Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Pulmonary infections remain a frequent and serious complication of human immunodeficiency virus type 1 (HIV-1)- related disease (1, 2). Alveolar macrophages (AM) are the predominant resident immune cell in the alveolar air space, and are targets for HIV-1 infection (3). AM are infected by HIV-1 in vivo (3, 4), and direct HIV-1 infection may contribute to the abnormalities of monocyte-macrophage effector cell function described in the setting of HIV-1 infection (5). HIV-1 infection of AM is generally latent, although productive HIV-1 replication can occur in vitro following stimulation with lipopolysaccharide or Mycobacterium avium complex (10) or in vivo during active pulmonary infection (11) without exhibiting cytopathic effects. As tissue macrophages are increasingly recognized as reservoirs of virus that contribute to the pathogenesis of acquired immunodeficiency syndrome (12), dissecting determinants of alveolar macrophage susceptibility to HIV-1 is of critical value in defining the role of HIV in the lungs.

Both the HIV and the simian immunodeficiency virus (SIV) employ chemokine receptors to mediate CD4-dependent entry into target cells (13). Isolates of HIV-1 exhibiting different cellular tropisms use various subsets of these chemokine receptors. CD4-dependent entry of monocytotropic (M-tropic) isolates of HIV-1 is generally mediated by the CCR5 chemokine receptor (13), whereas T-lymphocytotropic (T-tropic) isolates of HIV-1 use the CXCR4 chemokine receptor for entry (16).

Recent investigations suggest that cells of different tissue origin may exhibit different patterns of HIV-1 CD4 receptor-dependent coreceptor utilization (20). AM express CD4 receptors (24) and can be infected in vitro with HIV-1 (25). Recently, both CCR5 and CXCR4 receptor expression on human peripheral blood monocyte-derived macrophages has been reported (20). However, limited information is available regarding the specific expression and utilization of coreceptors that mediate entry of HIV-1 into human AM.

Characterizing the mechanisms of HIV-1 entry into AM may increase our understanding of viral pathogenesis and, importantly, help to identify novel therapeutic targets. The purpose of this study was to investigate and further define the role of -chemokine (CCR5 and CCR3) and -chemokine (CXCR4) receptors as coreceptors in the CD4 receptor-mediated entry of HIV-1 into human AM. To this end we used HIV-1 and SIV env-pseudotypes that allow for viral entry, and also rapid readout of long terminal repeat-driven chloramphenicol acetyl transferase (CAT) activity as a tool to characterize viral entry.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Bronchoscopy

Fiberoptic bronchoscopy was performed on consenting healthy nonsmoking adults, following protocols approved by the Beth Israel Deaconess Medical Center Institutional Review Board, Boston, MA. All individuals were without evidence of active pulmonary disease, had normal spirometry by pulmonary function testing, were without known risk factors for HIV infection, and were confirmed to be HIV-seronegative by enzyme-linked immunosorbent assay (ELISA). (ELISA was performed according to the manufacturer's instructions; Abbott Diagnostics, N. Chicago, IL.) Pulmonary immune cells were obtained by bronchoalveolar lavage (BAL) using a standard technique (26). Briefly, after topical 2% lidocaine anesthesia to the oropharynx, a fiberoptic bronchoscope was passed into the airways and wedged in a segment of the right middle lobe. BAL was performed by instilling six 50-ml aliquots of warm nonbacteriostatic 0.9% saline, followed by gentle suction after each aliquot was infused. The cells were separated from the pooled BAL fluid (BALF) by centrifugation at 100 × g for 10 min at 4°C, washed in cold RPMI-1640 media supplemented with 100 U/ml penicillin and 100 mg/ml streptomycin (Sigma, St. Louis, MO), and counted on a hemacytometer. Slides for cell morphology and differential determination were prepared by cytocentrifugation (Shandon, Pittsburgh, PA), stained by the modified-Giemsa method (Diff-Quik; Sigma), and examined by light microscopy.

AM Isolation

AM were isolated by adherence to six-well plastic-bottom tissue culture plates (2 or 3 × 10⁶ cells/well), washed to remove nonadherent cells, and then maintained in endotoxin- free complete culture media (RPMI-1640 supplemented with 10% heat-inactivated fetal calf serum; JRH Biosciences, Lanexa, KS), 100 U/ml penicillin, 100 mg/ml streptomycin, amphotericin-B, and 2 mM L-glutamine (Sigma). Isolation of AM by adherence yielded cells that were  98% viable as determined by trypan blue dye exclusion, that demonstrated > 95% positive nonspecific esterase staining, that were  96% phagocytic of 1.1 µm nonopsonized latex beads, and of which  95% were able to rosette and phagocytose antibody-coated human erythrocytes.

