Cell–cell fusion and internalization of the CNS-based, HIV-1 co-receptor, APJ

15 Pages·2003·0.51 MB·English

View metadata, citation and similar papers at core.ac.uk brought to you by CORE
provided by Elsevier - Publisher Connector
Available online at www.sciencedirect.com
R
Virology307(2003)22–36 www.elsevier.com/locate/yviro
Cell–cell fusion and internalization of the CNS-based,
HIV-1 co-receptor, APJ
Naiming Zhou,a Xuejun Fan,b Muhammad Mukhtar,a Jianhua Fang,a Charvi A. Patel,a
Garrett C. DuBois,b and Roger J. Pomerantza,*
aTheDorranceH.HamiltonLaboratories,ThomasJeffersonUniversity,JeffersonMedicalCollege,CenterforHumanVirology,DivisionofInfectious
Diseases,DepartmentofMedicine,1020LocustStreet,Suite329,Philadelphia,PA19107,USA
bDepartmentofMicrobiology,ThomasJeffersonUniversity,JeffersonMedicalCollege,CenterforHumanVirology,DivisionofInfectiousDiseases,
DepartmentofMedicine,1020LocustStreet,Suite329,Philadelphia,PA19107,USA
Received8July2002;returnedtoauthorforrevision3September2002;accepted16September2002
Abstract
APJ,amemberofthehumanGprotein-coupledseven-transmembranereceptorfamily,hasbeenshowntoserveasacoreceptorforthe
entryofhumanimmunodeficiencyvirustypeI(HIV-1)andsimianimmunodeficiencyvirus(SIV),anditisdramaticallyexpressedincentral
nervous system (CNS)-based cells. In this study, expression of APJ tagged with the green fluorescent protein (GFP) and a fluorescent
peptide, 5-carboxyfluorescein (5-CF) conjugated Apelin-13, were utilized for studying receptor internalization and recycling, in stably
expressingindicatorcells,humanneurons,primaryCNSmicrovascularendothelialcells(MVECs),andastrocytes.FusionoftheC-terminus
ofAPJtotheN-terminusofGFPdidnotalterreceptorligandbindingandfunctions,includingsignalingandinternalization.Using293cells
stablyexpressingAPJ-GFP,wedemonstratedthatrapidinternalizationoftheAPJreceptorwasinducedbystimulationwithApelin-36and
Apelin-13, in a dose-dependent manner. Furthermore, investigations showed that the internalized APJ was colocalized with transferrin
receptors,suggestingthattheinternalizationofAPJinducedbyApelinislikelytobeviaclathrin-coatedpits.Interestingly,wefoundthat
theinternalizedAPJmoleculeswererecycledtothecellsurfacewithin60minafterremovalofApelin-13,butmostoftheinternalizedAPJ
stillremainedinthecytoplasm,even2hafterwashoutofApelin-36.TheintactcytoplasmicC-terminaldomainwasfoundtoberequired
forligand-inducedAPJinternalization.HumanneuronsweredramaticallystainedbytheAPJ-bindingfluorescentpeptides.Primaryhuman
fetal astrocytes were less strongly labeled with 5-CF-Apelin-13, and in primary human CNS MVECs only weak distribution of green
fluorescencespecificforAPJinthecytoplasmwasobserved.Apelin-36blockedcellmembranefusionmostlyduetostericinterference,with
onlyaverymodesteffectonreceptorinternalization.TheCNSrepresentsauniquereservoirsiteforHIV-1.Assuch,moleculartherapeutics
andsmallmolecularinhibitorsofHIV-1entryviathisuniqueCNSreceptorarenowabletoberationallydesigned.
©2003ElsevierScience(USA).Allrightsreserved.
