54
Engagement of Adhesion Molecules (CDt8, CDtta, CD45, CD44, and CD58) Enhances Human Immunodeficiency Virus Type t Replication in Monocytic Cells through a Tumor Necrosis Factor-Modulated Pathway
Robin J. Shattock, Gian Paolo Rizzardi,* Peter Hayes, Divisions ofInfectious Diseases and Oncology, St. George's Hospital and George E. Griffin Medical School, London, United Kingdom
Engagement of monocytic cell membrane proteins was shown to enhance human immunodefi ciency virus type 1 (HIV-1) replication in monocytic cells. Cross-linking of CD18, CDlla, or CD45 by immobilized antibodies produced up to an ll-fold enhancement of HIV-1 release in the OM10.1 monocytic cell line in a tumor necrosis factor-a (TNF-a)-dependent manner. In addition, adhesion Downloaded from https://academic.oup.com/jid/article/174/1/54/959003 by guest on 29 September 2021 ofOM10.1 cells to immobilized intercellular adhesion molecule-I (ligand for CDl8/CDlla) induced similar TNF-a-dependent enhancement of HIV-l replication. After phenotypic differentiation of OMlO.l cells, engagement of cell membrane proteins CD18, CD1la, CD44, CD45, or CD58 by soluble antibodies enhanced HIV-1 replication in a TNF-a-dependent manner. These data suggest that cross-linkage of monocytic cell membrane proteins during cell-cell interaction and specifically during antigen presentation to CD4 T cells may enhance HIV-1 replication, facilitating infection of adjacent cells.
Human immunodeficiency virus type 1 (HIV-l)-infected HIV-1 replication in infected peripheral blood monocytes. We macrophages are a persistent reservoir ofvirus in tissue during have previously demonstrated for monocytes that adhesion per all stages ofHIV-I infection [1,2], and migration of such cells se, a phenotypic change associated with differentiation, to tis is likely to facilitate dissemination of HIV -1 throughout the sue culture plastic and endothelial cells enhances HIV -1 tran body. It is well established that replication of HIV-1 in mono scription [8]. In addition, our preliminary studies on cellular cytic cells is dependent on the state of cellular differentiation adhesion suggested that cross-linkage ofCD18 enhanced HIV [3] and correlates with nuclear translocation of the transcrip 1 replication in monocytic cells [9]. These observations have tional activator nuclear factor (NF)-KB and subsequent binding recently been confirmed by other studies [10, 11] in which ofthis factor to the enhancer region ofthe HIV -1 long terminal antibody blocking experiments demonstrated that CD 18 plays repeat sequence. Small numbers ofperipheral blood monocytes a role in the enhancement of HIV -1 replication induced by have been shown to harbor latent HIV-1 infection (positive for adhesion of infected monocytic cells to endothelium. Thus, proviral DNA without detectable release of HIV-l virions), engagement of monocyte adhesion molecules through interac while considerably larger numbers of latently infected macro tion of monocytes or macrophages with other cells clearly has phages have been demonstrated in lymph nodes and bronchoal the potential to enhance HIV -1 replication. veolar lavage fluid from lungs of asymptomatic HIV-1-in Monocyte adherence is known to induce mRNA production fected patients [4- 7]. Control ofHIV-1 production and release for several protooncogenes and monokines, including tumor from such cells is likely to play a crucial role in progression necrosis factor-a (TNF-a) and interleukin (IL)-8, both ofwhich ofHIV-1 disease. Thus, stimuli that induce monocyte differen contain NF-KB binding sites in their promoter regions [12, 13]. tiation and activation have the potential to enhance HIV-1 repli More specifically, engagement of CDI8/CDlla (lymphocyte cation. function-associated antigen [LFA]-I), CD58, CD44, and CD45 Infectious virus can be isolated from infected peripheral on monocyte membranes induces secretion ofIL-l, TNF-a, and blood monocytes on physical interaction with activated T cells with the exception of LFA-I, macrophage colony-stimulating [4, 5], suggesting that monocyte-T cell interactions induce factor (M-CSF), which all enhance HIV-l replication [14-17]. Thus, engagement of specific monocyte adhesion molecules has the potential to directly influence HIV -1 production in infected monocytes through NF-KB-dependent mechanisms. Received 20 November 1995; revised 5 March 1996. In this study, we have investigated the effects of engagement Financial support: Medical Research Council, UK (G9228196PB to R.J.S., P.H., G.E.G.), and Istituto Superiore di Sanita, Rome (G.P.R.). of specific monocyte adhesion molecules on HIV-1 replication Reprints or correspondence: Dr. Robin Shattock, Division ofInfectious Dis to define the mechanism by which monocyte interaction with eases, St. George's Hospital Medical School, Cranmer Terrace, SWl7 ORE, other cells might induce HIV -1 replication. London, UK. * Present affiliation: Istituto di Medicina Intema, Malettie Infettive e Immu nopathologia, Universita di Milano, Milan, Italy. Materials and Methods The Journal of Infectious Diseases 1996; 174:54-62 © 1996 by The University of Chicago. All rights reserved. Cell culture. The OMI0.1 cell line, a derivative of the HL60 0022-1899/96/7401-0007$01.00 myelomonocytic leukemia cell line containing a single integrated JID 1996; 174 (July) Monocyte Adhesion and HIV Replication 55
copy of HIV-I L A v provirus [18], was donated by S. Butera and specific cytokines. The following antibodies were used: anti-TNF T. Folks (COC, Atlanta). Cells were maintained in RPMI 1640 a, CB0006 (25 j.Lg/mL; Celltech, Slough, UK); anti-TNF-a, C168 supplemented with 2 roM glutamine, 100 U/mL penicillin, and 0.1 cA2 (25 j.Lg/mL, > 50X concentration required to neutralize 1 ng/ mg/mL streptomycin (Sigma, Poole, UK) and 5% heat-inactivated mL TNF as stipulated by manufacturer; Centocor, Malvern, PA); fetal calfserum (ICN Flow, High Wycombe, UK). For experimen anti-IL-Ia, S76/BM (111000, lOx concentration required to neu tation, cells were used in the log phase of growth and before use tralize 100 pg ofIL-la; National Institute for Biological Standards were washed three times with PBS to remove exogenous virus. and Control, Potters Bar, UK); anti-IL-I,B, S77/BM (1/500, 5X In some experiments, OMIO.l cells were treated with 10- 8 M concentration required to neutralize 100 pg of IL-I,B as stipulated 1,25 dihydroxycholecalciferol (1,25(OH)2D3) for 24 h before ex by supplier; National Institute for Biological Standards and Con perimentation. After 24 h, I,25(OH)2Drtreated cells were washed trol); and anti-M-CSF, BL-MCP (10 j.Lg/mL, concentration re three times in PBS directly before experimentation and resus quired to neutralize 5 ng of M-CSF as stipulated by supplier; IO~8 pended in fresh culture medium containing M 1,25(OHhD3. Genzyme, Boston). In some experiments, cells were incubated in Downloaded from https://academic.oup.com/jid/article/174/1/54/959003 by guest on 29 September 2021 Cross-linking of cell membrane proteins by immobilized anti the presence of 1 j.Lg/mL IL-l receptor antagonist (IL-IRA; Na bodies. Microtiter plates (3590; 96-well; Costar, Cambridge, tional Institute for Biological Standards and Control). MA) were coated with a primary Ftab"), fragment of goat mono Measurement of RT activity. RT activity was determined by clonal antibody (MAb) against the Fc portion of mouse immuno e2p]dTTP incorporation as described [20]. Samples of cell-free globulins by incubating antibody (50 p,L/well; 10 j.Lg/mL in PBS) tissue culture supernatant (5 j.LL) were added to 25 j.LL of RT at 4°C overnight. Plates were then washed twice with PBS, 50 j.LLI cocktail (4 j.Lg/mL polyadenylic acid, 1.26 j.Lg/mL pd[T] 12-18 [Oligo well of 1% bovine serum albumin (BSA) in PBS was added to (dT); Pharmacia, Milton Keynes, UK], 10 j.LCi/mL thymidine 5' each well, and the plates were incubated for 1 h at room tempera [a-32P]triphosphate triethylammonium salt [Amersham, Amer ture. This step blocked free protein binding sites not occupied by sham, UK], and 0.01% NP-40) and incubated at 37°C for 1.5 h. goat MAb. Plates were subsequently washed and incubated for 1 RT reaction mixture (5 j.LL) was then dotted onto DE81 paper h at room temperature with one of a panel of secondary mouse (Whatman, Maidstone, UK), air dried, and washed four times in MAbs (50 j.LLlwell; 10 j.Lg/mL) directed against specific cell mem 2X standard saline citrate and twice in absolute ethanol. 