Flow Cytometric Analysis of Chemokine Receptor Expression on AM

Analysis of chemokine receptor expression on macrophages was performed on specimens of BAL cell suspensions using an Epics Profile II Flow Cytometer (Coulter Electronics, Hialeah, FL). The instrument was calibrated daily with standardized fluorescent particles (Immunocheck; AMAC, Inc., Westbrook, ME). Using a laser power of 5.76 mW, two fluorescent signalsforward light scatter (for measuring cell size) and right-angle (side) scatter (for measuring cell granularity)were measured simultaneously in list-mode. Fluorescence was measured using the appropriate photomultiplier tubes and optical filters: a 525-nm bandpass filter (for fluorescein isothiocyanate [FITC] detection) with a 40-nm bandwidth, and a 575-nm bandpass filter (for phycoerythrin [PE] detection) with a 30-nm bandwidth. The fluorescence of the cells was expressed as the mean of the log fluorescence intensity of the cell population within the analyzed field, and the results were recorded as mean relative fluorescence units.

The AM were first identified on the basis of characteristic forward- and side (right angle)-scatter properties. The results of the histogram confirmed that the isolated cells were macrophages by positive staining for human leukocyte-associated antigen-DR (HLA-DR) using FITC- primary labeled murine antihuman HLA-DR antibody, and by negative staining for dual-labeled CD14/CD45 (AMAC, Inc.). This population was then examined for the binding of anti-CD4 receptor antibody, and the binding of anti-CCR5, -CCR3, and -CXCR4 receptor antibody. The murine antihuman CD4 receptor antibody was conjugated with PE (AMAC, Inc.). For the chemokine receptors, murine monoclonal antihuman CCR3 (7B11), CCR5 (2D7), and CXCR4 (12G5) were used as primary antibodies (LeukoSite, Inc., Cambridge, MA), and FITC-conjugated goat antimouse F(ab)₂ (Boehringer-Mannheim, Indianapolis, IN) was used as a secondary antibody. Briefly, 5 × 10⁵ BAL cells freshly isolated in Hanks' balanced salt solution containing calcium and magnesium were first incubated with the primary antibody (1:200 dilution) for 60 min, washed, and then incubated with the secondary antibody (1:500) for 30 min at 4°C. The cell pellet was resuspended in OptiLyse-C (AMAC, Inc.), a commercial fixative containing formalin, for 10 min and then analyzed. Each condition was prepared and analyzed in duplicate, and a minimum of 5,000 cells were counted for each sample. The proportion of cells staining positive and a mean log fluorescence value were determined for each population of cells.

Nonspecific binding of the FITC-conjugated secondary antibody was < 15%, and was subtracted from the mean fluorescence. For the purpose of reducing nonspecific and Fc receptor binding, all solutions contained 0.1% bovine serum albumin and nonimmune human serum, respectively.

Plasmids

The preparation of the CAT-expressing reporter proviral DNA (pHXB-CAT-env); T-tropic (HXB2), M-tropic (YU2), and dual-tropic (89.6) HIV-1 envelope expressors; and human CD4, CCR5, CCR3, and CXCR4 receptor expressors was performed as previously described (23, 27). The SIVpbj1.9 envelope expressor was constructed by polymerase chain reaction (PCR) amplification of the region spanning from nucleotides 6013 to 9072, using a pair of primers: 5'-CCC GGG GAT ATC TCT CGA GGT CGA CTG AAC CAT TTT GAT CCT CGC-3' and 5'-GCT CTA GAC TCG AGT ATT CAT AAA TTG ACC CTC AC-3'. After amplification at 94°C for 1 min, 52°C for 2 min, and 72°C for 3 min for 30 cycles, the amplified DNA was digested with EcoRV and XhoI and cloned into the same site of pCDNA3 (Invitrogen, Inc., Carlsbad, CA).

