Increased APOBEC3G Expression Is Associated With Extensive G-to-A Hypermutation in Viral DNA in Rhesus Macaque Brain During Lentiviral Infection

By Depboylu, Candan Eiden, Lee E; Weihe, Eberhard

Abstract APOBEC3G restricts retrovirus replication through inducing guanosine-to-adenosine (G-to-A) hypermutations in viral DNA. Its role in brain “intrinsic immunity” has not been elucidated nor has it been convincingly demonstrated which brain cell compartments produce this defense factor in human immunodeficiency virus (HIV) infection, acquired immunodeficiency syndrome (AIDS), and antiretroviral therapy. Here, we investigated by immunohistochemistry and in situ hybridization the cell-specific regulation of APOBEC3G in rhesus macaque brains during infection with simian immunodeficiency virus (SIV) clone deltaB670, a primate model of HIV disease. We found that APOBEC3G protein and mRNA were exclusively expressed by some perivascular macrophages throughout the brain of noninfected and asymptomatic SIV-infected monkeys. Depending on virus burden, APOBEC3G was induced in microglia/ macrophage-derived cells and T-lymphocytes in late-stage SIV infection. Intracellularly, APOBEC3G was found in the cytoplasm and/ or in the nucleus. APOBEC3G-positive cells were in close proximity to SIV gag-positive cells or were SIV-copositive. Induction of APOBEC3G was accompanied by G-to-A hypermutations in the gag and pol regions of retroviral DNA isolated from brain sections of AIDS- symptomatic monkeys. Although brain-directed treatment with antiretroviral 6-chloro-2′,3′-dideoxyguanosine suppressed brain SIV burden, encephalitis and reduced cerebral APOBEC3G synthesis hypermutations were still detectable. Upregulation of APOBEC3G may restrict spread of SIV in the brain and thus limit brain damage during lentiviral infection.

Key Words: Antiretroviral treatment, Hypermutations, Inflammatory infiltrates, Intrinsic immunity, Macrophages, Neuro-AIDS, T- lymphocytes.

INTRODUCTION

Infection with human immunodeficiency virus type-1 (HIV-1) leads to disease manifestations secondary to the profound immunodeficiency that develops in acquired immunodeficiency syndrome (AIDS). AIDS- related disease can result in encephalopathy, manifesting in cognitive and behavioral impairments and progressing to dementia and neurologic symptoms. Pathologically HIV-induced encephalopathy is characterized by encephalitis with the appearance of HIV-containing macrophage nodules, multinucleated giant cells, and inflammatory cell infiltrates in the brain (1), leading to severe tissue damage in the CNS.

In addition to innate and adaptive immune responses to retroviral infection, there is a third component of retroviral recognition and subsequent restriction that has recently been called “intrinsic immunity” (2). Human and nonhuman primate cells harbour at least 2 intrinsic intracellular resistance mechanisms that can suppress retroviral infection. The first is mediated by members of the apolipoprotein B mRNA-editing enzyme catalytic polypeptide-like (APOBEC) family of cytidine deaminases and was discovered through efforts to understand the role of the HIV-1 infectivity factor vif during viral infection (3). The second is mediated by tripartite interaction motif (TRIM) proteins and was revealed through studies of species-specific postentry blocks to HIV and simian immunodeficiency virus (SIV) infections (4). To our knowledge it is not known whether these intrinsic factors participate in the immune response of the brain to immunodeficiency virus infection.

The member of the APOBEC family APOBEC3G restricts the replication of retroviruses (3). It is incorporated into virions during viral replication and induces guanosine-to-adenosine (G-to- A) mutations in the viral DNA synthesized during reverse transcription (5).

Immunodeficiency viruses synthesize vif, which accelerates the proteasomal degradation of APOBEC3G, reducing incorporation of APOBEC3G into virions (6). This is a mechanism by which immunodeficiency viruses overcome the potent antiviral function of APOBEC3G. More recently, nonediting antiviral mechanism of APOBEC3G was additionally found (7). Lymphocytes and monocytes have been proposed to produce APOBEC3G (8). Although G-to-A hypermutation in the brain during HIV-1 infection has been described (9-11), no data exist about regulation of APOBEC3G in the brain during immunodeficiency virus infection. So far it has been reported that APOBEC3G is exclusively synthesized by neurons in the brain as demonstrated in pigtailed macaques (12). No or very low levels of APOBEC3G transcript expression were found in RNA preparations of the mouse (8) and human brain (13-15).