32p activity brane proteins. The use of a primary Ftab"), anti-mouse MAb was then determined using a direct beta plate reader (Canberra ensured that all secondary antibodies were bound to microtiter Packard, Pangbourne, UK). plates via their Fe portion, ensuring that no Fe portions were able Measurement of TNF-a and IL-1,B in culture supernatant. to interact with Fe receptors on the surface of monocytic cells. Supernatant TNF-a and IL-1,B levels were measured by ELISA The following secondary mouse MAbs were used in experi according to the manufacturer's protocols (R&D Systems, Oxon, ments: anti-C018 (MHM23), anti-CDlla (MHM24), anti-CDIlb UK). The lower limit of sensitivity of the TNF -a assay was 4.4 (M74l), anti-CDIlc (M732), anti-C045 (T29/33), anti-C04 pg/mL and of the IL-1,B assay was 0.3 pg/mL. Bioactive TNF (MT3l0), IgG 1a (X 931) and IgG2a (X 943) control antibodies concentrations were measured using the previously described bio (Dako, High Wycombe, UK), anti-intercellular adhesion mole assay involving the WEHI 164 cell line [20a]. 6 cule-I (ICAM-l; 84HlO), anti-major histocompatibility complex Flow cyto metry. Cells (10 ) were harvested for each test and (MHC)-II (DWHC54), anti-CD44 (F 10-44-2), anti-CD58 (BRIC pelleted by centrifugation at 400 g for 5 min. Cell pellets were 5) (Serotec, Oxford, UK), and anti-MHC-I (W6/32; American resuspended in 100 j.LL of PBS to which primary antibody specific Type Culture Collection, Rockville, MD). for cell surface adhesion molecules was added (1 j.Lg/test). Cells For some experiments, microtiter plates were coated with re were incubated for 30 min at 4°C and washed twice in PBS con combinant ICAM-I-Fc, consisting of the first three domains of taining 1% BSA. Fluorescein isothiocyanate-conjugated anti ICAM-I and the hinge, CH2, and CH3 domains of IgG1(Fc) as mouse antibody (5 j.LL) was added to cells and incubated for 30 described [19] (gift of Nancy Hogg, Imperial Cancer Research min at 4°C, and cells were subsequently washed twice in PBS Fund, London). Before use, microtiter plates were washed twice containing 1% BSA. Cells were fixed in 1 mL of 1% paraformalde with PBS. OMI0.1 cells were washed three times in PBS before hyde in PBS before analysis by flow cytometry in a FACStar addition to microtiter plates to remove all cell-free virus. Cells (Becton Dickinson, Cowley, UK). Results are given as a percent were resuspended in culture media (5 X 105 cells/mL), and 200 age of fluorescence-positive cells and as relative fluorescence in j.LLlwell was added to microtiter plates and incubated at 37°C in tensity calculated according to the following formula: relative flu 5% CO2. PMA stimulation (20 nM) of OMIO.l cells was used as orescence intensity = [mean fluorescence (sample) - mean a positive control for up-regulation of HIV -1 replication. Culture fluorescence (control)]/[mean fluorescence (control)]. supernatants were assayed for reverse transcriptase (RT) activity Statistical analysis. All data are expressed as mean ± SE and after 72 h of incubation. were compared by two-tailed Student's t test. Data were considered Engagement of cell membrane proteins by soluble antibody. statistically significant for P < .02. Specific antibody (1 j.Lg/mL) was added to OM 10.1 cells in microti ter plates (2 X 105 cells/200 j.LLlwell). Addition of antibody had no effect on cell viability, as assessed by Trypan blue exclusion. Results Cells were incubated at 37°C in 5% CO2 for 72 h, after which culture supernatants were assayed for RT activity. Cross-linkage of CD18, CD1ia, or CD45 enhances HIV-i Use of cytokine-neutralizing antibodies (NA4 To determine replication in OMiG.i cells. We examined the role of 11 the effects of induced cytokine secretion on HIV-1 replication, monocyte cell membrane proteins involved in adherence of experiments were done in the presence of NAs directed against monocytes to other cells with respect oftheir capacity to induce 56 Shattock ct al. JID 1996; 174 (July)
Cross-linking of CDI8, CDlla, and CD45 have been shown 2500 to induce TNF-ex and IL-I secretion, and immobilized antibody to CD45 also induces M-CSF secretion [14-16]. These mono 2000 kines enhance HIV-I transcription in monocytic cells [17]. Thus, studies were done with NAs directed against these mono 3 ~ kines to determine whether induced monokines mediated the g: 1500 effects of cross-linking CDI8, CDlla, or CD45 on HIV-I () .2:- replication (figure 2). These neutralization experiments clearly 'S: 13 1000 demonstrated that HIV -1 replication induced by immobilized
(A) neutralizing concentrations ofantibody to TNF-a demonstrated 120 that such enhancement of HIV-I replication was apparently mostly dependent on TNF-a secretion. 100 Engagement ofCDIS. CD l la. CD45. CD44. or CD58 with soluble antibody enhances HIV-I replication in differentiated OMIO.I cells. To determine whether engagement of cell 80 membrane adhesion molecule s modul ated HTV-I replication in E ~ more mature monocytic cells, OM I0.1 cells were differentiated 60 x u.. by exposure to IO- M I,25(OH hD) for 24 h before and during Z I- experiments. Incubation of OM I0.1 cells in 1,25(OHhD) had 40
