Table Of ContentSeptin 9 induces lipid droplets growth by a
phosphatidylinositol-5-phosphate and
microtubule-dependent mechanism hijacked by HCV
Abdellah Akil, Juan Peng, Mohyeddine Omrane, Claire Gondeau, Christophe
Desterke, Mickaël Marin, Hélène Tronchère, Cyntia Taveneau, Sokhavuth Sar,
Philippe Briolotti, et al.
To cite this version:
AbdellahAkil,JuanPeng,MohyeddineOmrane,ClaireGondeau,ChristopheDesterke,etal.. Septin9
induceslipiddropletsgrowthbyaphosphatidylinositol-5-phosphateandmicrotubule-dependentmech-
anism hijacked by HCV. Nature Communications, 2016, 7, 10.1038/ncomms12203. hal-01355643
HAL Id: hal-01355643
https://hal.science/hal-01355643
Submitted on 18 Nov 2016
HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est
archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.
Distributed under a Creative Commons Attribution| 4.0 International License
ARTICLE
Received31Aug 2015|Accepted 7 Jun2016 |Published 15 Jul2016 DOI:10.1038/ncomms12203 OPEN
Septin 9 induces lipid droplets growth by
a phosphatidylinositol-5-phosphate and
microtubule-dependent mechanism hijacked
by HCV
Abdellah Akil1,2,3,4,*, Juan Peng1,2,5,*, Mohyeddine Omrane1,2,5,*, Claire Gondeau6,7, Christophe Desterke8,
Mickae¨l Marin1,2, He´le`ne Tronche`re9, Cyntia Taveneau10, Sokhavuth Sar1,2, Philippe Briolotti6,7,
Soumaya Benjelloun3, Abdelaziz Benjouad4,11, Patrick Maurel6,7, Vale´rie Thiers12, Ste´phane Bressanelli10,
Didier Samuel1,2,5,13, Christian Bre´chot1,2,12 & Ama Gassama-Diagne1,2,5
The accumulation of lipid droplets (LD) is frequently observed in hepatitis C virus (HCV)
infection and represents an important risk factor for the development of liver steatosis and
cirrhosis. The mechanisms of LD biogenesis and growth remain open questions. Here,
transcriptomeanalysisrevealsasignificantupregulationofseptin9inHCV-inducedcirrhosis
comparedwiththenormalliver.HCVinfectionincreasesseptin9expressionandinducesits
assembly into filaments. Septin 9 regulates LD growth and perinuclear accumulation in a
manner dependent on dynamic microtubules. The effects of septin 9 on LDs are also
dependentonbindingtoPtdIns5P,which,inturn,controlstheformationofseptin9filaments
and its interaction with microtubules. This previously undescribed cooperation between
PtdIns5Pandseptin9regulatesoleate-inducedaccumulationofLDs.Overall,ourdataoffera
novelrouteforLDgrowththroughtheinvolvementofaseptin9/PtdIns5Psignallingpathway.
1INSERM,Unite´1193,F-94800Villejuif,France.2UniversityofParis-Sud,UMR-S1193,F-94800Villejuif,France.3LaboratoiredesHe´patitesVirales,
De´partementdeVirologie.InstitutPasteurduMaroc,BP20360Casablanca,Maroc.4Faculte´desSciences,LaboratoiredeBiochimie-Immunologie,Univ.
MohammedV,Rabat,Maroc.5DHUHepatinov,VillejuifF-94800,France.6INSERMU1183,InstituteofRegenerativeMedicineandBiotherapy,Universityof
Montpellier,34295Montpellier,France.7DepartmentofHepato-GastroenterologyA,HospitalSaintEloi,CHRU,34295Montpellier,France.8Universityof
Paris-Sud,-UFRmedecine-INSERMUMS33,Villejuif,France.9INSERMU1048,I2MCandUniversite´PaulSabatier,31432Toulouse,France.10Virologie
Mole´culaireetStructuraleCNRSUPR3296-INRAUsC1358,91198Gif-sur-Yvette,France.11Univ.InternationaledeRabat,SalaAlJadida,Maroc.12Institut
Pasteur,75724Paris,France.13AP-HPHoˆpitalPaul-Brousse,CentreHe´pato-Biliaire,VillejuifF-94800,France.*Theseauthorscontributedequallytothis
work.CorrespondenceandrequestsformaterialsshouldbeaddressedtoA.G.-D.(email:[email protected]).
NATURECOMMUNICATIONS|7:12203|DOI:10.1038/ncomms12203|www.nature.com/naturecommunications 1
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms12203
L
ipiddroplets(LDs)arehighlydynamicorganellesgenerated size and composition between organisms19,20. In drosophila and
from the endoplasmic reticulum (ER)1, which can remain mammalian septins, complexes contain in general three to four
attached to the ER or generate contacts with other different septins present in two copies. The core hetero-
organelles, including the mitochondria, lysosomes, peroxisomes oligomeric units of six to eight subunits serve to build higher
and Golgi2. LDs consist of a hydrophobic core of neutral lipids, order structures, including filaments and rings structures, which
which are primarily composed of triacylglycerols (TAGs) and areimportantforseptininvivobiologicalfunction20.Septinsare
cholesterol esters, surrounded by a monolayer of phospholipids implicated in multiple cellular functions, including cytokinesis,
and a large group of proteins that facilitate cellular signalling ciliogenesis, cell migration, vesicle trafficking, and cell
interactions, control the access of metabolic enzymes, and polarity21–23. Septins associate with the cytoskeleton and are
influence LD movement within the cell1,3. These proteins known as a fourth type of cytoskeletal structure23. Septins also
include predominantly the perilipin family (Plins) and proteins bind to membranes, specifically to PIs, providing membrane
of the rab family of GTPases1,4. There are five known perilipins stability and serving as diffusion barriers for membrane
(PLINs 1–5) expressed in hepatocytes, which play distinct roles. proteins20,21,23. Nevertheless, these septin/PI interactions have
PLIN2alsocalledADRP(adiposedifferentiation-relatedprotein) beenscarcelyinvestigated24–26andtheirlinkwithHCVlifecycle
isaprominentLDprotein,expressionlevelofwhichistiedtoLD and their potential role in LD biogenesis and growth remain
accumulation. PLIN3 initially described as a tail-interacting unclear. Septins are also involved in infection by pathogens,
protein of 47kDa (TIP47) is also ubiquitously expressed5. LD including bacteria, fungi23,27,28 and HCV virus29. Alteration in
growth and degradation are highly regulated processes and expressionprofileofseptinsarereportedindifferentcancersand
excessiveaccumulationofLDsinthelivercausessteatosis,which we determined that septin 9 is specifically upregulated in liver
may contributes to cirrhosis, the final stage for several chronic cancer30.
