Normal human thymus samples were discarded tissue from children (age range, 2 days to 7 years, n = 15) undergoing corrective cardiovascular surgery (Royal Children's Hospital, Melbourne, Australia) and were obtained with informed consent and under institutional guidelines. All experiments, excluding the immunohistochemistry, were conducted with this tissue. Infected thymus tissue for immunohistochemistry was obtained from HIV-1BaL-infected SCID-hu-thy-liv mice transplanted with human fetal liver and thymic tissue  (kindly supplied by Ramesh Akkina, Colorado State University, Fort Collins, USA). Infection of the SCID-hu-thy-liv thymus tissue was quantified in tissue digests by real time PCR (iCycler; Biorad, Hercules, CA) using previously described methods to detect full length HIV-1 DNA with primers specific for LTR and Gag .
Connective tissue was dissected from the human thymus samples and the thymus tissue disrupted with a scalpel blade prior to incubation with collagenase (1 mg mL-1, type II; Worthington Biochemical Corporation, Lakewood, NJ) and DNase (0.02 mg mL-1, grade II bovine pancreatic DNaseI; Worthington, Lakewood, NJ) in RPMI-1640 media (Gilbco/Invitrogen, Grand Island, NY) supplemented with 2% heat inactivated cosmic calf serum (HyClone, Logan, UT). Incubation was continued for 30 min at 37°C with intermittent agitation followed by 5 min at room temperature with constant agitation. To disrupt T cell-DC complexes, 100 mM EDTA was added (10 mM final concentration) to the digest, and incubation with agitation was continued for 5 min. The suspension was then passed through a nylon mesh to remove any remaining aggregates and/or stromal material. The resulting single cell suspension was subjected to Nycodenz (Axis-shield, Dundee, Scotland) density gradient centrifugation as previously described , with the exception that cells were resuspended in Nycodenz at a density of 1.070 g/mL, rather than 1.068 g/mL, as we found that this gave a greater DC yield. A low-density fraction (LDF) containing DC and a high-density fraction (HDF) were recovered. Immature double-negative (CD3-CD4-CD8-), double-positive (CD3loCD4+CD8+) and mature single-positive (CD3+CD4+CD8- or CD3+CD4-CD8+) thymocytes were isolated from the HDF using the monoclonal antibodies (mAbs); anti-HLA-DR-allophycocyanin-cychrome-7 (APC-Cy-7), anti-CD3-phycoerythrin (PE; BD Biosciences, Bedford, MA) and FACSAria cell sorting (BD Biosciences, Bedford, MA).
Phenotypic analysis of thymic DC subsets
Phenotypic analysis was performed on the enriched thymic DC population recovered from the Nycodenz LDF. Cells were immunostained with labelled mouse mAbs and incubated for 25 min at 4°C. The mAbs included anti-CD11c-APC, anti-CD123-PE, anti-HLA-DR-PE/ Peridinin Chlorophyll Protein Complex (perCP)/ APC-Cy7, anti-CD14-PE, anti-CD3-fluorescein isothiocyanate (FITC), anti-CD4 perCP, anti-CCR5 FITC, anti-CXCR4 PE and APC (BD Biosciences, San Jose, CA), anti-CD1c-FITC (Biosource International, Camarillo, CA), anti-CD83 PE, anti-CD86 APC, anti-DC-SIGN PE, anti-DEC-205 perCP-Cy5, anti-MR APC (Biolegend, San Diego, CA), and anti-M-DC8 (kindly supplied by Knut Schakel; Institute of Immunology, Technical University of Dresden, Germany). Cells labelled with anti-M-DC8 were washed and incubated with goat anti-mouse IgM-biotin (Chemicon, Boronia, Australia) for 20 min at 4°C and finally washed and incubated with streptavidin-APC (Becton Dickinson, Franklin Lakes, NJ, USA) for 25 min at 4°C.
Blood and thymic DC purification
For thymic DC purification, LDF cells were immunodepleted by magnetic cell sorting (Miltenyi Biotec, Bergisch Gladbach, Germany) using a cocktail of mAbs; anti-CD3 (OKT3), anti-CD15 (WEMG.I), anti-glycophorin A (GlyA; 10FM.N) and anti-CD19 (FMC63; a kind gift from Heddy Zola, Flinders Medical Centre, Adelaide, Australia), and anti-mouse IgG-coated magnetic microbeads (Miltenyi Biotec). For blood DC purification, PBMC isolated over Ficoll Hypaque gradients (Pharmacia, Uppsala, Sweden) from fresh buffy coats (Australian Red Cross Blood Service, Melbourne, Australia) were immunodepleted using the mAbs; anti-CD3 (OKT3), anti-CD11b (OKM1), anti-CD19 (FMC63) and anti-GlyA (10FM.N). The DC enriched populations were immunostained with sheep anti-mouse-FITC (Chemicon, Boronia, Australia) to identify any remaining cocktail-positive cells. After blocking with 10% normal mouse serum (Sigma, St. Louis, MO), cells were incubated with the mAbs; anti-CD11c-APC, anti-CD123-PE and anti-HLA-DR-APC-Cy7 (BD Biosciences). Using a FACSAria (BD Biosciences) we were able to sort two DC subpopulations by gating on total HLA-DR+ cells, in order to exclude any contaminating basophils/mast cells/natural killer cells, and then either CD11c+ mDC or CD123+ pDC. The number of isolated DC did not correlate with the thymus donor age and on average the recovery was 3 × 105 pDCs and 2 × 105 mDCs per 109 total thymic cells. The purity of sorted cells was always greater than 98% upon reanalysis (Figure 1).