In this study, we hypothesized that immunodeficiency virus- induced encephalitis may be associated with increased expression of APOBEC3G by brain resident and/or infiltrating immune cells. To test this hypothesis we examined by immunohistochemistry (IHC) and in situ hybridization (ISH) tissue sections of several brain regions of rhesus macaques infected with SIV as a model of human neuro-AIDS (16) and explored the effects of disease course. We analyzed whether APOBEC3G is increased in brain monocytes/macrophages and lymphocytes during SIV infection and whether APOBEC3G expression is related to viral burden and G-to-A hypermutation of SIV in the CNS. We also determined APOBEC3G status in the CNS of SIV-infected rhesus macaques with AIDS after treatment with 6-chloro-2′,3′- dideoxyguanosine (6-C1-ddG), previously shown to efficiently penetrate into the CNS and to be highly effective in reducing brain viral burden in SIV-infected rhesus macaques (17-19).

MATERIALS AND METHODS

Virus Stock, Inoculation Procedures, Antiretroviral Treatment, and Tissue Preparation

Procedures performed on rhesus macaques have been described previously (17, 18). Experiments involving the use of rhesus macaques were approved by the Animal Care and Use Committee of Bioqual, Inc., a National Institutes of Health-approved and Association for Assessment and Accreditation of Laboratory Animal Care-accredited research facility. All experiments were carried out under the ethical guidelines promulgated in the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Healthy juvenile rhesus macaques were inoculated intravenously with 10 rhesus infectious doses of cell-free SIV^sub deltaB670^. After inoculation, animals were monitored and examined for clinical evidence of disease. Blood and cerebrospinal fluid samples were obtained from the animals in short intervals. At time of death, 8 macaques exhibited clinical signs of AIDS (17, 18). Four age- matched, noninfected macaques were used as controls. Additionally, 8 SIV-infected monkeys that were treated with 6-Cl-ddG subcutaneously when their viral load was found to be > 100,000 virions/mL in plasma and >100 virions/mL in cerebrospinal fluid in more than 2 consecutive examinations were analyzed (17). One monkey received 6- Cl-ddG (200 mg/kg/day) for 3 weeks. The other 7 monkeys received 10 mg/kg/day 2′,3′-dideoxyinosine for 3 weeks for clinical stabilization and then 75 mg/kg/day of 6-Cl-ddG for 6 weeks. The vehicle for 2′,3′-dideoxyinosine administration was PBS and for 6- Cl-ddG administration was 70% propylene glycol/30% PBS. At death, anesthetized animals were perfused transcardially with PBS and formalin/PBS. Tissue specimens were obtained during necropsy and immersion-fixed overnight. Some blocks were cryopreserved in sucrose and snap-frozen in dry ice-cooled isopentane. Some blocks were postfixed in Bouin-Hollande solution and processed for paraffin embedding (17, 18).

Single and Double Immunohistochemistry

IHC was performed on deparaffinized tissue sections (7 [mu]m) or cryosections (14 [mu]m) from different brain regions including cortex, basal ganglia, thalamus, brainstem, and cerebellum using an antigen retrieval technique and visualized enzymatically with 3,3′- diaminobenzidine (Sigma, Deisenhofen, Germany) or enhanced by addition of ammonium nickel sulfate (Fluka, Buchs, Switzerland) or visualized by immunofluorescence, as previously described (17, 18). Additionally, IHC was performed on deparaffinized tissue sections from dorsal root ganglia and peripheral organs such as lymph node, adrenal gland, liver, intestine, and kidney. The antibodies used in this study are listed in Table 1. Recombinant human APOBEC3G protein (CEM15^sub E.coli,^ National Institutes of Health AIDS Research and Reference Program, Germantown, MD) was used for preabsorption controls of the antibodies against APOBEC3G. Double immunofluorescence was carried out for visualization of 2 different antigens in the same tissue section (17, 18). Serial immunostainings were carried out on adjacent sections. IHC signals were analyzed and documented with an Olympus AX70 microscope or with an Olympus Fluoview confocal laser scanning microscope (Olympus Optical, Hamburg, Germany).

