Receptor-mediated Endocytosis of CC-chemokines

Chemokines are chemotactic proteins which play a central role in immune and inflammatory responses. Chemokine receptors are members of the seven transmembrane G-protein coupled family and have recently been shown to be involved in the entry of human immunodeficiency virus (HIV) into target cells. To study chemokine endocytosis in detail we have used novel site-specific chemistry to make a fluorescently labeled CC-chemokine agonist (rhodamine-MIP-1α) and antagonist (NBD-RANTES). We have also generated a CHO cell line stably expressing a hemagglutinin-tagged version of the CC-chemokine receptor 1 (CCR1), and using these reagents we have examined the receptor-mediated endocytosis of CC-chemokines by confocal microscopy. Our studies reveal that the agonist was internalized and accumulated in transferrin receptor-positive endosomes whereas the antagonist failed to internalize. However, receptor-bound antagonist could be induced to internalize by co-administration of agonist. Analysis of receptor redistribution following chemokine addition confirmed that sequestration was induced by agonists but not by antagonists. Chemokines are chemotactic proteins which play a central role in immune and inflammatory responses. Chemokine receptors are members of the seven transmembrane G-protein coupled family and have recently been shown to be involved in the entry of human immunodeficiency virus (HIV) into target cells. To study chemokine endocytosis in detail we have used novel site-specific chemistry to make a fluorescently labeled CC-chemokine agonist (rhodamine-MIP-1α) and antagonist (NBD-RANTES). We have also generated a CHO cell line stably expressing a hemagglutinin-tagged version of the CC-chemokine receptor 1 (CCR1), and using these reagents we have examined the receptor-mediated endocytosis of CC-chemokines by confocal microscopy. Our studies reveal that the agonist was internalized and accumulated in transferrin receptor-positive endosomes whereas the antagonist failed to internalize. However, receptor-bound antagonist could be induced to internalize by co-administration of agonist. Analysis of receptor redistribution following chemokine addition confirmed that sequestration was induced by agonists but not by antagonists. INTRODUCTIONChemokines are a large family of chemotactic proteins which regulate leukocyte activation and recruitment to sites of inflammation. They can be divided into two main classes, the CXC- and the CC-chemokines based on the spacing of the first two cysteine residues. Both CXC- and CC-chemokines bind to seven transmembrane G-protein coupled receptors (GPCRs) 1The abbreviations used are: GPCRG-protein coupled receptorPCRpolymerase chain reactionHIVhuman immunodeficiency virusHAhemagglutininNBD7-nitrobenz-2-oxa-1,3-diazole-4-ylMIPmacrophage inflammatory polypeptideCHOChinese hamster ovary. which in most cases are promiscuous in that they will bind more than one ligand with high affinity (1Power C.A. Wells T.N.C. Trends Pharmacol. Sci. 1996; 17: 209-213Google Scholar). Although receptor-mediated endocytosis of certain GPCRs such as the β2-adrenergic receptor has been well documented, the endocytic pathways utilized by most GPCRs are still uncharacterized. Following stimulation by agonist, the β2-adrenergic receptor is rapidly phosphorylated by a specific GPCR kinase. This uncouples the interaction between the receptor and the G-protein and allows binding of β-arrestin which acts as an adapter to recruit the receptor into clathrin-coated vesicles (2Goodman O.B. Krupnick J.G. Santini F. Gurevich V.V. Penn R.B. Gagnon A.W. Keen J.H. Benovic J.L. Nature. 