Keywords:APJ;Chemokinereceptor;Internalization;Recycling;CNS;HIV-1
Introduction system (CNS), including regions of the hippocampus, stri-
atum, thalamus, cortex, and cerebellum (Edinger et al.,
APJ, also known as angiotensin receptor-like 1998;Matsumotoetal.,1996;O’Dowdetal.,1993).Recent
(AGTRL1), is a member of the G protein-coupled, seven- research utilizing immunocytochemical staining with a
transmembranereceptorfamilyandwasfirstidentifiedfrom polyclonal anti-APJ antibody and reverse transcriptase-
human genomic DNA by O’Dowd et al. (1993). Studies polymerasechainreaction(RT-PCR)showedthatAPJwas
with Northern blot analysis and in situ hybridization indi- expressed at a high level in neurons and oligodendrocytes
cated that APJ mRNA was detected in the central nervous and at lower levels in astrocytes, suggesting that the APJ
receptor may play certain important physiological roles in
*Correspondingauthor.Fax:(cid:1)1-215-503-2624. the CNS (Choe et al., 2000).
E-mailaddress:[email protected](R.J.Pomerantz). The endogenous peptidic ligand for APJ receptor, Ape-
0042-6822/03/$–seefrontmatter©2003ElsevierScience(USA).Allrightsreserved.
doi:10.1016/S0042-6822(02)00021-1
N.Zhouetal./Virology307(2003)22–36 23
lin, was first isolated from bovine stomach extracts (Tate- Although APJ is identified as a coreceptor for HIV-1
moto et al., 1998). Subsequently, the protein sequences of fusionandinfection,littleisknownabouttheinternalization
human, rat, and mouse Apelin pre-pro-proteins were de- andrecyclingofthisreceptor.Inthepresentstudy,weused
duced from the cDNAs (Habata et al., 1999; Hosoya et al., an APJ-GFP fusion protein and a 5-carboxyfluorescein (5-
2000; Lee et al., 2000; Tatemoto et al., 1998). The mature CF)-conjugated Apelin-13 to characterize the internaliza-
Apelin peptide consisting of 36 amino acid residues (Ape- tion and recycling of APJ receptor upon stimulation with
lin-36) and the shorter C-terminal peptide with 13 amino natural ligands, a phorbol ester (phorbol myristate acetate:
acidresidues(Apelin-13)weredemonstratedtoinducesig- PMA), and HIV-1 gp120s in transfected cells and primary
naling responses in APJ-transfected cells and neural cells humancells.Wealsoinvestigatedtheroleinternalizationof
(Choe et al., 2000; Hosoya et al., 2000). In a rat model, the APJ coreceptor plays in HIV-1 infection and in ligand
Apelins were found to play an important role in the hypo- inhibition of virus fusion and infection.
thalamicregulationofwaterintake(Leeetal.,2000;Reaux
et al., 2001; Taheri et al., 2002;), and to lower blood pres-
sure via a nitric oxide-dependent mechanism (Lee et al., Results
2000; Tatemoto et al., 2001).