Generation of Riboprobes, Radioactive In Situ Hybridization, and Combination of Immunofluorescence With In Situ Hybridization

Because monkey and human APOBEC3G are highly homologous, human- specific riboprobes were used to perform ISH on rhesus monkey tissue sections. A human APOBEC3G cDNA (accession number: NM021822) template of ~0.9 kilobase pair was amplified by polymerase chain reaction (PCR) from human spleen cDNA using the following primer set derived from the published sequence: 5′-TCA GAA ACA CAG TGG AGC GAA- 3′ and 5′-ATT TCC TGG GCA CAG CTG AA-3′. Generated cDNA was identical to the published APOBEC3G cDNA sequences of human and rhesus macaque (accession number: XM001094452). Using the online available nucleotide BLAST software (http://www.ncbi.nlm.nih.gov/ BLAST), minor inconsistent sequence homology existed only with human and macaque APOBEC3F cDNAs and not with other members of the APOBEC family. Amplified cDNA was subcloned into the pGEM-T Vector (Promega, Madison, WI). For ISH, 14-[mu]m-thick cryosections were cut from cryopreserved tissues throughout the brain and processed as reported previously (17, 18). Specific sense and antisense riboprobes were generated from linearized vector constructs by in vitro transcription using the appropriate RNA polymerases and [^sup 35^S]-UTP as label and applied on sections after limited alkaline hydrolysis. After application of [^sup 35^S]-riboprobes, radioactive signals were detected by autoradiography on Hyperfilm beta-max (Amersham Biosciences, Freiburg, Germany) for 1 to 4 days to estimate further exposure time when sections were coated with NTB-2 nuclear emulsion (Eastman Kodak, Rochester, NY). Immunofluorescence was performed in combination with radioactive ISH for visualization of an antigen with an RNA transcript in the same tissue section, as described previously (17, 18). Hybridized sections were analyzed and documented with an Olympus AX70 microscope. TABLE 1. Antibodies and Lectin Used in This Study

Detection of SIV Protein

A monoclonal antibody (Table 1) that was raised against gag of SIV^sub mac251^ was used to detect the cross-reacting gag from SIV^sub deltaB670^ (28). In some experiments gag was used for costaining with other antigens.

Isolation, Amplification and Sequencing of SIV DNA

Two to 3 (14 [mu]m) serial cryosections from basal ganglia of each monkey were carefully scraped from slides with a razor blade and pooled in PBS. After overnight digestion with proteinase K at 55[degrees]C and RNase (Sigma, Deisenhofen, Germany) digestion for 1 hour at 37[degrees]C, DNA was extracted using the DNeasy DNA Isolation Kit (Qiagen, Hilden, Germany). Fragments of ~0.7 kilobase pair from SIV gag and pol regions, respectively, were amplified by touchdown PCR with AmpliTaq DNA Polymerase (Applied Biosystems, Foster City, CA) and cloned into the TOPO TA-Cloning Vector pCR4 (Invitrogen, Karlsruhe, Germany). Nucleotide sequences of 20 independent clones for gag and pol per monkey, respectively, were analyzed using Hypermut software (30). This software which is available online (www.hiv.lanl.gov/HYPERMUT/hypermut.html) compares sequence sets to a reference sequence. The following primer mixtures were used: for gag, 5′-TTC ACG CAX AAX AXA AAG TXA-3′ (forward) and 5′-TTT TGA ATY AGC AGT GTT TGG-3′ (reverse); and for pol, 5-CXA TTC AAT TGT AGC AXX XAT-3′ (forward) and 5′-AAA AAT AGC TGG TGA YYY YTT- 3′ (reverse). Sequences susceptible to hypermutation (GA or GG in forward primers and TC or CC in reverse primers, respectively) are modified by incorporation of G+A (X) or C+T mixtures (Y).