substantial effects on expressi on of cell membrane proteins Downloaded from https://academic.oup.com/jid/article/174/1/54/959003 by guest on 29 September 2021 measured by flow cytometry. Expression of cell membrane 20 proteins CDI8, CDlla,CD I Tb, CD I lc, TCAM, MHC-I, and CD45 were all increased by exposure to 1,25(OH)2D3, while 0 expression ofMHC-Il was unaffected and expression of CD44 Control CD11a CD18 CD45 and CD58 were reduced (table I). OMIO.I cells differentiated by exposure to T,25(OHhD), while remaining non adherent when cultured in tissue culture (B) plastic, assumed an adherent phenotype to immobilized MAb 60 irrespective of specificity of the MAh. This was in contrast to nonst imulated OM 10.1 cells, which adhered only to immobi lized antibody directed against expressed cell membrane pro 50 teins. Basal levels of HTV-I release from OM I0.1 cells differ entiated in 1,25(OHhD3 were mod erately enhan ced with 40 E respect to nond ifferentiated OM I 0.1 cells (mean RT activity 0, :±: SE,427 :±: 10.6 and 215 :±: 12.4, respectively). Engagement ~30 of CD 18, CDlIa, and CD45 on differentiated OM 10.1 cells with soluble antibod y without cross-linkage greatly enhanced 20 HIV-1 repl ication over control cells incubated in 1,25(OH)2D3
10
0-+-...... ==---..,-- 2500 Control CD11a CD18 CD45
Figure 3. TNF-a (A) and interleukin (IL)-I,B (8) release from 2000 OMIO.I cells after cross-linkage of CDl la , CDI8, and CD45 with immobilized antibodies. Data are means of3 independentexperiments ~ (::!:SE). P < .005 compared with control. * ~ 1500 o ~ seen with any of the immobilized MAbs and isotype-matched ~ 1000 CIl controls that had been demonstrated to have no effect on HIV lcc I replication (data not shown). 500 Adhesion of OMIO.I cells to immobilized ICAM enhances HIV-I replication. Immobilized antibodi es were used in the above experiments to mimic the interaction between cell mem 0-+-----.- bran e adhesion molecules and their respective ligand s. To de Control 5 1 0.1 5 termine whether interaction with natural ligand would result ICAM-1-Fc ~g /ml in similar enhancement of HIV-I repl ication. OMIO.I cells ~ +anti-TNF were allowed to adhere to microtiter plates coated with titrated amounts of recombinant TCAM-I -Fc (figure 4). OMIO.I cells Figure 4.Effects of adhesion of OMI0.1 cells to immobilized in tercellular adhesion molecule (ICAM)- I on HlV-! replication. RT. readily adhered to these plates, and such adherence enhanced reverse transcriptase. Data are means of 3 independent experiments HIV-I replication up to 60-fold in wells coated with 5 I-lg/mL (::!: SE). * P < .01 compared with control, t P < .01 compared with ICAM-I -Fc. Furthermore, experiments done in the presence of adherence to rCAM-Fe (5 j.lg/mL) in absence of anti-TNF. JID 1996; 174 (July) Monocyte Adhesion and HIV Replication 59
3000 3500
3000 2500
"3.. "3.. 2500 !£ 2000 ~ ~ Q. ~ 2000 :; 1500 ~ 's :~ t5 1500 ;:. 1000 a: b: 1000 Downloaded from https://academic.oup.com/jid/article/174/1/54/959003 by guest on 29 September 2021 500 500
0 o ~ co ~ o .c co Control CD18 CD11a CD45 CD44 CD58 ~ ,- ,- ,- ,- '!1 ~ io = e ,- ~ 'E Q. 0 0 0 0 o o a o 0 Ei s < () o o II o () o o Q ~ ~ 0+ anti-TNF Illil+anti- MCSF
Figure 5. HIV-I expression after engagement of cell membrane Figure 6. Effects of neutralizing antibodies against TNF-a and proteins, by soluble antibody, expressed on 1,25 dihydroxycholecal macrophagecolony-stimulating factor (M-CSF) on HIV-I replication ciferol-treated OMIO.I cells. RT, reverse transcriptase; ICAM, inter in 1,25dihydroxycholecalciferol-treated OMIO.I cells after engage cellularadhesion molecule; MHC, major histocompatibility complex. ment of cell membrane proteins. Data are arithmetic mean of 6 inde Data are arithmetic means of 6 independent experiments (:tSE) . * P pendent experiments (:tSE). * P < .00001, ** P < .0000001 for < .00001 compared with control. anti-adhesion molecule + anti-TNF compared with corresponding anti-adhesion molecule alone. Experiments done in presence of M CSF represent means of2 independent experiments. RT, reverse tran scriptase. (figure 5). Furthermore, engagement of CD58 and CD44 with soluble antibody under such conditions also enhanced HIV-I replication in