1996; 383: 447-450Google Scholar). Several studies have shown that GPCRs are internalized via clathrin-coated vesicles, although this may not be a universal mechanism as alternative endocytic pathways have been described for a number of different ligands (3Raposo G. Dunia I. Delavier-Klutchko C. Kaveri S. Strosberg A.D. Benedetti E.L. Eur. J. Cell Biol. 1989; 50: 340-352Google Scholar, 4von Zastrow M. Kobilka B.K. J. Biol. Chem. 1992; 267: 3530-3538Google Scholar, 5Grady E.F. Slice L.W. Brant W.O. Walsh J.H. Payan D.G. Bunnett N.W. J. Biol. Chem. 1995; 270: 4603-4611Google Scholar, 6Fonseca M.I. Button D.C. Brown R.D. J. Biol. Chem. 1995; 270: 8902-8909Google Scholar, 7Zhang J. Ferguson S.S,G. Barak L.S. Menard L. Caron M.G. J. Biol. Chem. 1996; 271: 18302-18305Google Scholar, 8Garland A.M. Grady E.F. Lovett M. Vigna S.R. Frucht M.M. Krause J.E. Bunnett N.W. Mol. Pharmacol. 1996; 49: 438-446Google Scholar, 9Tolbert L.M. Lameh J. J. Biol. Chem. 1996; 271: 17335-17342Google Scholar, 10Berry S.A. Shah M.C. Khan N. Roth B.L. Mol. Pharmacol. 1996; 50: 306-313Google Scholar).Chemokines such as RANTES (regulated on activation normal T cell expressed and secreted), MIP-1α, and MIP-1β have been shown to be the major HIV-suppressive factors produced by CD8+ T cells (11Cocchi F. DeVico A.L. Garzino-Demo A. Arya S.K. Gallo R.C. Lusso P. Science. 1995; 270: 1811-1815Google Scholar), and a number of recent studies have gone on to show that chemokine receptors are co-receptors along with CD4 for the entry of HIV into cells (12Feng Y. Broder C.C. Kennedy P.E. Berger E.A. Science. 1996; 272: 872-877Google Scholar, 13Deng H. Liu R. Ellmeier W. Choe S. Unutmaz D. Burkhart M. Di Marzio P. Marmon S. Sutton R.E. Hill C.M. Davis C.B. Peiper S.C. Schall T.J. Littman D.R. Landau N.R. Nature. 1996; 381: 661-666Google Scholar, 14Dragic T. Litwin V. Allaway G.P. Martin S.R. Huang Y. Nagashima K.A. Cayanan C. Maddon P.J. Koup R.A. Moore J.P. Paxton W.A. Nature. 1996; 381: 667-673Google Scholar, 15Choe H. Farzan M. Sun Y. Sullivan N. Rollins B. Ponath P.D. Wu L. Mackay C.R. LaRosa G. Newman W. Gerard N. Gerard C. Sodroski J. Cell. 1996; 85: 1135-1148Google Scholar, 16Doranz B.J. Rucker J. Yi Y. Smyth R.J. Samson M. Peiper S.C. Parmentier M. Collman R.G. Doms R.W. Cell. 1996; 85: 1149-1158Google Scholar, 17Bleul C.C Farzan M. Choe H. Parolin C. Clark-Lewis I. Sodroski J. Springer T. Nature. 1996; 382: 829-833Google Scholar, 18Oberlin E. Amara A. Bachelerie F. Bessia C. Virelizier J-L. Arenzana-Seisdedos F. Schwartz O. Heard J-M. Clark-Lewis I. Legler D.F. Loetscher M. Baggiolini M. Moser B. Nature. 1996; 382: 833-835Google Scholar, 19Berson J.F. Long D. Doranz B.J. Rucker J. Jirik F.R. Doms R.W. J. Virol. 1996; 70: 6288-6295Google Scholar, 20D'Souza P.M. Harden V.A. Nature Med. 1996; 2: 1293-1300Google Scholar, 21Fauci A.S. Nature. 1996; 384: 529-534Google Scholar). This has raised the possibility of using chemokines therapeutically to block HIV entry. Receptor-mediated endocytosis of chemokines has not been studied in detail, but an understanding of this process is clearly important if we are to develop selective therapeutic agents which block HIV entry into cells. To this end we have studied the receptor-mediated endocytosis of CC-chemokines by CCR1 as a prototype for the CC-chemokine receptor family.RESULTS AND DISCUSSIONTo study CC-chemokine endocytosis we generated fluorescently labeled RANTES and MIP-1α using site-specific chemistry to attach the fluorophore specifically to the terminal amino group of the polypeptide chain (23Offord R.E. Gartner H. Wells T.N.C. Proudfoot A.E.I. Methods Enzymol. 1997; (in press)Google Scholar, 26Alouani S. Gaertner H.F. Mermod J-J. Power C.A. Bacon K. Wells T.N.C. Proudfoot A.E.I. Eur. J. Biochem. 