liver diseases and often evolves to hepatocellular carcinoma. Despite being involved in carcinogenesis, septin 9 expression
ChronicHCVinfectionisamajorpublichealthproblemanda andfunctionhavenotbeeninvestigatedinprecancerousdiseases
mainriskfactorforcirrhosis,andhepatocellularcarcinoma,thus such as cirrhosis. In this study, we investigate the expression of
remainingthemajorcauseofliverresectionandtransplantation6. septin 9 and the differential expression of the other septins in
There is no vaccine to protect against HCV, although major HCV-induced cirrhosis. We also demonstrate that septin 9
advances have been recently achieved regarding HCV infection regulates LD growth through binding to PtdIns5P in HCV-
treatment, and the combination of drugs, including protease infected cells. Of interest, we show that this new mechanism is
inhibitors,representsarealbreakthroughthatcurealmost90%of involved in lipid homoeostasis independently of HCV infection.
infected patients. However, these treatments have remained very
costly. Additionally, understanding the key mechanisms driving Results
HCV replication and persistence will have a profound impact, Transcriptomic analysis of septins in human cirrhosis. To
beyondHCV,fortheappraisalofthepathogenesisofthisfamily determinetheexpressionprofileofseptin9incirrhosis,weused
of viruses. Each step of the replication cycle of HCV is tightly theGSE14323dataset31,includingtranscriptomesofnormalliver
dependent on the host lipid metabolism, and liver steatosis (n¼19) and HCV-induced cirrhosis samples (n¼49). Data
associatedwiththeaccumulationofLDsisahallmarkofchronic revealed a significant upregulation (two-sided Student’s t-test,
HCV infection7,8. HCV and LD interactions are crucial for P¼0.012) of septin 9 transcriptional expression in cirrhosis
persistent viral propagation and virion production9. samples compared to that in normal liver samples (Fig. 1a).
HCVisanenvelopedpositivesingle-strandedRNAvirusfrom Septin 9 occupies a terminal position in an octameric septin
the Flaviviridae family. HCV genomic RNA encodes for a complex involved in the formation of higher order structures
polyproteinprecursorofB3,100aminoacidspost-translationally such as filaments and rings (Fig. 1b)32. According to the
cleavedbycellularandviralproteasesintothestructural(core,E1 importance of hetero-oligomerisation in the assemblies and
and E2), non-structural (NS) proteins (NS2, NS3, NS4A, NS4B, functions of septins23, we assessed the differential expression of
NS5AandNS5B)andp7protein8.Thecoreproteinplaysamajor septinsinHCV-inducedcirrhosisusingtheGSE14323datasetby
role in the virion assembly through interaction with LDs10. unsupervised principal component analysis (Fig. 1c–f). Sample
The proteins covering the LD surface and their link with HCV projection on the first factorial map showed a significant
life cycle have been characterized, whereas the potential role separation between the normal liver and HCV-induced
of phospholipids in these events remains unclear. The cirrhosis samples according to the first axis (P¼2.67E(cid:2)15)
composition of the phospholipid monolayer of LDs depends on (Fig. 1c). Learning machine algorithm of Random Forest
the cell type, but is mostly composed of phosphatidylcholine, confirmed a good separation of the normal liver and cirrhosis
phosphatidylethanolamine, and, to a lesser extent, samples. The classification error by class was low and especially
phosphatidylinositol (PtdIns) and lysophospholipids1,11,12. The null for the cirrhosis class after building 500 classification trees
phosphorylated derivatives of PtdIns (PIs) are important and multidimensional Scaling plot (MDSplot), confirming the
signalling molecules that are essential for a variety of cellular good separation of the samples after learning machine analysis
functions, including membrane remodelling and trafficking13. (Fig.1d;SupplementaryFig.1).RandomForestanalysispredicted
Although recent studies revealed the role of PtdIns 4-kinases that most septins are affected in HCV-induced cirrhosis when
(PI4K-IIIa and PI4K-IIIb) and their lipid products, PtdIns4P, comparedtonormalliver,whileseptin10waslessaffected.This
as critical regulators of the HCV life cycle8. The role of class II was confirmed by individual boxplot performed on each septin
phosphoinositide 3-kinase (PI3K) was also demonstrated14. (Fig. 1e, P value were calculated by two-sided Student t-test).