It is well-known that entry of human immunodeficiency Expression and function of APJ-EGFP and APJ329-
virustypeI(HIV-1)intotargetcellsrequiresbindingofthe EGFP
virus envelope glycoprotein (gp120) to CD4, followed by
interactionswithaseven-transmembranereceptorasaviral ToinvestigatetheinternalizationoftheAPJreceptor,we
coreceptor. All primary HIV-1 strains use the chemokine constructed two expression vectors, pAPJ-EGFP and
receptorCCR5(M-tropic),CXCR4(T-tropic),orboth(du- pAPJ329-EGFP,consistingofthehumanAPJreceptorand
al-tropic) as major coreceptors (Choe et al., 1996; Deng et aC-terminallytruncatedAPJ,respectively,bothfusedtothe
al., 1996; Doranz et al., 1996; Dragic et al., 1996; Feng et enhancedversionofthegreenfluorescentprotein(EGFP)at
al.,1996).RecentstudieshaveshownthattheAPJreceptor the C-terminus. The stably expressing 293 cell lines were
is employed by T-tropic, M-tropic, dual-tropic HIV-1 and established for characterizing the function of APJ-EGFP
simianimmunodeficiencyviruses(SIV)forsupportingenv- andAPJ329-EGFP.TodeterminetheeffectsofAPJ-EGFP
mediated membrane fusion or viral entry in vitro (Choe et and APJ329-EGFP on signal transduction in response to
al., 1998; Edinger et al., 1998; Puffer et al., 2000; Singh et Apelin-36, we measured intracellular Ca2(cid:1) mobilization in
al., 1999; Zhang et al., 1998). It was reported that some stably transfected 293 cells upon stimulation with 3 (cid:1)M
brain-derivedvirusesalsouseAPJasacoreceptor(Albright Apelin-36.ItwasdemonstratedthatAPJ-EGFPmaintained
et al., 1999; Shieh et al., 1998). With the detection of signaling activity comparable to wild-type APJ, whereas a
expressionofAPJintheCNS,theseobservationssuggested lower intensity Ca2(cid:1) response was observed for APJ329-
thattheAPJreceptormightplayanimportantroleinHIV-1 EGFP (Fig. 1a). The gene reporter fusion assay was also
infection and pathogenesis in the CNS. Apelin, the natural utilized to evaluate the HIV-1 coreceptor activity of both
ligand of APJ, has been found to be able to block the APJ-EGFP and APJ329-EGFP. Compared to APJ, both
activity of APJ as an HIV-1 coreceptor (Cayabyab et al., fusion proteins exhibited similarly high levels of luciferase
2000; Puffer et al., 2000; Zou et al., 2000). G-protein- activity (Fig. 1b), suggesting that the EGFP fusion protein
coupled receptors (GPCR), such as CCKAR (Tarasova et did not alter the HIV-1 coreceptor activity of APJ.
al.,1997),CXCR4(Amaraetal.,1997;Orsinietal.,1999;
Tarasovaetal.,1998),CCR5(Macketal.,1998),andCCR3 Ligand-induced internalization of APJ-EGFP
(Zimmermanetal.,1999),canbeinducedtoundergorapid
internalization upon binding of agonists. In general, such TheEGFPfusionproteinattheC-terminusmadeitpossible
ligand-induced receptor internalization regulates the num- toassesstheligand-inducedinternalizationoftheAPJreceptor
beroffunctionallyactiveGPCRsontheplasmamembrane. instablyexpressing293cellsbyfluorescencemicroscopy.In
PreviousstudieshaveshownthatentryofHIV-1intotarget theabsenceofpeptideligands,APJ-EGFPprimarilylocalized
cells is independent of the ability of the coreceptor to tothecellsurface(Fig.2a).Afterexposureto2(cid:1)MApelin-13
internalize (Aramor et al., 1997; Doranz et al., 1996; Hev- orApelin-36at37°Cfor30min,thereceptorwasdramatically
eker et al., 1998). The internalization of coreceptors con- redistributed in the cytoplasm with distinct perinuclear accu-
tributestochemokineinhibitionofvirusentry(Amaraetal., mulation. The localization of APJ-EGFP was not altered by
1997;Macketal.,1998;Signoretetal.,1997),buttransient 200 nM SDF-1(cid:2)(Fig. 2a), which in the same concentration
down-regulation of the coreceptor(s) may not be the most could induce significant internalization of the human chemo-
efficient way of blocking HIV-1 infection in vivo (Stauber kinereceptor,CXCR4(Fig.2b).