Reverse Transcriptase-PCR

Semiquantitative reverse transcriptase (RT)-PCR determination for CCR5, CCR3, and CXCR4 was performed on cytoplasmic RNA extracted from AM (28). RT-PCR reactions were performed in single tubes using rTtH polymerase according to the manufacturer's instructions (Perkin-Elmer, Inc., Branchburg, NJ). Briefly, serially diluted RNA for each chemokine receptor was reverse transcribed at 60°C for 1 h, followed by 30 cycles at 94°C for 1 min and 60°C for 1 min, and by final extension at 60°C for 7 min. Amplified DNA was analyzed on 2% agarose gel. The primers used were: for CCR5, 5'-CGG TCA CCT TTG GGG TGG TGA CAA GTG-3' and 5'-GTG CCT CTT CTT CTC ATT TCG ACA CCG-3'; for CCR3, 5'-CTG TCA CTT TTG GTG TCA TCA CCA GC-3' and 5'-TTG TAC TTT TTT TTA CTG GGG CAC CTC-3'; and for CXCR4, 5'-AGC TGT TGG CTG AAA AGC TGG TCT ATG-3' and 5'-GCG CTT CTG GTG GCC CTT GGA GTG TG-3'.

Preparation of Primate Immunodeficiency Virus env-Pseudotypes

CAT-expressing reporter viruses pseudotyped with different envelope proteins were prepared by cotransfecting 293T cells with 10 µg of pHXB-CAT-env together with 20 µg of vectors expressing T-tropic (HXB2), M-tropic (YU2), or dual-tropic (89.6) HIV-1, or SIVpbj1.9 envelope using a standard calcium phosphate (CaPO₄) precipitation method (23). As a control, CAT-expressing reporter viruses without an envelope protein were generated by cotransfecting 10 µg of pHXB-CAT-env and 20 µg of pCDNA3. Three days after transfection, supernatants of the transfected cultures were clarified by low-speed centrifugation at 1,500 rpm for 10 min, followed by filtration of the supernatants through a 0.2-µm Acrodisc 25 filter (Gelman Sciences, Ann Arbor, MI). Pseudotyped virions were quantitated on 1 ml of clarified supernatant by using an RT assay (27).

Infection of Cells with Viral env-Pseudotypes

For the viral pseudotype entry assay of AM, fresh cells were plated onto a six-well plate (2.5 × 10⁶ cells/well) and cultured overnight. Nonadherent cells were removed by washing twice with phosphate-buffered saline (PBS), and the remaining cells were maintained in complete culture media for 7 to 10 d, with a 3-d refeeding cycle. The cells were then washed three times with PBS and exposed to an equal amount of recombinant virus (corresponding to 25,000 cpm of RT activity) for 16 h at 37°C, washed, and then maintained in culture for 48 h at 37°C. Cell lysates were prepared as described previously and CAT enzymatic activity was determined for each sample as described (27).

As controls, coreceptor use for viral entry into transfected Cf2Th canine thymocytes was investigated. Briefly, 1 × 10⁶ Cf2Th cells grown in Dulbecco's modified Eagle's medium (DMEM) in 75-cm² culture flasks were transfected with 10 µg of human CD4 expression vector and 20 µg of human chemokine receptor DNA as described previously. After 8 h incubation, cells were washed three times with PBS and cultured overnight in 15 ml DMEM supplemented with 10% fetal bovine serum. Cells were then detached with 3 ml of 0.5 mM ethylenediamine tetraacetic acid in PBS, and 1 × 10⁵ Cf2Th cells/well were plated onto a six-well tissue culture plate. Individual wells were then exposed to viral env-pseudotypes corresponding to 25,000 cpm of RT activity. Three days after infection, cells were harvested and total cell lysate protein was quantitated with a Micro BCA kit (Pierce, Inc., Rockford, IL). Cell lysates equivalent to 20 µg protein from each sample were assayed for CAT activity (27) as described below.

Chemokine Inhibition Assays

Chemokine inhibition experiments were performed with AM in six-well plates (2.5 × 10⁶ cells/well). Cells were washed three times with PBS and preincubated in the presence and absence of regulated on activation, normal T cells expressed and secreted (RANTES), eotaxin (500 ng/ml), or SDF-1 (2.5 mg/ml) for 60 min at 37°C in 1 ml RPMI-1640 medium. The cells were then incubated with an equal volume of recombinant virus for 16 h at 37°C, washed, and maintained in culture for an additional 48 h at 37°C. The macrophages were then processed for CAT activity, as described below.

