Fig. 3f). Among the tested par genes, RNAi of par-5 causedthe strongest effects on intestinal morphology, with more irregularlyshaped endotubes and enterocytes (Fig. 3f).Overall, our results reveal that single depletion of PAR-5 and PAR-1,

but not of apical PAR proteins, predominantly affects positioning ofRAB-11-REs in enterocytes.

par-5 RNAi causes recycling endosome clustering aroundectopic F-actinPAR-5 served as an interesting proof-of-principle candidate forcoordinating the interplay between epithelial membrane trafficking,cytoskeleton and cell polarity. Its fly and mammalian orthologue iscritical for the localization of PAR3 (ref. 46) and PAR1 (refs 47,48).However, other roles in epithelial polarity remain to be determined.We thus validated and analysed the par-5 RNAi phenotype in moredetail. First, we confirmed the efficient depletion of PAR-5 byRNAi by western blot analysis (Supplementary Fig. S8a). Second,we excluded the possibility that the basal RAB-11-RE clusteringwas due to RNAi of both C. elegans 14-3-3 isoforms, par-5 (ftt-1) and ftt-2, which share 80% messenger RNA sequence identity.For this, we used an ftt-2 deletion strain49 (Supplementary Fig.S4a,b) as no viable, temperature sensitive par-5 strain existed42.Third, in non-transgenic worms, recycling endosomes labelled byendogenous RAB-11 also clustered peripherally on par-5 RNAi(Supplementary Fig. S4m), ruling out a possible epiphenomenon ofGFP::RAB-11 expression.We next asked whether peripherally clustered RAB-11-REs are

functionally perturbed, possibly causing accumulation of apicalmembrane proteins. We examined the localization of PEPT-1::DsRedand the ATP-binding cassette transporter PGP-1::GFP (ref. 50) relativeto endogenously labelled RAB-11-REs in control, arp-1 and par-5RNAi (Supplementary Fig. S5). In contrast to RNAi of arp-1 (dynactincomponent, positive control), PAR-5 depletion did not cause asignificant increase in apical membrane proteins within peripheralrecycling endosomes. Notably, we did not observe mis-sorting ofPEPT-1::DsRed or PGP-1::GFP to the basolateral membrane on arp-1or par-5 RNAi. These data suggest that alterations in the intracellularpositioning of RAB-11-REs do not necessarily perturb cargo traffickingthrough this compartment.

As 14-3-3! regulates several actin polymerization and depolymeriza-tion components51–53, we considered the possibility that the basal clus-tering of RAB-11-REs results from alterations of the actin cytoskeleton.First, we confirmed that PAR-5 or FTT-1, but not FTT-2, was requiredfor proper microfilament organization (Supplementary Fig. S4c,d).Second, we observed ectopic F-actin patches of varying spread anddensity at the basolateral cortex, rarely in the central cytoplasm,but not alongside the apical domain (Supplementary Fig. S4i,j,p,q).Virtually all (99%) RAB-11-RE clusters (n = 145, in 24 intestines)were located at, or around, ectopic F-actin patches. Unexpectedly,basal GFP::RME-1-labelled recycling endosomes (ref. 24) did notcluster at ectopic F-actin sites but vanished (Supplementary Fig. S4k,l).This interesting observation will require further investigations onthe role of PAR-5 for basal recycling endosomes. Third, staining forendogenous markers enabled us to confirm that par-5 RNAi-inducedectopic F-actin patches were predominantly enriched for clusteredRAB-11-REs (88%) and only moderately for RAB-5-labelled earlyendosomes (26%; Supplementary Fig. S4m–o). Only peripheralRAB-11-REs induced by par-5 RNAi, not by arp-1 RNAi, clusteredaroundprominent patches of ectopic F-actin (Fig. 4a–d). This indicatedthat peripherally clustered recycling endosomes are per se insufficientto cause ectopic F-actin accumulation, and that the ectopic F-actin mayprecede the clustering of RAB-11-REs. This possibility is consistentwith the finding that a small but significant fraction of ectopicF-actin patches (12%) was devoid of co-clustered RAB-11-REs. Next,we analysed the effects of PAR-5 depletion on apicobasal F-actinorganization and F-actin levels. Two line-based intensity quantificationapproaches (illustrated in Fig. 4e) revealed a !3.7-fold increase inmean fluorescence intensity at ectopic basal F-actin sites (Fig. 4f,g),and a moderate but significant reduction (1.2 fold) in mean peakintensities of apical F-actin (P = 3#10$6, Wilcoxon rank sum test)in par-5 RNAi when compared with control. This was coupled tomore irregularly shaped endotubes (compare Fig. 4a,b), suggestingthat PAR-5 depletion also altered apical F-actin organization. Althoughthe integral F-actin intensities of these line-based measures wereslightly increased in par-5 RNAi enterocytes (1.2-fold) (P = 0.013,Wilcoxon rank sum test), area-based intensity measurements revealedthat the total F-actin levels were not significantly changed (control,31 intestines; par-5 RNAi, 45 intestines; P = 0.92, Wilcoxon ranksum test). When compared with PAR-5 depletion, depletion ofthe actin depolymerizing factor (ADF)/cofilin orthologue UNC-60(ref. 54), a key F-actin regulator and target of 14-3-3! (ref. 51),caused RAB-11-RE clustering preferentially at apical ectopic mi-crofilaments (Supplementary Fig. S4r,s). Altogether, the effects ofPAR-5 depletion together with the ADF/cofilin RNAi phenotypeunderscore the importance of proper apicobasal F-actin organizationfor RAB-11-RE positioning.

RAB-11-RE clusters coincide with ectopic apical componentsWe next inspected whether other apical domain components wereenriched at the ‘ectopic sites’ with clustered GFP::RAB-11-REs andperipheral F-actin. The F-actin membrane linker ERM-1 (refs 25,27),the signalling adaptor and barbed-end capping protein EPS-8 (ref. 55),the intermediate filament protein IFB-2 (ref. 56) and the apicaljunction protein AJM-1 (ref. 57) all accumulated at the cortical siteswhere GFP::RAB-11-REs clustered (Fig. 5). "-tubulin also enriched



ART I C L E S

arp-

1 R

NA

i

par-

5 R

NA

i

GFP::RAB-11In

tens

ity (a

.u.)

Distance (pixels)

Bas

al F

-act

in p

eak

inte

nsiti

es,

mea

n gr

ey v

alue

(!10

0) (a

.u.) """

n = 40 m = 73

WT PTM

ean

grey

val

ue (!

100)

GFP::RAB-11F-actin

(g)

par-

5 R

NA

i

‘WT’

‘PT’

(f)

Nor

mal

ized

inte

nsity

ra

tios

(ect

opic

/api

cal)

"""

"""

"""

Intestines: FC, 21; par-5, 14; arp-1, 10

n.s.

"""

n.s. """

Inte

nsity

(a.u

.)

F-actin

F-actin F-actinC

ontr

ol R

NA

iGFP::RAB-11

F-actin F-actinGFP::RAB-11

F-actin F-actin

GFP::RAB-11F-actin

GFP::RAB-11F-actin

0

50

100

150

200

0 50 100 150 200 250 3000

50

100

150

200

Distance (pixels)0 50 100 150 200 250 300

Distance (pixels)0 50 100 150 200 250 300

Inte

nsity

(a.u

.)

0

50

100

150

200

FC par-5 arp-1 FC par-5 arp-1

F-actin GFP::RAB-11

0123456789

0

1

2

3

4

5

n = 31 m = 45

FC par-50

5

10

15

n = 23m = 37

n = 21m = 42

a

d

b c

e f g

n = intestines, m = line-based measures

Figure 4 PAR-5 depletion causes RAB-11-REs to cluster around ectopicF-actin at the basolateral cortex. (a–c) GFP::RAB-11-expressing andrhodamine phalloidin-stained intestines treated with control, par-5and arp-1 RNAi, as well as intensity diagrams corresponding to theindicated lines. To improve the visualization of ectopic F-actin, thebrightness and contrast of each grey image was increased. Scale bar,20 µm. (a) In control RNAi, rhodamine phalloidin-labelled F-actin isstrongly concentrated at the apical domain (white arrowhead) and notvisible at the basal domain or cell periphery. (b) On PAR-5 depletion,GFP::RAB-11-labelled recycling endosomes cluster peripherally aroundpatches of ectopic F-actin (red arrowheads) at the basolateral cortex.The endotube labelled by apical F-actin (white arrowhead) can becomemore dilated and irregularly shaped on par-5 RNAi. (c) In arp-1 RNAiintestines, peripheral GFP::RAB-11-labelled recycling endosomes do notcoincide with ectopic F-actin patches. (d) Comparison of mean peakintensity ratios of ectopic:apical F-actin (left) and ectopic (or basal):apical

GFP::RAB-11 (right) for line-based intensity measures collected fromcontrol, par-5 and arp-1 RNAi-treated intestines. When compared withcontrol and arp-1 RNAi, the ectopic:apical F-actin intensity ratio in par-5RNAi is increased threefold. The ectopic:apical GFP::RAB-11 intensityratios in par-5 and arp-1 RNAi are similarly increased. FC, feeding control.(e) Illustration of two methods of line-based intensity measurement ofrhodamine phalloidin-stained F-actin. (f) Lines perpendicular across controlRNAi and phenotypic par-5 RNAi intestines reveal a 3.8-fold increasein basal F-actin mean peak intensities. (g) Lines along selected regionsat the basal cell periphery of par-5 RNAi-treated intestines reveal a3.6-fold increase in mean basal F-actin intensities at sites with ectopicF-actin patches (‘PT’) when compared with those without (‘WT’). Individualmeasurements m are plotted as black dots; the grey line indicates themedian. Significance level, """P < 0.001; Wilcoxon rank sum test. Errorbars in d and f, s.e.m., derived using them values. The number of intestinesn in d,f and g derives from up to three experiments.

moderately at these sites (Fig. 5g–i), possibly accounting for themicrotubule arrays occasionally observed around recycling endosomeclusters (Supplementary Fig. S6g,n). In C. elegans one-cell embryos,proper cortical localization of anterior PAR proteins and regulationof the actomyosin meshwork are functionally interdependent9,58.Are apical PAR proteins also enriched at the ectopic sites onPAR-5 depletion? In control adult intestines, apical PAR proteinsconcentrated at the apical domain (PKC-3 or aPKC) and alongAJM-1-labelled apical junctions (PAR-6, PAR-3), but not at the

basolateral cortex (Supplementary Fig. S7b,h,n). Visualization ofendogenous PAR-3 by immunofluorescence proved, however, difficult,presumably owing to its low expression levels in adult intestines11.In contrast, over 83% of par-5 RNAi-treated intestines exhibitedectopically enriched PKC-3, PAR-6 and PAR-3 at the basolateralcortex adjacent to peripherally clustered GFP::RAB-11-REs andectopic AJM-1 (Supplementary Fig. S7e,k,q,t). These results showthat PAR-5 is critical to maintain apicobasal asymmetry of apicaldomain components.



ART I C L E S

Control par-5 RNAi par-5 RNAi par-5 RNAiControl Control

Control

Anti-AJM-1

par-5 RNAi

GL

GL

GL

GL

GFP::RAB-11 Anti-ERM-1 GFP::RAB-11 Anti-EPS-8 GFP::RAB-11 Anti-IFB-2

GFP::RAB-11Anti-!-tubulin

Anti-!-tubulinAnti-AJM-1

GFP::RAB-11Anti-!-tubulin

Anti-!-tubulin

12/12

10/12

18/18

39/39

10/13

100

83.3

100

100

77

ERM-1

EPS-8

IFB-2

AJM-1

!-tubulin

Component % (intestines) n intestines

24/24100F-actin

a

g h i

b c d e f

Figure 5 RAB-11-REs cluster at ectopic sites at the basolateralcortex that are enriched for various apical domain components.(a–h) GFP::RAB-11-expressing, control and par-5 RNAi-treated intestinesimmunostained for various apical cytoskeleton components, such asthe F-actin binding proteins ERM-1 (a–b) and EPS-8 (c,d) as well asintermediate filament protein IFB-2 (e,f). Images show single confocalmicroscopy sections with more peripheral (a) or medial (e) views ofthe intestines. GL, germline. Note that, in PAR-5-depleted intestinesonly, these apical cytoskeleton components co-accumulate ectopically atthe basolateral cortex at sites where GFP::RAB-11-labelled recyclingendosomes cluster. (g,h) Intestines immunostained for AJM-1 and

"-tubulin on control and par-5 RNAi (maximum intensity projections,seven consecutive confocal sections). (g) AJM-1 demarcates apicaljunctions, and "-tubulin-containing complexes are part of the apicalcytoskeleton, concentrating around the endotube. (h) In par-5 RNAiintestines, AJM-1 and "-tubulin-containing complexes localize ectopicallyto cortical sites with peripherally clustered GFP::RAB-11-labelledrecycling endosomes. (i) Mean percentage of par-5 RNAi-treatedintestines with peripherally clustered GFP::RAB-11-REs at which theindicated apical domain components co-accumulated ectopically. nderives from up to four independent experiments. Scale bar in enlargedbox of b, 10 µm; all others, 20 µm.

par-5 phenotype requires apical PARs and Rho GTPasesTo identify key factors required for ectopic site formation, wetested whether ectopic F-actin and recycling endosome clustersare prevented on double depletion of PAR-5 with regulators ofF-actin dynamics and cell polarity, such as Rho GTPases andapical PAR proteins2,10. To this end, we carried out double-feedingexperiments, with par-5 plus scrambled double-strandedRNA (dsRNA)(par-5+ feeding control) as reference. Notably, all double depletions(par-5+ second gene) significantly reduced the par-5 RNAi phenotypefrequency, that is the mean percentage of intestines with one ormore peripheral GFP::RAB-11-RE clusters (Fig. 6b). Among the testedRho GTPases, double depletion of PAR-5 with RHO-1 or CED-10(Rac1 orthologue) efficiently suppressed recycling endosome clustering(up to 49-fold), whereas double depletion of PAR-5 with CDC-42was less effective (2.7-fold suppression). Furthermore, intestinesin the reference sample exhibited more and larger ectopic F-actinpatches than those in the test samples (Fig. 6c–e). The modestclustering-suppression by CDC-42 co-depletion could be due toadditive phenotypic effects, as single depletion of CDC-42 also causedsmall ectopic F-actin patches co-clustered with GFP::RAB-11-REs(not shown). In contrast to cdc-42 RNAi, we did not observe ectopic

F-actin structures or recycling endosome clusters on RNAi of ced-10,rho-1, par-3 or par-6. However, depletion of CED-10 (Rac1) reducedthe fluorescence intensity of GFP::RAB-11 (not shown). Overall,these data demonstrate that Rho GTPases, specifically RHO-1 andCED-10 (Rac1), as well as apical PAR proteins, (1) are requiredfor the formation of both ectopic F-actin and RAB-11-RE clusterson PAR-5 depletion and (2) need PAR-5 to properly maintainapicobasal asymmetry.