RESULTS

Expression of APOBEC3G in Control Brain and Early-Stage SIV Infection

In control animals, APOBEC3G was exclusively expressed in some perivascular cells and was absent from other brain resident cells such as neurons, astrocytes, and microglia throughout the brain, including cortex, thalamus, basal ganglia, brainstem, and cerebellum. In the brains of asymptomatic SIV-infected rhesus monkeys (SIV,-AIDS), APOBEC3G was restricted to some perivascular cells (Fig. 1A, D, G, J), not significantly different from control. In both control and asymptomatic infected animals, colocalization studies identified these APOBEC3G-positive perivascular cells as CD68-positive perivascular macrophages.

Upregulation of APOBEC3G in the Brain in Late-Stage SIV Infection

A dramatic increase of APOBEC3G expression in the brain was seen in late-stage SIV infection (SIV,+AIDS) (Fig. 1B, E, H, K), as revealed by a panel of different rabbit polyclonal antibodies against APOBEC3G (Table 1), which all showed the same pattern. The mouse monoclonal antibody against APOBEC3G (Table 1) produced nonspecific stainings in given sections regardless of whether the sections were from Bouin-Hollande-fixed paraffin-embedded brain tissue blocks or fixed sucrose-protected cryotissue blocks. In contrast, the polyclonal antibodies against APOBEC3G produced clear signals in selective cell populations. In preabsorption controls with excess APOBEC3G protein, the immunohistochemical stainings only of the polyclonal antibodies were absent, demonstrating the specificity of these antibodies. Combination of immunofluorescence with radioactive ISH proved that increased APOBEC3G protein levels were due to increased mRNA synthesis (Fig. 2) and proved again the specificity of the protein staining. ISH with the riboprobe in sense orientation against APOBEC3G did not show any signals except background, confirming that our antisense probe was specifically detecting APOBEC3G mRNA.

FIGURE 1. Changes of APOBEC3G protein and mRNA expression in the basal ganglia after simian immunodeficiency virus (SIV) infection and antiretroviral treatment demonstrated by representative brightfield (A-F) and darkfield images (G-L) of immunohistochemistry and in situ hybridization for APOBEC3G, respectively. APOBEC3G protein (A, D) and mRNA (G, J) are detected only in some perivascular cells in the SIV, – acquired immunodeficiency syndrome (AIDS) group (D, J) and not in cells of the parenchyma (A, G). In the SIV,+AIDS group increased focal expression of APOBEC3G protein (B, E) and mRNA (H, K) is observed. Note that after 6-chloro-2′,3′- dideoxyguanosine (6-Cl-ddG) treatment (SIV,+AIDS,+ddG) APOBEC3G protein and mRNA expression are suppressed in the parenchyma (C, I) and restricted to cells in perivascular areas (F, L). Scale bars = 50 [mu]m.

FIGURE 2. APOBEC3G protein detected by Cy3-labeled secondary antibody (A) is colocalized with APOBEC3G mRNA detected by in situ hybridization with [^sup 35^S]-labeled riboprobes (silver grains; [B], brightfield; [C], darkfield) in a nodule (arrow) as well as in perivascular cuffs (arrowheads) on the same brain section of an acquired immunodeficiency syndrome (AIDS)-diseased monkey representative for the simian immunodeficiency virus (SIV), +AIDS group. Scale bars = 50 [mu]m.

FIGURE 3. Representative confocal double immunofluorescence of section (A-C) and adjacent section (D-F) from an acquired immunodeficiency syndrome (AIDS)-diseased monkey alternately costained for APOBEC3C and CD68 (A-C) and for APOBEC3G and simian immunodeficiency virus (SIV) gag (D-F), respectively. (A-C) APOBEC3G (red) is found in CD68-positive nodular, perinodular, and perivascular macrophages and multinucleated giant cells (green) besides APOBEC3G-negative/CD68-positive macrophages. (D-F) APOBEC3G (red)-expressing cells are close to (arrows) or even colocalized (arrowheads) with SIV gagpositive cells (green). C and F are merged images. Scale bars = 50 [mu]m.