etal.,1999).Certainviralenvelopegp120swerereportedto To further characterize internalization of APJ, cell-sur-
bind to CXCR4 and to induce rapid internalization of cell- face expression of APJ in stably expressing 293 cells upon
surfaceCXCR4,bothinthepresenceandintheabsenceof exposure to Apelins was analyzed by flow cytometry, with
CD4 (Tarasova et al., 1997). the anti-APJ monoclonal antibody, MB856. The internal-
24 N.Zhouetal./Virology307(2003)22–36
Fig.1.ExpressionandfunctionofAPJ-EGFPandAPJ329-EGFP.(a)IntracellularCa2(cid:1)mobilization.IntracellularCa2(cid:1)concentrationsinnontransduced293
cells,or293cellsstablyexpressingwild-typeormutantAPJ,weremeasuredinresponsetoApelin-36(3(cid:1)M).Thedatashownarerepresentativeofatleast
threeindependentexperiments.(b)CoreceptoractivityofAPJ.Targetcellswerecotransfectedwithplasmidsencodingcoreceptors,CD4,andluciferaseunder
thecontroloftheT7promoter.293effectorcellswereinfectedwithavacciniavirusencodingT7polymeraseandavacciniavirusencodingtheHIV-1
envelopeprotein(strain,89.6).Cellmixturesweremaintainedat37°Cfor4to6h.Cellfusionwasdeterminedbyaluciferaseassay.Valuesareexpressed
asluciferaseactivity(RLU/s).Allexperimentswererepeatedatleastthreetimes,andresultsareexpressedasthemeanvalue(cid:2)SE.
ization of APJ was induced rapidly upon treatment with transferrin and appeared as yellow spots (Fig. 3). In con-
Apelin-13andApelin-36.Thiseffectwasalreadydetectable trast, the cells treated with LysoTracker Red revealed no
5minafterApelinstimulationandreachedamaximumafter overlap of green and red colors. Taken together, these re-
30 min (Fig. 2c). APJ internalization was induced by Ape- sultsindicatethattheinternalizedAPJ-EGFPbyApelinsis
linsinadose-dependentmanner,andapproximately50%of associated with early endosomes.
cell-surfaceAPJwasinternalizedbyApelinsat2(cid:1)M(Fig.
2d).Apelin-13andApelin-36showednodifferenceindose-
Recycling of internalized APJ to the cell surface
dependence and kinetics in inducing APJ internalization.
The recovery of internalized APJ receptor to the cell
Colocalization of internalized APJ-EGFP with transferrin surface was investigated using the APJ-EGFP stably ex-
pressing 293 cells. After stimulation with 2 (cid:1)M Apelin-13
ToinvestigatetheintracellularrouteofinternalizedAPJ, or Apelin-36 at 37°C for 30 min, cells were washed thor-
the APJ-EGFP stably expressing 293 cells and CHO cells oughly with medium and incubated for the time periods
weretreatedwithhumantransferrincoupledtotetramethyl- indicated. To inhibit the de novo protein synthesis, cells
rhodamine, which marks early endosomes, or LysoTracker were always incubated in the presence of cycloheximide.
Red, which is taken up into lysosomes. In both 293 and TheresultsareshowninFig.4.AfterremovalofApelin-13,
CHOcells,uponstimulationwithApelin-13andApelin-36, the internalized APJ-EGFP was partly recycled to the cy-
the internalized APJ-EGFP was primarily colocalized with toplasmic membrane within 30 min of incubation. Further
Fig. 2. Internalization of APJ-EGFP upon stimulation with Apelins. (a) 293 cells stably expressing APJ-EGFP treated with only medium, Apelin-36,
Apelin-13, and SDF-1 as a control at 37°C for 40 min. (b) 293 cells stably expressing CXCR4-EGFP treated with medium only, SDF-1, Apelin-36, or
Apelin-13 at 37°C for 40 min. (c) Dependence of APJ internalization on concentrations of Apelins. The stably transfected 293 cells were treated with
increasingconcentrationsofpeptidefor30minat37°Cand,afterwashing,surfaceAPJwasdeterminebyFACS,asdescribedabove.(d)Timecourseof
APJinternalizationinducedbyApelins.Cellswereincubatedwith2(cid:1)MApelinsat37°Cfortheindicatedtimes,andafterwashing,wereanalyzedforAPJ
expressionasdescribedabove.Allexperimentswererepeatedatleastthreetimes,andresultsareexpressedasthemeanvalue(cid:2)SE.