Furthermore we demonstrated that PIs control epithelial These analyses showed that the expression of septin 7 and 2 is
morphogenesis and polarity15,16 and represent HCV cellular massively affected in these samples (Fig. 1d,e). A weak error on
targets17. misclassificationbyclass(confusionof1.67%)wasdetectedwhen
SeptinsformafamilyofGTP-bindingproteinscomposedof13 using the learning machine algorithm (Supplementary Fig. 1c).
members in mammals and conserved from yeast to humans18. Thus, we performed an unsupervised classification on these
Thecomplexityofthisgenefamilyisincreasedbytheexistenceof sampleswithpreviouslyusedseptinexpression(classificationtree
alternative splicing, which markedly increases the number of with Euclidean distance and complete linkage (Fig. 1f)). This
potential isoforms. Septins assemble into complexes variable in unsupervisedalgorithmperformedon58clinicalsamplesallowed
2 NATURECOMMUNICATIONS|7:12203|DOI:10.1038/ncomms12203|www.nature.com/naturecommunications
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms12203
b
9 GDPGDP 7 6 GTPGDP 2 2GDPGTP 6 7 GDPGDP 2
a
10.5
P value=0.012
c d
_9 10.2 4 HCcVo-cnitrrrohlosis P value= Septin_2
ptin 9.9 %) 2 2.670578e–15 Septin_7
Se 70 0 Septin_6
9.6 1. Septin_4
m 2 (1–2 SSeeppttiinn__98
HCV-cirrhosis Control Di–4 Septin_11
Septin_10
–6
–4 –2 0 2 4 6 0 5 10 15 20
Dim 1 (57.60%) Mean decrease accuracy
e
11.0 P value= 8 P value= 8 P value= 11.0 P value=
6.2e–14 8.3e–09 1.868e–12 10.5 2.32e–13
Septin_21100..50 Septin_4 76 Septin_6 76 Septin_7190..50
9.5
9.0
HCV-cirrhosis Control HCV-cirrhosis Control HCV-cirrhosis Control HCV-cirrhosis Control
P value= P value=0.158 P value=
8.0 2.78e–09 9.5 8.8 0.00022
Septin_8 77..50 Septin_10 98..50 Septin_11 88..40
6.5
8.0
7.6
HCV-cirrhosis Control HCV-cirrhosis Control HCV-cirrhosis Control
f
Color key and histogram
nt56
u4
o3
C2
1
6 7 8 9 10 11
Value Septin_2
Septin_7
Septin_9
Septin_10
Septin_11
Septin_4
Septin_6
Septin_8
Contol13Contol19Contol9Contol15Contol7Contol6Contol3Contol1Contol2Contol16Contol18Contol14Contol11Contol12Contol8Contol17Contol10Contol5Contol4Cirrhosis9Cirrhosis10Cirrhosis3Cirrhosis13Cirrhosis14Cirrhosis39Cirrhosis7Cirrhosis12Cirrhosis11Cirrhosis28Cirrhosis24Cirrhosis27Cirrhosis16Cirrhosis30Cirrhosis15Cirrhosis17Cirrhosis2Cirrhosis40Cirrhosis19Cirrhosis25Cirrhosis1Cirrhosis33Cirrhosis21Cirrhosis22Cirrhosis31Cirrhosis29Cirrhosis5Cirrhosis4Cirrhosis32Cirrhosis41Cirrhosis26Cirrhosis18Cirrhosis4Cirrhosis35Cirrhosis36Cirrhosis6Cirrhosis37Cirrhosis20Cirrhosis8Cirrhosis38Cirrhosis23
Normal liver HCV-cirrhosis
Figure1|Transcriptomicanalysisofseptin9inhumancirrhosis.(a)Rsoftwarewasusedtogenerateboxplotpresentsseptin9expressionincirrhosis
andinnormalliversamplesofGSE14323datasetandtocalculatementionedPvalueoftwo-sidedStudent’st-test.(b)Theoctamericcomplexofseptins
containsseptin2groupandseptin6groupmembers,septin7andseptin9whichcapsthetwoextremitiesoftherodshapedcomplex.(c)Rsoftwarewith
FactoMineRpackagewasusedtoobtainunsupervisedprincipalcomponentanalysisperformedonGSE14323withseptinmolecules:Pvalueobtainedon
firstprincipalaxisalloweddiscriminatingnormalliverfromcirrhosisinGSE14323.(d)Meandecreaseaccuracyofpredictvariablesduringthe‘Septin’
RandomForestlearningmachine.(e)RsoftwarewasusedtogenerateboxplotofseptinsallowstodiscriminatecirrhosisfromnormalliverwithGSE14323
andtocalculatementionedPvalueoftwo-sidedStudent’st-test.(f)Rsoftwarewasusedtoperformheatmapbyunsupervisedclassificationwithseptin
moleculesinGSE14323(classificationtreewithEuclideandistanceandcompletelinkage).
NATURECOMMUNICATIONS|7:12203|DOI:10.1038/ncomms12203|www.nature.com/naturecommunications 3
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms12203
the discrimination of two major groups of experiments with no treatment (Supplementary Fig. 3) and emphasized the role of
errorofmisclassificationbetweennormalliverandHCV-induced septin9inHCV-inducedLDaccumulation.Finally,werevealeda
cirrhosis samples. This result suggests that the expression of strongdecreaseofHCVgenomicRNAafterseptin9knockdown
septinproteinfamilyisderegulatedinHCV-inducedcirrhosisas (Fig.3i),suggestingacriticalroleofseptin9inHCVreplication.
compared with that in the normal liver. Wealsoanalysedthetranscriptlevelofthefivewell-characterized
variants of septin 9 (i1 to 5) after JFH-1 infection. Isoform 1
mRNAs were the most abundant, while isoform 3 mRNAs were
HCV assembles septin 9 which stabilizes microtubules.