CAT Assay

The efficiency of virus entry was measured in the infected cells by CAT activity (27). Three-day infected cells were detached and lysed by three freeze-thaw cycles in 0.25 M Tris-HCl (pH 8.0). The cell extracts were then heated to 60°C for 10-min to inactivate the endogenous acetylase. After clarification by a 10-min centrifugation at 13,000 × g, the amount of protein in the supernatant was quantitated by the Micro BCA kit according to the manufacturer's instructions (Pierce, Inc.). Lysates containing the equivalent of 25 µg of protein were used for the CAT enzymatic assay by incubating the samples for 3 h at 37°C in the presence of [¹⁴C]chloramphenicol and butyryl CoA (Promega, Inc., Madison, WI).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Participant Characteristics and Demographics

Bronchoscopies were performed on a total of eight healthy individuals. These included five males and three females with a mean age of 31.8 ± 8.8 yr (± SD). None had HIV-related risk factors, and all tested seronegative for antibodies to HIV by ELISA (data not shown). BALF return, cell yield, and characteristics for the healthy individuals were consistent with previous reports (data not shown) (29).

Flow Cytometric Analysis of Chemokine Receptor Expression on AM

Determination of chemokine receptor expression on AM, derived from a total of eight healthy individuals, was performed by flow cytometry analysis. A single population of cells was identified as macrophages by characteristic forward- and side-scatter properties, and 98% exhibited positive staining for HLA-DR. Representative flow cytometry histograms are presented in Figure 1. AM from all individuals demonstrated positive CD4-receptor staining. AM from all eight healthy individuals exhibited fluorescence surface intensity for CCR3, CCR5, and CXCR4. The levels for each of the chemokine receptors were low, as determined by the mean relative fluorescence values (data not shown).

View larger version (32K): [in this window] [in a new window]   Figure 1.   Flow cytometric histograms of AM from a healthy individual. Samples were prepared as detailed in MATERIALS AND METHODS. For each panel, the relative log fluorescence is plotted on the ordinate (FL1 = FITC and FL2 = PE), and the cell count is plotted on the abscissa. The panels demonstrate surface labeling for (A) FITC-conjugated isotype immunoglobulin G (IgG) control; (B) FITC-conjugated HLA-DR; (C) PE-conjugated isotype IgG control; (D) PE-conjugated CD4 receptor; (E) FITC-conjugated F(ab)2 secondary antibody control; (F) FITC-conjugated anti-CCR3 receptor; (G) FITC-conjugated anti-CCR5 receptor; and (H) FITC-conjugated anti-CXCR4 receptor. All samples were performed in duplicate. These histograms are representative of AM data from all eight individuals investigated, and were reproducible.

Detection of Chemokine Receptor Messenger RNA in AM

Consistent with the flow cytometry data, messenger RNA (mRNA) specific for CXCR4, CCR5, and CCR3 was isolated from the AM from all individuals assayed, as determined by RT-PCR (Figure 2). Of additional note, RT- PCR data demonstrated specific mRNA for the chemokine receptors Bonzo (30, 31) and BOB (32) in cultured AM (data not shown).

View larger version (24K): [in this window] [in a new window]   Figure 2.   Detection of chemokine receptor-specific mRNA in human AM. Quantitative RT-PCR was performed for CXCR4-, CCR5-, and CCR3-specific mRNA in AM using serial 10-fold dilutions. All total RNA samples were treated with ribonuclease-free deoxyribonuclease before complementary DNA synthesis amplification by RT-PCR, as described in MATERIALS AND METHODS. Amplified DNA was analyzed on 2% agarose gel.

HIV-1 Entry into AM

The observation that AM expressed, in addition to the CD4 receptor, the coreceptors CXCR4, CCR5, and CCR3 suggested that both M-tropic and T-tropic viruses could potentially enter these cells. To examine this possibility, 2.5 × 10⁶ AM were exposed to HIV-1 env-pseudotypes, and CAT enzymatic activity was measured. Incubation of AM with either dual-tropic (89.6) HIV-1 or M-tropic (YU2) HIV-1 env-pseudotypes produced high CAT enzymatic activity (Figure 3), suggesting efficient viral entry. However, CAT activity following incubation with the T-tropic (HXB2) HIV-1 env-pseudotype was minimal, indicating limited viral entry into macrophages despite robust CXCR4 expression.