26 N.Zhouetal./Virology307(2003)22–36
Fig.3.ColocalizationofinternalizedAPJwithtransferrin.ThestableAPJ-EGFPexpressing293andCHOcells(ingreen)werestainedwithtetramethyl-
rhodamine-transferrin(inred)orLysoTrackerRed(inred).ColocalizationofAPJ-EGFPwithtransferrinorLysoTrackerappearsasyellow.Thedatashown
arerepresentativeofatleastthreeindependentexperiments.
incubation resulted in full recovery of internalized APJ- Internalization of 5-CF-Apelin-13 in human astrocytes,
EGFPtothecellsurface,whereasafterwashoutofApelin- brain MVECs, and mature neurons
36, the internalized APJ-EGFP was still maintained in the
cytoplasm, even after 120 min of incubation. As reported previously (Choe et al., 2000); the APJ
receptor is expressed at high levels in neurons and oligo-
Role of the C-terminus of APJ in ligand-induced dendrocytes and at lower levels in astrocytes, but these
internalization primary cells are difficult to transfect with APJ-EGFP.
Thus,afluorescentpeptideligand,5-CF-conjugatedApelin-
We constructed a C-terminally truncated APJ, fused to 13,wasdevelopedtofurtherstudytheinternalizationofthe
EGFPatitsC-terminus,andtransfectedthemoietyinto293 APJ receptor in primary human cells and cultured neurons.
cells, for establishing a stable expressing cell line. The First, we examined the binding and internalization of this
cell-surface expression, and activity as a coreceptor for fluorescent peptide in APJ stably transfected 293 cells. As
HIV-1, of APJ329-EGFP was comparable to APJ-EGFP, shown in Fig. 6, the cell surface was stained with the
but intracellular calcium mobilization in response to Ape- fluorescent peptide after treatment with 5 (cid:1)M 5-CF-Ape-
lin-36 was much weaker than with APJ-EGFP. To investi- lin-13 at 4°C for 30 min. Of note, the green fluorescence
gate the role that the APJ C-terminus plays in ligand- was distributed in the cytoplasm upon incubation with
induced internalization, the stably expressing cells were 5-CF-Apelin-13at37°Cfor30min.Thesametreatmentof
plated onto two chamber slides overnight, and treated with C-terminally truncated APJ stably expressing 293 cells,
ligandsat37°Cfor30min.UnlikeinFig.2whichdemon- with5-CF-Apelin-13,resultedinnodistributionoffluores-
strates that exposure to Apelin-36 caused a dramatic inter- cent peptides in the cytoplasm. Cells lacking APJ, as neg-
nalization of APJ-EGFP, Apelin-36 did not affect cell- ativecontrols,routinelylackedfluorescenceinthesestudies
surface localization of APJ329-EGFP (Fig. 5). Of note, (not illustrated). These results indicated that 5-CF-Ape-
C-terminus truncation of CXCR4 also ablates internaliza- lin-13 binds to APJ and internalizes into the cytoplasm,
tion with SDF-1 (data not illustrated). together with the receptor.
N.Zhouetal./Virology307(2003)22–36 27
Fig.4.RecoveryofinternalizedAPJtothecellsurface.ThestableAPJ-EGFPexpressingcellswerefirstincubatedfor30minat37°Cwith2(cid:1)MApelin-13
orApelin-36.Afterinternalization,ligandswereremovedbywashingwithmedium,andthecellswerefurtherculturedinmediuminthepresenceof100
(cid:1)g/mlcycloheximideat37°CforvariousperiodsoftheindicatedtimesandthenexaminedforAPJexpression.Thedatashownarerepresentativeofatleast
threeindependentexperiments.
The primary CNS cells, human MVECs, astrocytes, and MVECs, little distribution of fluorescent peptide in the
cultured mature human neurons, differentiated from the cytoplasm was observed (Fig. 6).