almost undetectable (Supplementary Fig. 2b). Accordingly, the
Following the above described transcriptomic study, we
followingstudiesfocusedonseptin9isoform1(Septin9_i1)and
hypothesized thatHCV couldbe responsible of the upregulation
we used Huh7 cells stably expressing HCV full-length genomic
of septin 9 in cirrhosis. To explore this possibility, Huh7.5 cells
repliconfromgenotype1b(Huh7R)35.Thissystempermittedthe
were inoculated for 24, 48 and 72h with HCV JFH-1 (Japanese
efficient expression of all viral proteins and is a convenient and
fulminate hepatitis 1), a genotype 2a strain, which produces
valuable tool for molecular studies. A decrease in both the core
infectious particles in culture and recapitulates all steps of the
and envelope E2 proteins was observed in Huh7R cells
HCV life cycle33. Immunoblot data revealed a non-significant,
transfected with si3 (Supplementary Fig. 4a). By opposite both
time-dependent increase of septin 9 expression in non-infected
core and E2 expression increased in cells expressing septin 9_i1
cells, while septin 9 expression was significantly enhanced in
compared with control cells (Supplementary Fig. 4b). Later,
JFH-1-infectedcells(Fig.2a),withthehighestexpressionat72h
septin 9_i1 and the LD characteristics were assessed by
of infection (Fig. 2a) when septin 9 formed bundles of filaments
immunofluorescence analysis. Septin 9_i1 assembled in
surrounding the clusters of HCV core (Fig. 2b). Treatment of
filaments surrounding the large clusters of LD and transfection
Huh7.5 cells with two different septin 9 siRNAs (si1 and si3)
ofsi3inhibitedLDaccumulation(Fig.4a).Overall,profilesofLD
beforetheirinfectionstronglyreducedcoreexpression(Fig.2b,c).
distribution and changes in LD morphology (Fig. 4b,c) were
These data suggested that HCV-regulated septin 9 assembly and
similar to the results obtained using the JFH-1 strain (Fig. 3).
expression was required for core protein expression.
Septin 9_i1 filaments co-localized with endogenous septin 2
Thus,tofurthercharacterizetheseptin9filamentousstructure,
(Supplementary Fig. 5a), indicating that septin 9_i1 is integrated
we determined the presence of other septins. We focussed on
in a hero-oligomeric structure composed of other septins
septin 2, the major ubiquitously expressed septin and the most
consistent with data in Fig. 2. Moreover, the ratio of the
deregulated in HCV-induced cirrhosis (Fig. 1). In non-infected
intensity of exogenous septin 9_i1 revealed with an anti-V5 tag
cells, septin 2 formed short filaments that co-localized with
antibody to the intensity of endogenous septin 9 detected with
septin 9 (Fig. 2d,e). These filamentous structures containing
septin 9 antibody was 1.1 (Supplementary Fig. 5b). Thus, we
septin2andtheirco-localizationwithseptin9weresignificantly
concluded that septin 9_i1 expression was similar to that of
increaseduponJFH-1infection(Fig.2d–f).Depletionofseptin9
endogenous septin 9.
using si3 disrupted septin 2 filaments and reduced its staining
AlthoughthefirstevidenceforHCVinteractionwithLDswas
when compared with that in control cells in both non-infected
obtained from a study on the core protein, recent studies
and JFH-1-infected cells (Fig. 2d,e). These data were confirmed
indicated thatthe interactionof the HCV non-structuralprotein
by immunoblot analysis (Fig. 2g) and suggested the role of
NS5A with PLIN3 is essential for the recruitment of NS5A on
septin 9 in the formation of septin filamentous structure,
LDs to promote the release of HCV particles36,37. Interestingly,
consistent with previous reports32.
increases in PLIN3 and NS5A expression were observed in
Next, we assessed the effects of septin 9 expression on
Huh7R cells expressing septin 9_i1 (Supplementary Fig. 6a).
HCV-modulated microtubule (MT) network. Tubulin staining
These results were confirmed by immunoblot (Supplementary
indicated a strong MT network in HCV-infected cells
Fig. 6b) indicating that septin 9_i1 regulated LDs growth and
when compared with that of non-infected cells (Fig. 2h,i;
HCV protein expression.
Supplementary Fig. 2a). Importantly, MTs co-localized with
septin 9 filaments and si3 severely destroyed MT filaments
(Fig. 2h–j), indicating a regulatory role of septin 9 in
Septin 9 regulates neutral lipid metabolism. LD growth can be
HCV-dependent MT organization.
mediatedbyseveralmechanisms,but,typically,itssizeexpansion
involvedanincreaseintheneutrallipidcontent,whichessentially
Septin 9 regulates LD behaviour and HCV replication. The consistsofTAGandcholesterolester5.Therefore,weanalysedthe
accumulation of LDs is an important feature of HCV infection4. neutrallipidcontentofseptin9_i1-expressingcells.Transfection
Thus,tounderstatetheimportanceofseptin9inHCVlifecycle, ofseptin9_i1inHuh7RcellsinducedathreefoldincreaseofTAG
we investigated LDs. JFH-1 infection sharply increased LDs that andcorrelatedwithadecreaseofdiacylglycerol(DAGs)(Fig.4d).
formed clusters co-localized with the core protein around the Asaconsequence,theratioofTAGtoDAGwasgreatlyincreased
nucleus (Fig. 3a), while si3 treatment markedly decreased LDs (Fig. 4e). The fatty acid composition of TAG and DAG was also
(Fig.3b)andtheirco-localizationwithcore(Fig.3c,d).Atypical analysed. The composition of these two lipids was very similar
profile of LD distribution was presented (Fig. 3e,f). In non- and consisted mostly of C16 and C18 long chains (Fig. 4f).
infected cells, about 70% of LDs were present in the perinuclear Similar data were obtained using septin 9-depleted Huh7R cells
area and si3 treatment decreased the percentage to 55%. JFH-1 (Fig. 4g-i). A significant TAG decrease and an increase in DAG
infection increased the perinuclear LDs up to 92% and septin 9 wereobserved(Fig.4g),suggestingaconversionoftheTAGpool
depletion decreased this level to 62% (Fig. 3g). These changes in into DAGs.
