Cells T+ Both Resting and CD3-Stimulated CD4 Chemokine

Engagement of CD28 Modulates CXC Chemokine Receptor 4 Surface Expression in Both Resting and CD3-Stimulated CD4+ T Cells This information is current as of October 2, 2021. Paola Secchiero, Davide Zella, Sabrina Curreli, Prisco Mirandola, Silvano Capitani, Robert C. Gallo and Giorgio Zauli J Immunol 2000; 164:4018-4024; ;

doi: 10.4049/jimmunol.164.8.4018 Downloaded from http://www.jimmunol.org/content/164/8/4018

References This article cites 44 articles, 28 of which you can access for free at:

http://www.jimmunol.org/content/164/8/4018.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

by guest on October 2, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2000 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Engagement of CD28 Modulates CXC Chemokine Receptor 4 Surface Expression in Both Resting and CD3-Stimulated CD4؉ T Cells1

Paola Secchiero,2,3*† Davide Zella,2* Sabrina Curreli,* Prisco Mirandola,*† Silvano Capitani,† Robert C. Gallo,* and Giorgio Zauli‡

Optimal CD4؉ T cell activation requires the cooperation of multiple signaling pathways coupled to the TCR-CD3 complex and to the CD28 costimulatory molecule. In this study, we have investigated the expression of surface CXC chemokine receptor 4 .CXCR4) in enriched populations of CD4؉ T PBL, stimulated with anti-CD3 and anti-CD28 mAbs, immobilized on plastic) Anti-CD3 alone induced a progressive down-regulation of surface CXCR4, accompanied by a signiﬁcant decline in the entry of the HXB2 T cell line-tropic (X4-tropic) HIV-1 clone in CD4؉ T cells. Of note, this effect was strictly dependent on the presence in culture of CD14؉ monocytes. On the other hand, anti-CD28 alone induced a small but reproducible increase in the expression Downloaded from of surface CXCR4 as well as in the entry of HXB2 HIV-1 clone in resting CD4؉ T cells. When the two mAbs were used in combination, anti-CD28 potently synergized with anti-CD3 in inducing the expression of CD69 activation marker and stimulating .the proliferation of CD4؉ T cells. On the other hand, anti-CD28 counteracted the CXCR4 down-modulation induced by anti-CD3 The latter effect was particularly evident when anti-CD28 was associated to suboptimal concentrations of anti-CD3. Because CXCR4 is the major coreceptor for the highly cytopathic X4-tropic HIV-1 strains, which preferentially replicate in proliferating /CD4؉ T cells, the ability of anti-CD28 to up-regulate the surface expression of CXCR4 in both resting and activated CD4؉ T cells http://www.jimmunol.org provides one relevant mechanism for the progression of HIV-1 disease. The Journal of Immunology, 2000, 164: 4018–4024.

he 44-kDa homodimer CD28 is present on the surface of 5), including activation of c-Jun N-terminal kinase (JNK)4 (6–8), most CD4ϩ and some CD8ϩ T cells (reviewed in Refs. 1 and phosphatidylinositol 3-kinase (PI 3-K) (9–11). T and 2). Functional studies by numerous laboratories have Chemokines are a group of proteins that mediate directed leu- shown that binding of CD28 by its counterreceptors (CD80/B7.1, kocyte migration (reviewed in Ref. 12). Based on the locations of CD86/B7.2/B70, B7.3), expressed on the surface of APC, such as four cysteine residues that form disulfide bonds, these small (6–14 monocytes, dendritic cells, and activated B cells (2), determines kDa) basic substances have been classified into two main families, by guest on October 2, 2021 the fate of the T cell response (1–2). In fact, optimal CD4ϩ T cell CC or CXC chemokines. Both CC and CXC chemokines bind to activation requires a non-Ag-driven cosignal provided by CD28 in seven-transmembrane G protein-coupled receptors, which trans- addition to engagement of the TCR-CD3 complex with CD4 (2). duce signals through heterotrimeric G proteins (reviewed in Ref. CD28 plays a critical role in CD4ϩ T cell costimulation leading to 13). A number of studies demonstrated that the CCR5 (R5) and the production of cytokines (in particular IL-2) (3), and expression CXCR4 (X4) chemokine receptors act as major coreceptors for the of cytokine receptors, including the high affinity ␣-chain of the entry of macrophage-tropic and T cell line-tropic HIV-1 strains, IL-2R (4). Both events are required for the progression of T cell respectively, into target CD4ϩ cells (reviewed in Ref. 14). response. It has also been shown that CD28 engagement induces Most resting T lymphocytes show a functional expression of various biochemical events associated with signal transduction (2, surface CXCR4, which is down-regulated by exposure to its high affinity ligand, stromal derived factor-1␣, as well as by phorbol esters (15–18). Although earlier studies reported an up-regulation of the CXCR4 promoter and mRNA in the presence of mitogenic stimulation (19–21), subsequent studies have shown that mito-

*Institute of Human Virology, University of Maryland, Baltimore, MD 21201; †De- genic agonists induce the internalization of surface CXCR4 mol- partment of Morphology and Embryology, Human Anatomy Section, University of ecule (22–25). This issue is particularly relevant because it has ‡ Ferrara, Ferrara, Italy; and Institute of Normal Human Morphology, G. D’Annunzio been clearly demonstrated that HIV-1 preferentially replicates in University of Chieti, Chieti Scalo, Chieti, Italy activated/proliferating CD4ϩ T cells (26). Thus, the aim of this Received for publication September 20, 1999. Accepted for publication February 2, 2000. study was to analyze the expression of surface CXCR4 in resting ϩ The costs of publication of this article were defrayed in part by the payment of page CD4 T cells as well as in cells treated with anti-CD3 mAb alone charges. This article must therefore be hereby marked advertisement in accordance or in combination with anti-CD28 mAb, which mimics the phys- with 18 U.S.C. Section 1734 solely to indicate this fact. iological activation of CD4ϩ T cells (1–2). 1 This research was supported in part by the AIDS Project of the Italian Ministry of Health, and local funds of the G. D’Annunzio University of Chieti and University of Ferrara. 2 P.S. and D.Z. equally contributed to this study. 3 Address correspondence and reprint requests to Dr. Paola Secchiero, Institute of 4 Abbreviations used in this paper: JNK, c-Jun N-terminal kinase; CXCR, CXC che- Human Virology, University of Maryland Biotechnology Institute, 725 West Lom- mokine receptor; LTR, long terminal repeat; MFI, mean ﬂuorescence intensity; PI bard Street, Baltimore, MD 21201-1192. E-mail address: [email protected] 3-K, phosphatidylinositol 3-kinase.