DISCUSSIONHere, we carried out an in vivo RNAi screen in C. elegansintestine for genes maintaining apical enrichment of RAB-11-REs and apical membrane transport. In contrast to previousC. elegans RNAi screens for membrane-transport regulators26,40,three subcellular readouts were simultaneously monitored, that isPEPT-1::DsRed, GFP::RAB-11 and LROs. Using a semi-automatedworkflow with epifluorescence microscopy and custom-made imageanalysis, we employed advanced HCS technologies mainly used incell-based assays on live animals59. The primary screen with 356hits is a rich information resource, comprising phenotype profiles,biological functions, protein domain and orthologue annotations



ART I C L E S

par-5 + FC

par-5 + ced-10 (rac1)

F-ac

tinF-

actin

Tota

l int

estin

al p

atch

le

ngth

(pix

els)

Exp. 1 (3 samples)Exp. 2 (6 samples)

Double RNAi: par-5 + FC par-3 rho-1 cdc-42 ced-10

(rac1)par-6

a

c e

d

bDouble RNAi: par-5 +ref.

% s.e.m. nP

cdc-

42

ced-

10rh

o-1

FC par-

6

pkc-

3

par-

3PAR-5

RHO-1

PKC-3PAR-6

Actin!-tubulin

8576

Exp.963769

1,006738

<0.01<0.001<0.001

5.51.41.70.9

49.06.04.42.4

FCpkc-3

par-3

par-5 +

par-6

0

500

1,000

1,500

Intestines with ectopic F-actin =

Intestines imaged (n) =

6 28 6 1 8 2

3129 32 12 16 15

4416<0.010.61.6ced-106738<0.010.21.0rho-16798<0.016.418.2cdc-42

Figure 6 Formation of peripheral GFP::RAB-11-RE clusters and ectopicF-actin patches on depletion of PAR-5 depends on both Rho GTPases andapical PAR proteins. (a) In each experiment, we confirmed that all doubleRNAi samples—that is, reference (par-5+ feeding control bacteria (FC))and test (par-5 + apical par gene or rho GTPase gene) samples—havesimilar levels of PAR-5 depletion as indicated by similar PAR-5 to #-tubulinratios in total worm protein extracts. One representative western blot isshown. Concomitantly, we confirmed the efficient depletion of PKC-3,PAR-6 and RHO-1, as well as CDC-42 (GFP::CDC-42), in the respectivetest samples. Bands were cropped from three blots: #-tubulin, actinand RHO-1 from one; PAR-5 from a second; and PKC-3 and PAR-6from a third (see Supplementary Fig. S8 for uncropped western blotsand probing for GFP::CDC-42). (b) Epifluorescence-microscopy-basedquantification of the mean percentage of phenotypic worms with one ormore peripheral GFP::RAB-11-RE clusters after double RNAi with par-5.P -values relative to reference (ref.), Wilcoxon rank sum test. n, numberof analysed worms. In all test samples, recycling endosome clusteringwas suppressed. (c–e) Comparison of the phenotype strength in affectedintestines of reference and test samples. Epifluorescence images ofrhodamine phalloidin-stained, double-RNAi-treated intestines with ectopicF-actin patches. The dashed lines illustrate the different total ectopic F-actinpatch lengths per imaged intestine that result from the total number andlength of patches. Scale bars, 20 µm. (e) Box plots with overlaid individualdata points of two independent experiments (filled and open circles) of thetotal measured length of ectopic F-actin patches per phenotypic intestine.All test samples exhibit considerably reduced medians (indicated by greylines) of total ectopic patch length values. For this analysis, we selectedfrom each double-RNAi sample phenotypic worms with clustered recyclingendosomes—if these were present—or else we selected worms with defectivegermlines indicative of par-5 RNAi. Underneath the x axis, the numbers ofintestines with ectopic F-actin patches and the total number of analysedworms (n) are shown, with n deriving from one or two experiments.

(Supplementary Table S1). Functional predictions inferred fromhierarchical clustering for 22 genes were validated in secondary assays(Supplementary Table S3). Overall, both primary and secondary assaydata sets revealed numerous candidate genes with a potential role inepithelial cell polarity.We further garnered mechanistic insights into the regulation of

RAB-11-RE positioning and epithelial cell polarity. First, RAB-11-RE positioning requires the dynein–dynactin complexes, regulatorycomponents and kinesin-1. Therefore, our findings in C. elegansenterocytes suggest that their orthologues in higher-metazoanepithelia also mediate transport and apical enrichment of Rab11-REs.Unexpectedly, deletion of the myosin Vb (ref. 60) orthologue HUM-2

Downstream effectors modulating F-actin (such as ADF/cofilin* and profilin*) and cortical composition

Apical PARs PAR-3* PKC-3 PAR-6

Rho GTPases RHO-1

CED-10/Rac1 (CDC-42)

PAR-5

F-actin, apical PARs etc. (ADCs)

RAB-11-REsProper polarity

{ {ADCs (can be less concentrated)Disrupted

polarityClustered

RAB-11-REs

Wild type par-5 RNAi phenotype

Ectopic ADCs

Figure 7 Functional PAR-5 model. PAR-5 maintains apicobasal polarity byfunctioning as a regulatory hub for the coordinated interplay of Rho GTPases,apical PAR proteins and downstream effectors. Network componentsthat could require direct regulatory interaction with PAR-5, such asPAR-3 (ref. 46) and numerous F-actin-modulating downstream effectors(ADF/cofilin51,52, profilin62 and possibly orthologues of cofilin kinase52 andphosphatase53), are labelled by asterisks. Arrows indicate regulatory effectsthat can be mutual (double arrows). Other roles or effects, as well as crosstalkof network components, are not indicated. PAR-5 sustains proper F-actindynamics and asymmetry of F-actin and other apical domain components(labelled as ADCs), and allows for microtubule-dependent enrichment ofRAB-11-REs (wild type, left-hand cell, PAR-5 in grey). Single depletion ofPAR-5, and thus deregulation of this polarity-maintaining network, causesectopic sites at the basolateral cortex, comprising F-actin and other apicaldomain components, as well as strongly clustered RAB-11-REs (par-5 RNAiphenotype, right-hand cell, light grey for depleted PAR-5). Double depletionof PAR-5 with one of the Rho GTPases or apical PAR proteins suppressedboth ectopic F-actin and recycling endosome clustering efficiently (ormoderately in the case of CDC-42), indicating that ectopic F-actin and apicaldomain components induce the local clustering of RAB-11-REs (circulararrow). RAB-11-REs could also feed back on ectopic F-actin formation(dashed circular arrow), thereby enhancing the peripheral clustering. SeeDiscussion for further details.

did not significantly alter RAB-11-RE distribution (data not shown),which may be due to functional redundancy.Second, we revealed a previously unknown role of PAR-5 in

apicobasal polarity and RAB-11-RE positioning. Depletion of PAR-5induced the formation of ectopic sites at the basolateral cortex includingF-actin and other structural and regulatory apical domain components(ERM-1, EPS-8, IFB-2, "-tubulin, AJM-1 and the apical PAR proteinsPKC-3, PAR-6 and PAR-3), as well as clustered RAB-11-REs. Incontrast, depletion of apical PAR proteins or Rho GTPases did notcause prominent polarity- or organelle-specific positioning phenotypes.Thus, although seemingly robust, epithelial polarity relies on regulatoryhub proteins, such as PAR-5, that coordinate numerous componentsand functions simultaneously. For example, the mammalian PAR-5orthologue 14-3-3! regulates several key proteins for cytoskeletalorganization and polarity, such as Rho guanine nucleotide exchangefactors (GEFs) and GTPase-activator proteins (refs 61,62), PAR3(ref. 46) and components of the core machinery for F-actin dynamicsthat are downstream effectors predominantly of RhoA (RHO-1)and Rac1, but also Cdc42 (refs 45,62). These include ADF/cofilin51,profilin62, the ADF/cofilin regulatory kinase LIMK (ref. 52), which



ART I C L E S

is inhibited by PAR3 (ref. 63), and the ADF/cofilin phosphatase SSH(ref. 64). Importantly, double depletion of PAR-5 with each apicalPAR protein, RHO-1 or CED-10 (Rac1) strongly suppressed bothectopic F-actin patches and RAB-11-RE clusters, suggesting that allthese factors contribute interdependently to the formation of ectopicsites. By integrating published findings with our results, we propose thefollowingmodel for the role of PAR-5 in epithelial polaritymaintenance(Fig. 7). Under physiological conditions, PAR-5 constitutes a regulatoryhub (or nodal point) for a cell polarity-protein network including RhoGTPases (RHO-1, CED-10 (Rac1) and CDC-42), apical PAR proteins(PKC-3, PAR-6 and PAR-3) and F-actin modulating downstreameffectors. Such interactions are critical to maintain the low net levelsof F-actin and other apical domain components at the basolateralcortex. In the presence of PAR-5, transport and apical enrichment ofRAB-11-REs depends primarily on the microtubule-based machinery.par-5 depletion perturbs the cell polarity regulatory network: alteredF-actin dynamics and de-regulated activities of Rho GTPases, as well asectopically localized apical PAR proteins, then cause the formation ofthe ectopic sites at the basolateral cortex that is susceptible to capturingor forming ectopic microfilaments65–67.The prominent, peripheral clustering of RAB-11-REs around

ectopic F-actin patches could be accounted for by a number ofnon-exclusive mechanisms. First, altered F-actin dynamics and ectopicmicrofilaments could functionally interfere with proteins that mediatepositioning of RAB-11-REs, such as specific motor proteins orcytoskeletal adaptors. This interpretation is further supported by RNAiof unc-60 (ref. 54; ADF/cofilin, Supplementary Fig. S4r,s), which alsocaused ectopic microfilaments and co-clustered RAB-11-REs. Second,recycling-endosome-specific positioning factors could functionallydepend on regulatory interactions with PAR-5 and, on its depletion,cause recycling endosome clustering or fail to enrich recyclingendosomes apically (such as potentially dynein heavy chain andNUDEL (ref. 62; NUD-2)). Third, in the condition of altered F-actindynamics, clusteringRAB-11-REs and ectopicmicrofilament formationmay positively influence each other, thus exacerbating the phenotype.Several previously identified factors localized to Rab11-REs couldfacilitate this, such as the Rab11 effector NUF (ref. 68), actin nucleationfactors69, Cdc42 (ref. 70), PAR-6 (ref. 40) and microfilament tetheringmotors such as myosin Vb (ref. 60; HUM-2).It will be important for future research to further elucidate how

the polarity-maintaining protein network of which PAR-5 is acentral element functions in coordinating apicobasal polarity and thefunctional properties of membrane organelles such as RAB-11-REs.Our proof-of-principle analysis of PAR-5 demonstrates that the genesidentified in our screen can provide important molecular insights intothemechanisms underlying organelle positioning and apicalmembranetransport, as well as polarity maintenance. !

METHODSMethods and any associated references are available in the onlineversion of the paper at www.nature.com/naturecellbiology

Note: Supplementary Information is available on the Nature Cell Biology website

ACKNOWLEDGEMENTSWe thank various members of the HT-TDS including J. Wagner, M. Gierth and H.Grabner for assistance in development and robotic programming of automationsteps of the screening workflow. We are very grateful to M. Storch, U. Frömmel,

T. Döbel, L. Socher, S. Quaiser, S. Scheibe, F. Zakrezweski and D. Haase for theirassistance throughout the primaryHCS.We thankR. Schäfer for technical assistance.We are grateful to S. Eaton, K. Simons, E. Knust, E. Paluch, G. A. O’Sullivan,N. Goehring and B. Habermann for discussions and critical reading of themanuscript draft and to C. Eckmann for protocols and practical advice.

This study was supported by the German Federal Ministry of Education andResearch (InnoRegio Initiative, grant number 03I4035A; the Systems BiologyNetwork HepatoSys, grant number 0313082H, and the Virtual Liver initiative,www.virtual-liver.de), the EU sixth framework project ‘EndoTrack’ (FP6 GrantLSHG-CT-2006-019050) and the Max Planck Society’s inter-institutional initiative‘RNA interference’. J.F.W. was supported by a PhD fellowship of the BoehringerIngelheim Fonds.

AUTHOR CONTRIBUTIONSS.H. and M.Z. conceived the initial concept for the screen. S.H. establishedthe screening platform with support from J.F.W. S.H. and J.F.W. conductedthe genome-wide screen with the support of K.K., students from the TechnicalUniversity Dresden and staff members from the High-Throughput TechnologyDevelopment Studio (HT-TDS; see Acknowledgements). S.H. developed the imageanalysis software. J.F.W. conceived the embryonic lethal screen and conducted itwith support from S.H. and students. S.H., J.F.W. and M.Z. analysed the datawith the help of B.H., C.R.B. (general bioinformatics) and M.V. (hierarchicalclustering). J.F.W. carried out secondary assays with the support of B.O.F. J.F.W.conceived, carried out and analysed the experiments on PAR-5. G.M. contributedto quantifications and statistical analysis on PAR-5. J.F.W. and M.Z. wrote themanuscript with the contribution of S.H., G.M., C.R.B. and M.V.

COMPETING FINANCIAL INTERESTSThe authors declare no competing financial interests.

Published online at www.nature.com/naturecellbiologyReprints and permissions information is available online at www.nature.com/reprints

1. Apodaca, G. Endocytic traffic in polarized epithelial cells: role of the actin andmicrotubule cytoskeleton. Traffic 2, 149–159 (2001).

2. Iden, S. & Collard, J. G. Crosstalk between small GTPases and polarity proteins incell polarization. Nat. Rev. Mol. Cell Biol. 9, 846–859 (2008).

3. Knust, E. & Bossinger, O. Composition and formation of intercellular junctions inepithelial cells. Science 298, 1955–1959 (2002).

4. Goldenring, J. R. et al. Rab11 is an apically located small GTP-binding protein inepithelial tissues. Am. J. Physiol. 270, G515–G525 (1996).

5. Rodriguez-Boulan, E., Kreitzer, G. & Musch, A. Organization of vesicular traffickingin epithelia. Nat. Rev. Mol. Cell Biol. 6, 233–247 (2005).

6. Mellman, I. & Nelson, W. J. Coordinated protein sorting, targeting and distributionin polarized cells. Nat. Rev. Mol. Cell Biol. 9, 833–845 (2008).

7. Bryant, D. M. & Mostov, K. E. From cells to organs: building polarized tissue. Nat.Rev. Mol. Cell Biol. 9, 887–901 (2008).

8. Suzuki, A. & Ohno, S. The PAR–aPKC system: lessons in polarity. J. Cell Sci. 119,979–987 (2006).

9. Munro, E. PAR proteins and the cytoskeleton: a marriage of equals. Curr. OpinionCell Biol. 18, 86–94 (2006).

10. Nance, J. & Zallen, J. A. Elaborating polarity: PAR proteins and the cytoskeleton.Development 138, 799–809 (2011).

11. Achilleos, A., Wehman, A. M. & Nance, J. PAR-3 mediates the initial clusteringand apical localization of junction and polarity proteins during C. elegans intestinalepithelial cell polarization. Development 137, 1833–1842 (2010).

12. Totong, R., Achilleos, A. & Nance, J. PAR-6 is required for junction formation butnot apicobasal polarization in C. elegans embryonic epithelial cells. Development134, 1259–1268 (2007).

13. Daley, W. P. et al. ROCK1-directed basement membrane positioning coordinatesepithelial tissue polarity. Development 139, 411–422 (2012).

14. Etienne-Manneville, S. & Hall, A. Rho GTPases in cell biology. Nature 420,629–635 (2002).

15. Zerial, M. & McBride, H. Rab proteins as membrane organizers. Nat. Rev. Mol. CellBiol. 2, 107–117 (2001).

16. Lundquist, E. A. Small GTPases (January 17, 2006), WormBook, ed. The C. elegansResearch Community, WormBook, http://dx.doi.org/10.1895/wormbook.1.67.1,http://www.wormbook.org.