1995; 227: 328-334Google Scholar). Using this technique we conjugated rhodamine to MIP-1α and 7-nitrobenz-2-oxa-1,3-diazole-4-yl (NBD) to RANTES. These fluorescently labeled CC-chemokines retained their ability to bind specifically to CCR1 although the binding affinity was slightly reduced compared with the unmodified agonist (23Offord R.E. Gartner H. Wells T.N.C. Proudfoot A.E.I. Methods Enzymol. 1997; (in press)Google Scholar, 26Alouani S. Gaertner H.F. Mermod J-J. Power C.A. Bacon K. Wells T.N.C. Proudfoot A.E.I. Eur. J. Biochem. 1995; 227: 328-334Google Scholar). Prior to using these ligands to study endocytosis we assessed their biological activity in chemotaxis assays. Rhodamine-MIP-1α acted as a full agonist and induced maximal chemotaxis at a concentration of 100 nM (Fig. 1A). However, NBD-RANTES had no biological activity in a chemotaxis assay. We have previously shown that modification of the amino terminus of RANTES has a profound influence on its biological properties (25Proudfoot A.E.I. Power C.A. Hoogewerf A J. Montjovent M-O. Borlat F. Offord R.E. Wells T.N.C. J. Biol. Chem. 1996; 271: 2599-2603Google Scholar). Failure to cleave the initiating methionine from bacterially expressed RANTES (Met-RANTES) produced a protein which was devoid of agonist activity but was a potent antagonist. Similarly, conjugation of the fluorophore NBD to the amino terminus of RANTES also converted it from an agonist to an antagonist. As shown in Fig. 1B both Met-RANTES and NBD-RANTES can fully antagonize the chemotactic activity of RANTES on THP-1 cells.For endocytosis studies the human CCR1 cDNA was cloned by reverse transcriptase-PCR, and an HA epitope tag was placed at the extreme amino terminus. The tagged receptor was transfected into CHO cells, and lines stably expressing the receptor were cloned. Radioligand binding assays confirmed that the tagged CCR1 expressed in CHO cells (CHO-CCR1) retained high affinity ligand binding and ligand specificity (data not shown). In the initial experiments the fluorescent rhodamine-MIP-1α and NBD-RANTES were bound to CHO-CCR1 cells for 4 h at 4°C at a final concentration of 500 nM. Unbound ligand was washed off with ice-cold buffer prior to warming the cells to 37°C for timed periods up to 30 min (Fig. 2). The agonist, rhodamine-MIP-1α, was effectively internalized and accumulated in perinuclear vesicles (panels A-C) whereas the antagonist, NBD-RANTES, remained almost entirely on the cell surface (panels D-F). This is consistent with other GPCRs for which it has been shown that receptor internalization is dependent upon agonist stimulation and is inhibited by antagonists (4von Zastrow M. Kobilka B.K. J. Biol. Chem. 1992; 267: 3530-3538Google Scholar, 6Fonseca M.I. Button D.C. Brown R.D. J. Biol. Chem. 1995; 270: 8902-8909Google Scholar, 9Tolbert L.M. Lameh J. J. Biol. Chem. 1996; 271: 17335-17342Google Scholar, 10Berry S.A. Shah M.C. Khan N. Roth B.L. Mol. Pharmacol. 1996; 50: 306-313Google Scholar). However, since we have made NBD-RANTES we were able to perform the experiment to study the effect of agonist stimulation upon the receptor-mediated endocytosis of a fluorescent antagonist. Binding of NBD-RANTES in the presence of an equimolar concentration of RANTES showed that the antagonist could be induced to cluster and internalize in the presence of agonist (panels G-I). Although the internalized antagonist appeared in more peripheral endosomes rather than in the perinuclear structures in which the agonist accumulated, this observation nonetheless demonstrated that agonist-bound receptor could induce the sequestration of antagonist-occupied receptor. These findings suggest that activation of the GPCR kinase by agonist will induce phosphorylation, and consequently sequestration, of receptors occupied by agonists or antagonists. The significance of this relates to how a virus uses a GPCR to gain entry to a cell. These findings suggest that HIV would not have to act as an agonist to be internalized via CC-chemokine receptors provided there is sufficient chemokine agonist present to induce receptor internalization. This is consistent with recent findings, using chimeric receptors, that the regions of CCR5 required for viral entry and for chemokine signal transduction are distinct (27Atchison R.E. Gosling J. Montecarlo F.S. Franci C. Digilio L. Charo I.F. Goldsmith M.A. Science. 