these cells. MAbs to other adhesion molecules cause of the relative insensitivity of this assay, were too low and isotyp e-matched controls had no effect on HIV -I replica for accurate quantification. tion (data not shown for isotype controls). To determine whether induced HIV-I replication, observed Discussion on engagement of cell membrane proteins on differentiated OMIO.I cells, was dependent on cytokine production, experi This study demonstrates that cross-linkage of specific adhe ments were done in the presence ofNAs directed against TNF sion molecules (CDI8, CDlla, and CD45) by immobilized a and M-CSF (not included for CDIIa/CDI8) (figure 6). These neutralization experiments demonstrated that enhanced HIV-I 120 replication on engagement of CDI8, CDlla, CD44, CD45 , or CD58 was dependent on TNF-a secretion. Addition of poly myxin B (25 t-lg/mL) to any ofthe test conditions had no effect 100 on observed enhanced HIV-I replication (data not shown). Engagement ofCD18, CD45. CD44. or CD58 on differenti 80 ated OM10.1 cells with soluble antibody stimulates TNF-a E ~ secretion. Having demonstrated that NAs to TNF-a inhibited 60 u, enhancement of HIV-I replication in differentiated OMIO .I Z t- cells on engagement of CD 18, CD45, CD44, and CD58, we 40 sought to determine the effects of such engagement on TNF a secretion. Engagement of CD 18, CD45, CD44, and CD58 20 on differentiated OM 10.1 cells all enhanced TNF-a secretion (figure 7). No induction ofTNF-a secretion was seen with any 0 ofthe tested MAbs and isotype-matched controls that had been Control CD18 CD44 CD45 CD58 demonstrated to have no effect on HIV -I replication in differ entiated OMIO.I cells . Bioactivity of secreted TNF-a was as Figure 7, TNF-a secretion after engagementof cell membrane pro teins, by soluble antibody, expressed on 1,25 dihydroxycholecalcif sessed using the WEHI bioassay. Engagement ofCD18, CD45, erol-treated OMI0.1 cells. Data are arithmetic means of 3 indepen CD44, and CD58 all resulted in secretion oflevels ofbioactive dent experiments (:tSE). * P < .02, ** P < .005 compared with TNF that were higher than background (13.5 pg/mL) but, be- control. 60 Shattock et al. JlO 1996; 174 (July) antibodies enhances HIV-1 replication in infected OMlO.1 Furthermore, these data suggest that anti-TNF therapies may monocytic cells in a TNF-a-dependent manner. Such cross reduce HIV-1 replication in monocytic cells engaged in cell linkage may mimic clustering of these adhesion molecules by cell interaction through adhesion molecules. As macrophages respective ligands, thought to occur during strong adhesive form a major reservoir of HIV -1 within tissue, such treatments interaction between cells, for example between T cells and directed against TNF-a may delay HIV-l disease progression. macrophages during antigen presentation [22]. Furthermore, CDlla and CD18 are the a and (3 subunits of LFA-l, a experiments using immobilized ICAM-1, the ligand for CD181 member of the (32 integrin supergene family, which binds to 11a (LFA-I), demonstrated that engagement of LFA-I by im ICAM ligands [24]. LFA-l/ICAM interactions mediate the mobilized antibody reflected interaction with the physiologic binding of monocytes to leukocytes and endothelial cells [25, ligand and was dependent on ligand density. 26]. HIV-1 infection of mononuclear cells enhances CD 181
The ability ofsoluble antibodies to CD18, CD11a, and CD45 CDlla expression [27], while engagement ofCD18 on infected Downloaded from https://academic.oup.com/jid/article/174/1/54/959003 by guest on 29 September 2021 to enhance HIV-1 replication in differentiated OM 10.1 cells in T cells induces synergistic signals that enhance HIV-1 replica a TNF-dependent manner suggests that differences in maturity tion in response to T cell receptor stimulation [28]. Such obser may modulate responsiveness to receptor engagement. Further vations, together with data from this study, suggest that LFA more, HIV-1 replication was induced in differentiated OMlO.1 1 may play a key role in the regulation of HIV-1 replication. cells by CD44 and CD58 engagement, neither ofwhich induced It is of interest that LFA-l/ICAM-l interaction has been sug HIV-1 replication in unstimulated OM1 0.1 cells. The observa gested to be critical in the progression of a murine AIDS-like tion that cell surface expression of CD44 and CD58 was re infection [29]. duced in 1,25(OH)2Drtreated cells (table 1) suggests that ac Little is