17. Golachowska, M. R., Hoekstra, D. & van, I. S. C. Recycling endosomes in apicalplasma membrane domain formation and epithelial cell polarity. Trends Cell Biol.20, 618–626 (2010).

18. Shivas, J. M., Morrison, H. A., Bilder, D. & Skop, A. R. Polarity and endocytosis:reciprocal regulation. Trends Cell Biol. 20, 445–452 (2010).

19. Prekeris, R., Klumperman, J. & Scheller, R. H. A Rab11/Rip11 protein complexregulates apical membrane trafficking via recycling endosomes. Mol. Cell 6,1437–1448 (2000).



http://www.wormbook.org

http://www.nature.com/naturecellbiology/

http://www.nature.com/naturecellbiology

http://www.nature.com/reprints

http://dx.doi.org/10.1895/wormbook.1.67.1

ART I C L E S

20. Swiatecka-Urban, A. et al. Myosin Vb is required for trafficking of the cysticfibrosis transmembrane conductance regulator in Rab11a-specific apical recyclingendosomes in polarized human airway epithelial cells. J. Biol. Chem. 282,23725–23736 (2007).

21. Langevin, J. et al. Drosophila exocyst components Sec5, Sec6, and Sec15regulate DE-Cadherin trafficking from recycling endosomes to the plasma membrane.Dev. Cell 9, 355–376 (2005).

22. Ang, A. L. et al. Recycling endosomes can serve as intermediates during transportfrom the Golgi to the plasma membrane of MDCK cells. J. Cell Biol. 167,531–543 (2004).

23. Wu, S., Mehta, S., Pichaud, F., Bellen, H. & Quiocho, F. Sec15 interacts with Rab11via a novel domain and affects Rab11 localization in vivo. Nat. Struct. Mol. Biol. 12,879–885 (2005).

24. Chen, C. C. et al. RAB-10 is required for endocytic recycling in the Caenorhabditiselegans intestine. Mol. Biol. Cell 17, 1286–1297 (2006).

25. Zhang, H. et al. Apicobasal domain identities of expanding tubular membranesdepend on glycosphingolipid biosynthesis. Nat. Cell Biol. 13, 1189–1201 (2011).

26. Hermann, G. J. et al. Genetic analysis of lysosomal trafficking in Caenorhabditiselegans. Mol. Biol. Cell 16, 3273–3288 (2005).

27. Van Fürden, D., Johnson, K., Segbert, C. & Bossinger, O. The C. elegansezrin–radixin–moesin protein ERM-1 is necessary for apical junction remodellingand tubulogenesis in the intestine. Developmental Biol. 272, 262–276 (2004).

28. McGhee, J. D. The C. elegans intestine (March 27, 2007), WormBook, ed. TheC. elegans Research Community, WormBook, http://dx.doi.org/10.1895/wormbook.1.133.1 , http://www.wormbook.org.

29. McGhee, J. D. The C. elegans intestine. WormBook 1–36 (2007).30. Nehrke, K. A reduction in intestinal cell pHi due to loss of the Caenorhabditis

elegansNa+/H+ exchanger NHX-2 increases life span. J. Biol. Chem. 278,44657–44666 (2003).

31. Casanova, J. E. et al. Association of Rab25 and Rab11a with the apical recyclingsystem of polarized Madin–Darby canine kidney cells. Mol. Biol. Cell 10,47–61 (1999).

32. Kamath, R. S. & Ahringer, J. Genome-wide RNAi screening in Caenorhabditiselegans. Methods 30, 313–321 (2003).

33. Blum, H. Models for the Perception of Speech and Visual Forms 362–380 (MITPress, 1967).

34. Thomas, P. et al. PANTHER: a browsable database of gene products organizedby biological function, using curated protein family and subfamily classification.Nucleic Acids Res. 31, 334–341 (2003).

35. Johnson, S. C. Hierarchical clustering schemes. Psychometrika 32,241–254 (1967).

36. McGary, K. L., Lee, I. & Marcotte, E. M. Broad network-based predictabilityof Saccharomyces cerevisiae gene loss-of-function phenotypes. Genome Biol. 8,R258 (2007).

37. Kardon, J. R. & Vale, R. D. Regulators of the cytoplasmic dynein motor. Nat. Rev.Mol. Cell Biol. 10, 854–865 (2009).

38. Horgan, C. P., Hanscom, S. R., Jolly, R. S., Futter, C. E. & McCaffrey, M. W.Rab11-FIP3 links the Rab11 GTPase and cytoplasmic dynein to mediate transportto the endosomal-recycling compartment. J. Cell Sci. 123, 181–191 (2010).

39. Arimoto, M. et al. The Caenorhabditis elegans JIP3 protein UNC-16 functionsas an adaptor to link kinesin-1 with cytoplasmic dynein. J. Neurosci. 31,2216–2224 (2011).

40. Balklava, Z., Pant, S., Fares, H. & Grant, B. D. Genome-wide analysis identifiesa general requirement for polarity proteins in endocytic traffic. Nat. Cell Biol. 9,1066–1073 (2007).

41. Guo, S. & Kemphues, K. J. par-1, a gene required for establishing polarity inC. elegans embryos, encodes a putative Ser/Thr kinase that is asymmetricallydistributed. Cell 81, 611–620 (1995).

42. Morton, D. et al. The Caenorhabditis elegans par-5 gene encodes a 14-3-3 proteinrequired for cellular asymmetry in the early embryo. Dev. Biol. 241, 47–58 (2002).

43. Jin, J. et al. Proteomic, functional, and domain-based analysis of in vivo 14-3-3binding proteins involved in cytoskeletal regulation and cellular organization. Curr.Biol. 14, 1436–1450 (2004).

44. Pozuelo Rubio, M. et al. 14-3-3-affinity purification of over 200 humanphosphoproteins reveals new links to regulation of cellular metabolism, proliferationand trafficking. Biochem. J. 379, 395–408 (2004).

45. Angrand, P. O. et al. Transgenic mouse proteomics identifies new 14-3-3-associatedproteins involved in cytoskeletal rearrangements and cell signaling. Mol. CellProteomics 5, 2211–2227 (2006).

46. Hurd, T. et al. Phosphorylation-dependent binding of 14-3-3 to the polarityprotein Par3 regulates cell polarity in mammalian epithelia. Curr. Biol. 13,2082–2090 (2003).

47. Benton, R. & St Johnston, D. Drosophila PAR-1 and 14-3-3 inhibit Bazooka/PAR-3 to establish complementary cortical domains in polarized cells. Cell 115,691–704 (2003).

48. Kusakabe, M. & Nishida, E. The polarity-inducing kinase Par-1 controlsXenopus gastrulation in cooperation with 14-3-3 and aPKC. EMBO J. 23,4190–4201 (2004).

49. Berdichevsky, A., Viswanathan, M., Horvitz, H. R. & Guarente, L. C. elegans SIR-2.1interacts with 14-3-3 proteins to activate DAF-16 and extend life span. Cell 125,1165–1177 (2006).

50. Sato, T. et al. The Rab8 GTPase regulates apical protein localization in intestinalcells. Nature 448, 366–369 (2007).

51. Gohla, A. & Bokoch, G. M. 14-3-3 regulates actin dynamics by stabilizingphosphorylated cofilin. Curr. Biol. 12, 1704–1710 (2002).

52. Birkenfeld, J., Betz, H. & Roth, D. Identification of cofilin and LIM-domain-containing protein kinase 1 as novel interaction partners of 14-3-3!. Biochem. J.369, 45–54 (2003).

53. Nagata-Ohashi, K. et al. A pathway of neuregulin-induced activation ofcofilin-phosphatase Slingshot and cofilin in lamellipodia. J. Cell Biol. 165,465–471 (2004).

54. Ono, K., Parast, M., Alberico, C., Benian, G. & Ono, S. Specific requirement for twoADF/cofilin isoforms in distinct actin-dependent processes in Caenorhabditis elegans.J. Cell Sci. 116, 2073–2085 (2003).

55. Croce, A. et al. A novel actin barbed-end-capping activity in EPS-8 regulates apicalmorphogenesis in intestinal cells of Caenorhabditis elegans. Nat. Cell Biol. 6,1173–1179 (2004).

56. Hüsken, K. et al. Maintenance of the intestinal tube in Caenorhabditiselegans : the role of the intermediate filament protein IFC-2. Differentiation 76,881–896 (2008).

57. Koppen, M. et al. Cooperative regulation of AJM-1 controls junctional integrity inCaenorhabditis elegans epithelia. Nat. Cell Biol. 3, 983–991 (2001).

58. Cowan, C. R. & Hyman, A. A. Asymmetric cell division in C. elegans : cortical polarityand spindle positioning. Annu. Rev. Cell Dev. Biol. 20, 427–453 (2004).

59. Boutros, M. & Ahringer, J. The art and design of genetic screens: RNA interference.Nat. Rev. Genetics 9, 554–566 (2008).

60. Provance, W. et al. Myosin-Vb functions as a dynamic tether for peripheral endocyticcompartments during transferrin trafficking. BMC Cell Biol. 9, 44 (2008).

61. Zenke, F. T. et al. p21-activated kinase 1 phosphorylates and regulates 14-3-3binding to GEF-H1, a microtubule-localized Rho exchange factor. J. Biol. Chem.279, 18392–18400 (2004).

62. Meek, S. E., Lane, W. S. & Piwnica-Worms, H. Comprehensive proteomicanalysis of interphase and mitotic 14-3-3-binding proteins. J. Biol. Chem. 279,32046–32054 (2004).

63. Chen, X. & Macara, I. G. Par-3 mediates the inhibition of LIM kinase 2 toregulate cofilin phosphorylation and tight junction assembly. J. Cell Biol. 172,671–678 (2006).

64. Kligys, K. et al. The slingshot family of phosphatases mediates Rac1 regulationof cofilin phosphorylation, laminin-332 organization, and motility behavior ofkeratinocytes. J. Biol. Chem. 282, 32520–32528 (2007).

65. Rollason, R., Korolchuk, V., Hamilton, C., Jepson, M. & Banting, G. ACD317/tetherin-RICH2 complex plays a critical role in the organization of thesubapical actin cytoskeleton in polarized epithelial cells. J. Cell Biol. 184,721–736 (2009).

66. Torkko, J. M., Manninen, A., Schuck, S. & Simons, K. Depletion of apical transportproteins perturbs epithelial cyst formation and ciliogenesis. J. Cell Sci. 121,1193–1203 (2008).

67. Desclozeaux, M. et al. Active Rab11 and functional recycling endosome are requiredfor E-cadherin trafficking and lumen formation during epithelial morphogenesis. Am.J. Physiol. Cell Physiol. 295, C545–C556 (2008).

68. Cao, J., Albertson, R., Riggs, B., Field, C. M. & Sullivan, W. Nuf, a Rab11 effector,maintains cytokinetic furrow integrity by promoting local actin polymerization. J. CellBiol. 182, 301–313 (2008).

69. Schuh, M. An actin-dependent mechanism for long-range vesicle transport. Nat. CellBiol. 13, 1431–1436 (2011).

70. Rodal, A. A., Motola-Barnes, R. N. & Littleton, J. T. Nervous wreck and Cdc42cooperate to regulate endocytic actin assembly during synaptic growth. J. Neurosci.28, 8316–8325 (2008).



http://www.wormbook.org



DOI: 10.1038/ncb2508 METHODS

METHODSNematode strains. Worm cultures and genetic crosses were carried out accordingto standard protocols71. The expression plasmids used to generate intestine-specificmarker strains were based on the pAZ110 tagging plasmid backbone72 in whichwe introduced a !3 kb sequence of the pept-1 promoter derived from the pKN111plasmid provided by Keith Nehrke30. In the multiple cloning site following thepromoter we inserted a coding sequence for an amino- or carboxy-terminalfluorescent protein tag such as GFP (derived from pPD95.79, A. Fire plasmids)or DsRed (amplified from pDsRed-N1, Clontech), as well as a PCR-amplified,genomic coding sequence of an organelle or membrane marker. All transgeneinsertions were created by biolistic transformation of unc-119 mutant worms(DP38 (unc-119(ed3) III ); ref. 72). To increase RNAi sensitivity, some singletransgenic strains were crossed into the rrf-3 deficient strain NL2099 (rrf-3(pk1426)II ; ref. 73) provided by the Caenorhabditis Genetic Center (CGC), which is fundedby the US National Institutes of Health (NIH) National Centre for ResearchResources.

The double transgenic, primary screening strain MZE1 (unc-119(ed3)III;cbgIs91[pPept-1:pept-1::DsRed;unc-119(+)];cbgIs98[pPept-1:GFP::rab-11.1;unc-119(+)]) resulted from the crossing of two marker strains, MZE91R expressingPEPT-1::DsRed (which in addition carries a deletion for rrf-3) (unc-119(ed3)III; cbgIs91[pPept-1:pept-1::DsRed;unc-119(+)]; rrf-3(pk1426) II ) and MZE98expressing GFP::RAB-11 (unc-119(ed3) III; cbgIs98[pPept-1:GFP::rab-11.1;unc-119(+)]). The double transgenic strains for secondary assay analysis resultedfrom crossing MZE91R with other GFP–organelle marker strains generatedaccording to the description above or provided by the CGC. The secondaryassay strains are MZE2 (unc-119(ed3) III; cbgIs91[pPept-1:pept-1::DsRed;unc-119(+)]; cbgIs99[pPept-1:GFP::rab-5;unc-119(+)]), MZE3 (unc-119(ed3) III;cbgIs91[pPept-1:pept-1::DsRed;unc-119(+)]; cbgIs100[pPept-1:amans-2::GFP;unc-119(+)]), MZE4 (unc-119(ed3) III; cbgIs91[pPept-1:pept-1::DsRed;unc-119(+)];cbgIs103[pPept-1:SP12::GFP;unc-119(+)]) (AMANS-2::GFP and SP-12::GFP con-structs were derived from plasmids provided by M. Rolls; ref. 74; Harvard MedicalSchool) and MZE 5 (unc-119(ed3) III; cbgIs91[pPept-1:pept-1::DsRed;unc-119(+)];pwIs50[lmp-1::GFP +Cb-unc-119(+)]). The original LMP-1::GFP-expressingRT258 strain75 used to generate MZE5 was provided by the CGC.

The rrf-3 deficient, GFP::RAB-11-expressing strain MZE98R (unc-119(ed3)III; cbgIs98[pPept-1:GFP::rab-11.1;unc-119(+)]; rrf-3(pk1426) II ) was used fortotal worm protein extracts on single or double RNAi with par-5, and forimmunofluorescence. The ftt-2-deletion strain MT14355 (ftt-2(n4426) X.; ref. 76)was obtained from the CGC (deletion confirmed by PCR). MZE6 (unc-119(ed3) III;cbgIs98[pPept-1:GFP::rab-11.1;unc-119(+)]); hum-2(ok596)) results from crossingMZE98R and RB801, a hum-2-deletion strain provided by the CGC (deletionconfirmed by PCR). The PGP-1::GFP strain was provided by K. Sato77 (GunmaUniversity). The RT348 strain expressing GFP::RME-1 was provided by B. Grant75(Rutgers University). The GFP::CDC-42-expressing strain TH51 (3-9) (unc-119(ed3)III; ddEx14[pie-1p::GFP::R07G3.1(genomic);unc-119(+)]) was provided bytheHyman laboratory (Max Planck Institute ofMolecular Cell Biology andGenetics,MPI-CBG).