View larger version (95K): [in this window] [in a new window]   Figure 3.   Entry assay of HIV-1 env-pseudotypes into AM. AM were incubated with the indicated HIV-1 env-pseudotypes for 3 d, cell lysates were prepared, and CAT enzymatic activity in the lysates was determined by thin layer chromatography; mock indicates CAT enzymatic activity after infection of CAT-expressing reporter viruses without envelope protein.

SIV Entry into AM

Incubation of AM with the SIVpbj1.9 env-pseudotype demonstrated elevated CAT enzymatic activity similar to that observed with the M-tropic or dual-tropic HIV-1 env-pseudotypes (Figure 3).

Chemokine Inhibition of HIV-1 Pseudotype Entry into AM

To further elucidate the use of chemokine receptors for viral entry in AM, we performed a series of blocking experiments using cognate chemokine ligands. Preincubation of AM with RANTES, a ligand for both CCR5 and CCR3, blocked entry of the M-tropic (YU2) HIV-1 env-pseudotype (Figure 4A), as evidenced by minimal CAT activity. As expected, preincubation of AM with SDF-1 (a ligand for CXCR4) or with eotaxin (a ligand specific for CCR3) did not inhibit CAT activity following incubation with the M-tropic (YU2) HIV-1 pseudotype. The observed differences upon incubation with RANTES versus eotaxin indicated that entry of the M-tropic (YU2) viral pseudotype into AM was mediated by CCR5 rather than CCR3.

View larger version (33K): [in this window] [in a new window]   Figure 4.   Effect of chemokines on HIV-1 env-pseudotype entry into AM. AM were preincubated with the indicated chemokines before challenge with YU2 (upper panel) or SIVpbj1.9 (lower panel) env-pseudotypes (A and B, respectively). Cell lysates were prepared from the challenged cells, and CAT enzymatic activity in the lysates was analyzed by thin layer chromatography.

Chemokine Inhibition of SIV Pseudotype Entry into AM

As noted in Figure 3, SIVpbj1.9 env-pseudotype entered AM efficiently. To further elucidate the specific coreceptors mediating SIVpbj env-pseudotype entry into AM, we measured the entry-dependent CAT activity of this env-pseudotype in the presence and absence of ligands for the chemokine receptors. In the absence of chemokines, the SIVpbj env-pseudotype entered AM (Figure 4B). Preincubation of the AM with RANTES (a specific ligand for CCR5 and CCR3), SDF-1 (specific for CXCR4), or eotaxin (specific for CCR3) individually did not affect CAT activity after incubation with the SIVpbj1.9 env-pseudotype. These observations suggest that virus entry was mediated by each of the tested chemokine receptors (CCR5, CCR3, CXCR4), or possibly by an alternative receptor. This observation was similar to the effect observed with the dual-tropic (89.6) HIV-1 env-pseudotype.

Entry of HIV-1 Pseudotypes into Cf2Th Canine Thymocytes

As controls to verify the coreceptor use of each HIV-1 env-pseudotype, different combinations of the CD4 receptor and chemokine receptor molecules were transiently expressed in Cf2Th canine cells by plasmid DNA transfection. Native Cf2Th cells do not express detectable amounts of endogenous CD4 or chemokine receptor molecules (23). These receptor-expressing Cf2Th cells were then challenged with the HIV-1 or SIV-1 pseudotypes.

As expected, expression of the CD4 receptor alone did not produce detectable CAT activity after incubation of the Cf2Th cells with any of the HIV-1 env-pseudotypes, indicating that a coreceptor or alternative receptor is required (Figure 5). Consistent with previous reports (16), the T-tropic HIV-1 (HXB2) env-pseudotype exhibited CAT activity only in hCD4/hCXCR4-coexpressing Cf2Th cells, whereas the M-tropic HIV-1 (YU2) env-pseudotype exhibited strong CAT activity in hCD4/hCCR5- and hCD4/hCCR3-coexpressing Cf2Th cells (Figure 5). Entry of the dual-tropic HIV-1 (89.6) env-pseudotype into the cells was efficiently mediated by CCR5, CXCR4, and CCR3. Interestingly, the SIVpbj1.9 env-pseudotype entered the cells that were expressing not only CCR5 or CCR3 but also CXCR4, a pattern of viral entry similar to that of the dual-tropic HIV-1 pseudotype (data not shown). These studies confirmed the integrity and specificity of the reagents used to characterize HIV and SIV entry into AM, as described previously.