Materials and Methods Radioactivity incorporated into DNA was measured using an automated Cells liquid scintillation counter. Results were expressed as arithmetic mean cpm of triplicate cultures. PBMC were isolated by Ficoll-Hypaque density-gradient centrifugation (Pharmacia, Uppsala, Sweden) of heparinized leukocyte units obtained HIV-1 infection assay from 14 healthy adult donors, who gave their informed consent to this research according to the Helsinki declaration of 1975. Enriched popula- At 48 h post-seeding, cells were infected with HIV-1 (HXB2 clone; Ad- tions of resting CD4ϩ T cells were isolated by immunomagnetic negative vanced Biotechnologies, Columbia, MD; multiplicity of infection of 0.01) selection with Dynabeads M450 (Dynal, Polyscience, PA). For this pur- for 3 h and then washed three times with PBS. Viral stocks were treated before use with RNase-free DNase I (Boehringer Mannheim, Indianapolis, pose, we used a cocktail of mAbs against CD19 and CD20, present on B ϩ cells; CD16 on granulocytes and NK; CD56 and CD57, on NK; CD14 on IN) to remove contaminating DNA. Fourteen hours postinfection, CD4 T monocytes; and CD8 on T cells (all mAb were from Immunotech-Coulter, cells were collected, resuspended in proteinase K-lysis buffer, and allowed Marseilles, France). The final cultures of CD4ϩ T cells thus obtained were to incubate at 56°C for 60 min, then at 98°C for 20 min. Serial dilutions of always Ͼ80% pure, as determined by staining performed with a combi- cell lysates were subjected to HIV-1 DNA PCR by using the following nation of FITC-conjugated anti-CD3 plus PE-conjugated anti-CD4 mAb primers designed based on previously published sequences (29): 5Ј primer, (Becton Dickinson, San Jose, CA), followed by flow cytometry analysis. 5Ј-TCTCTCTGGTTAGACCAGATCTG; 3Ј primer, 5Ј-ACTGCTA The major contaminating cell population was represented by CD14ϩ GAGATTTTCCACACTG. These primers, which amplify a 180-bp frag- monocytes, whose number was between 5 and 10% in the different cell ment in the LTR R/U5 region, were designed to detect early steps in re- preparations, as evaluated by forward-side scatter analysis and staining verse transcription (30). Samples were subjected to 40 cycles of with PE-conjugated anti-CD14 (Becton Dickinson), followed by flow amplification (95°C for 1 min, 50°C for 1 min, and 72°C for 1 min). cytometry. Negative controls were represented by samples containing buffer only or In some experiments, PBMC from the same donors were divided in two uninfected cells. The PCR products were separated on a 2% agarose gel, 32 aliquots and subjected to the immunomagnetic negative selection proce- transferred to a nylon membrane, hybridized with an P-labeled oligonu- dure, as described above, either in the absence or presence of anti-CD14 cleotide probe (5Ј-CTCAATAAAGCTTGCCTTGAGTGCTTCAAGTA Downloaded from mAb. When anti-CD14 mAb was not included in the cocktail of mAbs, the GTGTGTGC) against an internal sequence of the HIV PCR product, and number of CD14ϩ monocytes recovered in the final cell population ranged analyzed after exposure to x-ray film. To normalize for the quantity of between 20 and 30%. When anti-CD14 mAb was included in the cocktail DNA in each sample, ␤-globin PCR was conducted (␤-globin primers; of mAbs, the number of CD14ϩ monocytes was comprised between 5 and Stratagene, La Jolla, CA). ␤-Globin PCR products were visualized under 10%. After the immunomagnetic negative selection containing anti-CD14 UV light after staining of agarose gels with ethidium bromide. Each sample mAb, one-half of the cell preparation was seeded in tissue flasks for3hat was amplified in duplicate or triplicate.

37°C to eliminate residual monocytes by plastic adherence. After this step, http://www.jimmunol.org/ the number of CD14ϩ monocytes was lowered to Ͻ1%. At the end of Statistical analysis puriﬁcation, cells were resuspended in AIM-V serum-free medium (Life Ϯ Technologies, Grand Island, NY) at 1.8 ϫ 106 cells/ml and seeded (0.6 The results were expressed as means SDs of three or more experiments ml/well) in 48-well ﬂat-bottom plates, coated as described below. performed in duplicate. Statistical analysis was performed using the two- tailed Student’s t test. Adherence of Abs and proteins to microtiter plates Anti-CD3 (IgG2a-clone ϫ 35; Immunotech-Coulter), anti-CD28 (IgG1; Results Becton Dickinson), and isotype-matched irrelevant mAbs (Becton Dickin- Stimulation with either anti-CD3 mAb or anti-CD28 mAb alone son) as well as full-length synthetic HIV-1 Tat protein (86 aa, derived from shows striking differential effects on surface CXCR4 in enriched the HIV-1 T cell line-tropic BH10 strain; Technogen, Caserta, Italy) (27)