Feeding clones for RNAi. The genome-wide RNAi screen was carried out by thefeedingmethod described previously78 using the Ahringer feeding library79 providedby Source BioScience LifeSciences UK Limited with 16,757 transformed Escherichiacoli clones (strain HT115). The clone-set used for the emb screen comprised1,037 selected feeding clones of this RNAi feeding library and was provided byT. Hyman (MPI-CBG). The unc-60a/b feeding clone was a gift from Carrie Cowan(Research Institute of Molecular Pathology, Vienna). The negative control forRNAi was an HT115 feeding clone transformed with L4440 vector encoding forscrambled dsRNA without a corresponding target in the worm genome (referredto as feeding control). Self-cloned RNAi constructs using the L4440 feeding vectorinclude F07F6.4 (the positive RNAi control in the primary screen), ced-10/rac1,rho-1 (two constructs) and nud-2. Generally, we subcloned !1 kilobase (kb) ofgenomic DNA consisting mostly of exons into L4440 using the following primers:F07F6.4-FW-Bgl-II 5%-CATAGATCTGAAATCTTCTTCCT-3%; F07F6.4-REV-Hind-III 5%-CATAAGCTTCTCCATTTGCTTTC-3%; ced-10-FW-Nhe1 5%-GACG-CTAGCGTATTCGACAACTACTCAGCAAATG-3%; ced-10-REV-Xma1 5%-GACC-CCGGGCATCTTAGAGCACCGTACACTTG-3%; rho-1-FW1-Nhe1 5%-GACGCT-AGCCACATAGAGATAGATGGCTGCG-3%; rho-1-Rev1-Xma1 5%-GACCCCGGG-CACAATTTCTTCATCCAATCCGC-3%; rho-1-FW2-Nhe1 5%-GACGCTAGCTCT-GCAGGTCGAACTTGCTCTATG-3%; rho-1-REV2-Xma1 5%-GACCCCGGGCATC-TCTCAATTCGGTGCTACTGC-3%; nud-2 FW-Nhe1 5%-GACGCTAGCGATCGAC-AGGACGATTCCCGTG-3%; nud-2 REV-Xma1 5%-GACCCCGGGGTTAAGCCCG-TGTCGTTGTAAGATG-3%.

PCR validation of bacterial clones. To confirm the sequence of identified RNAiclones from the feeding library79, we isolated the L4440 plasmidDNA and sequencedits dsRNAcoding regionwith a forward (5%-GACCGAGCGCAGCGAGTCAGTG-3%)or reverse (5%-CACTGGCCGTCGTTTTACAAC-3%) sequencing primer.

Single- and double-RNAi feeding experiments. Single bacterial clones wereinoculated in 4ml Luria broth (LB) medium (plus 100 µgml$1 ampicillin (Amp)and 5 µgml$1 tetracycline (Tet)), grown for 8 h to overnight at 200 r.p.m., andthe bacteria pelleted and resuspended in &500 µl NGM+ (nematode growthmedium containing 100 µgml$1 Amp, 5 µgml$1 Tet and 1mM isopropyl-$-d-thiogalactoside) of which&150 µl were seeded on 6 cmNGM+ plates. Seeded plateswere dried under an airflow hood and induced in the dark for 8 h to overnight atroom temperature. For double-RNAi experiments with par-5, we chose a feedingclone ratio of 1:3 (one part (par-5) plus two parts (second gene x)). Applied to thereference sample (par-5+ FC), this 1:3 ratio resulted in a par-5 RNAi phenotypefrequency of !50%, which enabled detection of suppression as well as exacerbationof the par-5 RNAi phenotype on double RNAi. The feeding clones were inoculatedin an overnight culture of 20ml (par-5) or 5ml (clone x) LB medium (plus100 µgml$1 Amp and 5 µgml$1 Tet). The bacteria were pelleted and resuspended in2.5ml (par-5) or 0.4ml (clone x) NMG+medium. For OD600 measurements, theseresuspended bacteria were diluted 1:80 in 1ml NGM+ medium. The OD600 ratio ofpar-5 to clone x was determined and used to calculate the ‘volume x’ of clone x fora 1:3 mixture with the par-5 bacteria (one part (50 µl par-5 bacteria) plus two parts(OD600-corrected ‘volume x’ of ‘clone x’ bacteria)). These volumes were multipliedby the number of feeding plates used per double-RNAi sample, that is at least bytwo. Per 6 cm NGM+ feeding plate, &150 µl of the final 1:3 double-RNAi mixturewas seeded. One plate or half of the plates of each sample with adult, double-RNAi-treated worms was used to produce total worm protein extracts to confirm similarlevels of PAR-5 depletion by western blot (Fig. 6a and Supplementary Fig. S8b,c).The other plate or half of the plates was examined by epifluorescence microscopyto determine the par-5 RNAi phenotype frequency, that is the mean percentages ofintestines with one or more peripheral GFP::RAB-11 recycling endosome clusters.

Immunofluorescence stainings on dissected intestines. Approximately 30–40young adult worms were picked and transferred to a 10 µl droplet of cutting buffer(1#M9 or 1#PBS buffer containing 0.001% Tween-20 + 0.25mM levamisole(($)-tetramisole hydrochloride, Sigma-Aldrich) on a coverslip for decapitation.Dissected worms were transferred from cutting buffer into a 0.2ml Thermotube(‘PCR tube’, a thin-walled tube with flat or domed cap, ABgene), left to settle bygravity or spun down in a microcentrifuge (2,000g , 30 s) and the buffer exchangedfor fixation buffer (either 1.25–4% freshly diluted paraformaldehyde (PFA) (in1#PBS, 0.001% Tween-20) for subsequent F-actin staining with fluorescentphalloidins, or ice-cold 100% methanol (containing 0.001% Tween-20)) to fixthe samples for 10–15min at room temperature or 5min on dry ice, respectively.Methanol fixation of dissected worms was preferentially used for all primaryantibodies except anti-CASP, which gave better staining results on PFA fixation.PFA fixation, which is obligatory for fluorescent phalloidin staining of F-actin,diminished the apical enrichment of GFP::RAB-11-labelled recycling endosomes.

After PFA fixation, samples were either permeabilized for 10min in 1#PBS+0.1% Tween-20 (for some antibodies with 0.01% Tween-20 or Triton X-100), ordirectly blocked in blocking solution (1#PBS, 0.001% Tween-20, 2% BSA) on arotor wheel for 1 h with three buffer changes. Post blocking, samples were incubatedovernight at 4 'C on a rotor wheel with primary antibody diluted in blocking buffer(1#PBS, 0.001-0.01% Tween-20, 2% BSA; 50–100 µl added per PCR tube).

Primary antibodies for immunofluorescence and dilutions: anti-CASP (1:2,000)(gift from Sean Munro, Cambridge; ref. 80); anti-ERM-1 (1:200) (gift from OlafBossinger, Rheinisch-Westfälische Technische Hochschule Aachen; ref. 81); anti-RAB-11 and anti-RAB-5 (1:500–1:1,000) (gift fromAnne Spang, University of Basel;ref. 82); MH33 (anti-IFB-2, 1:100) andMH27 (anti-AJM-1, 1:500) (the monoclonalantibodies MH33 and MH27 developed by R. H. Waterstone were obtained fromthe Developmental Studies Hybridoma Bank developed under the auspices of theUS National Institute of Child Health and Human Development and maintained bytheUniversity of Iowa, Department of Biology); anti-#-tubulin (cloneDM1A, 1:500,Sigma-Aldrich), anti-"-tubulin (1:30,000; ref. 83); anti-EPS-8 (1:50) (gift from PierPaolo Di Fiore, Italian FIRC Institute of Molecular Oncology–European Instituteof Oncology Campus, Milan, Italy; ref. 84); anti-PKC-3 (1:1,000, PKC! (C-20)sc-216, Santa Cruz Biotechnology). Anti-PAR-6 (1:5,000) and anti-PAR-3 (serum1:1,500–1:3,000) were gifts from T. Hyman (MPI-CBG; both described in ref. 85).

Before addition of the secondary antibodies, samples were washed 3#20min atroom temperature in washing buffer (1#PBS, 0.001–0.01%Tween-20). Alexa Fluorconjugated secondary antibodies (Alexa Fluor 405/488/555/647 goat anti-mouse orgoat anti-rabbit IgG (H+ L), Molecular Probes) were diluted 1:500 in blockingsolution and PCR tubes were incubated for 2 h at room temperature on a rotorwheel, followed by 3#20min washes at room temperature. For staining of F-actinwith fluorescent phalloidin (rhodamine phalloidin (Cytoskeleton) or Alexa Fluor488 phalloidin (Molecular Probes)), this was diluted 1:140 in 1#PBS+ 0.001%Tween-20 and applied during secondary antibody incubation or after PFA fixationfor 1–1.5 h in the dark. After washing 3# 20min, dissected worms were left tosettle, the supernatant removed and theworms transferredwith 10–20 µl Vectashield(Vector Labs) to coverslips. The coverslips were adhered with 2% agarose pads andsealed with nail polish.

NATURE CELL BIOLOGY


METHODS DOI: 10.1038/ncb2508

Antibodies for western blot analysis and confirmation of RNAi efficiency.Western blot analysis was carried out with total worm protein extracts afterseparation by SDS–PAGE. RNAi-treated, young adult worms were removed fromfeeding plates either by picking 50 worms or by washing off the plate with 0.5–1ml1#M9 buffer + 0.001% or 0.01% Tween-20, transferred to 1ml reaction tubes,left to settle for 5min on ice by gravity or centrifuged for 1min at 400g , andwashed two or three times to remove bacteria. The volume over the settled wormswas reduced to 20 or 40 µl (marked on the tube), supplemented with Laemmlibuffer, put at 95 'C for 2min, sonicated for 10min, put at 95 'C for 2min andcentrifuged at 20,000g for 10min to pellet residual debris before loading. Antibodiesused for western blot probing included anti-PAR-5 antibodies (gifts from AndyGolden42 (US National Institute of Diabetes and Digestive and Kidney Diseases)and T. Hyman85 (MPI-CBG), 1:3,000 and 1:5,000, respectively), mouse anti-actin(clone C4; Chemicon,Millipore) used at 1:40,000 and anti-RHO-1 (ref. 86; gift fromT. Hyman (MPI-CBG)). For CDC-42, RAC-1 and PAR-3, no suitable antibodiesfor western blot probing were available. Therefore, cdc-42 RNAi efficiency wasassessed bymeans of the GFP::CDC-42-expressing strain TH51 (gift fromT.Hyman(MPI-CBG)) and anti-GFP probing (antibody generated in house by DavidDrechsel (MPI-CBG); Supplementary Fig. S8c). Efficient rac-1 RNAi correlatedwith reduced GFP::RAB-11 fluorescence intensities (our unpublished observations).Efficient PAR-3 depletion correlated with previously described phenotypes, thatis perturbed germline morphologies and egg laying defects (judged by differentialinterference contrast microscopy). Furthermore, AJM-1 staining revealed perturbedspermatheca morphology in PAR-3-depleted worms, in line with previous reports87.

Confocal laser scanning microscopy of worms during secondary assays andpar-5 RNAi phenotype analysis. Approximately !30 live RNAi-treated wormswere collected manually from feeding plates, paralysed in 10mM levamisole andmounted on a 2% agarose pad. Fluorescent images were obtained using a Zeiss LSMMeta or Zeiss LSM DuoScan microscope (Carl Zeiss MicroImaging) with a #63numerical aperture 1.4 Oil Plan-Apochromat objective or a#40 numerical aperture0.45 water emulsion Plan-Apochromat objective with a differential interferencecontrast prism, respectively. Signals after excitation with 405 nm and band-pass420–480 filter emissionwere used to identify broad-spectrumLRO autofluorescence.DsRed and GFP signals were recorded in a sequential manner by frame orline-scan. DsRed was excited before GFP at 543 nm (LSM Meta) or 561 nm (LSMDuoScan) and emission collected with band-pass 585–615 (LSMMeta) or long-pass575 (LSM DuoScan) filters. GFP signals were measured with 488 nm excitationand band-pass 505–530 emission filters. Imaging settings were adjusted for eachsample to avoid pixel saturation. The settings were carefully tested with singleand double transgenic lines to exclude bleed-through events between GFP andDsRed. Only DsRed or GFP signals that did not correlate with autofluorescencesignals were considered as specific. Images were processed and analysed usingFiji (http://fiji.sc/wiki/index.php/Fiji), an image-processing package based on NIHImageJ. If indicated, brightness and contrast were adjusted for better visibility usingPhotoshop (Adobe Systems).

Epifluorescence microscopy in secondary assays and par-5 RNAi analysis.To analyse selected RNAi clones for LRO phenotypes, we used the epifluorescentAxio Imager Z1 microscope (Carl Zeiss MicroImaging; filter set, excitation 365 nm,beam splitter FT 395, emission band-pass 445/50). The par-5 RNAi phenotypefrequency on double RNAi was validated by counting phenotypic worms withone or more GFP::RAB-11-labelled recycling endosome cluster per intestine anddetermination of the percentage of phenotypic worms per sample. Worms fromeach double-RNAi feeding plate sample were washed three times in 1#M9 (+0.01%Tween-20), resuspended in 250 µl, of which 120 µl was transferred into separate wellson a ‘µ-Slide 8 well’ (ibiTreat, ibidi). Before phenotypic analysis, live worms wereparalysed by adding 30 µl 100mM sodium azide (yielding 20mM final NaN3) toprevent twitching. Wild type and par-5 phenotypic worms were counted manuallyusing the Zeiss Axiovert 200M (Carl Zeiss MicroImaging), an inverted wide-fieldmicroscope, equipped with a SPOT RT camera (Diagnostic Instruments) and fittedwith a Zeiss Plan-Apochromat #10 numerical aperture 0.45 objective. For analysisof rhodamine phalloidin-stained F-actin, the Zeiss ECPlan-Neofluar#40 numericalaperture 0.75 objective was used.

Intensity profile extraction. To compute the ectopic:apical intensity ratios ofrhodamine phalloidin-stained F-actin and GFP::RAB-11 in control, par-5 and arp-1RNAi-treated intestines, we collected intensity profiles along 20-pixel-wide linesdrawn perpendicularly to the worm intestines in Fiji (http://fiji.sc/wiki/index.php/Fiji). Both actin and GFP::RAB-11 intensity profiles were measured on the sameline. Each profile contains a set of (xi, yi) couples, indicating the position along theline and the associated intensity value. To assess differences in the mean peripheral(ectopic) and apical F-actin or GFP intensities, we acquired the line-based intensityprofiles (or ‘intensity profiles’) only at sites where the peripheral patches were seenat the outmost margin of the enterocytes or intestine and optically well separatedfrom the actin-rich apical domain. Furthermore, regions with more out-of-focus

light or close-by, overly bright tissues (such as pharynx or muscles) were excluded.Intensity profiles were acquired with the same criteria in control intestines lackingectopic patches and peripherally clustered recycling endosomes.

