Emergence of CD4 independence envelopes and astrocyte infection in R5 simian-human immunodeficiency virus model of encephalitis

Abstract

Human immunodeficiency virus type 1 (HIV-1) infection in the central nervous system (CNS) is characterized by replication in macrophages or brain microglia that express low levels of the CD4 receptor and is the cause of HIV-associated dementia and related cognitive and motor disorders that affect 20 to 30% of treatment-naive patients with AIDS. Independent viral envelope evolution in the brain has been reported, with the need for robust replication in resident CD4(low) cells, as well as CD4-negative cells, such as astrocytes, proposed as a major selective pressure. We previously reported giant-cell encephalitis in subtype B and C R5 simian-human immunodeficiency virus (SHIV)-infected macaques (SHIV-induced encephalitis [SHIVE]) that experienced very high chronic viral loads and progressed rapidly to AIDS, with varying degrees of macrophage or microglia infection and activation of these immune cells, as well as astrocytes, in the CNS. In this study, we characterized envelopes (Env) amplified from the brains of subtype B and C R5 SHIVE macaques. We obtained data in support of an association between severe neuropathological changes, robust macrophage and microglia infection, and evolution to CD4 independence. Moreover, the degree of Env CD4 independence appeared to correlate with the extent of astrocyte infection in vivo. These findings further our knowledge of the CNS viral population phenotypes that are associated with the severity of HIV/SHIV-induced neurological injury and improve our understanding of the mechanism of HIV-1 cellular tropism and persistence in the brain.

Importance: Human immunodeficiency virus type 1 (HIV-1) infection of astrocytes in the brain has been suggested to be important in HIV persistence and neuropathogenesis but has not been definitively demonstrated in an animal model of HIV-induced encephalitis (HIVE). Here, we describe a new nonhuman primate (NHP) model of R5 simian-human immunodeficiency virus (SHIV)-induced encephalitis (SHIVE) with several classical HIVE features that include astrocyte infection. We further show an association between severe neuropathological changes, robust resident microglia infection, and evolution to CD4 independence of viruses in the central nervous system (CNS), with expansion to infection of truly CD4-negative cells in vivo. These findings support the use of the R5 SHIVE models to study the contribution of the HIV envelope and viral clades to neurovirulence and residual virus replication in the CNS, providing information that should guide efforts to eradicate HIV from the body.

FIG 1

Types of neuropathologic lesions in the brains of subtype B R5 SHIV_SF162P3N-infected macaques. Shown are SIV⁺ cells as revealed by in situ hybridization in cortical perivascular histolytic lesions with MNGCs (1), thalamic expansive lesions with white matter damage (2), and vacuolated chronic burnt-out lesions (3). Subsets of CD163⁺, Mac387⁺, and CD68⁺ macrophages are demonstrated by IHC in the insets, with SIV (green), CD68 (red), and glut-1 (blue) shown in the confocal inset.

FIG 2

Phylogenetic tree analysis of CSF and blood env sequences in subtype B R5 SHIVE macaques. Neighbor-joining phylogenetic trees of env V3-to-V5 sequences from the CSF and blood plasma rooted to HIV-1SF162 are shown, with bootstrap values of >70% indicated at the appropriate nodes. Genetic-distance scales representing the number of nucleotide substitutions per site between env sequences (0.005) are shown at the bottom of each tree. *, sequence obtained by conventional PCR.

FIG 3

ET94 Env analysis. (A) Phylogenetic tree and Highlighter plot analyses of the env sequences from the CSF, cerebellum, and occipital and temporal regions of ET94. A neighbor-joining tree of the env V1-to-V5 region rooted to HIV-1SF162 is shown, with bootstrap values of >70% indicated at the appropriate nodes. All sequences were obtained by SGA, and the genetic distance, representing the number of nucleotide substitutions per site between env sequences, is scaled at the bottom (0.005). A Highlighter plot was generated at www.hiv.lanl.gov, and the sequence identifiers for the Env clones used in functional studies are indicated on the y axis. The positions of nucleotide base transitions and transversions in the Highlighter plots are indicated by short colored-coded bars. Tics that are bracketed represent G-to-A changes, while those in circles represent G-to-A changes in a sequence consistent with an APOBEC signature. (B and C) sCD4 sensitivity and infection of primary macrophages and CD4^low (B) and CD4-negative (C) cells of pseudoviruses expressing Envs from the different brain regions and the CSF compartment. (B)The numbers in parentheses indicate the number of clones analyzed per compartment. (C) The level of infection of CD4-negative cells is expressed as a percentage of that of cells expressing both CD4 and CCR5. P values by Mann-Whitney t tests that approach significance (<0.1) are shown, with asterisks indicating those that reached statistical significance (P < 0.05). (D) Comparison of the abilities of the three individual CSF Env clones of ET94 to mediate infection of primary macrophages and CD4^low cells. The error bars in panels B, C, and D indicate means and standard deviations (SD) of 2 or 3 independent experiments.

FIG 4

CD4-independent Envs in the brains and CSF of R5 SHIV_SF162P3N SHIVE macaques. Infection mediated by Envs amplified from the brains and CSF of the five SHIV_SF162P3N SHIVE macaques in Cf2Th.CCR5 was assessed, with the level of infection in CD4-negative cells expressed as a percentage of that of Cf2Th.CD4^hi.CCR5 cells that express both receptors. The dashed lines delineate 20% CD4-independent entry efficiency, and the insets show absolute relative light units (RLU) in Cf2Th.CD4^hi.CCR5 cells of select CD4i Env populations from ET94 and DG08. The error bars indicate means and SD of 2 or 3 independent experiments.

FIG 5

Phylogenetic tree analysis of env sequences in the blood and brains of subtype C SHIVE macaques. Neighbor-joining phylogenetic trees of env V3-to-V5 sequences from the plasma, CSF, cerebellum, and occipital and parietal regions of the brains of SHIVC-infected macaques GC98 and FI08 rooted to ZM109F.PB4 are shown, with bootstrap values of >70% indicated at the appropriate nodes. Genetic distances representing the number of nucleotide substitutions per site between env sequences (0.005) are indicated at the bottom of each tree. *, sequence obtained by SGA.

FIG 6

CD4-independent Envs in the brains of subtype C SHIVE macaques GC98 and FI08. Env clones amplified from the cerebellum (blue) and occipital (purple) and parietal (green) regions and from the CSF (red) were assessed for the ability to enter Cf2th.CCR5 cells. Infectivity in these CD4-negative cells is expressed as a percentage of that of cells expressing both receptors. The dashed lines delineate 20% CD4-independent entry efficiency, and the insets show absolute RLU in Cf2Th.CD4^hi.CCR5 cells of pseudovirions expressing representative Env populations with >20% CD4i entry efficiency. The data are the means and SD of 2 or 3 independent experiments.

FIG 7

Amino acid sequence alignment of CD4-dependent and CD4i envelope clones from the CSF of ET94 (A), GC98 (B), and FI08 (C). The full-length gp120 of the CD4-dependent clone 2 and the CD4i clone 20 of ET94, as well as the V1-to-V5 gp120 of CD4-dependent clones 4 and 6 and CD4i clones 20 and 7 from GC98 and FI08, respectively, are shown. The dashes indicate amino acid identity, and changes altering potential sites of N-linked glycosylation are boxed.

FIG 8

Astrocyte infection in the brains of subtype C SHIVE macaque GC98 (A) and FI08 (B). Immunohistochemistry and in situ hybridization staining of brain sections were carried out (SIV ISH and IHC for GFAP).