ϩ by guest on October 2, 2021 were resuspended in PBS containing 0.1% BSA (Sigma, St. Louis, MO). CD4 T cells Forty-eight-well flat-bottom polystyrene plates (Costar, Cambridge, MA) The surface CXCR4 levels were initially evaluated in enriched were coated overnight at 4°C with anti-CD3, anti-CD28, isotype-matched ϩ irrelevant mAb (IgG1 and IgG2a; Immunotech-Coulter), Tat, BSA, used populations of CD4 T PBL, before and after seeding in serum- alone or in combination, at the concentrations indicated in the text. Plates free medium on wells coated with BSA, anti-CD3, anti-CD28, or were then rinsed with AIM-V serum-free medium (Life Technologies) to isotype-matched irrelevant mAbs. The dose of proteins or Abs remove nonadherent proteins, and medium was immediately added to the reported in all experiments is that added overnight to the plates to plates after the final wash. coat the wells and does not represent the amount of proteins bound FACS evaluation of surface markers to each well. This has been estimated to be 1–4% of the dose The surface expression of CXCR4, CCR5, and CD69 was evaluated by added (28, 30). direct staining with the PE-conjugated anti-CXCR4 (PharMingen, San Di- Surface CXCR4 was restricted to 20–40% of freshly isolated ego, CA; clone 12G5), FITC-conjugated anti-CCR5 (R&D Systems, Min- CD4ϩ T cells. Upon culture at 37°C in BSA-treated control wells, neapolis, MN; clone 45502.111), and PC5-conjugated anti-CD69 (Immu- CXCR4 expression increased rapidly, with the percentage of ϫ 5 notech-Coulter) mAbs. Briefly, aliquots of 3 10 cells were stained with CXCR4ϩ cells raising up to 90% after 4–24 h in culture, persisting 5 ␮l of each mAb in 200 ␮l of PBS containing 2% FCS, at 4°C for 30 min. Nonspecific fluorescence was analyzed by using isotype-matched controls. to high levels for up to 72 h (Fig. 1A). Interestingly, when cells After staining procedures, samples were analyzed using FACSCalibur were seeded on plates coated with anti-CD3 mAb alone (1 ␮g/ (Becton Dickinson). In the experiments performed with increasing concen- Ͻ ϩ ϩ ϩ well), a significant ( p 0.01) decrease in the number of CXCR4 trations of monocytes, CD4 T lymphocytes and CD14 monocytes were cells was noticed from 24 h onward (Fig. 1A). Maximal levels of identified by forward-side scatter analysis and by double staining with PE-conjugated anti-CXCR4 ϩ FITC-conjugated anti-CD4 (Becton Dick- CXCR4 inhibition were observed at 48 h of culture (Fig. 1A), as inson for CD4ϩ T lymphocytes) or PE-conjugated anti-CXCR4 ϩ FITC- also shown by the analysis of CXCR4 MFI (Fig. 1B). On the other conjugated anti-CD14 (for monocytes). Data collected from 10,000 cells hand, the percentage of CXCR4ϩ cells did not significantly ( p Ͼ are presented as either percentage of positive cells or mean fluorescence 0.1) differ between plates coated with BSA or anti-CD28 mAb. intensity (MFI) values. However, with respect to control cells treated with BSA, anti- [3H]Thymidine incorporation assay CD28 mAb induced a small but reproducible ( p Ͻ 0.05) increase Cell proliferation was evaluated by [3H]thymidine incorporation assay, of CXCR4 surface expression, evaluated as MFI, between 24 and performed as described previously (28). CD4ϩ T cells were first seeded in 48 h of culture (Fig. 1C). 48-well flat-bottom plates coated with BSA, anti-CD3 (1 ␮g/well), anti- In parallel, we have analyzed the surface expression of CD69 CD28 (1 ␮g/well), anti-CD3 ϩ anti-CD28 (1 ␮g each/well) mAbs at the 6 early activation marker (31), which was confined to 10–30% of concentration of 1.8 ϫ 10 cells/ml (0.6 ml/well), as described above. ϩ Thirty-six hours postseeding, aliquots of 0.15 ml were harvested and freshly isolated CD4 T cells. Surface CD69 expression was tran- seeded in 96-well tissue culture plates (Costar), supplemented with 1 ␮Ci siently up-regulated by seeding the cells in serum-free BSA- [3H]thymidine (6.7 Ci/mmol; DuPont, Boston, MA) for additional 12 h. treated cultures at 37°C, reached maximal expression (50–90% of 4020 CD28 UP-REGULATES CXCR4 SURFACE EXPRESSION Downloaded from

FIGURE 2. Kinetics of CD69 surface expression in CD4ϩ T cells seeded on plates coated with BSA, anti-CD3 (1 ␮g/well), or anti-CD28 (1 ␮g/well) mAb, immobilized on plastic. Data are expressed as means Ϯ SD of four independent experiments, in which the surface expression of CD69 http://www.jimmunol.org/ was analyzed by FACS and reported as percentage of positive cells in cultures performed for the time indicated.

CD69 at 24 h (Fig. 2). No signiﬁcant differences in CXCR4 and CD69 surface expression were noticed between cells seeded on plates coated with BSA or irrelevant mAb. Therefore, in next ex-