To identify the cells with upregulation of APOBEC3G expression in late-stage SIV disease, double immunofluorescence and serial staining of adjacent sections from different brain regions for APOBEC3G and established markers for various brain resident cells and for immune cells invading the brain were performed. Colocalization of APOBEC3G with the microglia/macrophage activation marker CD68 revealed APOBEC3G expression in perivascular and parenchymal activated microglia/macrophages, multinucleated giant cells, and macrophage nodules (Fig. 3A-C). Some CD68-positive monocytes attached to endothelial cells also expressed APOBEC3G in the brains of AIDS-symptomatic monkeys (Fig. 4A-C), demonstrating APOBEC3G synthesis during infiltration through the blood-brain barrier. It is very likely that these cells were truly adherent and in the process of infiltrating into the brain because they were not washed away by the transcardial perfusion before or during fixation procedures as previously demonstrated and quantitated by us for the same animals in the current study (18). There were also APOBEC3G- positive cells attached to ependymal surfaces at the ventricular sites. Some CD3-positive T-lymphocytes were APOBEC3G-positive (Fig. 4D-F). We did not observe APOBEC3G expression in neurons stained for the neuronal marker neuronal nuclear protein (Fig. 5A-C). APOBEC3G was absent from glial fibrillary acidic protein-positive astrocytes (Fig. 5D-F), 2′,3′-cyclic nucleotide 3′-phosphodiesterase-positive oligodendrocytes (Fig. 5 G-I), von Willebrand factor-positive endothelial cells and CD20-positive B-lymphocytes. APOBEC3G- positive cells were in close proximity to SIV gag-positive cells or even coincided with them in the brain of AIDS-diseased monkeys (Fig. 3D-F). At the blood-brain barrier, APOBEC3G was present in SIV gag- positive as well as in SIV gag-negative infiltrating cells (Fig. 4G- I). The subcellular localization of APOBEC3G comprised cytoplasm and nucleus.

FIGURE 4. Representative confocal double immunofluorescence demonstrating synthesis of APOBEC3G (red) in CD68-positive (green) perivascular macrophages (arrowhead) and endothelium-adherent infiltrating monocytes (arrows) (A-C) and in a CD3-positive (green) lymphocyte (D-F) in the brain of a monkey of the simian immunodeficiency virus, +acquired immunodeficiency syndrome group (SIV,+AIDS). Note that not all brain-infiltrating APOBEC3C-positive (red) monocytes (arrowheads) and lymphocytes (arrow) are SIV gag- positive (green) ([G-I], insets in [G-I]). C, F, and I are merged images. Scale bars = (A-C, G-I) 15 [mu]m; (D-F) 20 [mu]m.

FIGURE 5. Analyses of cellular localization of APOBEC3G in the brain of monkeys from the simian immunodeficiency virus, +acquired immunodeficiency syndrome group (SIV,+AIDS). Representative confocal double immunofluorescence for APOBEC3G with established markers for brain resident cells demonstrating a strict lack of APOBEC3G expression in neuronal nuclear protein (Neun)-positive neurons (A- C), glial fibrillary acidic protein (GFAP)-positive astrocytes (D- F) and 2′,3′-cyclic nucleotide 3′-phosphodiesterase (CNPage)- positive oligodendrocytes (G-I) in SIV encephalitis. C, F, and I are merged images. Scale bars = 30 [mu]m. Effect of Antiretroviral Treatment with 6-CI-ddG on APOBEC3C in Late-Stage SIV Infection

To determine whether APOBEC3G biosynthesis is related to active viral replication, APOBEC3G expression was analyzed in the brain of SIV-infected monkeys treated with the antiviral 6-Cl-ddG in late- stage SIV infection (SIV,+AIDS,+ddG) (17, 18). We and others have previously demonstrated that 6-Cl-ddG effectively suppressed viral replication in the brain (17-19). Brain APOBEC3G synthesis was markedly reduced by 6-Cl-ddG (Fig. 1C, F, I, L). Only some perivascular macrophages were APOBEC3G-positive in the brain of SIV,+AIDS,+ddG animals (Fig. 1F, L). The number of SIV gag-positive cells, macrophage nodules, multinucleated giant cells, and inflammatory infiltrates was markedly reduced in the brains of 6-Cl- ddG-treated monkeys compared with nontreated AIDS-diseased monkeys (Table 2).