thelevelanddistributionofLDswereaccompaniedbya2.5-fold Diacylglycerol acyltransferase-1 (DGAT1) is one of the two
increaseofLDsizeinJFH-1-infectedcells,whilesi3decreasedthe knownDGATenzymesthatcatalysethefinalstepintriglyceride
LD size by half (Fig. 3h). Similar changes were observed for LD biosynthesis38.TheroleofDGAT1inthetraffickingoftheHCV
number (Fig. 3h). PLIN2 is among the best studied protein core protein to LDs and its involvement in the production of
associated with LDs and is considered as a marker of LDs34. infectious HCV particles were reported39,40. Thus, we analysed
The strong increase in PLIN2 expression and its co-localization DGAT expression and we observed a significant increase in the
with core observed in JFH-1-infected cells were abolished by si3 total intensity of the DGAT1 signal in septin 9_i1-expressing
4 NATURECOMMUNICATIONS|7:12203|DOI:10.1038/ncomms12203|www.nature.com/naturecommunications
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms12203
a b
Non infected JFH-1 Non infected JFH-1
Time (h) 244872 244872 Septin 9 Septin 9 Core Merge
Septin 9 50 ntrol
o
Core 20 C
Actin
37
9 Septin 9 Septin 9 Core Merge
No infected n
JFH-1 pti
metryactin 86 Sept*in 9 * si3 Se
o9/ 4
Densitseptin 02 c sJiF sHe-p1tin 9
(h) 24 48 72 Control si1 si3
75
Septin 9
d
Septin 9 / septin 2
Non infected JFH-1 Core 20
50
Actin
ol
ntr
o e g
C
Non infected JFH-1
Non infected JFH-1
Control si3septin9 Control si3septin9
n 9 Septin 9 Septin 9 Septin 9 Septin 9 Ctrl si3 Ctrl si3
epti Septin 2 40
3 S Septin 9 70
si
Core 20
f Septin 2 Septin 2 Septin 2 Septin 2 Actin
n 37
atioRr) 1 *** Septin 2
h ColocalisSindex (ep0t.in50 9N /o m inicferocttuebduleJ (FMHT-s1) Merge Merge Merge Merge Densitometryseptin 2/actin 0000....186420 *
Ctrl si3 Ctrl si3
Non infected JFH-1
Non infectedJFH-1
ol
ntr i Non infected JFH-1
o
C
Control si3septin9 Control si3septin9
Septin 9 Septin 9 Septin 9 Septin 9
9
n
pti
e
S
3 MTs MTs MTs MTs
si
j
n Merge Merge Merge Merge
atioRr) 1 ***
olocalisindex ( 0.05
C Non infected JFH-1
Figure2|Septin9increasesinJFH-1-infectedcellsandregulatesseptin2andmicrotubulesfilaments.(a)Immunoblotofseptin9andcoreinHuh7.5
cellsinfectedornotwithJFH-1for24,48and72h.Actinwasusedasaloadingcontrol.Thebargraphpresentsthedensitometryanalysisofthe
immunoblotsfromthreeindependentexperiments.(b)Huh7.5cellstransfectedwithnon-targeting(control)orseptin9siRNA(si3)for24htheninfected
ornotwithHCVJFH-1for72hthenstainedforseptin9(green)andcore(red).(c)Immunoblotofseptin9andcoreinHuh7.5cellstransfectedwith
non-targeting(control)orseptin9siRNA(si1orsi3)theninfectedwithHCVJFH-1for72h.(d)Huh7.5treatedasinbandstainedforseptin9(green)and
septin2(red).(e)Dotrectanglesindarepresentedinhighermagnification.(f)BargraphshowsPearson’scorrelationcoefficient(Rr)ofseptin9and
septin2calculatedin30cellsfrom2independentexperiments.(g)Immunoblotofseptin2,septin9andcoreinHuh7.5cellstreatedasdescribedinb.
Actinwasusedasaloadingcontrol.Bargraphpresentsseptin2expressionfromthreeindependentexperiments.(h)Huh7.5cellsweretreatedas
describedinbandstainedformicrotubules(MTs)withbtubulin(red)andseptin9(green).(i)Dotsquaresinhpresentinhighermagnification.(j)Bar
graphshowsPearson’scorrelationcoefficient(Rr)analysisofseptin9andMTscalculatedin30cellsfromtwoindependentexperiments.Valuesare
means±s.e.m.Student’st-testwasused.*Po0.05,***Po0.0001.Scalebar,10mm.