1996; 274: 1924-1926Google Scholar). It may also explain the observation that in certain macrophage cultures addition of CC-chemokines actually enhances rather than inhibits HIV replication (28Schmidtmayerova H. Sherry B. Bukrinsky M. Nature. 1996; 382: 767Google Scholar).Fig. 2Receptor-mediated endocytosis of fluorescent chemokines. Chemokines were bound to the surface of CHO-CCR1 cells at 4°C. Unbound chemokine was washed off, and cells were either fixed immediately or rapidly warmed to 37°C for 15 or 30 min prior to fixation. Following fixation the cells were examined by confocal microscopy. Panels A, D, and G show chemokine distribution with no warmup, panels B, E, and H show distribution after a 15-min warmup, and panels C, F, and I show distribution after a 30-min warmup. Rhodamine-MIP-1α (panels A, B, and C), NBD-RANTES (panels D, E, and F), and NBD-RANTES plus RANTES (panels G, H, and I). (Bar = 10 μm).View Large Image Figure ViewerDownload (PPT)To define the endocytic compartment into which rhodamine-MIP-1α was being delivered we performed dual-labeling studies with the transferrin receptor. Rhodamine-MIP-1α was bound to CHO-CCR1 cells at 4°C, and receptor-bound ligand was subsequently allowed to internalize for 30 min at 37°C. Cells were fixed, permeabilized, and stained using an antibody to the transferrin receptor (Fig. 3). Analysis of the cells by confocal microscopy revealed that at 4°C, the rhodamine-MIP-1α decorated the plasma membrane (panel B) whereas the transferrin receptor staining was both on the plasma membrane and in numerous endocytic vesicles (panel A). Following a 30-min warm up the rhodamine-MIP-1α was internalized into perinuclear endosomal vesicles (panel D) and showed an exact co-localization with the transferrin receptor (panel C). Since it is well documented that the transferrin receptor is a marker for the clathrin-coated pit endocytic pathway it is reasonable to conclude that the CCR1 is also internalized via this mechanism.Fig. 3Co-localization of rhodamine-MIP-1α with the transferrin receptor. Rhodamine-MIP-1α was bound to CHO-CCR1 cells at 4°C. Unbound chemokine was washed off, and the cells were either fixed immediately or warmed to 37°C for 30 min prior to fixation. Fixed cells were permeabilized and stained with a monoclonal antibody to the transferrin receptor (H68.4) followed by a goat anti-mouse-fluorescein isothiocyanate antibody. Dual staining of the transferrin receptor (panels A and C) and rhodamine-MIP-1α (panels B and D) were analyzed by confocal microscopy. Panels A and B show distribution at 4°C, and panels C and D after a 30-min warmup to 37°C. (Bar = 10 μm).View Large Image Figure ViewerDownload (PPT)As a final study we decided to examine the receptor redistribution following addition of agonist or antagonist. For these experiments various chemokines were added to CHO-CCR1 cells at 37°C for 30 min after which time the cells were washed, fixed, and permeabilized, and the CCR1 receptor was visualized by staining with the anti-HA monoclonal antibody (Fig. 4). The various chemokines were added at a concentration of either 50 nM or at 1 μM. Without chemokine addition the receptor was predominantly seen on the plasma membrane (panel A) and addition of the antagonists