Cytodyn Incorporated (Santa Fe, New Mexico) provided the anti-LFA-1 monoclonal antibody (Cytolin®). HIV-SF162 and HIV-IIIB were obtained from the NIH AIDS Reference and Reagent Program. These two viruses were chosen because they utilize either CCR5 or CXCR4 as a co-receptor for viral entry (SF162 and IIIB respectively). HIV-AC225 is a primary isolate from a recently infected individual. It is predicted to be a CCR5 utilizing isolate based on the sequence of the V3 region of its envelope. All three viruses were propagated on CD8-depleted PBMC from the same donor. Culture media was prepared using RPMI supplemented with Hepes, Penicillin/Streptomycin, L-glutamine, and 10% fetal calf serum.
Twenty-five subjects (13 HIV-1 negative and 12 HIV-1 positive) were enrolled in a blood draw only study. To be included in the study, HIV-1 infected subjects had to be in the asymptomatic stage of infection with a CD4 T cell count greater than 350 cells/mm3, an HIV-1 plasma RNA viral load less than 100,000 copies/ml, and not be on antiretroviral therapy. All subjects signed informed consent as approved by the Massachusetts General Hospital (MGH) Human Subjects Committee.
Peripheral blood samples were obtained by venipuncture into tubes containing Acid Citrate Dextrose (ACD). Plasma was separated from whole blood using centrifugation. Peripheral blood mononuclear cells (PBMC) were obtained by density gradient centrifugation (FICOLL, Sigma). HIV-1 serostatus was confirmed at the initial visit in all subjects. T cell subset analysis was performed at every visit for all subjects. HIV-1 RNA testing was performed at the initial visit for all subjects, and then only for HIV-1 infected subjects at subsequent visits. Assays were performed in the Clinical Laboratory at MGH.
To determine whether LFA-1 MAb could bind to HIV-1, a virus capture assay was performed similar to what has previously been described [21–23]. Briefly, ninety-six well plates (Nunc) were coated with 5ug/ml of the anti-LFA-1 antibody, PBS, or 1 ug/ml mixture of monoclonal antibodies specific to HIV-1 gp120, for 2 hours at room temperature. The gp120-specific antibody mixture contained three human monoclonal antibodies, 17b, A32, and EH21. These antibodies bind to discontinuous epitopes, are known to cross-react with envelope glycoproteins from multiple Clade B isolates, and likely bind to monomeric gp120 [24–27]. After blocking with PBS containing 4% whey, undiluted virus stocks were plated in triplicate, and incubated at room temperature for 2 hours. Wells were washed with PBS containing 0.05% Tween20, and harvested with 100ul PBS containing 0.5% TritonX 100. This supernatant was kept at 4°C overnight then diluted to 1:500 in PBS containing lysing solution. The presence of virus in these supernatants was determined using a p24 ELISA, as per the manufacturers instructions (HIV-1 p24 ELISA, Zeptometrix).
Isolation of cell subsets
PBMC were depleted of CD8+ cells by magnetic bead based separation using Dynabead CD8 (Invitrogen) as per the manufacturers instructions. This resulted in >95% depletion of CD8+ cells. CD4+ cells were isolated from PBMC by positive selection using paramagnetic beads coupled to anti-CD4 monoclonal antibodies (Dynabead CD4, Invitrogen). This isolation was performed according to the manufacturers instructions and resulted in a population of cells containing >95% CD4+ cells.
CTL lysis of activated CD4+ T cells
To determine whether LFA-1 MAb abrogated CTL lysis of activated uninfected CD4+ cells, we first obtained CD4+ target cells using magnetic positive enrichment, as described above. These cells were activated for three days at 37°C with 1.25 ug/ml PHA (phytohemagglutinin) in media containing 50U/ml IL-2. Autologous CD4+ cell depleted PBMC were used as effector cells. These cells were incubated with 10 ug/ml of LFA-1 MAb or PBS, as a control, on ice for 30 minutes then added without washing to the target cells. The activated CD4+ target cells were differentiated from effector cells via staining with 1uM CFSE. Effector and target cells were combined at a ratio of 25:1. After a 4 hour co-culture at 37°C, all cells were stained with 1 ug/ml 7AAD for 20 minutes at room temperature. The cells were then washed in PBS containing 2 ug/ml Actinomycin D and 1% fetal calf serum. Prior to acquisition, the cells were fixed in Actinomycin D containing buffer with 1% formaldehyde. The frequency of CFSE + 7AAD + cells (dead targets) was determined by flow cytometry on a BD LSRII. Single stained controls were used to set the gates. Percent cytotoxicity was calculated using the following formula: 100*(%sample lysis-%target cell only lysis)/(100-% target cell only lysis).
Identifying cells that bind LFA-1 MAb
To determine the cell subsets that bind to this anti-LFA-1 antibody, flow cytometry was performed. PBMC were either left inactivated or were activated for 1 hour at 37°C with 10 ng/ml of PMA (phorbol 12-myristate 13-acetate) and 5 ug/ml Ionomycin. The cells were then incubated with 24 ug/ml of LFA-1 MAb for 20 minutes at 4°C. This concentration of anti-LFA-1 was determined by titration on PBMC using flow cytometry. After washing in PBS containing 1% fetal calf serum, cells were stained with a FITC-labeled anti-mouse IgG for 20 minutes at 4°C. The various cell subsets were identified using fluorochrome-labeled antibodies to discriminate T cells, B cells, monocytes, and dendritic cells (CD3 PE, CD4 Q605, CD8 APC-Cy7, CD14 Pacific Blue, CD19 PE-Cy5, CD11c APC). The cells were washed and fixed in 1% formaldehyde prior to acquisition on a BD LSR-II. Gating was performed manually to identify CD3+ CD19- T cells, CD19 + CD3- B cells, and CD19-CD3- monocytes and dendritic cells. The T cell subset was further differentiated into CD4+ and CD8+ cells. Dendritic cells were defined as CD14- CD11c+, while monocytes were defined as CD14+ CD11c+. The percent of LFA-1 MAb + cells within each cell subset is reported.