periments, we only used BSA as a negative control. by guest on October 2, 2021

CD3- and CD28-induced modulation of surface CXCR4 correlates with entry of the HXB2 X4-tropic strain In next experiments, primary CD4ϩ T cells were cultured for 48 h in BSA- or anti-CD3- or anti-CD28-coated plates, and then incu- bated with HXB2 inoculum for 3 h. After additional 14 h of culture in fresh medium, samples were analyzed by PCR for the presence/ amount of viral DNA as a measurement of viral entry. Semiquan- titative PCR of strong-stop DNA (with LTR R/U5 primers), an early product of reverse transcription, revealed significant lower levels of proviral DNA in cells seeded on CD3-coated plates with respect to those seeded on control (BSA-coated) plates (Fig. 3). On FIGURE 1. Kinetics of CXCR4 surface expression in CD4ϩ T cells seeded on plates coated with BSA, anti-CD3 (1 ␮g/well), or anti-CD28 (1 ␮g/well) mAb, immobilized on plastic. A, Data are expressed as mean Ϯ SD of four independent experiments, in which the surface expression of CXCR4 was analyzed by FACS and reported as percentage of positive cells in cultures performed for the time indicated. B and C, MFI of surface CXCR4 analyzed at 48 h in CD4ϩ T cells seeded on plates coated with BSA, anti-CD3 (B), or anti-CD28 mAb (C). A representative of four separate experiments is shown. B and C, Dotted lines, staining with irrelevant isotype-matched mAb; thin lines, BSA-treated cells; thick line, anti-CD3- FIGURE 3. Effect of anti-CD3 and anti-CD28 on infection of CD4ϩ T treated (B) or anti-CD28-treated (C) cells. Horizontal axis, surface CXCR4 cells by X4-dependent HIV-1 HXB2. Primary CD4ϩ T cells were cultured expression detected by PE fluorescence intensity; vertical axis, relative cell in BSA- or anti-CD3 (1 ␮g/well)- or anti-CD28 (1 ␮g/well)-coated plates number. for 48 h and then infected with HIV-1 for 3 h. Five-fold serially diluted aliquots (lanes 1–5) were amplified with primers for the LTR R/U5 region,

ϩ as described in Materials and Methods, and analyzed for signal after South- CD69 cells) at 4 h, and rapidly declined thereafter (Fig. 2). Anti- ern blot hybridization. A ␤-globin amplification was used to confirm com- CD3 induced a persistent and significant ( p Ͻ 0.01) increase in the parability of the samples (the ethidium bromide-stained agarose gel of PCR ϩ number of CD69 cells with respect to BSA from 24 h onward. In products is shown). The data are representative of three separate turn, anti-CD28 alone induced only a transient up-regulation of experiments. The Journal of Immunology 4021 Downloaded from

FIGURE 4. Expression of CXCR4 (A) and CD69 (B) surface markers and [3H]thymidine incorporation (C)in CD4ϩ T cells seeded on plates coated with BSA, anti-CD3 (1 ␮g/well), anti-CD28 (1 ␮g/well), or anti-CD3 (1 ␮g/ well) plus anti-CD28 (1 ␮g/well) mAbs. A, Data are reported as means Ϯ SD of ﬁve independent experiments, in http://www.jimmunol.org/ which the levels of surface CXCR4 were analyzed by FACS and reported as percentage of positive cells in cultures performed for the time indicated. B, Data are expressed as means Ϯ SD of ﬁve independent experiments, in which the levels of surface CD69 were analyzed by FACS and reported as MFI evaluated after 24 h of culture. C, Data are reported as means Ϯ SD of three independent experiments, in which the [3H]thymidine incorporation was measured by liquid scintillation, and results were expressed as arithmetic mean cpm of triplicate cultures. by guest on October 2, 2021

the other hand, cells seeded on CD28-coated plates showed levels CD28 counteracts the CD3-mediated down-regulation of surface of proviral DNA reproducibly higher than those found in both CXCR4 expression in enriched CD4ϩ T cells CD3- and BSA-coated plates. These ﬁndings are in agreement with previous data from our group (30) and other groups (22–24), who In the following experiments, cells were seeded on plates coated have demonstrated that the susceptibility of CD4ϩ T cells to highly with high concentrations (1 ␮g/well) of anti-CD3 plus anti-CD28 ϩ cytopathic (X4-tropic) HIV-1 strains is directly related to the lev- mAbs, a treatment mimicking the in vivo activation of CD4 T els of CXCR4 exposed on the cell surface. cells by APC (1, 2). Of note, in cells costimulated by anti-CD3 ϩ 4022 CD28 UP-REGULATES CXCR4 SURFACE EXPRESSION

FIGURE 6. CXCR4 surface expression in CD4ϩ T cells cultured with increasing concentrations of CD14ϩ monocytes. Cells were seeded on plates Downloaded from coated with BSA, anti-CD3 (1 ␮g/well), anti-CD3 plus anti-CD28 (1 ␮g/well, each), or anti-CD3 (1 ␮g/well) plus HIV-1 Tat protein (0.4 ␮g/well). CD4ϩ T lymphocytes were distinct from CD14ϩ monocytes on the basis of their forward-side scatter features and bright expression of surface CD4 (CD4ϩ T cells) vs dim expression of surface CD4 and bright expression of surface CD14 (monocytes). Data are expressed as means Ϯ SD of three to ﬁve independent http://www.jimmunol.org/ experiments, in which the levels of surface CXCR4 were analyzed by FACS at 48 h and expressed as percentage of positive cells.

tion of CD69 activation marker. In fact, the levels of CD69 MFI were signiﬁcantly ( p Ͻ 0.05) higher in CD4ϩ T cells plated on anti-CD3 ϩ anti-CD28 than those observed in cells seeded on BSA, anti-CD3, or anti-CD28 (Fig. 4B). In parallel, the effect of anti-CD3 and anti-CD28, used alone or in combination, was evaluated on CD4ϩ T cell proliferation. As shown by guest on October 2, 2021 in Fig. 4C, most primary CD4ϩ T cells seeded for 36 h on BSA or anti-CD28 mAb remained quiescent. Anti-CD3 mAb alone induced a