View larger version (51K): [in this window] [in a new window]   Figure 5.   Effect of chemokine coreceptor expression on the entry of HIV-1 env-pseudotypes into transfected Cf2Th canine thymocytes. Cf2Th canine thymocytes were transfected with plasmids expressing the CD4 receptor and specific chemokine receptors, and then incubated with HIV-1 env-pseudotypes as indicated. CAT enzymatic activity in the lysates was analyzed by thin layer chromatography.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Macrophages are increasingly recognized as important cells in the pathogenesis of HIV-1 disease. Macrophage-tropic HIV-1 isolates are primarily responsible for the initiation of HIV-1 infection in vivo (33), and macrophages are an important source of productive HIV-1 infection in the setting of active Mycobacterium tuberculosis and Pneumocystis carinii pneumonia (11, 12) or during the HIV-1 viremia that characterizes the late clinical stages of HIV-1-related disease (12). In addition, elevated levels of HIV-1 are detected in the lungs of persons coinfected with M. tuberculosis or P. carinii (11, 12), although the source of the virus has not been established. Elucidating the pathogenesis of HIV-1 infection of macrophages may lead to novel therapeutic treatment strategies targeted at reducing, in macrophages, the HIV-1 replication and elevated HIV-1 viremia that characterize advanced HIV-related disease. Further, a better understanding of HIV-1 pathogenesis in the lungs may improve therapeutic options for reducing the pulmonary complications of HIV infection.

The importance of specifically investigating AM is supported by the observed differences in the behavior of tissue macrophages compared with peripheral blood monocytes or monocyte-derived macrophages with regard to HIV-1 disease. AM are infected with HIV-1 in vivo (3), but whether these cells originate from the migration of HIV-infected monocytes from the peripheral blood, or whether AM become infected in situ remains controversial. Peripheral blood monocyte-derived macrophages and AM exhibit differential susceptibility to in vitro infection by monocytotropic HIV-1 isolates (36). Spontaneous release of HIV p24 antigen or RT activity differs when comparing monocytes with AM (37). Anticryptococcal activity of HIV-1-infected AM differs from that of HIV-1-infected peripheral blood monocytes or peritoneal macrophages (38). Further, the virus in HIV-1-infected monocytes may evolve independently from the virus in HIV-1-infected AM (11). These observations suggest that different cell or host factors contribute to the pathogenesis of HIV-1 infection of macrophages, and provide the rationale for investigating tissue-differentiated AM as they relate to HIV-1 infection.

The mechanism of HIV-1 infection of AM, however, is incompletely understood. The CD4 glycoprotein is a major cellular surface receptor for HIV-1, and is expressed at low levels on normal human AM (24), as was confirmed in the current study. Recombinant soluble CD4 or anti-CD4 antibody completely inhibits laboratory-passaged T-tropic HIV-1 infection in vitro (39), but this does not appear to occur in vivo, on the basis of clinical trials. The observation that RANTES, MIP-1, and MIP-1 inhibited productive HIV-1 infection of cells indicates that these chemokine receptors function as coreceptors for HIV-1 entry (40). However, the specific role of these receptors in the mechanisms of HIV-1 infection of AM has not been completely defined.

In the current study, in addition to surface expression of the CD4 receptor, AM from healthy nonsmoking individuals expressed specific mRNA for all three chemokine receptors as determined by RT-PCR, as well as surface CCR5, CCR3, and CXCR4 protein, as determined by flow cytometry using specific antibodies. However, whereas M-tropic (YU2) HIV-1 and dual-tropic (89.6) HIV-1 env-pseudotypes entered cells efficiently, the T-tropic (HXB2) HIV-1 pseudotype did not enter AM, as determined by CAT reporter gene enzymatic activity. Coincubation of AM with RANTES, which is specific for CCR5 and CCR3, significantly blocked entry of the M-tropic (YU2) pseudotype into AM. Coincubation with the CXCR4-specific ligand SDF-1, or the CCR3-specific ligand eotaxin, did not block the entry of the M-tropic (YU2) env-pseudotype. Thus, AM from healthy nonsmoking adults express CCR5, CCR3, and CXCR4 on the cell surface; but only M-tropic and dual-tropic, not T-tropic, HIV-1 env-pseudotypes were able to enter the cells efficiently.