known about the physiologic function of CD45, quisition of HIV-1 enhancement through these membrane although evidence suggests that it is involved in cell-cell inter molecules most likely reflects changes in their ability to gener actions [30, 31], and thus far the only known natural ligand to ate intracellular signaling rather than quantitative increase in CD45 is CD22 expressed on B cells [31]. CD45 is a protein receptor engagement. It cannot be excluded that cell-bound tyrosine phosphatase [32]; it is known to activate tyrosine ki antibody may have enhanced HIV -1 replication through cross nases able to induce NF-KB binding to HIV-l long terminal linkage of Fe receptors; however, this is unlikely since soluble repeat, which provides a potent stimulus for HIV-1 replication. antibodies, which were ofthe same isotype, to other expressed CD44 is known to facilitate binding to hyaluronate, fibronectin, adhesion molecules did not induce enhanced replication of laminin, and collagen and has proposed functions in extracellu HIV-1. lar matrix binding, cell migration, adhesion, and lymphocyte The observation that engagement of CDlla, CDI8, CD44, homing [33]. CD58 is the cell surface molecule that is the CD45, and CD58 induced IL-1 and TNF secretion is in ligand for CD2 expressed on T cells. CD2-CD58 interaction agreement with previous studies [13, 15]. However, data for is important in generating synergistic activation signals for T CD11a and CDl8 are in contrast with a previous report sug cell activation on T cell receptor recognition ofspecific antigen gesting that engagement of these molecules neither induced and induces HIV-1 replication in infected T cells [28]. All TNF secretion nor enhanced HIV-1 replication [23]. However, three molecules are thought to be involved in monocyte-T the same group and others have demonstrated that blockade of cell aggregation [34], which promotes CD18/ICAM interaction CD 18 ligation using soluble antibody inhibits enhanced HIV [34-36]. Thus, their engagement may enhance HIV-l replica 1 replication in monocytic cells induced on adherence to endo tion via promotion of LFA-I/ICAM interaction. thelial cells [10, 11]. Differences in epitope specificity between T cell adhesion to macrophages is triggered by recognition of antibodies to CD 18 may account for such discrepancies. In specific antigen, inducing enhanced adhesion through multiple deed, we have also observed that the antibody used by Fan et ligand interactions. Such adhesion is likely to generate cross al. [10] (TSl/18) does not induce HIV-l replication in OMlO.l linking ofLFA-l, CD58, and possibly CD44 and CD45, all of cells (data not shown), suggesting that this antibody does not which have the potential to induce HIV -1 replication in infected mimic ligand binding, contrary to the observed effects ofMAbs monocytic cells, as we have shown here. These findings were used in this study (MHM23) and immobilized ICAM-I. obtained using a monocytic cell line and may not parallel the Engagement of CDlla, CDI8, CD44, CD45, or CD58 was effects of adhesion molecule ligation on HIV-1 replication in shown to enhance both HIV -1 replication and TNF secretion, primary macrophages or the in vivo significance ofsuch effects. and demonstration that the former was neutralized by MAb to However, data presented herein support previous observations TNF strongly suggests that TNF is crucial in the induction of that CD 18 ligation during cell-cell interaction with activated such enhanced HIV-1 replication. In addition to the observed T cells induces HIV -1 replication in infected peripheral blood secretion of soluble TNF, intracellular and membrane-bound monocytes from asymptomatic patients [4, 5]. TNF may have contributed to the enhancement ofHIV -1 repli Recent studies have suggested that HIV-1 production from cation. Induction of TNF secretion by adhesion molecule liga dendritic cells may also occur only on interaction with T cells tion on monocytic cells may contribute to enhanced plasma [37, 38], suggesting that adhesion molecule engagement may levels of TNF observed in vivo in HIV-1-infected patients. also have a role in regulation of HIV-1 production in such JID 1996; 174 (July) Monocyte Adhesion and HIV Replication 61 cells. Should such a mechanism occur in vivo, it would have 11. Gilles P, Lathey J, Spector S. Replication of macrophage-tropic and T-ce1l major implications for HIV -1 pathogenesis [39]. HIV-1 dissem tropic strains