NATURECOMMUNICATIONS|7:12203|DOI:10.1038/ncomms12203|www.nature.com/naturecommunications 5
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms12203
cells. Moreover, DGAT1 co-localized with septin 9_i1 filaments, MTsarerequiredforseptin9-inducedLDbehaviour.Different
as shown at the highest magnification (Fig. 4j). Together, these reports indicated that MTs play a role in the dynamics and
data indicated that septin 9 could also participate in LD growth growth of LDs41,42. Furthermore, septins interact with MTs and
by regulating lipid biosynthesis. are essential for MT-dependent cell processes12,23,43. Therefore,
a b
JFH-1
Non infected
LDs LDs Core Merge y
Control ence intensitD per cell186400000 *** **
eptin 9 LDs LDs Core Merge Fluorescof L 2N00onC itnrflecstie3dCJtFrlH-s1i3
S
3
si
c d
LDs Core Merge 60
Control sity (a.u.)6420000 0 1 2 3 4 sation index 000...8641 *
n 9 LDs Core Merge Inten40 olocali 0.20
pti 20 C Ctrl si3
Se 0 JFH-1
3 0 1 2 3 4
si Distance (µm)
e f
Non infected JFH-1
Periphery Control si3 Septin 9 Control si3 Septin 9
Perinuclear Fluorescence intensityof LDs (a.u.) 864200000 864200000 864200000 864200000
Distance (µm) 0 10 20 0 10 20 0 10 20 0 10 20
g h
Perinuclear Periphery LD size LD number
% of LDs intensities150000 Ctrl si3 Ctr*l**si3 % of LDs intensities150000 Ctrl si3 Ctr*l**si3 µ2LD size (m) 210 Ctrl*si3 Ctr*l*si3 LD number per cell112050500000 Ctrl*si3 Ctr*l*si3
Non infected JFH-1 Non infected JFH-1 Non infected JFH-1 Non infected JFH-1
i
ptin 9NA (%)1186200000 RNA (%)1120860000 **
SemR 4200 CV 4200
0 H 0
Ctrl si3 Ctrl si3
Figure3|Septin9regulatescoreandLDsaccumulationinJFH-1infectedcellsandvirusreplication.(a)Huh7.5cellsweretransfectedwith
non-targeting(controlorCtrl)orseptin9siRNA(si3)for24htheninfectedwithJFH-1forfurther72handstainedforcore(green)andLDs(red).
(b)LDsfluorescenceintensityin30cellsfromtwoexperimentsperformedasdescribedina.(c)Dotsquaresinaarepresentedinhighermagnification.
Therightpanelspresentgreen,red,lineprofileblotsofthepinklines.(d)BargraphsshowPearson’scorrelationcoefficient(Rr)forco-localizationbetween
coreandLDsof30cellsfromtwoexperiments.(e)Representationoftheperipheral(black)andperinuclear(red)regionsofthecell.(f)Radialprofileplots
showingLDdistributionbetweentheperipheral(black)andtheperinuclearregions(red)ofcellsindicatedwiththewhitearrowsina.(g)Quantificationof
LDsinperinuclearandperipheralregionsin30cellsfromtwoindependentexperimentsperformedasdescribedina.(h)LDsize(left)andLDnumber
(right)in30cellsfromtwoindependentexperimentsdoneasdescribedina.(i)Huh7.5cellsweretransfectedwithnon-targeting(control)orseptin9
siRNA(si3)for24handinfectedwithJFH-1asinafor72h.Cellswereanalysedforseptin9mRNAandHCVRNAlevelbyqRT–PCR.Bargraphsshow
resultsfrom3independentexperiments.Valuesaremeans±s.e.m.Student’st-testwasused.*Po0.05,**Po0.001,***Po0.0001.Scalebar,10mm.
6 NATURECOMMUNICATIONS|7:12203|DOI:10.1038/ncomms12203|www.nature.com/naturecommunications
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms12203
a b
Perinuclear
V5 tag LDs Merge
Periphery
Perinuclear
Empty vector 100 *** ***
(EV) % of LDsntensities 84260000
i 0
Septin 9_i1 Ctrl si3 EV I1
(I1) 100 Periphery
% of LDsntensities 86420000 *** ***
Septin 9 LDs Merge i 0 Ctrl si3 EV I1
c
Control e 2 ***
(Ctrl) LDs siz2µm) ( 10 ***
Ctrl si3 EV I1
si3 Septin 9 er 300 ***
(si3) Ds numbper cell210000 ***
L 0
Ctrl si3 EV I1
d e f 18-18-18 TAG 16-16-14
TAG DAG Ratio of 16-16-16
(nmol per 106 cells) (nmol per 106 cells) TAG/DAG 18-18-16 18-16-16
60 4 60
*
40 3 40 EV Septin 9_i1
2
20 1 20 DAG 18-18
0 0 0 18-18 16-16
EV Septin 9_i1 EV Septin 9_i1 EV Septin 9_i1
16-18
EV Septin 9_i1
g h i
TAG DAG Ratio of TAG
18-18-18 16-16-16
(nmol per 106 cells) (nmol per 106 cells) TAG/DAG
20 10 15 18-18-16 18-16-16
* 10
Control Septin 9_siRNA
10 5
5 18-18 DAG 16-16
0 0 0 16-18
Control Septin 9 Control Septin 9 Control Septin 9
siRNA siRNA siRNA
Control Septin 9_siRNA
j
DGAT1 / V5 tag DGAT1
y
sit 100 ***
EV enu.) 80
nta.
e i1 ( 60
cencGAT 40
Septin 9_i1 uoresof D 200
Fl EV Septin 9_i1
Figure4|Septin9regulatesTAGandDAGlevelsinHuh7Rcells.(a)Huh7Rcellstransfectedwithemptyvector(EV)orseptin9_i1for48hwerestained
forLDs(red)andV5tag(green).Cellstransfectedwithnon-targeting(control)orseptin9siRNA(si3)for48hwerestainedforLDs(red)andseptin9
(green).(b)Schematicrepresentationshowstheperinuclearandperipheralregionsofthecell.BargraphsrepresentLDsintensityintheperinuclearand
peripheralregionof36cellsfromfiveindependentexperiments.(c)LDsizeandLDnumberin36cellsfrom5independentexperiments.(d–f)Huh7Rcells
weretransfectedwitheitherEVorseptin9_i1cDNAsthenanalysedfortriacylglycerol(TAG)anddiacylglycerol(DAG).(d)Datarepresentnmolper106
cells.(e)RatioofTAGtoDAG.(f)DistributionoffattyacidspeciesinTAGandDAGaccordingtocarbonnumberintheacylgroup.Resultsareobtained
from3independentsexperiments.(g–i)Huh7Rcellstransfectedwithnon-targetingorseptin9siRNAfor72hwereanalysedforTAGandDAGasind–f.