NBD-RANTES (panels B and C) or Met-RANTES (panels D and E) did not induce any significant receptor redistribution. However the agonists MIP-1α (panels F and G) and RANTES (panels H and I) both induced receptor sequestration, although MIP-1α appeared more effective in this respect than RANTES. Even at 50 nM, MIP-1α induced most of the CCR1 to redistribute from the plasma membrane to perinuclear vesicles, and at 1 μM receptor down-regulation was almost complete.Fig. 4CC-chemokine receptor 1 redistribution following addition of chemokine agonists and antagonists. CHO-CCR1 cells were incubated for 30 min at 37°C with a range of CC-chemokines. Following this period the cells were fixed and permeabilized, and the HA-tagged receptor was detected with the anti-HA antibody 12CA5 followed a biotinylated rabbit anti-mouse antibody and a streptavidin-Texas Red conjugate. Panel A, no chemokine addition; panel B, 50 nM NBD-RANTES; panel C, 1 μM NBD-RANTES; panel D, 50 nM Met-RANTES; panel E, 1 μM Met-RANTES; panel F, 50 nM MIP-1α; panel G, 1 μM MIP-1α; panel H, 50 nM RANTES; panel I, 1 μM RANTES.View Large Image Figure ViewerDownload (PPT)These observations are the first detailed description of chemokine and chemokine receptor endocytosis, and they show that CCR1 is likely to be internalized via clathrin-coated pits. We demonstrate that both ligand and receptor accumulate in perinuclear endosomes and that receptor internalization is dependent upon agonist stimulation. Provided that all CC chemokine receptors behave like CCR1, does this give us any insight into how we might use chemokines as therapeutic agents to prevent HIV entry into cells? The protective effects of chemokines may be due to two factors. The chemokine may act as a competitive inhibitor preventing binding of the virus to the receptor or it may down-regulate surface receptors so that there are no receptors available for the virus. If the second explanation were true then antagonists, which do not induce receptor sequestration, would not be expected to block HIV entry. Recently published data (29Arenzana-Seisdedos F. Virelizier J-L. Rousset D. Clark-Lewis I. Loetscher P. Moser B. Baggiolini M. Nature. 1996; 383: 400Google Scholar) and our own studies 3Simmons, G., Clapham, P. R., Picard, L., Offord, R. E., Rosenkilde, M. M., Schwartz, T. W., Buser, R., Wells, T. N. C., and Proudfoot, A. E. I. (1997) Science, in press. confirm that chemokine antagonists are effective at inhibiting HIV entry into target cells. Consequently our study strongly suggests that the observed HIV suppressive effects of chemokines are due to competitive inhibition for receptor binding and not due to receptor down-regulation. These findings have significant implications for the design and discovery of novel therapeutic agents targeted at blocking HIV entry into cells. INTRODUCTIONChemokines are a large family of chemotactic proteins which regulate leukocyte activation and recruitment to sites of inflammation. They can be divided into two main classes, the CXC- and the CC-chemokines based on the spacing of the first two cysteine residues. Both CXC- and CC-chemokines bind to seven transmembrane G-protein coupled receptors (GPCRs) 1The abbreviations used are: GPCRG-protein coupled receptorPCRpolymerase chain reactionHIVhuman immunodeficiency virusHAhemagglutininNBD7-nitrobenz-2-oxa-1,3-diazole-4-ylMIPmacrophage inflammatory polypeptideCHOChinese hamster ovary. which in most cases are promiscuous in that they will bind more than one ligand with high affinity (1Power C.A. Wells T.N.C. Trends Pharmacol. Sci. 1996; 17: 209-213Google Scholar). Although receptor-mediated endocytosis of certain GPCRs such as the β2-adrenergic receptor has been well documented, the endocytic pathways utilized by most GPCRs are still uncharacterized. Following stimulation by agonist, the β2-adrenergic receptor is rapidly phosphorylated by a specific GPCR kinase. This uncouples the interaction between the receptor and the G-protein and allows binding of β-arrestin which acts as an adapter to recruit the receptor into clathrin-coated vesicles (2Goodman O.B. Krupnick J.G. Santini F. Gurevich V.V. Penn R.B. Gagnon A.W. Keen J.H. Benovic J.L. Nature. 