Intensity profile quantification. The same procedure was applied for bothF-actin and GFP::RAB-11 intensity quantification (implemented in MATLAB,MathWorks). Intensity profiles were first normalized to the same unit length alongthe x axis, to compensate for different intestinal widths, and then equalized along they axis through baseline subtraction, to correct for different background intensities. Amedian filter was carried out on the normalized curves and peaks were detected witha peak detection algorithm in three different regions: from left and right each 25%covering the ‘basal’ (or ectopic) parts of the profile, and 50% covering the central,apical part of the profile. Peaks were identified as local maxima in one of these threeareas of the smoothed normalized profile. We subsequently manually controlledthe correct assignment of apical versus ectopic peaks. Peak identification enabledthe extraction of quantitative features such as position, intensity and ectopic:apicalintensity ratios. Peak position indicates the relative position of each peak on the xaxis (intestinal width); peak intensity indicates the absolute marker intensity; theintensity ratio indicates the ratio of intensity in the basal, ectopic peaks to the central,apical ones. To determine the width and intensity of each peak in a robust way, wefitted them with a peak fitting procedure based on Gaussian fitting. The extractedvalues were then averaged and compared across the different knockdown conditionsto quantify apical and basal mean peak intensities, and their ratio.

Total actin quantification. The total amount of F-actin was assessed throughregion-of-interest-based quantification on rhodamine phalloidin-stained intestinesof selected control and par-5 RNAi-treated, phenotypic animals. The area ofapproximately two to three segments of each imaged intestine considered for theintensity profile analysis was traced, and the mean actin intensity determined usingFiji software (http://fiji.sc/wiki/index.php/Fiji). Out-of-focus areas and areas closeto overly exposed tissues were excluded from analysis.

Statistical analysis. The statistical significance of intensity measures (Fig. 4d,f,g)and phenotype frequencies (Fig. 6b) was assessed withWilcoxon rank sum tests. Thesignificance of functional enrichments in the hit set (Fig. 2c) was assessed with thehypergeometric distribution. Results are expressed as mean± s.e.m.

71. Brenner, S. The genetics of Caenorhabditis elegans. Genetics 77, 71–94 (1974).72. Praitis, V., Casey, E., Collar, D. & Austin, J. Creation of low-copy integrated transgenic

lines in Caenorhabditis elegans. Genetics 157, 1217–1226 (2001).73. Simmer, F. et al. Genome-wide RNAi of C. elegans using the hypersensitive rrf-3

strain reveals novel gene functions. PLoS Biol. 1, E12 (2003).74. Rolls, M. M., Hall, D. H., Victor, M., Stelzer, E.H. & Rapoport, T. A. Targeting of rough

endoplasmic reticulum membrane proteins and ribosomes in invertebrate neurons.Mol. Biol. Cell 13, 1778–1791 (2002).

75. Chen, C. C. et al. RAB-10 is required for endocytic recycling in the Caenorhabditiselegans intestine. Mol. Biol. Cell 17, 1286–1297 (2006).

76. Berdichevsky, A., Viswanathan, M., Horvitz, H. R. & Guarente, L. C. elegans SIR-2.1interacts with 14-3-3 proteins to activate DAF-16 and extend life span. Cell 125,1165–1177 (2006).

77. Sato, T. et al. The Rab8 GTPase regulates apical protein localization in intestinalcells. Nature 448, 366–369 (2007).

78. Timmons, L., Court, D. L. & Fire, A. Ingestion of bacterially expressed dsRNAs canproduce specific and potent genetic interference in Caenorhabditis elegans. Gene263, 103–112 (2001).

79. Kamath, R. S. & Ahringer, J. Genome-wide RNAi screening in Caenorhabditiselegans. Methods 30, 313–321 (2003).

80. Gillingham, A. K., Pfeifer, A. C. & Munro, S. CASP, the alternatively spliced productof the gene encoding the CCAAT-displacement protein transcription factor, is a Golgimembrane protein related to giantin. Mol. Biol. Cell 13, 3761–3774 (2002).

81. Van Fürden, D., Johnson, K., Segbert, C. & Bossinger, O. The C. elegansezrin–radixin–moesin protein ERM-1 is necessary for apical junction remodellingand tubulogenesis in the intestine. Dev. Biol. 272, 262–276 (2004).

82. Poteryaev, D., Fares, H., Bowerman, B. & Spang, A. Caenorhabditis elegans SAND-1is essential for RAB-7 function in endosomal traffic. EMBO J. 26, 301–312 (2007).

83. Hannak, E., Kirkham, M., Hyman, A. A. & Oegema, K. Aurora-A kinase isrequired for centrosome maturation in Caenorhabditis elegans. J. Cell Biol. 155,1109–1116 (2001).

84. Croce, A. et al. A novel actin barbed-end-capping activity in EPS-8 regulates apicalmorphogenesis in intestinal cells of Caenorhabditis elegans. Nat. Cell Biol. 6,1173 (2004).

85. Hoege, C. et al. LGL can partition the cortex of one-cell Caenorhabditis elegansembryos into two domains. Curr. Biol. 20, 1296–1303 (2010).

86. Schonegg, S. & Hyman, A. A. CDC-42 and RHO-1 coordinate acto-myosincontractility and PAR protein localization during polarity establishment in C. elegansembryos. Development 133, 3507–3516 (2006).

87. Aono, S., Legouis, R., Hoose, W. & Kemphues, K. PAR-3 is required forepithelial cell polarity in the distal spermatheca of C. elegans. Development 131,2865–2874 (2004).

NATURE CELL BIOLOGY


http://fiji.sc/wiki/index.php/Fiji




Correction notice Nature Cell Biology (2012) | doi:10.1038/ncb2508 Caenorhabditis elegans screen reveals role of PAR-5 in RAB-11-recycling endosome positioning and apicobasal cell polarity Julia Franziska Winter, Sebastian Höpfner, Kerstin Korn, Benjamin O. Farnung, Charles R. Bradshaw, Giovanni Marsico, Michael Volkmer, Bianca Habermann & Marino Zerial In the version of this Article initially published online, PEPT-2::DsRed should have been PEPT-1::DsRed in Supplementary Fig. S1d. In Supplementary Fig. 3e, some text elements in the column for gene annotations were erroneous or lost. The legend for Supplementary Fig. S6 should have referred to !-tubulin rather than "-tubulin. These errors have been corrected.


S U P P L E M E N TA RY I N F O R M AT I O N

WWW.NATURE.COM/NATURECELLBIOLOGY 1

DOI: 10.1038/ncb2508

Figure S1 GFP::RAB-11 labelled endosomes share characteristics with Rab11-endosomes of higher metazoan epithelia. (a-f) Confocal images of L4 to young adult (YA) stage worms of the screening strain. (a,b) In L4 stage animals, GFP::RAB-11 labels REs enriched along the apical domain (arrow in a) as well as bulkier organelles that enrich subapically (dashed arrow in a) and become less apparent in YA stage animals (b). (c, d) Fractions of the apical membrane protein PEPT-1::DsRed can be observed in the subapical, bulkier GFP::RAB-11 positive REs (dashed arrows). This accounts for on average 20% (n=11/54) of L4 to YA-stage MZE1 worms. For endogenous RAB-11 labelled REs see Supplementary Information, Fig. S5a online. (e) Depletion of !-tubulin abrogates apical enrichment, and GFP::RAB-11-REs scatter. (f) Depletion of SEC-15, a Rab11-effector and component of the exocyst complex implicated in basolateral vesicle transport and tethering,

increases the apical enrichment of GFP::RAB-11-REs. (g-l) GFP::RAB-11 expressing intestines immunostained for CASP (g-i), or RAB-5 (j-l) to visualise the distinct distribution of REs versus Golgi and early endosomes, respectively. GFP::RAB-11-REs tend to be more apically enriched than early endosomes, in particular when compared to Golgi bodies, which are more evenly distributed. Note that the endotube is not a round, but rather folded tubing that can alternate along the vertical and longitudinal axes of the intestine. Thus, depending on the angle of optical sectioning of the endotube, the gut lumen’s width and endotube morphology can vary. The optical section in (g) cuts the endotube at the level of the enclosing apical membrane at which GFP::RAB-11-labelled REs accumulate. The optical section in (j-l) cuts a flatly folded endotube more centrally, showing parts of its lumen. Scale bars indicate 20 "m.

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

!" #$ %"&"'"

/012345 &*+,(%&'(6!"#$$%&'()) /012345

7,*+1281+8439:8;<==31>1*+4258",?<218.)

(")

!"#

$$%&'

())

!"#$%%@&,8

&'($%)%@&,

*" +"

AB8;+4?1

C&;+4?1

!"#

$$%&'

())

!"#$$%&'())

#D#E()$$F;%1G

,-



2 WWW.NATURE.COM/NATURECELLBIOLOGY


GFP::RAB-11 (REs)

autofluorescent LROs

6 diffuse GFP::RAB-11 7 apical morphology changed

14 less AF10 more GFP::RAB-11

9 less GFP::RAB-11

sec-15 (exocyst complex) pbs-4 (protein metabolism)

krs-1 (t-RNA-synthetase) pbs-5 (protein metabolism)

control RNAi

control RNAi

Membrane traffic24%

Unknown12%

Signaling11%

Carbohydrate metabolism9%

Mitochondrial function

7%

Cytoskeleton associated7%

Lipid metabolism7%

Transport4%

RNA processing 4%

Translation18%

Membrane traffic15%

Protein Metabolism14%

RNA processing10%

DNA binding

7%Cytoskeletonassociated

7%

Lipid metabolism5%

Mitochondrial fct.

4%Signaling4%

Unknown3%

other 13%

Biological functions in the GWS hit set (147 hits) Biological functions in the emb hit set (209 hits)

b

Identity of other* functions in decreasing order of frequency (each <3%): GWS-others* = Metabolism, DNA binding, Protein Metabolism, Transcription, Extracellular Matrix, Translation, Nuclear Transportemb-others* = Transcription, Carbohydrate Metabolism, Nuclear Transport, Extracellular Matrix, Transport, Metabolism

c

Num

ber o

f gen

es

Worm length as an automated measure for RNAi efficiency of emb and ste genes in the GWS

Average of total worm length per experiment (AU)

0

130

260

390

520

650

780

910

1040

1170

1300

1430

1560

1690

1820

1950

2080

2210

100

200

300

400

500

600

% of genes previously reported to cause emb or ste phenotypes

> 50%41-50%31-40%21-30%11-20%<10%

aWinter et al., supplementary Figure S2

F08F8.2 (HMG-CoA-reductase)

other* 15%

Figure S2 Remaining phenotypic defect categories, biological functions of primary screen hits, and automated assessment of RNAi efficiency in the GWS. (a) Complementary to Figure 2a: Remaining 5 of 14 phenotypic defect categories. For each phenotypic category, a representative image of a primary screen RNAi experiment is shown to indicate the characteristic changes of marker distribution or intensity compared to control RNAi. An altered morphology or distribution of apically enriched REs (category 7) is predominantly caused by exocyst complex components. The strongest increase or gain of RAB-11::GFP-intensity (category 10) is caused by depletion of genes encoding t-RNA-synthetases. Reduced fluorescence intensities for the individual markers (categories 9, 14) occur frequently upon depletion of genes required for protein metabolism. Scale bar in (a) is 50 µm, and 20 µm for the close-up. (b) Comparison of biological functions present in the individual hit sets obtained in the GWS (147 hits) and the

emb-screen (209 hits). Absolute contributions of biological functions to each hit set are shown. The GWS hit set comprises mostly ‘membrane trafficking’ genes (24%) and genes of yet unknown biological functions (in total 12%, of which 30% are conserved) with a potential role in epithelial membrane trafficking and cell polarity. The emb- screen hit set comprises more genes of core biological functions such as ‘translation’ and ‘protein metabolism’ that affect general biological processes. ‘Membrane trafficking’ is the second most abundant biological function (15%) in the emb hit set. (c) Worm length as an automated measure of RNAi efficiency in the GWS for emb and ste genes. The average total worm length per well was computed by automated image analysis for each RNAi experiment in the GWS. This measure results from the number of worms per well and their average length. Short worm length values occurred frequently upon RNAi of established emb or ste genes (acc. to WormBase release WS152).




Phenotype profiles (14 categories):

3210Phenotypic strength:

yyp p ( g )

5) s

catte

red

6) d

iffus

e 7

) api

cal m

. cha

nged

8) a

ccum

ulat

ed

10)

exc

ess

11)

enl

arge

d

1) ‘

dots

’ 2

) fuz

zy

3) l

ess

9) l

ess

12)

diff

use

13)

lum

inal

1

4) le

ss

4) b

asal

PEPT-1::DsRed

GFP::RAB-11 AF/LROs

a c

b

Group assigned by SECOND-ARY ASSAYS

PRIMARY SCREEN

Functional predictions for co-clustering genes

Winter et al., Supplementary Figure S3

Secondary assay group

Lysosomal route & integrity

AP-3 complex, HOPS-complex, glo-4

2

Secretion arf-3, KDEL-R, sec-24.2 (COP-II)

4

Endocytic & degradative routes

rab-5, rab-7, ESCRT-, v-H-ATPase (V0+1) complex components

3

Microtubule cytoskeleton and dynein/dynactin components

Endosome positioning & transport

5

Predominantly affectingRAB-11 REs?

Exocyst complex 6

1Core function affecting various (traffic) processes

Genes for transcription, protein- or lipid metabolism

Functionally known genes with characteristic phenotype profiles

Figure S3 Hierarchical clustering reveals groups of genes with similar phenotype profiles and allows functional predictions for uncharacterised genes. a-c (a) Compact hierarchical clustering dendrogram of phenotype profiles of 189 primary-screen candidates, excluding all candidate genes implicated in biosynthesis and metabolism (such as ‘translation’ or ‘protein metabolism’, see Supplementary Information, Table S1 online). Phenotype strength ranges from 1 (weak, light grey) to 3 (strong, black). Along the x-axis of the dendrogram, the length of edges indicates the correlation between clusters. Coloured frames highlight clusters with functionally characteristic phenotype profiles upon RNAi of established trafficking regulators, e.g. gene implicated in secretion (green) and minus-end directed microtubule-based transport (blue), or degradation (orange). Functionally yet unknown,

co-clustering genes are predicted to exert a similar function. (b) Established traffic regulators selected from framed clusters (in a) with the inferred, functional predictions for co-clustering genes. (c) Following secondary assay analysis of 60 selected genes, these were assigned to one of the six groups as indicated by coloured bars. The matching colours of bars (c) and frames (a) indicate confirmed functional predictions. d and e Two hierarchical clustering dendrograms generated for the selected subset of 189 genes (d) and for the full set of 356 hits (e). The clustering dendrograms indicate for each candidate the gene name, biological function and the description either from NCBI KOG (K, version 090828), Wormbase (W, release WS204), or InterPro Domain (I, version 22.0) in that particular priority. The colour coding of the secondary assay groups corresponds to the colours used in Figure S3c.