Preparation and characterisation of HIV-1 stocks
HIV-1 viruses were generated by transfection of 293T cells with either X4 or R5 viruses [pDRNL4-3-nef/EGFP or pDRNL(AD8)-nef/EGFP respectively] (kindly supplied by Damien Purcell, The University of Melbourne, Melbourne, Australia). Supernatants were centrifuged, filtered through 0.45 μm pore-size filters, concentrated by ultra-centrifugation over a 20% sucrose gradient and stored at -80°C. The 50% tissue culture infective doses of the virus stocks was evaluated by limiting dilution on PHA (10 μg/mL; Murex, Kent, UK) stimulated PBMCs.
Infection with and transfer of HIV-1
PHA-stimulated PBMC (aPBMC; positive control for productive infection), unstimulated PBMC, thymocytes and DC subpopulations were either mock infected with media alone or infected with viral supernatants at a multiplicity of infection of 0.1 at 37°C in RC-10 (RPMI-1640 supplemented with 10% (vol/vol) cosmic calf serum, 100 U/mL penicillin, 100 μg/mL streptomycin, 2.9 mg/mL L-glutamine (Gilbco/Invitrogen, Grand Island, NY)). Following 2 h of culture, the cells were washed thoroughly to remove unbound virus. Cells were cultured at 37°C in round-bottom 96-well microtitre plates at a concentration of 105 cells/100 μL RC-10. Thymocytes and PBMC were cultured with IL-2 (10U/mL; Roche Diagnostics, Indianapolis, IN), while DC were cultured with IL-3 (10 ng/mL; R&D Systems Inc, Minneapolis, MN) and GM-CSF (40 ng/mL; R&D Systems Inc), which have previously been reported to increase DC survival . In some experiments cells were treated with the nucleoside reverse transcriptase inhibitor azidothymidine (0.1 μM). In experiments designed to detect transfer of HIV-1, blood or thymic DC were pulsed with virus for 2 h as described above and following 24 hours of culture, unstimulated mock infected PBMC or CD3hi/CD3lo thymocytes were added to blood or thymic DC respectively. Cells were harvested 5 days post infection and the number of productively infected (EGFP+) cells detected using flow cytometry. To confirm transfer, in some experiments the pDC-CD3hi thymocyte co-cultures were additionally immunostained with anti-CD3-PE at day 5 post infection.
Sections (5 μm) of cryopreserved OCT-embedded thymus fragments were analysed by immunohistochemistry. All incubations were performed at room temperature in a humidified chamber. The sections were exposed to 0.3% hydrogen peroxide solution to neutralize endogenous peroxidases and then incubated with blocking buffer (10% normal goats' or fetal bovine serum) for 15 min followed with IgG1 or the mAbs; CD83, CD40 (diluted 1:200; AbD Serotec, Raleigh, NC), CD123 (diluted 1:30; BD Biosciences), HLA-DR (diluted 1:160), p24 (diluted 1:200) and CD68 (diluted 1:400; Dako, Glostrup, Denmark) for 1 h. After rinsing in PBS, the sections were exposed to biotinylated-mAb (Vectastain, Vector Laboratories Inc., Burlingame, CA) for 30 min, rinsed with PBS and then incubated with Steptavidin-HRP (Dako) for 30 min. Immunostaining was revealed using 3,3'-Diaminobenzidine substrate solution according to the manufacturer's guidelines (Dako). Sections were counterstained with haematoxylin and blued with Scott's tap water to enhance nuclear definition. Finally, sections were dehydrated through 4 changes of alcohol (70%, 95% and 2 × 100%), cleared in 3 changes of xylene and mounted with DePeX (Merck, Darmstadt, Germany).
Flow cytometry was performed using a FACSCalibur (Becton Dickinson) and results were analysed with Weasel software (Walter and Elisa Hall Institute, Melbourne, Australia).
Statistical analyses were performed with the Wilcoxon paired sign rank sum test or Mann Whitney test using GraphPad Prism (GraphPad software, La Jolla, CA). A p value of less than 0.05 was considered significant.