1996; 383: 447-450Google Scholar). Several studies have shown that GPCRs are internalized via clathrin-coated vesicles, although this may not be a universal mechanism as alternative endocytic pathways have been described for a number of different ligands (3Raposo G. Dunia I. Delavier-Klutchko C. Kaveri S. Strosberg A.D. Benedetti E.L. Eur. J. Cell Biol. 1989; 50: 340-352Google Scholar, 4von Zastrow M. Kobilka B.K. J. Biol. Chem. 1992; 267: 3530-3538Google Scholar, 5Grady E.F. Slice L.W. Brant W.O. Walsh J.H. Payan D.G. Bunnett N.W. J. Biol. Chem. 1995; 270: 4603-4611Google Scholar, 6Fonseca M.I. Button D.C. Brown R.D. J. Biol. Chem. 1995; 270: 8902-8909Google Scholar, 7Zhang J. Ferguson S.S,G. Barak L.S. Menard L. Caron M.G. J. Biol. Chem. 1996; 271: 18302-18305Google Scholar, 8Garland A.M. Grady E.F. Lovett M. Vigna S.R. Frucht M.M. Krause J.E. Bunnett N.W. Mol. Pharmacol. 1996; 49: 438-446Google Scholar, 9Tolbert L.M. Lameh J. J. Biol. Chem. 1996; 271: 17335-17342Google Scholar, 10Berry S.A. Shah M.C. Khan N. Roth B.L. Mol. Pharmacol. 1996; 50: 306-313Google Scholar).Chemokines such as RANTES (regulated on activation normal T cell expressed and secreted), MIP-1α, and MIP-1β have been shown to be the major HIV-suppressive factors produced by CD8+ T cells (11Cocchi F. DeVico A.L. Garzino-Demo A. Arya S.K. Gallo R.C. Lusso P. Science. 1995; 270: 1811-1815Google Scholar), and a number of recent studies have gone on to show that chemokine receptors are co-receptors along with CD4 for the entry of HIV into cells (12Feng Y. Broder C.C. Kennedy P.E. Berger E.A. Science. 1996; 272: 872-877Google Scholar, 13Deng H. Liu R. Ellmeier W. Choe S. Unutmaz D. Burkhart M. Di Marzio P. Marmon S. Sutton R.E. Hill C.M. Davis C.B. Peiper S.C. Schall T.J. Littman D.R. Landau N.R. Nature. 1996; 381: 661-666Google Scholar, 14Dragic T. Litwin V. Allaway G.P. Martin S.R. Huang Y. Nagashima K.A. Cayanan C. Maddon P.J. Koup R.A. Moore J.P. Paxton W.A. Nature. 1996; 381: 667-673Google Scholar, 15Choe H. Farzan M. Sun Y. Sullivan N. Rollins B. Ponath P.D. Wu L. Mackay C.R. LaRosa G. Newman W. Gerard N. Gerard C. Sodroski J. Cell. 1996; 85: 1135-1148Google Scholar, 16Doranz B.J. Rucker J. Yi Y. Smyth R.J. Samson M. Peiper S.C. Parmentier M. Collman R.G. Doms R.W. Cell. 1996; 85: 1149-1158Google Scholar, 17Bleul C.C Farzan M. Choe H. Parolin C. Clark-Lewis I. Sodroski J. Springer T. Nature. 1996; 382: 829-833Google Scholar, 18Oberlin E. Amara A. Bachelerie F. Bessia C. Virelizier J-L. Arenzana-Seisdedos F. Schwartz O. Heard J-M. Clark-Lewis I. Legler D.F. Loetscher M. Baggiolini M. Moser B. Nature. 1996; 382: 833-835Google Scholar, 19Berson J.F. Long D. Doranz B.J. Rucker J. Jirik F.R. Doms R.W. J. Virol. 1996; 70: 6288-6295Google Scholar, 20D'Souza P.M. Harden V.A. Nature Med. 1996; 2: 1293-1300Google Scholar, 21Fauci A.S. Nature. 1996; 384: 529-534Google Scholar). This has raised the possibility of using chemokines therapeutically to block HIV entry. Receptor-mediated endocytosis of chemokines has not been studied in detail, but an understanding of this process is clearly important if we are to develop selective therapeutic agents which block HIV entry into cells. To this end we have studied the receptor-mediated endocytosis of CC-chemokines by CCR1 as a prototype for the CC-chemokine receptor family.