1) Opt-2 dots

2) Opt-2 fuzzy

3) Opt-2 loss

4) Rab-11 basal

5) Rab-11 scattered

6) Rab-11 diffuse

7) Rab-11 apically changed

8) Rab-11 aggregated

9) Rab-11 loss

10) Rab-11 excess

11) AF aggregated

12) AF diffuse

13) AF luminal

14) AF loss

2nd assays

OMIM Disease

pgp-2 Transport member of the ABC transporter, orthologous to human MDR1 (ABCB1) (W)

apd-3 Membrane traffic Vesicle coat complex AP-3, delta subunit (K)

apb-3 Membrane traffic adaptin, orthologous to the beta3 subunit of adaptor protein complex 3 (AP-3) (W)

aps-3 Membrane traffic adaptin (AP-3) (W)

apm-3 Membrane traffic Clathrin-associated protein medium chain (K)

C28H8.4 Membrane traffic ER lumen protein retaining receptor (K)

Y37D8A.22 Membrane traffic Vacuole membrane protein VMP1 (K)

T08D2.1 Membrane traffic emp24/gp25L/p24 family of membrane trafficking proteins (K)

K08E3.5 Carbohydrate metabolism UDP-glucose pyrophosphorylase (K)

F07F6.4 Membrane traffic Predicted GTPase-activating protein (K)

H43E16.1 Unknown Unnamed protein (K)

arf-3 Membrane traffic GTP-binding ADP-ribosylation factor Arf1 (K)

F13B9.1 Unknown immunoglobulin/MHC-like domain, glutamine/asparagine-rich domain (W)

sec-24.2 Membrane traffic Vesicle coat complex COPII, subunit SEC24/subunit SFB2 (K)

T09E8.3 Membrane traffic ER vesicle integral membrane protein involved in establishing cell polarity, signaling and protein degradation (K)

phy-2 Carbohydrate metabolism Prolyl 4-hydroxylase alpha subunit (K)

gbf-1 Signaling SEC7-like (I)

uso-1 Membrane traffic Target SNARE coiled-coil region (I)

tnt-2 Cytoskeleton associated Troponin (K)

F40G9.1 Unknown Sec20 (I)

M57.2 Lipid metabolism Protein geranylgeranyltransferase type II, alpha subunit (K)

lpin-1 Lipid metabolism Protein involved in plasmid maintenance/nuclear protein involved in lipid metabolism (K)

lips-8 Lipid metabolism Triacylglycerol lipase (K)

srw-103 Signaling 7-transmembrane olfactory receptor (K)

T06H11.2 Unknown T06H11.2 gene (K)

srj-19 Signaling 7-transmembrane olfactory receptor (K)

nhr-80 Signaling Hormone receptors (K)

Y49E10.21 Unknown Y49E10.21 gene (K)

T14G10.7 Lipid metabolism GPI transamidase complex, GPI17/PIG-S component (K)

tba-4 Cytoskeleton associated Alpha tubulin (K)

lis-1 Cytoskeleton associated WD40 repeat-containing protein (K)

Y19D2B.1 Cytoskeleton associated Alpha tubulin (K)


tbb-1 Cytoskeleton associated Beta tubulin (K)

F47B10.9 Unknown Kelch related (I)

C45E5.2 Carbohydrate metabolism Chitinase (K)

tbb-2 Cytoskeleton associated Beta tubulin (K)

dyrb-1 Cytoskeleton associated Dynein-associated protein Roadblock (K)

F55C12.1 Membrane traffic Rab11-binding domain s FIP domains C-terminal (I)

dnc-2 Cytoskeleton associated Putative dynamitin (K)

dlc-1 Cytoskeleton associated Dynein light chain type 1 (K)

cap-1 Cytoskeleton associated F-actin capping protein, alpha subunit (K)

dyci-1 Cytoskeleton associated Cytoplasmic dynein intermediate chain (K)

cap-2 Cytoskeleton associated F-actin capping protein, beta subunit (K)

dnc-1 Cytoskeleton associated Microtubule-associated protein dynactin DCTN1/Glued (K)


C23G10.8 Unknown Unnamed protein (K)

dylt-1 Cytoskeleton associated Dynein light chain (K)

K08F9.3 Carbohydrate metabolism Glycoside hydrolases family 18s catalytic domain gene (I)

nhr-227 Signaling Nuclear hormone receptors ligand-bindings core,Zinc fingers nuclear hormone receptor-type (I)

paa-1 Signaling Protein phosphatase 2A regulatory subunit A and related proteins (K)

W09D10.5 Carbohydrate metabolism Galactosyltransferases (K)

F17E9.3 Unknown Unnamed protein (K)

gei-4 Cytoskeleton associated Unnamed protein (K)

spl-1 Lipid metabolism Glutamate decarboxylase/sphingosine phosphate lyase (K)

M01A10.3 Carbohydrate metabolism Oligosaccharyltransferase, delta subunit (ribophorin II) (K)

H37N21.1 Signaling Protein kinases core,Tyrosine protein kinase (I)

dhc-1 Cytoskeleton associated Dyneins, heavy chain (K)

nud-1 Cytoskeleton associated Nuclear distribution protein NUDC (K)

arp-1 Cytoskeleton associated Actin and related proteins (K)

par-5 Signaling Multifunctional chaperone (14-3-3 family) (K)

F25B4.6 Lipid metabolism Hydroxymethylglutaryl-CoA synthase (K)

F08F8.2 Lipid metabolism 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) reductase (K)

H41C03.1 Membrane traffic Phosphatidylinositol transfer protein SEC14 and related proteins (K)

T09A5.11 Carbohydrate metabolism Oligosaccharyltransferase, beta subunit (K)

ril-1 Unknown none available (K)

str-171 Signaling 7-transmembrane olfactory receptor (K)

B0280.11 Lipid metabolism Protein-tyrosine phosphatases receptor/non-receptor type (I)

Y43F4B.5 Carbohydrate metabolism Phosphoglucomutase/phosphomannomutase (K)

par-6 Signaling Cell polarity protein PAR6 (K)

par-3 Signaling orthologous to mammalian atypical PKC isotype-specific interacting protein (ASIP) (W)

R10D12.12 Carbohydrate metabolism Predicted glycosyltransferase (K)

let-92 Signaling Serine/threonine protein phosphatase 2A, catalytic subunit (K)

smgl-1 Unknown Uncharacterized conserved protein (Neuroblastoma-amplified protein) (K)

rab-10 Membrane traffic GTP-binding protein SEC4, small G protein superfamily, and related Ras family GTP-binding proteins (K)

C06A5.1 Unknown Uncharacterized conserved protein (K)

fat-6 Lipid metabolism Fatty acid desaturase (K)

pept-1 Transport H+/oligopeptide symporter (K)

ddr-1 Signaling Discoidin domain receptor DDR1 (K)

ZK637.1 Transport Synaptic vesicle transporter SV2 (major facilitator superfamily) (K)

T20D3.8 Lipid metabolism N-acetylglucosaminyltransferase complex, subunit PIG-C/GPI2, required for phosphatidylinositol biosynthesis (K)

F40F11.2 Unknown F40F11.2 gene (K)

T12A2.2 Carbohydrate metabolism Oligosaccharyltransferase, STT3 subunit (K)

ZK930.1 Membrane traffic Protein kinase containing WD40 repeats (K)

F15B10.1 Carbohydrate metabolism UDP-N-acetylglucosamine transporter (K)

syn-3 Membrane traffic SNARE protein SED5/Syntaxin 5 (K)

glf-1 Carbohydrate metabolism H04M03.4 encodes a UDP-galactopyranose mutase (K)

K02B12.3 Membrane traffic Prolactin regulatory element-binding protein/Protein transport protein SEC12p (K)

Y43H11AL.2 Transport Predicted membrane protein (K)

hgrs-1 Membrane traffic Membrane trafficking and cell signaling protein HRS, contains VHS and FYVE domains (K)

vps-34 Membrane traffic Phosphatidylinositol 3-kinase VPS34, involved in signal transduction (K)

F59F3.4 Transport F59F3.4 gene (K)

rab-5 Membrane traffic GTPase Rab5/YPT51 and related small G protein superfamily GTPases (K)

T15B7.2 Signaling Protein tyrosine phosphatase-like protein PTPLA (contains Pro instead of catalytic Arg) (K)

C33H5.18 Lipid metabolism CDP-diacylglycerol synthase (K)

F29G6.3 Unknown Unnamed protein (K)

gspd-1 Carbohydrate metabolism Glucose-6-phosphate 1-dehydrogenase (K)

C05C8.7 Carbohydrate metabolism Mannose-6-phosphate isomerase (K)

chc-1 Membrane traffic Vesicle coat protein clathrin, heavy chain (K)

fasn-1 Lipid metabolism putative fatty acid synthase, orthologous to human FASN (W)

sec-23 Membrane traffic Vesicle coat complex COPII, subunit SEC23 (K)

sca-1 Transport Ca2+ transporting ATPase (K)

vab-10 Cytoskeleton associated Dystonin, GAS (Growth-arrest-specific protein), and related proteins (K)

gob-1 Carbohydrate metabolism Unnamed protein (K)

vps-16 Membrane traffic Vacuolar assembly/sorting protein VPS16 (K)

W09G3.6 Membrane traffic WD40 repeat-containing protein (K)

Y57A10A.27 Unknown Y57A10A.27 gene (K)

prl-1 Signaling Protein tyrosine phosphatase IVA1 (K)

C55A1.9 Unknown Unnamed protein (K)

glo-4 Membrane traffic RCC1 domain (K)

Y17G9B.1 Unknown Unnamed protein (K)

R08D7.7 Carbohydrate metabolism Sugar (pentulose and hexulose) kinases (K)

haf-4 Transport Peptide exporter, ABC superfamily (K)

C32E8.9 Lipid metabolism Enoyl-CoA hydratase (K)

snap-1 Membrane traffic NSF attachment protein (I)

msp-152 Cytoskeleton associated Major sperm protein domain (K)

T07A9.9 Transport GTP-binding protein CRFG/NOG1 (ODN superfamily) (K)

bicd-1 Cytoskeleton associated Microtubule-associated protein Bicaudal-D (K)

rack-1 Signaling G protein beta subunit-like protein (K)

eat-3 Membrane traffic mitochondrial dynamin-related protein, orthologous to human OPA1 (W)

Y42G9A.4 Lipid metabolism Mevalonate kinase MVK/ERG12 (K)

rmd-5 Unknown Uncharacterized conserved protein (K)

fat-7 Lipid metabolism Fatty acid desaturase (K)

vha-6 Membrane traffic Vacuolar H+-ATPase V0 sector, subunit a (K)

unc-32 Membrane traffic ATPases V0/A0 complexs 116kDa subunit (I)

vha-4 Membrane traffic Vacuolar H+-ATPase V0 sector, subunit c'' (K)

T09A5.7 Unknown Mitochondrial distribution/morphology family 35/apoptosis (I)

T01H10.8 Membrane traffic BEACH domain-containing protein, orthologous to the mammalian LYST (W)

vha-1 Membrane traffic Vacuolar H+-ATPase V0 sector, subunits c/c' (K)

dnj-25 Membrane traffic Auxilin-like protein and related proteins containing DnaJ domain (K)

C37A5.1 Transport Bestrophin (Best vitelliform macular dystrophy-associated protein) (K)

gop-1 Unknown Protein of unknown function FPL (I)

vps-41 Membrane traffic Vacuolar assembly/sorting protein VPS41 (K)

vps-33.1 Membrane traffic Vacuolar sorting protein VPS33/slp1 (Sec1 family) (K)

vps-18 Membrane traffic rtholog of the vacuolar sorting protein VPS18/Vps18p (W)

let-767 Lipid metabolism 17 beta-hydroxysteroid dehydrogenase type 3, HSD17B3 (K)

C47G2.5 Signaling SIT4 phosphatase-associated protein (I)

cmd-1 Signaling Calmodulin and related proteins (EF-Hand superfamily) (K)

vps-39 Membrane traffic Vacuolar assembly/sorting proteins VPS39/VAM6/VPS3 (K)

vps-11 Membrane traffic Zinc fingers RING-type,Vacuolar protein sorting-associated protein 11 (I)

K02D10.5 Membrane traffic SNAP-25 (synaptosome-associated protein) component of SNARE complex (K)

acs-1 Lipid metabolism Long chain fatty acid acyl-CoA ligase (K)

T10H9.3 Membrane traffic SNARE protein Syntaxin 18/UFE1 (K)

sre-55 Signaling Sre G protein-coupled chemoreceptor (K)

ZC15.1 Unknown Zinc fingers CCHC-type (I)

srg-4 Signaling Receptor-like protein, Srg family (K)

dlk-1 Signaling Serine/threonine protein kinase, contains leucine zipper domain (K)

ZC449.3 Signaling Mitogen-activated protein kinase (MAPK) kinase MKK4 (K)

hipr-1 Membrane traffic Actin-binding protein SLA2/Huntingtin-interacting protein Hip1 (K)

sptl-1 Lipid metabolism Serine palmitoyltransferase (K)

F37B12.3 Lipid metabolism F37B12.3 gene (K)

vps-20 Membrane traffic Uncharacterized conserved protein predicted to be involved in protein sorting (K)

aps-1 Membrane traffic Clathrin adaptor complex, small subunit (K)

T05H4.7 Carbohydrate metabolism Predicted chitinase (K)

F49C12.12 Unknown Unnamed protein (K)

vha-10 Membrane traffic Vacuolar H+-ATPase V1 sector, subunit G (K)

vha-15 Membrane traffic Vacuolar H+-ATPase V1 sector, subunit H (K)

C49H3.8 Cytoskeleton associated Actin/actin-like,Heat shock protein 70s conserved site (I)

nud-2 Cytoskeleton associated LIS1-interacting protein NUDE (K)

vps-28 Membrane traffic Vacuolar sorting protein VPS28 (K)

vha-8 Membrane traffic Vacuolar H+-ATPase V1 sector, subunit E (K)

vha-12 Membrane traffic Vacuolar H+-ATPase V1 sector, subunit B (K)

vha-9 Membrane traffic Vacuolar H+-ATPase V1 sector, subunit F (K)

vha-2 Membrane traffic Vacuolar H+-ATPase V0 sector, subunits c/c' (K)

vha-5 Membrane traffic Vacuolar H+-ATPase V0 sector, subunit a (K)

T27F7.1 Membrane traffic Vacuolar sorting protein VPS24 (K)

rab-7 Membrane traffic Ras-related GTPase (K)

ist-1 Signaling PH and PTB domain-containing insulin receptor substrate (IRS) homolog (W)

vps-32.1 Membrane traffic Protein involved in glucose derepression and pre-vacuolar endosome protein sorting (K)

pod-2 Lipid metabolism Acetyl-CoA carboxylase (K)

R03E1.2 Membrane traffic ATPase membrane sector associated protein (K)

vha-17 Membrane traffic Vacuolar H+-ATPase V0 sector, subunit M9.7 (M9.2) (K)

vps-32.2 Membrane traffic Snf7 (I)

vha-13 Membrane traffic Vacuolar H+-ATPase V1 sector, subunit A (K)

F42C5.10 Unknown F42C5.10 gene (K)

F25H2.4 Unknown Uncharacterized conserved protein encoded by sequence overlapping the COX4 gene (K)

C45G9.5 Unknown C45G9.5 gene (K)

flh-3 Unknown Zinc fingers FLYWCH-type (I)

E03H4.8 Membrane traffic Vesicle coat complex COPI, beta' subunit (K)

sec-10 Membrane traffic Exocyst subunit - Sec10p (K)

sec-15 Membrane traffic Exocyst complex, subunit SEC15 (K)

sec-8 Membrane traffic Exocyst complex subunit Sec8 (K)

exoc-8 Membrane traffic Exocyst complex subunit (K)

B0041.3 Carbohydrate metabolism Predicted peptidoglycan-binding protein, contains LysM domain (K)

sec-6 Membrane traffic Exocyst complex subunit SEC6 (K)

H24G06.1 Signaling WD40 repeat protein (K)

unc-116 Cytoskeleton associated Kinesin (SMY1 subfamily) (K)

par-1 Signaling Serine/threonine protein kinase (K)

B0361.8 Carbohydrate metabolism Glycosyltransferase (K)

Gene name!"#$%&'()#*+,!-./"0/**/0#1"!23*,!4!2"5,!.+./*#'-6'17#89#,

Biological function

Colour-coded complex components with established roles in membrane traffic (Table S2)

Gene annotations

%+*,"'3('!-*#+'()+"#:"!;#<=>$#?@A#B?C#D+"*'!-#EFEGHGIC#J!";K/*+#BJC#"+4+/*+#J8HELIC!"#$-(+"M"!#%!;/'-#B$C#D+"*'!-#HH7EI#'-#()/(#3/"(',24/"#3"'!"'(0

6'12"+#89.N#O'+"/",)',/4#,42*(+"'-1#!:#/#*2K*+(#!:#PGF#3"';/"0#O=8#)'(*#+Q,42.'-1#1+-+*&'()#R,!"+S#:2-,('!-*#/::+,('-1#/44#K'!4!1',/4#3"!,+**+*#B*3+,':'+.#'-#T/K4+#8PI7

d

Figure S3 continued




Figure S3 continued

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

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

!!1#!$%&'("!2*++!!