(j)Huh7RcellstransfectedwithEVorseptin9_i1werestainedforV5tag(red)anddiacylglycerolacyltransferase-1(DGAT1)(green).Whitesquares
indicatetheareashowninhighermagnification.BargraphrepresentsDGAT1fluorescenceintensityin50cellsfromtwoindependentexperiments.
Valuesaremeans±s.e.m.Student’st-testwasused.*Po0.05,***Po0.0001.Scalebar,10mm.
NATURECOMMUNICATIONS|7:12203|DOI:10.1038/ncomms12203|www.nature.com/naturecommunications 7
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms12203
wepostulatedthatMTscouldplayaroleinseptin9-inducedLD production of PtdIns5P (ref. 48). First, the effect of YM201636
growth. Consequently, we performed a nocodazole washout was verified by PtdIns5P immunostaining in Huh7R cells
(Fig. 5). Treatment of Huh7R cells expressing septin 9_i1 with (Fig. 7a). Subsequently, the cells were transfected either with an
nocodazole resulted in the dispersion of LDsthroughout the cell empty vector or a septin 9_i1 construct and treated with
anddecreasedtheirsize.Onehourafternocodazoleremoval,LDs YM201636. Moreover LD size and level increased in cells
startedtogatheraroundthenucleusandincreasedinsize.These expressing septin 9_i1 and were surrounded by septin 9_i1
changes were accompanied by the partial recovery of the filamentsasdescribedinFig.4.YM201636treatmentprofoundly
microtubule filamentous structure (Fig. 5a). Finally, LDs were reduced the LD size (Fig. 7b,c and Supplementary Fig. 8a) and
restoredtotheirinitialsizeandrelocatedaroundthenucleus3h decreased TAGs concomitantly with an increase of DAGs
after nocodazole removal (Fig. 5b–d). Taken together, these (Supplementary Fig. 8b–d). Interestingly, YM201636 treatment
results indicate that MT dynamics are required for septin 9 to disrupted septin 9_i1 (Fig. 7b; Supplementary Fig. 8a) and
induce LD growth and perinuclear clustering. MT filaments and prevented both co-localization (Fig. 7c).
Furthermore, PtdIns5P addition to Huh7R cells transfected with
septin9siRNApartlyrescuedthedisruptionofMTfilamentsand
PtdIns5P binds to septin 9 and promotes LD growth. To
LD clustering caused by septin 9 depletion (Supplementary
furtheridentifythemechanismunderlyingseptin9assemblyand
Fig. 9). Thus, these data demonstrated that PtdIns5P is a crucial
its role on LD, we assessed the importance of phosphoinositides
regulator of septin 9 structural and functional features.
(PI). The phospholipid monolayer that surrounds LDs plays a
Interestingly our data revealed that both septin 9 and
role in the regulation of LDs morphology. Phosphatidylcholine
HCV-regulated PtdIns5P levels. Indeed infection of Huh7.5 cells
stabilizesthetensionandreducesLDgrowth,whilephosphatidic
using JFH-1 particles increased PtdIns5P (Supplementary
acid contributes to LD expansion44. PtdIns was also detected
Fig.10a).MoreoverPtdIns5PincreasedinHuh7Rcellscompared
on LD surface11, however, the presence and the role of its
with naive Huh7 cells and transfection of the later cells with
phosphorylated derivatives, PIs, on LD growth have not been
septin 9_i1 cDNA significantly enhanced the PtdIns5P signal
investigated.Therefore,weidentifiedthePIsthatbindtoseptin9
(Supplementary Fig. 10b,c).
by performing a protein–lipid overlay assay using a glutathione-
S-transferase (GST)-tagged recombinant protein (Fig. 6a). Data
fromtheassayrevealedthatseptin9boundspecificallytomono-
Septin 9 PBR is required for its assembly and LD growth.
phosphorylated PtdIns (PtdIns3P, PtdIns4P and PtdIns5P) and
SeptinsbindPIsinvitroviathepolybasicregion(PBR)neartheir
the strongest interaction was observed with PtdIns5P while no
N terminus26,49. The conserved sequence encompassing residues
signalwasdetectedusingGSTproteinaloneasanegativecontrol
aa289toaa294isboxedinFig.8a.Thus,asanotherapproachto
(Fig. 6a), supporting the specificity of the observed binding
assesstheimportanceofPIforseptin9structuralandfunctional
signals.
features,wegeneratedaPBR-deletedmutant(septin9_del1)from
Furthermore, we showed that PtdIns3P, PtdIns4P and
the cDNA of V5-tagged septin 9_i1. The recombinant proteins
PtdIns5P promoted a significant increase in LD size compared
from the entire septin 9_i1 and the deleted construct were
with non-treated cells, whereas PtdIns(4,5)P2 evaluated as a
produced in Escherichia coli, purified to near homogeneity, and
non-binding septin PI, had no significant effect (Fig. 6b,c).
analysed by SDS gel electrophoresis (Fig. 8b). Coomassie blue
PtdIns5P was the most efficient (Fig. 6b,c), consistent with the
staining revealed amajorband ofthe expected molecular weight
results in Fig. 6a. Moreover, septin 9 assembled surrounding the
(70kDa)forbothseptin9_i1andthedeletedmutant,whilesome
LDs that became more visible upon the addition of PtdIns5P
lower molecular weight species co-purified with septin 9_i1
(Fig. 6b). Overall, these data revealed the crucial role of mono-
(Fig. 8b). Immunoblot (Fig. 8c) confirmed that all detected
phosphorylatedPIsintheexpansionofLDsandtheorganization
protein bands were recognized by a V5-tag antibody. Purified
ofseptin9high-orderstructures.Thus,wefocusedonPtdIns5P,
recombinantproteinswereusedtoperformalipidoverlayassay.