NT2cell-line,werealsousedforstudyinginternalizationof
this fluorescent peptide. After incubation with 5 (cid:1)M fluo- Effects of PMA and the HIV-1 envelope glycoprotein
rescent Apelin-13 at 37°C for 40 min, astrocytes were gp120 on APJ internalization
clearlybutrelativelyweaklylabeledwithgreen5-CF-Ape-
lin-13inthecytoplasm,whileadensegreenfluorescencein IthasbeenshownthatCXCR4andCD4,butnotCCR5,
the entire cell was observed in differentiated NT2 cells, areinternalizedfromthecellsurfaceinresponsetophorbol
which grow in culture in densely packed multicellular ester stimulation (Orsini et al., 1999; Signoret et al., 1997,
“balls” (Mukhtar and Pomerantz, 2000). In human brain 1998).WeexaminedtheeffectofPMAoninternalizationof
28 N.Zhouetal./Virology307(2003)22–36
Fig.5.EffectoftheC-terminaltruncationofAPJonreceptorinternalization.293cellsstablyexpressingAPJ329-EGFPweretreatedwithApelin-36(3(cid:1)M)
at37°Cfor40min.CellswerethenanalyzedforAPJ329-EGFPinternalizationviafluorescencemicroscopy.Thedatashownarerepresentativeofatleast
threeindependentexperiments.
the APJ receptor. The stimulation of stably expressing 293 Since Apelin-36 was demonstrated to induce APJ inter-
cellswith5(cid:1)MPMAat37°Cfor30and60minresultedin nalization and prevent recycling to the cell surface, and to
apparent receptor internalization of CXCR4-EGFP, but further confirm the role of receptor internalization in inhi-
PMA treatment at the same concentration did not affect bition of HIV-1 entry, we transfected 293-CD4 cells with
APJ-EGFP localization on the cell surface, as illustrated in APJ and APJ329 for cell fusion assays to study the effects
Fig. 7a. Thus, APJ is similar to CCR5 and not CXCR4 in of Apelin-36 on inhibition of cell membrane fusion. As
this respect. shown in Fig. 8b, cell membrane fusion was inhibited by
We also examined the potential of different HIV-1 en- additionofApelin-36inaconcentration-dependentmanner
velopeproteingp120stowardinducingAPJinternalization. forbothwild-typeAPJandtheC-terminallytruncatedAPJ,
We treated the 293 cells, which stably express APJ-EGFP, and fusion inhibition by Apelin-36 was somewhat, but not
withrecombinantgp120sfromthedual-tropicisolates89.6 significantly, more efficient in wild-type APJ transfected
andSF2,T-tropicisolateMN,andM-tropicisolateBal,but cells, as compared to APJ329-transfected cells.
as indicated in Fig. 7b, none of these viral isolates’ enve-
lopes induced APJ internalization, as compared to the
strong internalization of APJ induced by treatment with
Discussion
Apelin-36 (3 (cid:1)M).
TheAPJreceptoriswidelyexpressedinthehumanCNS
Role of APJ internalization in ligand blocking of HIV-1 (Choe et al., 2000; Edinger et al., 1998; Matsumoto et al.,
infection 1996; O’Dowd et al., 1993). APJ and its natural ligand,
Apelin, may play an important role in the hypothalamic
Asdemonstratedabove(Fig.2),theabilityofApelin-13 regulation of water intake in the endocrine axis (Lee et al.,
to induce receptor internalization showed no difference as 2000; Reaux et al., 2001; Taheri et al., 2002), and in low-
compared to Apelin-36. To determine the activity of Ape- ering blood pressure via a nitric oxide-dependent mecha-
lin-13andApelin-36inblockingHIV-1entry,wefirstused nism(Leeetal.,2000;Tatemotoetal.,2001).Recentlythe
a gene reporter fusion assay to examine the ability of both APJ receptor has been identified as a coreceptor for HIV-1
peptidestowardinhibitionofcell–cellfusion,involvedwith and SIV for supporting env-mediated membrane fusion or
the APJ receptor and HIV-1 gp120. Before initiation of viral entry in vitro and possibly in vivo (Choe et al., 1998;
89.6
fusion, the target cells were incubated in the presence or Edinger et al., 1998; Puffer et al., 2000; Singh et al., 1999;
absence of the test Apelin at 37°C for 30 min, resulting in Zhangetal.,1998).Inthisstudy,ouraimwasthemolecular
sufficient APJ receptor molecules on the cell surface for characterization of internalization of APJ receptor in trans-
internalization. Nonetheless, the fusion data showed that fectedcellswithAPJC-terminallyfusedwithEGFP,andin
treatmentwithApelin-36ledtoblockingofcell–cellfusion human astrocytes, CNS MVECs and cultured human neu-
inadose-dependentmanner,whereasadditionofApelin-13 rons with a fluorescent peptide ligand. We also evaluated
resulted in no inhibitory activity (Fig. 8a), suggesting that the role of APJ internalization in ligand inhibition of virus
receptor internalization may not associate with ligand inhi- entry.