of human immunodeficiency virus type 1 is augmented by macrophage-endothelial cell contact. J Virol 1995;69:2133-9. ination to T cells as a direct result of cell-cell interaction and 12. Kasahara K, Strieter R, Chensue S, Standiford T, Kunkel S. Mononuclear signaling during antigen presentation would have the potential cell adherence induces neutrophil chemotactic factor/interleukin-8 gene to cause depletion ofCD4 T cells in an antigen-specific manner. expression. J Leukoc BioI 1991;50:287-95. Such a scheme of T cell depletion would model the reported 13. Sporn SA, Eierman OF, Johnson CE, et al. Monocyte adherence results sequential loss in antigen specificity seen in HIV -1-infected in selective induction of novel genes sharing homology with mediators of inflammation and tissue repair. J Immunol 1990; 144:4434-41. patients [40, 41]. 14. Couturier C, Haffner-Cavaillon N, Weiss L, Fischer E, Kazatchkine M. While further work is needed to determine the role of adhe Induction of cell interleukin I through stimulation of the adhesion sion molecules in regulation of HIV-1 replication in vivo, data promoting proteins LFA-1 (C011a/C018) and CR3 (CD11b/C018) of human monocytes. Eur J ImmunoI1990;20:999-1005. from this study suggest that agents that interfere with engage Downloaded from https://academic.oup.com/jid/article/174/1/54/959003 by guest on 29 September 2021 ment ofmonocytic adhesion molecules, principally LFA-l, and 15. Gruber M, Webb 0, Gerrard T. Stimulation of human monocytes via therapies directed to reduce or antagonize the action of TNF CD45, C044, and LF A-3 triggers macrophage colony-stimulating factor production. J Immunol 1992; 148:1113-8. a may reduce virus production in HIV-1-infected monocytic 16. Webb 0, Shimizu Y, Seventer G, Shaw S, Gerraerd T. LFA-3, C044, and cells. C045: physiologic triggers of human monocyte TNF and IL-l release. Science 1990;249:1295-7. 17. Butera S. Interrelationships between cytokines and the human immunode Acknowledgments ficiency virus. In: Griffin GE, ed. Cytokines in infection. London: Bai1 liere Tindall, 1994: 109 - 26. We thank Salvatore Butera and Thomas Folks (CDC, Atlanta) 18. Butera ST, Perez VL, Wu BY, Nabel GJ, Folks TM. Oscillation of the for the OMID.I cells; Nancy Hogg (Imperial Cancer Research human immunodeficiency virus surface receptor is regulated by the state Fund, London) for ICAM-I-Fc; Celltech (Berkshire, UK) and Cen of viral activation in a C04 cell model of chronic infection. J Virol tocor (Malvern, PA) for MAB against TNF-a; National Institute 1991; 65:4645-53. for Biological Standards and Control (Hertfordshire, UK) for anti 19. Cabanas C, Hogg N. Ligand intercellular adhesion molecule 1 has a neces bodies against IL-Ia, IL-I,B, and recombinant IL-IRA; and Gary sary role in activation ofintegrin lymphocyte function-associated mole cule 1. Proc Nat! Acad Sci USA 1993; 90:5838-42. Nabel, Salvatore Butera, Thomas Folks, and Nancy Hogg for help 20. Potts B. Mini reverse transcriptase (RT) assay. In: Aldovini A, Walker B, ful critical discussion and insight. eds. Techniques in HIV research. New York: Stockton Press, 1990: 103-6. 20a.Espevik T, Nissen-Meyer J. A highly sensitive cell line, WEK1 164 sub References clone 13, for measuring cytotoxic factor/tumour necrosis factor from human monocytes. J Immuno1 Methods 1986; 95:99-105. 1. Gendelman H, Orenstein J, Baca L, et al. The macrophage in the persis tence and pathogenesis of HIV infection. AIDS 1989;3:475-95. 21. Butera ST, Roberts BO, Leung K, Nabel GJ, Folks TM. Tumor necrosis 2. Embretson J, Zupancic M, Beneke J, et al. Analysis ofhuman immunode factor receptor expression and signal transduction in HIV -I-infected ficiency virus-infected tissues by amplification and in situ hybridization cells. AIDS 1993; 7:911-8. reveals latent and permissive infections at single cell resolution. Proc 22. Dustin ML. Two-way signalling through the LF A-I lymphocyte adhesion Nat! Acad Sci USA 1993; 90:357 -61. receptor. Bioessays 1990; 12:421-7. 