!!3#!456(""!65+52!!

!!7#!456(""!+85''%9%)!

!!!:#!456(""!);--.+%!

!!<#!456(""!5&;85220!8=5>?%)!

!!@#!456(""!5??9%?5'%)!

!!A#!456(""!2*++!!

!"B#!456(""!%C8%++!!

!""#!DE!5??9%?5'%)!

!",#!DE!);--.+%!

!"1#!DE!!2.F;>52!

!"3#!DE!2*++!!

!! !,>)!5++50+!

!! !GHIH!J;+%5+%!

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`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a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aa('0&%!a>(-;>?%9!!MN#!!9&>(@!!!!!!!!$9*'%;>!H%'56*2;+F!!!!!!!,:Q!&9*'%5+*F%!9%?.25'*90!8*F&2%CY!+.6.>;'!4$K@V$QHJ<!!MN#!!&6+(1!!!!!!!!$9*'%;>!H%'56*2;+F!!!!!!!,BQ!&9*'%5+*F%Y!9%?.25'*90!+.6.>;'!6%'5!'0&%!$QH_1V$\$1!!MN#!!+>9(7!!!!!!!!4KD!&9*8%++;>?!!!!!!!!!!!QF522!>.82%59!9;6*>.82%*&9*'%;>!M+>4K$#!QHE!!MN#!!=;+(7!!!!!!!!JKD!6;>);>?!!!!!!!!!!!!!!L;+'*>%!L3!!MN#!!E7,S:P",!!!!!$9*'%;>!H%'56*2;+F!!!!!!!\6;`.;';>(8*>b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b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a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`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c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a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c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c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d!&9*+'5';8!58;)!&=*+&=5'5+%+!!MN#!!_B113P1!!!!!!H%'56*2;+F!!!!!!!!!!!!!!!W=;5F;>%!&09*&=*+&=5'%(9%`.;9;>?!%>/0F%!!MN#!!+9^("B1!!!!!!Q;?>52;>?!!!!!!!!!!!!!!!!<('95>+F%F695>%!*2-58'*90!9%8%&'*9!!MN#!!WB:L""P,!!!!!\>T>*^>!!!!!!!!!!!!!!!!!!WB:L""P,!?%>%!MN#!!E73J"P:!!!!!!ZC'958%22.259!H5'9;C!!!!!H.8;>V52&=5('%8'*9;>!!MN#!!U<7_@DP7!!!!!ZC'958%22.259!H5'9;C!!!!!U<7_@DP7!?%>%!MN#!!>=9(@B!!!!!!!Q;?>52;>?!!!!!!!!!!!!!!!!L*9F*>%!9%8%&'*9+!!MN#!!+9b("A!!!!!!!Q;?>52;>?!!!!!!!!!!!!!!!!<('95>+F%F695>%!*2-58'*90!9%8%&'*9!!MN#!!E1,_:P1!!!!!!4KD!&9*8%++;>?!!!!!!!!!!!\7!+>4K$(5++*8;5'%)!4KD!+&2;8;>?!-58'*9!!MN#!!U3AZ"BP,"!!!!\>T>*^>!!!!!!!!!!!!!!!!!!U3AZ"BP,"!?%>%!MN#!!W"3O"BP<!!!!!X;&;)!F%'56*2;+F!!!!!!!!!O$I!'95>+5F;)5+%!8*F&2%CY!O$I"<V$IO(Q!8*F&*>%>'!!MN#!!EB@J",P"!!!!!W95>+25';*>!!!!!!!!!!!!!!W%'95'9;8*&%&';)%!W$4("YQ;?>52!9%8*?>;';*>!&59';82%+!Q4$<,!+.6.>;'+!MI#!!))9("!!!!!!!!Q;?>52;>?!!!!!!!!!!!!!!!!J;+8*;);>!)*F5;>!9%8%&'*9!JJ4"!!MN#!!aN:1<P"!!!!!!W95>+&*9'!!!!!!!!!!!!!!!!Q0>5&';8!]%+;82%!'95>+&*9'%9!Q[,!MF5b*9!-58;2;'5'*9!+.&%9-5F;20#!!MN#!!2;>(,:!!!!!!!W95>+89;&';*>!!!!!!!!!!!!a;>8!-;>?%9+!S,L,(2;T%!MI#!!!W,BJ1P@!!!!!!X;&;)!F%'56*2;+F!!!!!!!!!K(58%'02?2.8*+5F;>02'95>+-%95+%!8*F&2%CY!+.6.>;'!$IO(SVO$I,!!MN#!!9&5("!!!!!!!!JKD!6;>);>?!!!!!!!!!!!!!!Q;>?2%(+'95>)%)!JKD(6;>);>?!9%&2;85';*>!&9*'%;>!D!M4$D#!!MN#!!E3BE""P,!!!!!\>T>*^>!!!!!!!!!!!!!!!!!!E3BE""P,!?%>%!MN#!!W",D,P,!!!!!!S596*=0)95'%!F%'56*2;+F!!G2;?*+588=5902'95>+-%95+%Y!QWW1!+.6.>;'!!MN#!!aNA1BP"!!!!!!H%F695>%!'95--;8!!!!!!!!!$9*'%;>!T;>5+%!8*>'5;>;>?!RJ3B!9%&%5'+!!MN#!!E"7_"BP"!!!!!S596*=0)95'%!F%'56*2;+F!!\J$(K(58%'02?2.8*+5F;>%!'95>+&*9'%9!!MN#!!+0>(1!!!!!!!!H%F695>%!'95--;8!!!!!!!!!QKD4Z!&9*'%;>!QZJ7VQ0>'5C;>!7!!MN#!!?2-("!!!!!!!!S596*=0)95'%!F%'56*2;+F!!LB3HB1P3!%>8*)%+!5!\J$(?5258'*&095>*+%!F.'5+%!!MN#!!E,"J7P7!!!!!!JKD!6;>);>?!!!!!!!!!!!!!!$*20>.82%*';)%!T;>5+%!1e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c(DW$5+%![B!+%8'*9Y!+.6.>;'!HAP<!MHAP,#!!MN#!!]&+(1,P,!!!!!H%F695>%!'95--;8!!!!!!!!!Q>-<!MI#!!]=5("1!!!!!!!H%F695>%!'95--;8!!!!!!!!![58.*259!Lc(DW$5+%!["!+%8'*9Y!+.6.>;'!D!!MN#!!E31J,P"!!!!!!W95>+89;&';*>!!!!!!!!!!!!S082;>YS082;>+!K('%9F;>52YW95>+89;&';*>!9%?.25'*9!8082;>!MI#!!E3,S7P"B!!!!!\>T>*^>!!!!!!!!!!!!!!!!!!E3,S7P"B!?%>%!MN#!!52=(:!!!!!!!!H%'56*2;+F!!!!!!!!!!!!!!!J%2'5("(&099*2;>%(7(8596*C025'%!)%=0)9*?%>5+%!!MN#!!E,7L,P3!!!!!!\>T>*^>!!!!!!!!!!!!!!!!!!\>8=5958'%9;/%)!8*>+%9]%)!&9*'%;>!%>8*)%)!60!+%`.%>8%!*]%925&&;>?!'=%!SGf3!?%>%!!MN#!!S37OAP7!!!!!!\>T>*^>!!!!!!!!!!!!!!!!!!S37OAP7!?%>%!MN#!!&%&'("!!!!!!!W95>+&*9'!!!!!!!!!!!!!!!!LcV*2;?*&%&';)%!+0F&*9'%9!!MN#!!=;+(:!!!!!!!!JKD!6;>);>?!!!!!!!!!!!!!!L;+'*>%+!L1!5>)!L3!!MN#!!E,AO:P1!!!!!!\>T>*^>!!!!!!!!!!!!!!!!!!\>>5F%)!&9*'%;>!!MN#!!86&("!!!!!!!!W95>+89;&';*>!!!!!!!!!!!!S4Z_!6;>);>?!&9*'%;>V$1BB!5>)!9%25'%)!WDa!a>(-;>?%9!&9*'%;>+!!MN#!!U71O@DXP,!!!!H;'*8=*>)9;52!-.>8';*>!!!KDJLg.6;`.;>*>%!*C;)*9%).8'5+%Y!KJ\EDAV1ATJ5!+.6.>;'!!MN#!!?+&)("!!!!!!!S596*=0)95'%!F%'56*2;+F!!O2.8*+%(:(&=*+&=5'%!"()%=0)9*?%>5+%!!MN#!!=;+(33!!!!!!!JKD!6;>);>?!!!!!!!!!!!!!!L;+'*>%!L,_!!MN#!!&6+(,!!!!!!!!$9*'%;>!H%'56*2;+F!!!!!!!,BQ!&9*'%5+*F%Y!9%?.25'*90!+.6.>;'!6%'5!'0&%!$QH_<V$QH_"BV$\$"!!MN#!!=;+(37!!!!!!!JKD!6;>);>?!!!!!!!!!!!!!!L;+'*>%+!L1!5>)!L3!!MN#!!-2=(1!!!!!!!!\>T>*^>!!!!!!!!!!!!!!!!!!a;>8!-;>?%9+!EXURSL('0&%!MI#!!!]=5("!!!!!!!!H%F695>%!'95--;8!!!!!!!!![58.*259!Lc(DW$5+%![B!+%8'*9Y!+.6.>;'+!8V8e!!MN#!!2%'(<""!!!!!!W95>+89;&';*>!!!!!!!!!!!!*9'=*2*?!*-!KGW"Y!'=%!8*>+%9]%)!8*9%!8*F&*>%>'!*-!'=%!F.2';+.6.>;'!SS43VKGW!8*F&2%C!MR#!!!&6+(3!!!!!!!!$9*'%;>!H%'56*2;+F!!!!!!!,BQ!&9*'%5+*F%Y!9%?.25'*90!+.6.>;'!6%'5!'0&%!$QH_,V$4Z"!!MN#!!ZB1L3P@!!!!!!H%F695>%!'95--;8!!!!!!!!![%+;82%!8*5'!8*F&2%C!SG$IY!6%'5e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`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aS"7P"!!!!!!!\>T>*^>!!!!!!!!!!!!!!!!!!a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aS33AP1!!!!!!Q;?>52;>?!!!!!!!!!!!!!!!!H;'*?%>(58';]5'%)!&9*'%;>!T;>5+%!MHD$N#!T;>5+%!HNN3!!MN#!!)>b(,7!!!!!!!H%F695>%!'95--;8!!!!!!!!!D.C;2;>(2;T%!&9*'%;>!5>)!9%25'%)!&9*'%;>+!8*>'5;>;>?!J>5h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c(DW$5+%!["!+%8'*9Y!+.6.>;'!O!!MN#!!]=5("7!!!!!!!H%F695>%!'95--;8!!!!!!!!![58.*259!Lc(DW$5+%!["!+%8'*9Y!+.6.>;'!L!!MN#!!W,<E<P"!!!!!!H%F695>%!'95--;8!!!!!!!!![58.*259!+*9';>?!&9*'%;>![$Q,3!!MN#!!S3AL1P@!!!!!!S0'*+T%2%'*>!5++*8;5'%)!!D8';>V58';>(2;T%YL%5'!+=*8T!&9*'%;>!<B+!8*>+%9]%)!+;'%!MI#!!>.)(,!!!!!!!!S0'*+T%2%'*>!5++*8;5'%)!!XIQ"(;>'%958';>?!&9*'%;>!K\JZ!!MN#!!]&+(,@!!!!!!!H%F695>%!'95--;8!!!!!!!!![58.*259!+*9';>?!&9*'%;>![$Q,@!!MN#!!]=5(@!!!!!!!!H%F695>%!'95--;8!!!!!!!!![58.*259!Lc(DW$5+%!["!+%8'*9Y!+.6.>;'!Z!!MN#!!]=5(",!!!!!!!H%F695>%!'95--;8!!!!!!!!![58.*259!Lc(DW$5+%!["!+%8'*9Y!+.6.>;'!_!!MN#!!]=5(A!!!!!!!!H%F695>%!'95--;8!!!!!!!!![58.*259!Lc(DW$5+%!["!+%8'*9Y!+.6.>;'!E!!MN#!!]=5(,!!!!!!!!H%F695>%!'95--;8!!!!!!!!![58.*259!Lc(DW$5+%![B!+%8'*9Y!+.6.>;'+!8V8e!!MN#!!]=5(7!!!!!!!!H%F695>%!'95--;8!!!!!!!!![58.*259!Lc(DW$5+%![B!+%8'*9Y!+.6.>;'!5!!MN#!!%'9("!!!!!!!!4KD!&9*8%++;>?!!!!!!!!!!!4KD(6;>);>?!&9*'%;>!S\O_$"V_4\KG!M44H!+.&%9-5F;20#!!MN#!!E1<_",P1!!!!!X;&;)!F%'56*2;+F!!!!!!!!!E1<_",P1!?%>%!MN#!!_B1:"P@!!!!!!S596*=0)95'%!F%'56*2;+F!!O208*+02'95>+-%95+%!!MN#!