bition of HIV-1 entry, at least for APJ as a coreceptor. The green fluorescent protein (GFP) has been widely
N.Zhouetal./Virology307(2003)22–36 29
Fig.6.InternalizationofthefluorescentApelin-13inhumanprimaryCNScellsandculturedneurons.Thehumanastrocytes,MVECs,andhumancultured
neurons were treated with 5 mM 5-CF-Apelin-13 at 37°C for 60 min. The cells were then analyzed by fluorescence microscopy. The data shown are
representativeofatleastthreeindependentexperiments.
used to construct chimeric proteins to study their localiza- investigatethefateofinternalizedAPJ,thecellsexpressing
tion, distribution, and function in different systems. Previ- APJ-EGFP were costaining with tetramethylrhodamine-
ousstudieshavedemonstratedthatfusingGPCRwithGFP transferrinorLysoTrackerRed.Wefoundthattheinternal-
at the C-terminus led to a fully functional CCKAR (Tara- ized APJ molecules were colocalized with transferrin, a
sova et al., 1997), (cid:3)-Adrenergic receptor (Barak et al., marker for early endosomes (Ghosh and Maxfield, 1995),
2
1997), and CXCR4 (Tarasova et al., 1998). For this study, suggesting that the internalization of APJ induced by Ape-
we constructed a chimeric APJ C-terminally fused with linsislikelytobevia clathrin-coatedpits.Previous studies
EGFP and established stably expressing cell lines. Com- have shown that the internalized cholecystokinin receptors
pared to the wild-type APJ receptor, APJ-EGFP was found (Tarasova et al., 1997) and CCR5 (Ghosh and Maxfield,
to have normal signaling and HIV-1 gp120-induced cell 1995) accumulate in early endosomes and recycle with
membrane fusion. The expression of APJ-EGFP allows us almost100%efficiencytothecellsurface.CXCR4hasbeen
to easily detect the receptor moiety under fluorescence mi- found to target to lysosomes after SDF-1-induced internal-
croscopy, for further characterizing receptor internalization izationandtobepoorlyrecycledtothecellsurface(Signo-
and recycling. ret et al., 2000; Tarasova et al., 1998). Here we found that
In response to Apelins, the cell-surface APJ-EGFP was although Apelin-13 and Apelin-36 have the same potential
rapidlyinternalizedinadose-dependentmanner.Tofurther in inducing receptor internalization, internalized APJ mol-
Fig.7.EffectofphorbolesterandHIV-1gp120onAPJinternalization.ThestablyexpressingCXCR4-EGFPandAPJ-EGFP293cellswereincubatedwith
5(cid:1)MPMA(a),andAPJ-EGFP293cellswereinducedwith50(cid:1)g/mlHIV-1gp120sorApelin-36(3(cid:1)M)(b),at37°Cfor60min.Thedatashownare
representativeofatleastthreeindependentexperiments.