3. Griffin G, Leung K, Folks T, Kunkel S, Nabel G. Activation of HIV 23. Fan ST, Mackman N, Cui MZ, Edgington TS. Integrin regulation of an gene expression during monocyte differentiation by induction of NF inflammatory effector gene. Direct induction of the tissue factor pro KB. Nature 1989; 339:70-3. moter by engagement of beta 1 or alpha 4 integrin chains. J Immunol 4. Schrier R, McCutchan J, Wiley C. Mechanisms of immune activation of 1995; 154:3266-74. human immunodeficiency virus in monocytes/macrophages. J Virol 24. de Feugerolles A, Springer T. Intracellular adhesion molecule 3, a third 1993;67:5713-20. adhesion counter receptor for lymphocyte function-associated molecule 5. Mikovits J, Lohrey N, Schulof R, Courtless J, Ruscetti F. Activation of 1 on resting lymphocytes. J Exp Med 1992; 175:185-90. infectious virus from latent human immunodeficiency virus infection of 25. Butcher E. Leukocyte-endothelial cell recognition: three (or more) steps monocytes in vivo. J Clin Invest 1992;90:1486-91. to specificity and diversity. Cell 1991; 67: 1033-6. 6. Embretson J, Zupancic M, Ribas J, et al. HIV infection is active and 26. Patarroyo M, Prieto J, Beatty PG, Clark EA, Gahmberg CG. Adhesion progressive in lymphoid tissue during the clinically latent stage of dis mediating molecules of human monocytes. Cell Immunol 1988; 113: ease. Nature 1993; 362:359-62. 278-89. 7. Lebargy F, Branellec A, Deforges L, Bignon J, Bemaudin JF. HIV-1 in 27. Palmer S, Hamblin AS. Increased CD II/CD18 expression on the periph human alveolar macrophages from infected patients is latent in vivo but eral blood leucocytes of patients with HIV disease: relationship to dis replicates after in vitro stimulation. Am J Respir Cell Mol Bioi 1994; ease severity. Clin Exp Immunol 1993;93:344-9. 10:72-8. 28. Diegel ML, Moran PA, Gilliland LK, et al. Regulation of HIV production 8. Shattock R, Friedland J, Griffin G. Release of human immunodeficiency by blood mononuclear cells from HIV-infected donors: II. HIV-1 pro virus by THP-1 cells and human macrophages is regulated by cellular duction depends on T cell-monocyte interaction. AIDS Res Hum Re adherence and state of activation. J Virol 1993;67:3569-75. troviruses 1993; 9:465-73. 9. Shattock R, Griffin G. Cellular adherence enhances HIV replication in 29. Makino M, Yoshimatsu K, Azuma M, et al. Rapid development of murine monocytic cells. Res Viro11994; 145:139-45. AIDS is dependent ofsignals provided by CD54 and CD 11a. J Immunol 10. Fan ST, Hsia K, Edgington T. Upregulation of human immunodeficiency 1995; 155:974-81. virus->l in chronically infected monocytic cell line by both contact with 30. King PO, Batchelor AH, Lawlor P, Katz DR. The role of CD44, CD45, endothelial cells and cytokines. Blood 1994; 84: 1567 - 72. C045RO, CD46, and C055 as potential anti-adhesion molecules in- 62 Shattock et a!. JID 1996; 174 (July)
volved in the binding of human tonsillar T cells to phorbol 12-myristate 36. Pardi R, Inveradi L, Bender J. Regulatory mechanisms in leukocyte adhe 13-acetate-differentiated U-937 cells. Eur J ImmunoI1990;20:363-8. sion: flexible receptors for sophisticated travelers. Immunol Today 1992; 31. Stamenkovic I, Sgroi D, Aruffo A, Sy M, Anderson T. The B lymphocyte 13:224-30. adhesion molecule CD22 interacts with leukocyte common antigen 37. Pope M, Betjes M, Romani N, et a!. Conjugates of dendritic cells and CD45RO on T cells and 0'2-6 sialyltransferase, CD75, on B cells. Cell memory T lymphocytes from skin facilitate productive infection with 1991; 66: 1133-44. HIV-1. Cell 1994; 78:389-98. 32. Walton K, Dixon J. Protein tyrosine phosphatases. Annu Rev Biochem 38. Cameron P, Lowe M, Crowe S, et a1. Susceptibility of dendritic cells to 1993; 62: 101-20. HIV-I infection in vitro. J Leukoc Bioi 1994;56:257-65. 33. Lesley J, Hyman R, Kincade P. CD44 and its interaction with extracellular 39. Pauza C. HIV persistence in monocytes leads to pathogenesis and AIDS. Cell Immunol 1988; 112:414-24. matrix. Adv Immunol 1993; 54:271-335. 40. Clerici M, Stocks N, Zajac R, et al. Detection of three distinct patterns of 34. Lorenz H, Harper T, Lagoo A, Baur A, Eger G, Kaldden J. CD45 mono T helper cell dysfunction in asymptomatic, human immunodeficiency clonal antibody-induced cell adhesion in peripheral blood mononuclear virus--seropositive patients. J Clin Invest 1989; 84: 1892-9.
cells via lymphocyte-associated antigen-I (LFA-I) and intracellular ad Downloaded from https://academic.oup.com/jid/article/174/1/54/959003 by guest on 29 September 2021 41. van Noesel C, Gruters R, Terpstra F, Schellekens P, van Lier R, Miedema hesion molecule I (ICAM-l). Cell Immunol 1993; 147:110~28. F. Functional and phenotypic evidence for a selective loss of memory 35. Mackay C, ImhofB. Cell adhesion in the immune system. Immunol Today T cells in asymptomatic human immunodeficiency virus-infected men. 1993; 14:99-102. J Clin Invest 1990; 86:293-9.