Brain frontal and occipital lobes, cerebellum, together with spleen and LN samples were obtained at autopsy and kept frozen at -80°C. Tissue samples were provided by the National NeuroAIDS Tissue Consortium (NNTC) and by the University of Edinburgh Brain Bank. HIV-1 envelopes described here were amplified from subjects 7766, 6568, CA110, 10017 and NA20; heterosexual patients with HIV associated dementia (HAD), HIV encephalitis (HIVE) or cognitive impairment all of whom progressed to AIDS and died. Table 1 lists the five patients included for this study. Samples from three subjects provided by the NNTC are from patients treated extensively by HAART (7766, 6568 and 10017), while the fourth (CA110) had no reported use. NA20 was described previously and was a subject from before the HAART era [12, 13]. We have increased the number of functional envelopes amplified from this subject.
Nucleic acid extraction and PCR amplification of env from single molecule templates
Total DNA was purified from tissues using the QIAamp DNA Mini kit (Qiagen), according to the manufacturer's instructions. DNA was eluted in nuclease-free water and frozen immediately at -20°C for storage until analysis.
Sequences covering rev and env genes were amplified by nested PCR from proviral and circular (episomal) forms of viral DNA present in brain and spleen or LN tissue. The presence of rev in cis upstream from envelope is important for efficient envelope expression and production of high titer env+ pseudovirions. PCRs were carried out using serially diluted tissue DNA samples so that envelopes subsequently cloned were derived from single genomes when 30% or less of PCR reactions were positive. PCRs used high fidelity DNA polymerases (Platinum®Taq DNA Polymerase; Invitrogen Inc., or Phusion™ DNA Polymerase; Finnzymes Inc.). PCRs were set up as described previously  using the following primers. For proviral amplification, outer primers were RevenvA (5'-TAGAGCCCTGGAAGCATCCAGGAAG-3') and EnvN (5'-CTGCCAATCAGGGAAGTAGCCTTGTGT-3'). Outer primers for episomal DNA amplification were RevenvA and LAI (5'-GCGCTTCAGCAAGCCGAGTCCT-3') . The inner primers were the same for both proviral and episomal rev-env amplification as follows; RevenvBTOPO (5'-CACCTAGGCATCTCCTATGGCAGGAAGAAG-3') and Env-lo (5'-GTTTCTTCCAGTCCCCCCTTTTCTTTTAAAAAG-3'; ). The PCR products in positive wells at endpoint dilutions were purified from a 0.8% crystal violet stained agarose gel using a QIAquick Gel Extraction Kit (Qiagen).
All envelopes from subjects CA110, 7766, 6568 and 10017 were amplified and cloned using limiting dilution protocols [2, 78]. NA20 envelopes, (including previously described B59, B76, B501, LN8, LN10, LN14 and LN16 [12, 13] and newly amplified envelopes, (23-14-2, 23-14-3, 23-14-4, 23-14-9, 23-15-18, 23-15-23, 23-15-28, 23-16-38, 23-17-52 and 23-17-54, 23-15-28) were cloned from proviral DNA not rigorously diluted to endpoint due to limited amounts of DNA.
Envelope cloning and sequencing
Purified PCR products were cloned into pcDNA™ 3.1D/V5-His-TOPO® (pcDNATM3.1 Directional TOPO® Expression Kit; Invitrogen Inc). Env+ plasmids were transformed into competent E. coli (TOP10; Invitrogen Inc.). Colonies were screened for correct rev-env insertions by PCR using a universal T7 Promoter primer (5'-TAATACGACTCACTATAGGG-3') and M5-R (5'-CCAGCTGGGGCACAATAATGTATGGGAATTGG-3' (a primer that hybridizes within our insert); ) using Go Taq® Green Master Mix (Promega Inc.). Plasmid DNA was purified using QIAprep Miniprep Kit (Qiagen) and sequenced by Genewiz Inc.