Induction of soluble antiviral factors
To induce the production of an antiviral soluble factor, ninety-six well round bottom plates (BD Falcon) were coated with 1 ug/ml of LFA-1 MAb or PBS as a control and stored up to 1 week at 4°C. On the day of the assay, plates were warmed to room temperature and 2 × 105 PBMC were added to each well. After incubation at 37°C for 24 hours, the plates were centrifuged at 1700 rpm for 7 minutes to pellet the cells and the supernatant was removed and stored at −20°C prior to use in subsequent assays.
To test the direct effect of LFA-1 MAb on HIV-1 replication, the antibody was incubated with virus then a viral replication assay was performed. One hundred TCID50 of each virus stock (HIV-SF162, HIV-IIB, HIV-AC225) was incubated for 4 hours at 37°C with serial 10-fold dilutions of LFA-1 MAb. This mixture was then used to infect CD8-depleted PBMC from a single HIV-1 negative donor that had been activated for 3 days with 1.25 ug/ml of PHA in RPMI supplemented with 50U IL-2. Infected cells were maintained at 37°C for 7 days. The degree of viral replication was determined by measuring the concentration of p24 in the supernatant using an ELISA, as per the manufacturers instructions (HIV-1 p24 ELISA, Zeptometrix).
To test the effect of induced soluble factors on HIV-1 replication, a viral replication assay was performed in the presence of supernatant from PBMC treated with LFA-1 MAb or PBS, as described above. PHA activated CD8-depleted PBMC from a single HIV-1 negative donor were resuspended in supernatants diluted 1:2 with culture media, then infected with 10 TCID50 of HIV-SF162 and incubated at 37°C for 7 days. The concentration of p24 was then measured using an ELISA as per the manufacturers instructions (HIV-1 p24 ELISA, Zeptometrix).
Several assays were used in an attempt to determine the identity of the soluble factor. We started by testing for the presence of cytokines and chemokines that are known to have anti-HIV activity. The concentrations of MIP1-alpha, MIP1-beta, and RANTES were determined using a multiplex bead based ELISA system (Flowcytomix, Bender). Twenty-five microliters of supernatant from PBMC cultures incubated with either LFA-1 MAb or PBS (see above) were mixed with antibody-labeled beads, and biotin-conjugated secondary antibodies. After 2-hours of incubation, the beads were pelleted by centrifugation and incubated with PE-Streptavidin for 1 hour. The beads were then washed and resuspended in assay buffer containing 1% formaldehyde to ensure the samples were not infectious. Preliminary experiments verified that the use of formaldehyde did not interfere with the measurement of the analytes. The fluorescent signal from the beads was detected using a BD LSR-II flow cytometer. The resulting data was analyzed using the Flowcytomix software provided by the manufacturer. The concentration of each chemokine was determined by comparison to a standard curve generated from recombinant chemokines provided by the manufacturer.
An ELISA was performed to determine the concentration of alpha-defensins 1, 2 and 3 (Human HNP 1–3, Hycult Biotech) and Interferon-alpha (Human IFN-alpha pan ELISA kit, MAbtech). One hundred microliters of supernatant from PBMC cultures incubated with LFA-1 MAb or PBS were diluted 1:2 in dilution buffer and the assay was performed in duplicate according to each manufacturer’s instructions. The captured analytes were detected using a biotinylated secondary detection antibody followed by Streptavidin-peroxidase, and developed using a TMB substrate. Serial dilutions of the standards provided by each manufacturer were run in duplicate to create a standard curve from which the concentration of alpha defensins or IFN-alpha was calculated.
To broaden our search for the identity of the soluble factor, we utilized an antibody array (G2000 Human Cytokine Array, RayBioTech). This assay is similar to a standard sandwich ELISA, but it is performed on a glass slide and utilizes a fluorescent readout. This array can detect 174 different analytes including common cytokines, chemokines, and growth factors. A full list of the analytes can be found in Additional file 1. Supernatant from PBMC cultures incubated with either LFA-1 MAb or PBS (see above) was sent to RayBiotech for testing. Since the cell culture media we used in this assay contained serum, which includes many of the analytes being measured, a media only control was also tested. The mean signal intensity of LFA-1 MAb treated supernatants was compared to the mean signal intensity of PBS treated supernatants after subtracting out the signal intensity of the media control. A Wilcoxon sign rank test was used to determine whether the mean signal intensity was higher in LFA-1 MAb treated supernatants.
A follow-up experiment was performed to determine the concentration of Epidermal Growth Factor (EGF) in the LFA-1 MAb and PBS treated supernatants. In this quantitative ELISA, 100ul of supernatant was diluted 1:2 in dilution buffer and the assay was performed in duplicate as indicated by the manufacturer (EGF Human ELISA kit, MAbtech). Duplicate wells of serial dilutions the EGF standard were run in parallel with the samples to generate a standard curve from which the concentration of EGF was calculated.