N.Zhouetal./Virology307(2003)22–36 31
Fig.8.Apelin-dependentinhibitionofcellmembranefusionwiththeHIV-189.6dual-tropicisolate.(a)ActivityofApelin-36andApelin-13inblocking
APJ-associatedcell–cellfusion.(b)EffectofApelin-36oninhibitionofcell-membranefusionincellsstablyexpressingwild-typeandC-terminallytruncated
APJ.Targetcellswerecotransfectedwithplasmidsencodingwild-typeAPJorthemutantAPJ329,CD4,andluciferaseunderthecontroloftheT7promoter.
293effectorcellswereinfectedwithavacciniavirusencodingT7polymerase,andavacciniavirusencodingtheHIV-1envelopeglycoprotein(strain,89.6).
Beforefusion,thetargetcellswereincubatedinthepresenceorabsenceofthetestApelinfor30min.Cellmixturesweremaintainedat37°Cfor4to6h.
Fusionwasdeterminedbyaluciferaseassay.Valuesareexpressedaspercentagesofluciferaseactivity.Allexperimentswererepeatedatleastthreetimes,
andresultsareexpressedasthemeanvalue(cid:2)SE.
eculesbytreatmentwithApelin-13wererecycledtothecell It has been shown previously that serine and threonine
surface within 60 min, while APJ receptors internalized by residuesintheC-terminaldomainofGPCRsarecriticalfor
Apelin-36 still remained in the cytoplasm even after a 60 ligand-inducedinternalization(Benyaetal.,1993;Innamo-
minincubation.Ithasbeenfoundthattherangeofrecycling rati et al., 1999; Maestes et al., 1999; Pizard et al., 1999;
times for other GPCRs is between 20 to 60 min (Ashworth Rothetal.,1997).C-terminallytruncatedCXCR4,CCKBR,
et al., 1995; Marchese and Benovic, 2001; Tarasova et al., and IL-8 receptor B were found to be expressed and func-
1997).Inourrecyclingexperiments,weincubatedthecells tionaswild-type,buttheligand-inducedinternalizationwas
at 37°C for 120 min after removal of Apelin-36, but most severely impaired (Amara et al., 1997; Lamey et al., 2002;
internalized receptors still remained in the cytoplasm. Maestesetal.,1999;Signoretetal.,1997).Toexaminethe
Mack et al. (1998) have demonstrated that aminooxy- function of the APJ C-terminus in receptor internalization,
pentane-RANTES (AOP-RANTES) efficiently induces in- we constructed and APJ mutant with deletion of the C-
ternalizationofCCR5and,unlikewild-typeRANTES,pre- terminusandcreatedastablyexpressingcellline.Wefound
vented recovery of CCR5 to the cell surface. The that the mutant APJ in stable cells appeared the same as
mechanism through which Apelin-36 and AOP-RANTES wild-type by expression levels and coreceptor activity, but
prevented recycling of internalized receptors to the cell the signaling in response to stimulation with Apelins was
surface remains unclear. It has been established that recy- severely impaired, and ligand-induced internalization was
cling of internalized receptors needs to be proceeded by totally blocked. These results are in agreement with previ-
receptor dephosphorylation (Grady et al., 1995 Krueger et ous studies with CCR5 (Alkhatib et al., 1997; Prado et al.,
al., 1997). Signoret et al. (2000) found that sustained bind- 1996). However, further studies will be needed to identify
ingofAOP-RANTESpreventsCCR5recycling.Therecep- thedeterminantsintheC-terminaldomainofAPJinvolved
tor-bound Apelin-36 was demonstrated to be barely disso- in ligand-induced internalization.
ciated from the APJ receptor (Hosoya et al., 2000). It is CD4 and CXCR4, but not CCR5, have been found to
possible that the bound Apelin-36 may block dephosphor- undergorapidinternalizationuponphorbolesterstimulation
ylation of the internalized APJ and recycling to the cell (Amara et al., 1997; Kraft et al., 2001, Orsini et al., 1999;
surface. Pelchen-Matthewsetal.,1993;Signoretetal.,1997,1998).