Up to 32 endpoint rev-env clones were sequenced for each sample and analyzed phylogenetically to establish the population diversity. Envelope sequences were analyzed for variation likely to impact on phenotypes e.g. N283, variable loop length, V1-V2, V3 and V1-V5 loop charge, PNGSs as well as for mutations likely to render envelopes non-functional e.g. deletions, premature stop codons, loss of a conserved cysteine involved in disulphide bonding etc. Non-functional envelopes with stop codons or deletions were not included in the analyses.
The nucleotide sequences of novel envelopes reported here have been assigned GenBank accession numbers JN786685-JN786871.
Complete gp160 env nucleotide sequences were assembled and aligned using Clustal ×  with manual adjustment. All positions with an alignment gap of one or more nucleotides were excluded.
Phylogenetic and molecular evolutionary analyses were conducted using MEGA version 5 . Maximum likelihood phylogenetic trees were generated using General Time Reversible Substitution Model using a discrete Gamma distribution with 5 rate categories and by assuming that a certain fraction of sites are evolutionary invariable for subjects CA110, 7766 and 6568 and without invariant sites for subjects NA20 and 10017. Bootstrap analyses on 1,000 replicates was used to assess the robustness of the tree. Significant (≥ 70%) bootstrap values were assigned to internal tree nodes. Reference sequences representing three HIV-1 group M subtype B http://www.hiv.lanl.gov/ envelopes (FR.83. HXBc2. K03455; TH.90.BK132.AY173951;US.98.1058 11.AY331295) were used as outgroups.
293T cells were used to prepare env+ pseudovirions by transfection. HeLa TZM-bl  were used to evaluate env+ pseudovirion infectivity titers and neutralization. Pseudovirion infectivity was also evaluated on CD4+ CXCR4+ CCR5- HeLa HIJ cells to monitor CXCR4-use. HeLa TZM-bl cells express high levels of CD4, CCR5 and CXCR4 and contain HIV-inducible β-galactosidase and luciferase reporter genes. 293T cells, TZM-bl cells, and HIJ cells , were maintained in Dulbecco's modified Eagle's medium (DMEM, Gibco-Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS) and were cultured as previously described [12, 66, 68].
Macrophage cultures were prepared from elutriated monocytes [12, 66, 68], which were provided by the University of Massachusetts Center for AIDS Research Elutriation Core. The elutriated monocytes were cultured for 5 days in DMEM medium containing 10% human plasma (HP) for differentiation before setting up for infection. Alternatively, macrophages were prepared from Ficoll-purified white blood cells from whole blood by as described previously . On the day prior to infection, the macrophages were washed and resuspended in DMEM medium containing 10% HP and cultured in 48-well tissue culture plates (1.25 × 105 cells/well/0.5 ml).
Production and infectivity assays of env
Env+ pseudovirions were prepared by cotransfection of env+ pTOPOenv vector with env- pNL4.3Δenv construct that carried a premature stop codon in env  into 293T cells using calcium phosphate. Cell free supernatants were harvested after 48 h culture and frozen at -152°C prior to experimental analysis.
Env+ pseudovirions were titrated on HeLa TZM-bl cells, HeLa HIJ cells and on macrophages. For HeLa TZM-bl and HeLa HIJ cells, 2 × 104 cells/0.5 ml were added to each well on a 48 well plate the day prior to virus titration. Virus titers were determined as described previously (Peters 2004, 2006). Briefly, 100 μl of serially diluted viral supernatants in DMEM media (10% FBS) were added to cells in duplicate and incubated for 3 hours. 0.4 ml of DMEM (10% FBS) was added to each well and cultures incubated for 48 h (Hela TZM-bl) or 72 h (HIJ). TZM-bl cells were then fixed in 0.5% gluteraldehyde in PBS and β-galactosidase X-gal substrate added. HeLa HIJ cells were fixed in cold methanol:acetone 1:1, washed and immunostained for p24 using monoclonal antibodies 38:96K and EF7 (UK Centre for AIDS Research), followed by an anti-mouse IgG-β-galactosidase conjugate and X-gal substrate (0.5 mg/ml X-gal, 3 mM potassium ferrcyanide, 3 mM potassium ferrocyanide, 1 mM magnesium chloride).