FIGURE 5. Dose response for CD3-induced down-regulation of CXCR4 surface expression in CD4ϩ T cells cultured for 48 h in wells coated with increasing concentrations of anti-CD3 mAb (0.1 ng to 1 ␮g/ well) plus BSA or anti-CD28 mAb (1 ␮g/well) or HIV-1 Tat protein (0.4 ␮g/well). A, Data represent the means Ϯ SD of four separate experiments and are expressed as percentage of positive cells. B and C, The MFI of surface CXCR4 was analyzed at 48 h in cultures treated with anti-CD3 plus anti-CD28 mAbs (B) or anti-CD3 plus anti-Tat (C) in comparison with anti-CD3 mAb alone. A representative of four separate experiments is shown. Dotted lines, staining with irrelevant isotype-matched mAb; thin lines, anti-CD3-treated cells; thick lines, anti-CD3 plus anti-CD28-treated cells (B) or anti-CD3 mAb plus HIV-1 Tat (C). Horizontal axis, surface CXCR4 expression detected by PE fluorescence intensity; vertical axis, relative cell number. anti-CD28 mAbs, surface CXCR4 expression showed a progres- FIGURE 7. Kinetics of CCR5 surface expression in CD4ϩ T cells sive decline reaching the lowest values at 48 h after the beginning treated with BSA, anti-CD3 mAb (1 ␮g/well), anti-CD28 mAb (1 ␮g/well), of the treatment (Fig. 4A). However, at this time point (48 h), the ␮ ϩ HIV-1 Tat protein (0.4 g/well) immobilized on plastic alone or in com- number of CXCR4 cells in well coated with anti-CD3 plus anti- bination. Data are expressed as means Ϯ SD of four independent experi- CD28 was significantly ( p Ͻ 0.01) higher than that observed in ments, in which the surface expression of CCR5 was analyzed by flow well coated with anti-CD3 alone (Fig. 4A). On the other hand, cytometry in cultures performed for the time (h) indicated, and expressed anti-CD3 ϩ anti-CD28 synergized in inducing maximal stimula- as percentage of positive cells. The Journal of Immunology 4023 significant ( p Ͻ 0.01) increase of [3H]thymidine uptake, with respect cells in bone marrow (37). Consistently, we and other authors (22– to cells seeded on BSA or anti-CD28. The combination of immobi- 24) have shown that freshly isolated PBL CD4ϩ T cells express lized anti-CD3 ϩ anti-CD28 mAbs potently stimulated the prolifer- low levels of surface CXCR4 and that the receptor is rapidly ex- ation of CD4ϩ T cells to levels significantly greater ( p Ͻ 0.01) than pressed on the cell surface upon culture at 37°C. those observed in cells seeded on anti-CD3 alone (Fig. 4C). In this study, we have dissected for the first time the individual The up-regulation of CXCR4 mediated by CD28 mAb was rem- role of CD3 and CD28 in modulating CXCR4 expression. Coli- iniscent of a similar effect observed in the presence of extracellular gation of CD3 and CD28 is required for optimal stimulation of HIV-1 Tat protein immobilized on plastic (30). To further inves- CD4ϩ T cells (1, 2), as also documented in this study by their tigate these similarities, anti-CD3 mAb was titered down to 0.1 ability to up-regulate CD69 activation marker and to induce a sig- ng/well in the absence or presence of predetermined optimal con- nificant increase of [3H]thymidine uptake with respect to anti-CD3 centrations of anti-CD28 mAb (1 ␮g/well) or Tat protein (0.4 ␮g/ alone. However, the effects of anti-CD3 and anti-CD28 on surface well) (Fig. 5A). The CD3 mAb-induced CXCR4 down-regulation CXCR4 were divergent. In fact, anti-CD3 mAb induced an abrupt was dose dependent, with significant ( p Ͻ 0.01) decrease of sur- down-regulation of surface CXCR4 in CD4ϩ T cells that required face CXCR4 being observed for concentrations of anti-CD3 mAb the presence in culture of at least 5–10% CD14ϩ monocytes. On ranging between 1000 and 10 ng/well at 48 h. Anti-CD28 mAb and the other hand, anti-CD28 mAb alone induced a small but repro- extracellular Tat were able to efficiently ( p Ͻ 0.05) counteract the ducible up-regulation of CXCR4 surface expression in resting suppression of surface CXCR4 induced by 10–1000 ng/well (for CD4ϩ T cells. When cells were maximally stimulated by the com- anti-CD28) and 100–100 ng/well (for Tat) of anti-CD3 mAb (Fig. bination of anti-CD3 plus anti-CD28 mAbs, they showed interme- 5A). This effect was particularly evident when anti-CD28 mAb diate levels of surface CXCR4, with respect to cells treated with (Fig. 5, A and B) or extracellular HIV-1 Tat (Fig. 5, A and C) was single mAb. Of note, the changes in surface expression levels of Downloaded from combined to 100 ng/well of anti-CD3 mAb. CXCR4 showed a clear correlation with the ability of HXB2 X4- tropic HIV-1 strain to infect CD4ϩ T cells, as also described in The CD3-mediated down-modulation of surface CXCR4 in previous studies (30, 36, 37). However, we have not ruled out the CD4ϩ T cells requires the presence in culture of CD14ϩ possibility that both CD3 and CD28 activation may interfere with monocytes HIV-1 replication also at a postentry step, an