These studies demonstrate that AM express both -chemokine receptors as well as -chemokine receptors on the surface. The possibility that other CXCR4-expressing cells provided contaminating signals was largely excluded by employing two independent methods of detection of these receptors. The flow cytometry excluded cells such as CD4+ T or B lymphocytes on the basis of size, granularity, and absence of cell surface expression of HLA-DR. The RT-PCR used adherent AM cultured in the absence of interleukin-2 to assay for chemokine-specific mRNA, thus eliminating the possibility of contaminating nonadherent T or B lymphocytes.

Although human AM expressed CCR3 and CXCR4 in addition to CCR5, the CCR5 chemokine receptor was preferentially used as a coreceptor for CD4 receptor- dependent HIV-1 viral entry. This conclusion is supported by the observation that entry of the M-tropic (YU2) HIV-1 pseudotype into AM was significantly impaired in the presence of RANTES (which interacts with both CCR5 and CCR3), but was not impeded by either eotaxin (a chemokine that interacts with the CCR3 receptor) or SDF-1 (a specific ligand for CXCR4). In comparison, the entry of the SIVpbj1.9 pseudotype into AM was not blocked by RANTES, eotaxin, or SDF-1. Together, these data indicate that the entry of primate virus into host cells may depend on the type of host cell. Whether this cell-specificity reflects the absence of essential associated proteins, limitations in coreceptor conformational changes, or the use of alternative receptors remains to be determined.

In the current study, the mode of coreceptor utilization for the entry of these env-pseudotype viruses was confirmed using receptor-transfected Cf2Th canine thymocytes, and the findings were consistent with previous reports (17). In contrast to AM, entry of the M-tropic (YU2) HIV-1 pseudotype into these cells employed either CCR3 or CCR5, whereas entry of the T-tropic (HXB2) HIV-1 pseudotype used CXCR4, and the dual-tropic (89.6) HIV-1 pseudotype was mediated by CCR5, CCR3, or CXCR4. Previous reports described the entry of most SIV isolates into target cells as using CCR5 or BOB exclusively as coreceptors (17). In the current study, we observed that entry of the SIVpbj1.9 pseudotype was mediated by CCR5, CCR3, or CXCR4, similar to that seen with the dual-tropic (89.6) HIV-1 pseudotype. A possible explanation for the observed differences between SIVpbj1.9 and other SIV isolates may relate to amino acid sequence differences in the gp120 V3 region, an important genetic element responsible for the interaction with coreceptors (41). Further, our data suggest that primate viral entry into target cells is both receptor- and cell-specific.

Certain limitations need to be acknowledged in our study. Because the provirus of each reporter virus is env-defective and is competent only for a single cycle of virus replication, the intracellular level of CAT activity is a direct reflection of virus entry only. Thus, the behavior of the env-pseudotype may not represent the behavior of the intact virus. However, a recent report that RANTES, but not MIP-1 or MIP-1, suppressed in vitro HIV infection of normal human AM (42) supports the findings in our study, and suggests that the env-defective virus is a useful model. In addition, our observation that the T-tropic (HXB2) HIV-1 env-pseudotype did not efficiently enter AM is similar to that reported for monocytes (43), and supports the concept of HIV-1 macrophage tropism. Finally, due to limited access to specific reagents, the role and potential contribution of other receptors such as CCR2b, Bonzo, and BOB were not specifically investigated.

In summary, our data demonstrate that AM from healthy adults express the CD4 receptor and low levels of the CCR3, CCR5, and CXCR4 surface receptors. However, despite the presence of the CXCR4 receptor, only M-tropic or dual-tropic HIV-1 pseudotypes entered these cells efficiently. Furthermore, viral entry into AM appeared to employ the CCR5 chemokine receptor preferentially. Thus, the mechanisms for HIV entry appear more complex than the simple coexpression of CD4 receptors together with CCR3, CCR5, and CXCR4 chemokine receptors. These data suggest that CD4 receptor-dependent HIV-1 infection of AM likely involves additional factors or cofactors that mediate or determine viral entry, an issue critical to the design of therapeutics that would protect these host-defense cells from infection and thereby limit a pulmonary viral reservoir in the host.