Macrophages were seeded in 48 well plates (1.25 × 105 cells/0.5 ml/well) the day prior to infection. Macrophages were pretreated with 0.1 ml DEAE dextran (10 μg/ml) in DMEM medium containing 10% HP for 30 min at 37°C before virus supernatants were added and spinoculating for 45 minutes in a benchtop centrifuge . Infected macrophages were incubated for a further 3 h at 37°C before the addition of 0.4 ml of DMEM (10% HP) and incubating at 37°C for seven days. Macrophages were then fixed and immunostained for p24 as described for HIJ cells. DEAE dextran and spinoculation enhance virus infectivity by approximately 20-fold by increasing attachment  and entry . Infection following this procedure does not bypass the requirement of CD4 and CCR5 for infection, which remains sensitive to entry inhibitors. Thus, macrophage infection conferred by envelopes described here was inhibited by maraviroc (not shown). Since env+ pseudovirions are capable of only a single round of replication, we were able to estimate the number of focus-forming units (FFU) by counting individual or small groups of infected blue-stained cells by light microscopy. Average numbers of FFUs/ml were then calculated. All values represent the averages of at least two independent experiments, each done in duplicate and using macrophages from different donors. Error bars in figures were calculated from replicate wells of both experiments.
Each set of macrophage, TZM-bl and HIJ infections included several control env+ pseudovirions including NL4.3 (X4), JR-CSF and NA420 LN40 (non-mac-tropic R5 envelopes), JR-FL and NA420 B33 (mac-tropic R5 envelopes).
Inhibition and neutralization assays
Inhibition and neutralization assays for soluble CD4 (sCD4), maraviroc and mab b12 were carried out in 96 well plates as described previously using HeLa TZM-bl cells as target cells  For maraviroc, cells were treated with 2-fold dilutions in 50 μl for 30 minutes before adding an equal volume containing 200 FFU of pseudovirions. For sCD4 and b12, 50 μl samples of serially diluted sCD4 were mixed with 50 μl env+ pseudovirions carrying 200 FFU at 37°C for 1 h and added to HeLa TZM-bl cells. To evaluate residual infectivity, medium was removed and 100 μl of medium without phenol red added. Cells were then fixed and solubilized by adding 100 μl of Beta-Glo (Promega Inc.). Luminescence was then read in a BioTek Clarity luminometer.
A robust semiparametric regression model  implemented by R package drc  was used to model dose response relationships between the macrophage infectivity (as a percentage of TZM-bl infectivity) and IC50 concentration of maraviroc, sCD4 and b12. The dosages that caused 50% inhibition (IC50) were estimated as well as their 95% confidence intervals (not shown). Two parameter log-logistic regressions were used for one envelope, FL11-1-249 for b12 inhibition due to over-fitting of semiparametric approach in the neighbor area of IC50. When there was no apparent inhibition or inhibition failed to reach 100%, the model-fitting algorithm either did not converge or reported an extrapolating estimate out of the experimental ranges with a very wide confidence interval. For these inhibitions, the IC50 estimates were winsorized by defining them manually from Excel plotted graphs. Two tailed, nonparametric Mann Whitney tests were used to evaluate whether there exists statistically significant differences between distributions of IC50s of maraviroc, sCD4 and b12 for envelopes from brain and from LN/spleen. Two tailed, non-parametric Spearman tests were used to evaluate whether there exists monotonic correlation between macrophage infectivity and IC50s for maraviroc, sCD4 and b12 (Figure 5). Two tailed, nonparametric Mann Whitney tests were also used to test for significant differences in V1-V5 positive charge, length and number of potential N-linked PNGSs sites. These tests were carried out using Graphpad Prism 5 for Mac OSX.