issue that deserves Previous studies investigating the role of mitogenic stimulation on further investigation. Finally, a striking parallelism was observed http://www.jimmunol.org/ CXCR4 surface expression were performed using either PBMC or between anti-CD28 mAb and a synthetic peptide corresponding to ϩ partially purified CD4 T cells (19, 21–24). Similarly, the data Tat protein of the BH10 X4-tropic HIV-1 strain, as both molecules illustrated above were obtained using enriched (Ͼ80% pure) were able to up-regulate surface CXCR4. ϩ ϩ CD4 T cells, which contain a fraction of residual CD14 mono- The cell surface expression of CXCR4 appears to be regulated cytes (Յ10% in the different cell preparations). To ascertain the both by gene expression (19, 20), and by continual recirculation of potential role of these accessory cells, the effect of anti-CD3, used the receptor between the cell surface and endosomal compartments alone or in combination with anti-CD28 or HIV-1 Tat protein, was (18, 23, 25, 38). It has been shown that anti-CD3 down-regulates ϩ next evaluated in CD4 T cell preparations containing increasing surface CXCR4 through a protein kinase C-dependent pathway ϩ by guest on October 2, 2021 concentrations of CD14 monocytes. When the percentage of (22, 23, 25). Although we have not addressed the molecular mech- ϩ CD14 monocytes was lowered to Ͻ1% by immunomagnetic neg- anisms underlining the ability of CD28 to counteract the CD3- ative selection and plastic adherence, anti-CD3 mAb (1 ␮g/well) mediated CXCR4 down-regulation, it is well established that, upon ϩ was unable to down-regulate ( p Ͼ 0.1) surface CXCR4 in CD4 interaction with its counterreceptors predominantly expressed on ϩ T cells (Fig. 6). The possibility that the unresponsiveness of CD4 T APC (2), CD28 triggers two main intracellular signals involving cells was due to the adherence procedure itself rather than to the JNK (6–8) and PI 3-K (9–11). Interestingly, extracellular Tat is ϩ monocyte depletion was ruled out, analyzing CD4 T cell prepara- also able to activate both JNK (39) and PI 3-K (40) in lymphoid tions from the same donors progressively enriched in monocytes. The CD4ϩ T cells. It is, therefore, possible that both anti-CD28 and ϩ enrichment in CD14 cells (reaching 20–30% of the total cell pop- extracellular Tat activate common intracellular pathways to elicit ulation) significantly ( p Ͻ 0.05) increased the ability of anti-CD3 to the up-regulation of surface CXCR4 in CD4ϩ T cells. ϩ down-regulate surface CXCR4 expression in CD4 T cells (Fig. 6). Although it has been shown that in vitro HIV-1 can use several chemokine receptors to enter CD4ϩ cells, CCR5 and CXCR4 are Anti-CD28 mAb and extracellular Tat do not significantly affect ϩ the most important coreceptors for HIV-1 infection in vivo (14). the CCR5 surface expression in enriched CD4 T cells In the course of HIV-1 infection, R5 viral strains predominate In the last group of experiments, the effect of anti-CD3 mAb, anti- during the early phase of infection, while dual-tropic and X4 viral CD28 mAb, and Tat, used alone or in combination, was evaluated strains predominate during disease progression to AIDS (reviewed on the surface expression of CCR5. In agreement with previous in Ref. 14). Moreover, evidence showing that deletion in the CCR5 studies (32–34), we found that surface CCR5 was expressed by a gene does not protect against infection by X4-tropic strains (41), ϩ minority of freshly purified resting CD4 T cells and it was sub- and the fact that patients heterozygous for the D32 CCR5-harbor- stantially unaffected by any of the combinations used to stimulate ing virus with an X4 phenotype have a poor prognosis (42, 43), the cells for up to 72 h of culture (Fig. 7). strengthen the role of CXCR4 as a main HIV-1 coreceptor. In agreement with previous data demonstrating that CXCR4 and Discussion CCR5 are differentially expressed in T lymphocytes (32–34), we In contrast to most chemokine systems, which recruit cells to sites found that the CD28- and Tat-mediated up-regulation of CXCR4 of inflammation, the stromal derived factor-1/CXCR4 system is surface expression was not accompanied by significant modifica- believed to localize cells to hemopoietic and lymphoid organs (15, tions of surface CCR5 chemokine receptor in CD4ϩ T cells. Thus, 35, 36). In fact, transgenic mice overexpressing high levels of hu- CD3/CD28 costimulation of CD4ϩ T cells may exert selective man CXCR4 on the membrane of CD4ϩ T cells present a disturbed pressure in favor of X4-tropic isolate production through a com- lymphocyte trafficking, mainly characterized by a severe loss of bination of mechanisms, including production of high levels of CD4ϩ T lymphocytes from peripheral blood and homing of these ␤-chemokines (21), lack of induction, or even down-regulation 4024 CD28 UP-REGULATES CXCR4 SURFACE EXPRESSION