.)/)"/*0)H28IFJ,+K8;1LH*G4254;;458?2H<=;LH3H<2(LHG1G4;8,*",?98.M8L

1&2#2$&-*#"(3/-4&2/

52#2367-2,),"-208#)9"-2082/)/4:";&4'"):4*+#&:'),"62#):"&/"0)0+6*/)"46*((&-"<=*+#)">?@

.)/)"*//24*4&2/:

F1;L2,=+,H*;81,+K128N2H>8@-'I8O/!8PO:8012;,H*8QRQSTSU:87H2>'4;18P7:821314;187.TQBU:H28I*+12#2H8FH>4,*8PI:8012;,H*8TT9QU8,*8+K4+8=42+,L<3428=2,H2,+5

",?<218.M1$8V,1242LK,L438L3<;+12,*?8HN84338M6W8=2,>4258V-.8K,+;9

)




95% of intestines with ectopic, basolateral F-actin (n=71/75)

0% of intestines with ectopic F-actin (n=0/84)

Winter et al., Figure S4

GL

GL

**

a b

c d

Anti-RAB-11 Anti-RAB-11

control RNAi !"#$% RNAi

F-actin F-actin

n=5/5 intestines with apical enriched RAB-11

n=6/6 intestines with peripherally clustered RAB-11

control RNAi !"#$% RNAi

!""#$%#&# !""#$%#&#

!""#$%#&# !""#$%#&#

*

Figure S4 Ectopic F-actin upon PAR-5 (FTT-1) depletion as well as upon cofilin depletion interferes with RAB-11 RE positioning and causes clustering of RAB-11-REs. (a-d) PAR-5 (FTT-1), but not FTT-2, is required for RE-positioning and proper F-actin organisation. (a, b) Control RNAi intestines of the ftt-2 deletion strain MT14355 exhibit apically enriched RAB-11-REs and normal apico-basal F-actin asymmetry. (c, d) Par-5 RNAi causes RAB-11-RE-clustering and a frequent and strong F-actin phenotype in enterocytes and germlines (GL). Arrowheads mark ectopic, peripheral F-actin. Asterisks mark gut lumen. (e-l) Rhodamine phalloidin stained, RNAi treated intestines with GFP::RAB-11 labelled REs (e,f,i,j), or basal, GFP::RME-1 labelled REs (g,h,k,l). For improved visualisation, brightness and contrast were increased in the grey-scale images. (e-h) In control RNAi, F-actin concentrates at the apical domain (e,g, medial sections) and is barely detectable at the basolateral cortex (f,h, peripheral sections). (g,h) GFP::RME-1 labelled basal REs enrich basolaterally in control RNAi. (i-l) Upon PAR-5 depletion, GFP::RAB-11 labelled REs cluster peripherally around or closely to ectopic F-actin structures at the basolateral

cortex (i,j), whereas basal, GFP::RME-1 labelled REs do not cluster at ectopic F-actin sites but virtually disappear (k,l, n=10 intestines). (m-o) Par-5 RNAi treated, non-transgenic intestines (NL2099 strain) show that ectopic F-actin (green) is more frequently and strongly surrounded or covered by endogenously labelled RAB-11-REs than by endogenous RAB-5-EEs (both in red, compare close-ups). (o) Mean percentage of all ectopic F-actin patches enriched for endogenously labelled RAB-11 or RAB-5 endosomes (analysed from n intestines with spatially well-defined ectopic F-actin patches, with n deriving from 2 independent experiments). Errorbars indicate s.e.m. and were derived using the m values. (p-s) Both RNAi of par-5 and of unc-60 (ADF-cofilin) cause ectopic microfilaments and clustered RAB-11-REs in enterocytes, but of different morphology and distributions. RAB-11-RE-clusters accumulate peripherally upon par-5 RNAi (p,q), and more often apically upon unc-60 RNAi (r,s). Brightness and contrast increased of peripheral grey-scale F-actin images of q and s. The scale bar indicates 10µm in the grey-scale images of m,n, as well as in p-s, and 20µm in all other images.




GFP::RAB-11GFP::RAB-11

!"#$% RNAi; peripheralnF-actin

Anti-RAB-5

mF-actin

Anti-RAB-11

!"#$% RNAi; peripheral

Winter et al., Figure S4 continued

F-actin

F-actin F-actin

F-actin F-actin

F-actinAnti-RAB-5

!"#$% RNAi; medial

&'($)* RNAi; medial!"#$% RNAi; medial

F-actin

GFP::RAB-11F-actin

GFP::RAB-11

p q r s

GFP:RME-1!"#$% RNAi; peripheral

econtrol RNAi; medial

fcontrol RNAi; peripheral

F-actin GFP:RME-1 F-actin


F-actin

GFP::RAB-11F-actin

GFP::RAB-11

&'($)* RNAi, peripheral

F-actin

GFP::RAB-11F-actin

GFP::RAB-11

F-actin

GFP::RAB-11F-actin

GFP::RAB-11

Anti-RAB-11

ji !"#$% RNAi; medial

g


h

GFP::RAB-11F-actin

F-actin F-actinGFP::RAB-11F-actin

k l

F-actin F-actincontrol RNAi; medialGFP:RME-1 GFP:RME-1

control RNAi; peripheral

26%n=16m=54

o

RAB-5

20

40

60

80

100

88%n=20 m=80

mea

n %

of

ecto

pic

F-ac

tinpa

tche

s e

nric

hed

for

RAB-11

n = # of intestines m = # of F-actin patches

F-actin F-actin

Figure S4 continued




020406080100120140160180200

0 5 10 15 20 25 30

Anti-RAB-11

PEPT-1::DsRed

PEPT-1::DsRed Anti-RAB-11

Anti-RAB-11

0

20

40

60

80

100

120

140

160

0 5 10 15 20 25 30

0

50

100

150

200

250

0 5 10 15 20 25 30

Anti-RAB-11


controlRNAi

'()#*RNAi

)'(#+RNAi

controlRNAi

PGP-1::GFP Anti-RAB-11

0 5 10 15 20 250

20

40

60

80

100

120

140

160

Anti-RAB-11

PGP-1::GFP

PEPT-1::DsRed

PEPT-1::DsRed

'()#*RNAi

0

50

100

150

200

250

PGP-1::GFP

Anti-RAB-11

0 5 10 15 20 25 30

)'(#+RNAi

0 5 10 15 20 25 300

50

100

150

200

250

PGP-1::GFP Anti-RAB-11

inte

nsity

(a.u

.)

distance (Um)

a

b

c

d

e

f

Intensity Profiles % of intestines with apical membrane proteinpresent in RAB-11-REs:

39% (n=7/18)

70% (n=16/23)

17% (n=3/18)

6% (n=1/18)

74% (n=20/27)

0% (n=0/14)

inte

nsity

(a.u

.)

distance (Um)

inte

nsity

(a.u

.)

distance (Um)

inte

nsity

(a.u

.)

distance (Um)

inte

nsity

(a.u

.)

distance (Um)

inte

nsity

(a.u

.)

distance (Um)

Figure S5 Apical membrane proteins PEPT-1::DsRed and PGP-1::GFP accumulate in basally mislocalised RAB-11-REs upon depletion of the dynactin-component ARP-1, but only weakly upon depletion of PAR-5. Line-based intensity measures (left: indicated by transparent lines on images, right: corresponding intensity profiles) on confocal images across intestines of worms expressing PEPT-1::DsRed (a-c) and PGP-1::GFP (d-f) immunostained for endogenous RAB-11. (a,d) In control RNAi intestines, PEPT-1::DsRed and PGP-1::GFP reside predominantly in the apical membrane, but small amounts are also present in RAB-11-labelled REs (in 39% (n=7/18) and 6% (n=1/18) of control RNAi intestines, respectively).

(b,e) Upon depletion of ARP-1, minus-end directed endosome transport is perturbed. Peripherally displaced RAB-11-REs contain significantly more apical membrane proteins PEPT-1::DsRed or PGP-1::GFP than REs in control conditions (in 70% (n=16/23) and 74% (n=20/27) of intestines with arp-1 RNAi phenotype). Note that relatively more PEPT-1::DsRed than PGP-1::GFP accumulates in peripheral REs. (c,f) Compared to arp-1 RNAi, depletion of PAR-5 causes little (PEPT-1::DsRed) to no (PGP-1::GFP) accumulation of apical membrane protein in peripherally clustered RAB-11-REs. For all samples, n derives from three independent experiments. Scale bars indicate 10"m.




Figure S6 Microtubule cytoskeleton organisation is subtly altered at sites of peripherally clustered REs in par-5 RNAi treated intestines. (a-c) Control RNAi intestine imaged at both peripheral (top) and medial (bottom) planes. In control RNAi, GFP::RAB-11-labelled REs are not peripherally clustered at the basal domain. Microtubules are distributed throughout the cell. (d-g) In par-5 RNAi,

the microtubule organisation proximal to peripherally clustered REs appears to be subtly changed, with microtubules orienting towards the clusters (g). (h-n) Similar changes were also observed in non-transgenic intestines (NL2099 strain), immunostained for !-tubulin and endogenous RAB-11 upon control (h-j) and par-5 RNAi (k-n). Scale bars indicate 20"m, and in g and n 10"m.

7,*+1281+8439:8;<==31>1*+4258",?<218.W

-2/462#"ABC&

!"#$%"ABC&

!"#$$%&'())!"#$$%&'()) >12?1 >12?1* + - , ) (

$

0

/

>12?1&*+,(X(+<Y<3,* >12?1&*+,(%&'())' & % ! #

&*+,(X(+<Y<3,* &*+,(%&'())

-2/462#"ABC&

!"#$%"ABC&

&*+,(X(+<Y<3,* &*+,(X(+<Y<3,*

=12,=K1243

>1G,43

>1G,43

>1G,43

=12,=K1243

=12,=K1243

=12,=K1243

=12,=K1243





control RNAia d

!"#$% RNAi

v

componentPKC-3PAR-6PAR-3

% (intestines)83% 86%85%

intestinesn=10/12n=19/22n=29/34

Mean percentage of )'(#+ RNAi treated intestines with peripherally clustered REs co-accumulating with

Anti-AJM-1Anti-PKC-3GFP::RAB-11

b c

Anti -PKC-3Anti-AJM-1max. int. prj.

e f

g h i j k l

m n o p q r

s t u

Anti-AJM-1Anti-PKC-3GFP::RAB-11 Anti-PKC-3

Anti-AJM-1max. int. prj.

Anti-AJM-1Anti-PAR-6GFP::RAB-11 Anti-PAR-6 Anti-AJM-1



Anti-AJM-1Anti-PAR-3

GFP::RAB-11

Anti-PAR-3 Anti-AJM-1Anti-AJM-1Anti-PAR-3

GFP::RAB-11

Anti-PAR-3 Anti-AJM-1

Figure S7 Ectopic localisation of apical PAR proteins at sites of clustered REs. (a to g’’) Immunofluorescence stainings of GFP::RAB-11 expressing intestines. In control RNAi intestines, PKC-3 (b), PAR-6 (h), and PAR-3 (n) are more concentrated along the apical domain. PAR-6 and PAR-3 localise to AJM-1 labelled apical junctions. In PAR-5 depleted intestines, PKC-3 (e), PAR-6 (k), and PAR-3 (q,t) accumulate at ectopic, cortical sites positive for AJM-1 and peripherally clustered REs (see arrowheads). Although immunofluorescence signals of the endogenous proteins could

be weak, such as in the case of PAR-3 (n,q,t), the apical signals were sensitive to RNAi of pkc-3, par-6, or par-3, respectively. In phenotypic par-5 RNAi enterocytes, the apical concentration of AJM-1, PKC-3, PAR-6 and PAR-3 can be reduced. (v) Mean percentage of par-5 RNAi treated intestines with peripheral RE-clusters at which apical PAR-proteins co-accumulated ectopically. The number of analysed intestines n derives from up to four independent experiments. Scale bars indicate 20"m.





Single RNAi

FC

a

Double RNAi (uncropped Western Blot of Figure 6a).

Double RNAi: par-5 +

cdc-

42

ced-

10

rho-

1

FC par-

6

pkc-

3

par-

3

b

par-5

FC par-5

actin

PAR-5(A. Golden)

!-tubulin

!-tubulin

RHO-1

actin

PAR-5(C. Hoege)

Uncropped Western Blot, complementing Fig. 6a. Total worm protein extracts of GFP::CDC-42 expressing strain

c

GFP(-CDC-42)

PAR-5(C. Hoege)

RNAi:

par-5 +

cdc-

42

FC cdc-

42

FC

actin

PAR-5(C. Hoege)

RNAi:

par-5 +

cdc-

42

FC cdc-

42

FC

GFP(-RAB-11)

FC par-5

50

64

36

98

148


PKC-3

PAR-6

cdc-

42

ced-

10

rho-

1

FC par-

6

pkc-

3

par-

3


cdc-

42

ced-

10

rho-

1

FC par-

6

pkc-

3

par-

3

kDa

kDa

kDakDa

kDa

kDa

kDa kDa

Figure S8 Full scans of Western blots. (a) Total worm protein extracts after treatment with single RNAi of feeding control (FC) versus par-5. Whereas PAR-5 is efficiently depleted, the probing for actin, GFP::RAB-11, and a-tubulin reveal no significant differences between control and par-5 RNAi samples. (b) Full scans of Western blots cropped for Figure 6a. Total worm protein extracts after treatment with double RNAi of par-5 + FC (the reference sample) versus double RNAi of par-5 + an apical par- gene or a rho GTPase-gene (the test samples). The left two images show different exposures of the same immuno-probings (primary antibodies indicated at the right margin of the upper image). The right image shows one exposure of anti-PKC-3 and PAR-6 probing. Importantly, the total worm protein extracts of both reference and test samples have similar levels of PAR-5 depletion,

i.e. similar PAR-5/a-tubulin or PAR-5/actin ratios. This shows that the suppression of the clustering phenotype in the test samples is real and not due to insufficient par-5 RNAi. Efficient depletion of RHO-1, PKC-3, and PAR-6 is visualized in the appropriate double RNAi samples. Reduced PAR-6 levels upon pkc-3 RNAi were observed in three independent experiments. Ced-10 (rac1) RNAi is coupled to reduced GFP::RAB-11 intensities (not shown). Extracts in a and b were prepared from MZE98R worms. (c) Full scan of Western blot complementing the data shown in Figure 6a. Since no specific CDC-42 antibody was available, the efficacy of CDC-42 depletion in single and double cdc-42 RNAi was confirmed by using total worm protein extracts of GFP::CDC-42 expressing strain treated with single RNAi (left two lanes) or double RNAi with par-5 (right two lanes).




Supplementary Tables

Supplementary Table S1 Hit list of the primary HCS with a total of 356 hits and their phenotype profiles. Gene annotations comprise Wormbase descriptions (release WS 204), NCBI KOGs, Protein ID of the highest metazoan orthologue, InterPro domains (v.22.0), HMMerThread domains, OMIM disease information, sequencing results, as well as secondary assay results for 60 genes. For 92 of the genes in common with the rme-screen of Balklava et al. 2007, we included the published information on rme, sec, or both phenotypes. The second worksheet lists the 14 phenotypic defect categories and provides more detailed descriptions of these.

Supplementary Table S2 Identified protein complexes with established regulatory roles in membrane trafficking exhibit characteristic RNAi phenotype profiles.

Supplementary Table S3 Phenotypic and functional profiling of secondary assay genes. Abbreviations: AF=autofluorescence, REs=apical recycling endosomes, EEs=early endosomes, ER= endoplasmatic reticulum, LEs=late endosomes, LROs=lysosome-related organelles. Note that candidates are assigned to a specific functional group based on the most pronounced RNAi-induced phenotypic changes, although their overall RNAi phenotype is often more complex. For secondary assay image examples see Figure 3. (*) Knockdown affects also LRO positioning, integrity, or sorting of AF material. (w) ‘Worm only’ means that there is no gene with more than 50% identity in sequence similarity in any non-worm species, whereas all other candidates are conserved up to humans (listed known or characterized orthologues acc. to WormBase (release WS227)).

Supplementary Note 1Supplementary Note 1 provides information on all key methods applied in the primary RNAi high-content screen and for its analysis. These primary screen methods comprise:1. Automated bacteria processing 2. Nematode processing 3. Automated imaging of worms 4. Software-based image analysis for phenotype identification and assignment of phenotype profiles5. Determination of sequence hits6. Functional annotation of candidates7. Orthology Information 8. Detection of biological process terms overrepresented in the dataset9. Hierarchical clustering


Caenorhabditis elegans screen reveals role of PAR-5 in RAB-11 … · Caenorhabditis elegans screen...

Documents

Transcript of Caenorhabditis elegans screen reveals role of PAR-5 in RAB-11 … · Caenorhabditis elegans screen...