whichwasthemosteffectiveonLDs.PtdIns5Pspecificallybinds
Septin9_i1boundPtdIns3P,PtdIns4P,andPtdIns5Pasexpected,
to the plant homeodomain (PHD) finger found in many
although a signal was observed with phosphatidic acid and
chromatin-remodelling proteins, and the PHD domain of the
PtdIns(3, 5)P2. The lipid-binding signal was strongly reduced
tumour suppressor inhibitor of growth (ING2) was proposed to
for septin 9_del1 (Fig. 8d). Subsequently Huh7R cells35 were
function as a receptor of PtdIns5P (ref. 45). We used the
transfected with the septin 9_i1, septin 9_del1 cDNAs, or empty
biotinylated GST-tagged PHD domain of ING2 as a probe
vector (EV).Septin 9_i1 formed filaments whereas septin 9_del1
to visualize cellular PtdIns5P, as previously described46. As
could not form filaments (Fig. 8e). Expression of septin 9_i1
expected, compared with untreated cells, PtdIns5P addition
increased LD in the perinuclear region compared to
strongly increased both the PHD signal and the LD size and
EV-transfected cells (Fig. 8e,f). In contrast, septin 9_del1
confirmed that addition of exogenous PtdIns5P increased its
expression significantly reduced the total level of LDs and their
intracellular level (Supplementary Fig. 7a–c). To validate the
perinuclear distribution (Fig. 8e,f). These changes in the LD
role of PtdIns5P on LDs, we used another approach based on
distribution were accompanied by significant decrease of the LD
IpgD, the virulence factor of Shigella flexneri, which is a
size and number (Fig. 8g). Furthermore septin 9_i1 filament
PtdIns(4,5)P2-4-phosphatase responsible for the profound
structures co-localized with MTs. By contrast, the mutant septin
increase of PtdIns5P in S. flexneri-infected cells47. Transfection
9_del1 lost the filamentous structure and co-localization with
of Huh7R cells with GFP-tagged-IpgD cDNA (GFP-IpgD)
MTs.ThePearsonco-localizationcoefficient(Rr)betweenseptin
increased PtdIns5P signal (Supplementary Fig. 7d) and LD size
9 and MTs dropped from 0.56±0.01 in cells expressing septin
and accumulation (Fig. 6d).
9_i1to0.17±0.03incellsexpressingseptin9_del1andtheRGB
line profile showed a similarity between MTs and septin 9_i1
PIKfyveregulatesseptin9filamentsandLDaccumulation.To cellular distribution (Fig. 8h). These results indicate that PBR is
furtherstudy theinteraction betweenseptin 9 and PtdIns5Pand necessaryfortheformationofseptin9filamentousstructuresand
its importance in LD accumulation, we used YM201636, a LD growth. Moreover, PBR is involved in MTs organization
selective inhibitor of FYVE finger containing phosphoinositide and highlighted the critical role of septin 9 in cytoskeleton
kinase (PIKfyve), a major contributor to the intracellular organization.
8 NATURECOMMUNICATIONS|7:12203|DOI:10.1038/ncomms12203|www.nature.com/naturecommunications
ARTICLE
NATURECOMMUNICATIONS|DOI:10.1038/ncomms12203
a
Septin 9_i1
Time after nocodazole washout (WO)
Control
0 1 h 3 h
MTs MTs MTs MTs
LDs LDs LDs LDs
MTs/LDs MTs/LDs MTs/LDs MTs/LDs
V5 tag V5 tag V5 tag V5 tag
Merge Merge Merge Merge
b c d
Perinuclear Periphery LD size
Periphery % of LDsintensities 0.15 ******* % of LDsintensities 000...642 ******* µ2LD size (m) 12 ** **
Perinuclear 0 0 0
Ctrl 0 1 2 Ctrl 0 1 2 Ctrl 0 1 2
WO (h) WO (h) WO (h)
Septin 9_i1 Septin 9_i1 Septin 9_i1
Figure5|Septin9regulatesLDsinmicrotubules-dependentmanner.(a)Huh7Rcellstransfectedwithseptin9_i1weretreatedwithnocodazole(33mM)
for2hat37(cid:2)Candplacedonicefor1h.Thencellswerewashedfivetimeswithice-coldculturemediumtoremovethenocodazoleandmovedat37(cid:2)Cfor
indicatedtimeinthefigurebeforestainingforV5tag(green),LDs(red)andmicrotubules(grey)withbtubulin.(b)Schematicrepresentationshowsthe
perinuclearandperipheralregions.(c)PercentageofLDsintensityintheperinuclearandperipheralregionsof20cellsfromtwoindependentexperiments.
(d)LDsizeanalysedin20cellsfromtwoindependentexperiments.Valuesaremeans±s.e.m.Student’st-testwasused.*Po0.05,**Po0.001,
***Po0.0001.Scalebar,10mm.
Septin 9 regulates LD growth in naive cells. Unravelling this significantly increases LD size comparing to EV-transfected
new role of septin 9 in LD behaviour led us to examine its cells (Supplementary Fig. 11a,b). Additionally, neutral lipids
relevance independently of HCV infection. We performed a set analysis revealed an increase TAG in septin 9_i1-transfected
of experiments using naive Huh7 cells. Septin 9_i1 expression cells (Supplementary Fig. 11c,d). Expression of GFP-IpgD
NATURECOMMUNICATIONS|7:12203|DOI:10.1038/ncomms12203|www.nature.com/naturecommunications 9
Description:4 Faculté des Sciences, Laboratoire de Biochimie-Immunologie, Univ. Mohammed V, Rabat, Maroc. 5 DHU Hepatinov, Villejuif Foti, M., Audhya, A. & Emr, S. D. Sac1 lipid phosphatase and Stt4 phosphatidylinositol 4-kinase