(44) of ␤-chemokine (CC) receptors and sustained expression of tor CXCR4/fusing on various leukocyte subsets: rapid internalization and recy- CXCR4. cling upon activation. J. Immunol. 160:1522. ϩ 19. Carroll, R. G., J. L. Riley, B. L. Levine, Y. Feng, S. Kaushal, D. W. Ritchey, The entry of X4-tropic HIV-1 into CD4 T cells is strictly re- W. Bernstein, O. S. Weislow, C. R. Brown, E. A. Berger, et al. 1997. Differential lated to the expression levels of surface CXCR4 (23–25, 29, this regulation of HIV-1 fusion cofactor expression by CD28 costimulation of CD4ϩ T cells. Science 276:273. study), while critical steps in the HIV-1 cycle, such as integration 20. Moriuchi, M., H. Moriuchi, W. Turner, and A. S. Fauci. 1997. Cloning and and initiation of transcription, depend on cell activation (26, 29). analysis of the promoter region of CXCR4, a coreceptor for HIV-1 entry. J. Im- Even if X4-tropic strains are able to infect resting lymphocytes munol. 159:4322. 21. Riley, J. L., R. G. Carroll, B. L. Levine, W. Bernstein, D. C. St Louis, expressing CXCR4, proviral integration and subsequent HIV-1 O. S. Weislow, and C. H. June. 1997. Intrinsic resistance to T cell infection with replication will not occur in the absence of cell activation. The HIV type 1 induced by CD28 costimulation. J. Immunol. 158:5545. existence of an inverse correlation between the CD3-mediated ac- 22. Jourdan, P., C. Abbal, N. Nora, T. Hori, T. Uchiyama, J. P. Vendrell, J. Bousquet, N. Taylor, J. Pene, and H. Yssel. 1998. IL-4 induces functional cell-surface ex- tivation of T lymphocytes and the surface expression of CXCR4 pression of CXCR4 on human T cells. J. Immunol. 160:4153. ϩ suggests that the activation of CD4 T cells by TCR/CD3 would 23. Bermejo, M., J. Martin-Serrano, E. Oberlin, M.-A. Pedraza, A. Serrano, B. Santiago, not be an advantage to the propagation of X4-tropic strains in CD4 A. Caruz, P. Loetscher, M. Baggiolini, F. Arenzana-Seisdedos, and J. Alcamı`. 1998. Activation of blood T lymphocytes down-regulates CXCR4 expression and interferes lymphocytes because of CXCR4 down-regulation. However, our with propagation of X4 HIV strains. Eur. J. Immunol. 28:3192. data have demonstrated for the first time that both CD28 ligands, 24. Cole, S. W., B. D. Jamieson, and J. A. Zack. 1999. cAMP up-regulates cell surface expression of lymphocyte CXCR4: implications for chemotaxis and as well as HIV-1 Tat protein, play a major role in facilitating HIV-1 infection. J. Immunol. 162:1392. infection with X4-tropic strains of HIV-1 in at least two ways: 1) 25. Peacock, J. W., and F. R. Jirik. 1999. TCR activation inhibits chemotaxis toward by inducing the full activation/proliferation of resting CD4ϩ T stromal cell-derived factor-1: evidence for reciprocal regulation between CXCR4 and the TCR. J. Immunol. 162:215. cells, and 2) by sustaining the surface expression of CXCR4. Due 26. Vrelizier, J. L. 1989. Cellular activation and human immunodeficiency virus in- Downloaded from to the primary role of CD28 as costimulatory molecule (2), its fection. Curr. Opin. Immunol. 2:409. ability to synergize with CD3 in stimulating the proliferation of 27. Ivey-Hoyle, M., J. S. Culp, M. A. Chaikin, B. D. Hellmig, T. J. Matthews, ϩ R. W. Sweet, and M. Rosemberg. 1991. Envelope glycoproteins from biologi- CD4 T cells and to antagonize the CD3-mediated down-regula- cally diverse isolates of immunodeficiency viruses have widely different affinities tion of surface CXCR4 most likely has a relevant functional sig- for CD4. Proc. Natl. Acad. Sci. USA 88:512. nificance for the progression toward AIDS, facilitating the spread- 28. Zauli, G., D. Gibellini, C. Celeghini, C. Mischiati, A. Bassini, M. La Placa, and S. Capitani. 1996. Pleiotropic effects of immobilized vs soluble recombinant ing of X4-tropic strains. HIV-1 Tat protein on CD3-mediated activation, induction of apoptosis and HIV-1 ϩ LTR transactivation in purified CD4 T lymphocytes. J. Immunol. 157:2216. http://www.jimmunol.org/ 29. Zack, J. A., S. J. Arrigo, R. R. Weitsman, A. S. Go, A. Haislip, and L. S. Y. Chen. References 1990. HIV-1 entry into quiescent primary lymphocytes: molecular analysis re- 1. Jenkins, M. K. 1994. The ups and downs of T cell costimulation. Immunity 1:443. veals a labile, latent viral structure. Cell 61:213. 2. Rudd, C. E. 1996. Upstream-downstream: CD28 cosignaling pathways and T cell 30. Secchiero, P., D. Zella, S. Capitani, R. C. Gallo, and G. Zauli. 1999. Immobilized HIV-1 Tat protein up-regulates the expression of surface CXCR4 in resting function. Immunity 4:527. ϩ 3. Powell, J. D., J. A. Ragheb, S. Kitagawa-Sakakida, and R. H. Schwartz. 1998. CD4 T cells. J. Immunol. 162:2427. Molecular regulation of interleukin-2 expression by CD28 co-stimulation and 31. Perfetto, S. P., T. E. Hickey, P. J. Blair, V. C. Maino, K. F. Wagner, S. Zhou, anergy. Immunol. Rev. 165:287. D. L. Mayers, D. St Louis, C. H. June, and J. N. Siegel. 1997. Measurement of 4. Cerdan, C., Y. Martin, M. Courcoul, H. Brailly, C. Mawas, F. Birg, and D. Olive. CD69 induction in the assessment of immune function in asymptomatic HIV- 1992. Prolonged IL-2 receptor ␣/CD25 expression after T cell activation via the infected individuals. Cytometry 30:1. 32. Bleul, C. C., L. Wu, J. A. Hoxie, T. A. Springer, and C. R. Mackay. 1997. The

adhesion molecules CD2 and CD28: demonstration of combined transcriptional by guest on October 2, 2021 and post-transcriptional regulation. J. Immunol. 149:2255. HIV coreceptors CXCR4 and CCR5 are differentially expressed and regulated on 5. Nunes, J. A., Y. Collette, A. Truneh, D. Olive, and D. A. Cantrell. 1994. The role human T lymphocytes. Proc. Natl. Acad. Sci. USA 94:1925. of p21ras in CD28 signal transduction: triggering of CD28 with antibodies, but 33. Qin, S., J. B. Rottman, P. Myers, N. Kassam, M. Weinblatt, M. Loetscher, not the ligand B7-1, activates p21ras. J. Exp. Med. 180:1067. A. E. Koch, B. Moser, and C. R. Mackay. 1998. The chemokine receptors 6. Su, B., E. Jacinto, M. Hibi, T. Kallunki, M. Karin, and B. Neriah. 1994. JNK is CXCR3 and CCR5 mark subsets of T cells associated with certain inflammatory involved in signal integration during costimulation of T lymphocytes. Cell 77:727. reactions. J. Clin. Invest. 101:746. 7. Barz, C., T. Nagel, K. E. Truitt, and J. B. Imboden. 1998. Mutational analysis of 34. Lee, B., M. Sharron, L. J. Montaner, D. Weissman, and R. W. Doms. 1999. CD28-mediated costimulation of Jun-N-terminal kinase and IL-2 production. Quantification of CD4, CCR5, and CXCR4 levels on lymphocyte subsets, den- J. Immunol. 161:5366. dritic cells, and differentially conditioned monocyte-derived macrophages. Proc. 8. Kempiak, S. J., T. S. Hiura, and A. E. Nel. 1999. The Jun kinase cascade is Natl. Acad. Sci. USA 96:5215. 35. Aiuti, A., I. J. Webb, C. Bleul, T. Springer, and J. C. Gutierrez-Ramos. 1997. The responsible for activating the CD28 response element of the IL-2 promoter: proof ϩ ␬ chemokine SDF-1 is a chemoattractant for human CD34 hematopoietic progen- of cross-talk with the I B kinase cascade. J. Immunol. 162:3176. ϩ 9. Pages, F., M. Ragueneau, R. Rottapel, A. Truneh, J. Nunes, J. Imbert, and itor cells and provides a new mechanism to explain the mobilization of CD34 D. Olive. 1994. Binding of phosphatidylinositol-3-OH kinase to CD28 is required progenitors to peripheral blood. J. Exp. Med. 185:111. for T-cell signalling. Nature 369:327. 36. Kitchen, S. G., and J. A. Zack. 1997. CXCR4 expression during lymphopoiesis: 10. Truitt, K. E., C. M. Hicks, and J. B. Imboden. 1994. Stimulation of CD28 triggers implications for human immunodeficiency virus type 1 infection in the thymus. an association between CD28 and phosphatidylinositol 3-kinase in Jurkat T cells. J. Virol. 71:6928. 37. Sawada, S., K. Gowrishankar, R. Kitamura, M. Suzuki, G. Suzuki, S. Tahara, and J. Exp. Med. 179:1071. ϩ 11. Edmead, C. E., Y. I. Patel, A. Wilson, G. Boulougouris, N. D. Hall, S. G. Ward, A. Koito. 1998. Disturbed CD4 T cell homeostasis and in vitro HIV-1 suscep- and D. M. Sansom. 1996. Induction of activator protein (AP)-1 and nuclear fac- tibility in transgenic mice expressing T cell line-tropic HIV-1 receptors. J. Exp. tor-␬B by CD28 stimulation involves both phosphatidylinositol 3-kinase and Med. 187:1439. acidic sphingomyelinase signals. J. Immunol. 157:3290. 38. Tarasova, N. I., R. H. Stauber, and C. J. Michejda. 1998. Spontaneous and ligand- 12. Baggiolini, M. 1998. Chemokines and leukocyte traffic. Nature 392:565. induced trafficking of CXC-chemokine receptor 4. J. Biol. Chem. 273:15883. 13. Premack, B. A., and T. J. Shall. 1996. Chemokine receptors: gateway to inflam- 39. Kumar, A., S. K. Manna, S. Dhawan, and B. B. Aggarwal. 1998. HIV-Tat protein mation and infection. Nat. Med. 2:1174. activates c-Jun N-terminal kinase and activator protein-1. J. Immunol. 161:776. 14. Moore, J. P., A. Trkola, and T. Dragic. 1997. Co-receptors for HIV-1 entry. Curr. 40. Borgatti, P., G. Zauli, M. L. Colamussi, D. Gibellini, M. Previati, L. Cantley, and Opin. Immunol. 9:551. S. Capitani. 1997. Extracellular HIV-1 Tat protein activates phosphatidylinositol 3- and ϩ 15. Bleul, C. C., R. C. Fuhlbrigge, J. M. Casasnovas, A. Aiuti, and T. A. Springer. Akt/PKB kinases in CD4 T lymphoblastoid Jurkat cells. Eur. J. Immunol. 27:2805. 1996. A highly efficacious lymphocyte chemoattractant, stromal cell-derived fac- 41. Michael, N. L., J. A. Nelson, V. N. Kewalramani, G. Chang, S. J. O’Brien, tor 1 (SDF-1). J. Exp. Med. 184:1101. J. R. Mascola, B. Volsky, M. Louder, G. C. White, D. R. Littman, et al. 1998. 16. Amara, A., S. L. Gall, O. Schwartz, J. Salamero, M. Montes, P. Loetscher, Exclusive and persistent use of the entry coreceptor CXCR4 by human immunode- M. Baggiolini, J. L. Virelizier, and F. Arenzana-Seisdedos. 1997. HIV coreceptor ficiency virus type 1 from a subject homozygous for CCR5 ␦32. J. Virol. 72:6040. down-regulation as antiviral principle: SDF-1␣-dependent internalization of the 42. Bratt, G., E. Sandstrom, J. Albert, M. Samson, and B. Wahren. 1997. The influ- chemokine receptor CXCR4 contributes to inhibition of HIV replication. J. Exp. ence of MT-2 tropism on the prognostic implications of the ␦32 deletion in the Med. 186:139. CCR-5 gene. AIDS 11:1415. 17. Signoret, N., J. Oldrige, A. Pelchen-Matthews, P. J. Klasse, T. Tran, L. F. Brass, 43. Michael, N. L., G. Chang, L. G. Louie, I. R. Mascola, D. Dondero, D. L. Birx, M. M. Rosenkilde, T. W. Schwartz, W. Holmes, W. Dallas, et al. 1997. Phorbol and H. W. Sheppard. 1997. The role of viral phenotype and CCR-5 gene defects esters and SDF-1 induce rapid endocytosis and down-modulation of the chemo- in HIV-1 transmission and disease progression. Nat. Med. 3:338. kine receptor CXCR4. J. Cell Biol. 139:651. 44. Loetscher, P., M. Seitz, M. Baggiolini, and B. Moser. 1996. Interleukin-2 regu- 18. Foster, R., E. Kremmer, A. Schubel, D. Breitfeld, A. Kleinschmidt, C. Nerl, lates CC chemokine receptor expression and chemotactic responsiveness in T G. Bernhardt, and M. Lipp. 1998. Intracellular expression of the HIV-1 corecep- lymphocytes. J. Exp. Med. 184:569.