Guanosine-to-Adenosine Hypermutations in the Brain During SIV Infection

To analyze whether upregulated APOBEC3G may have functional activity on SIV, viral gag and pol regions were PCR-based amplified from RNase-treated DNA extracted from basal ganglia cryosections. Twenty clones were sequenced for gag and pol from each SIV-infected monkey. Hypermutated sequences were defined as having >5% of the total guanosines mutated to adenosine. No PCR signal could be achieved for gag and pol from noninfected monkey brains. Hypermutated clones were detected in only 1 animal from 5 asymptomatically SIV-infected monkeys (2 of 20 for gag and 3 of 20 for pol). In contrast, hypermutated clones for gag and pol were found in all 8 untreated AIDS-diseased monkeys (3-5 of 20 for gag and pol, respectively). Hypermutated clones were isolated in 4 of 8 antiretrovirally treated monkeys (2-4 of 20 for gag and pol, respectively). Most G-to-A hypermutations were found in GG/GGG sequences, with no significant preference in gag or pol regions and no significant preference for an individual monkey group. See Table 2 for details.

TABLE 2. Summary of Results for Viral Protein Detection, Signs of Encephalitis, Biosynthesis of APOBEC3G and Mutational Analysis

Expression of APOBEC3G Outside the Brain in Late-Stage SIV Infection

To explore whether APOBEC3G expression is restricted to monocytes/ macrophages and T-lymphocytes in the periphery in late-stage SIV infection, we performed APOBEC3G immunostaining in peripheral nervous tissue (dorsal root ganglia) and peripheral lymphoid (lymph node, intestine mucosa, and liver) and nonlymphoid organs (kidney and adrenal gland) from AIDS-diseased monkeys. APOBEC3G was expressed in isolectin RCA-120-labeled macrophages but not in RCA- 120-positive endothelial cells and RCA-120-positive ganglion cells in dorsal root ganglia (Fig. 6A). SIV gag-positive cells were APOBEC3G-copositive or surrounded by APOBEC3G-positive cells in the dorsal root ganglia (Fig. 6B). In lymph nodes, APOBEC3G was expressed in a subpopulation of isolectin-positive mononuclear cells including macrophages and dendritic-like cells (Fig. 6C), as well as in a subpopulation of CD3-positive T-lymphocytes (Fig. 6D), but was devoid in CD20-positive B-lymphocytes (Fig. 6E). APOBEC3G was expressed in some SIV-infected cells in lymph nodes (Fig. 6F). As demonstrated by single stainings, APOBEC3G-immunoreactive cells exhibiting characteristics of macrophages, dendritic-like cells, and lymphocytes occurred in the kidney, liver, adrenal gland, and small intestine during SIV disease (Fig. 7C, D, F-H). Hepatocytes surrounding the central veins were also APOBEC3G-positive in the liver (Fig. 71). The polyclonal antibodies were specifically detecting APOBEC3G in the peripheral tissues as the immunohistochemical stainings were absent in preabsorption controls with excess recombinant APOBEC3G protein as demonstrated for kidney tissue from an control animal (Fig. 7A, B) and in adrenal gland tissue from an AIDS-diseased animal (Fig. 7D, E).

FIGURE 6. Representative confocal double immunofluorescence on serial sections through the lumbar dorsal root ganglion (A, B) and axillary lymph node (C-F) from a monkey of the simian immunodeficiency virus, + acquired immunodeficiency syndrome (SIV,+AIDS) group stained for APOBEC3C with established cellular markers and SIV. (A) APOBEC3G (red) is predominantly expressed in cells of mononuclear origin, including macrophages and multinucleated giant cells, as revealed by isolectin histochemistry (RCA-120; green). Isolectin-positive endothelial cells and ganglion cells with their processes (green) are not stained for APOBEC3G (red). (B) SIV gag-positive cells (green) are APOBEC3G-positive or surrounded by APOBEC3G-positive cells (red) in the dorsal root ganglion. (C, D) A subpopulation of RCA-120-positive macrophages and dendritic-like cells (green) and CD3-positive T-lymphocytes (green) are stained for APOBEC3G (red). CD20-positive B-lymphocytes (green) lack APOBEC3G expression (red) (D). (E) Only a subpopulation of SIV gag-positive cells is APOBEC3G-positive (red). Single colored images in the left and middle columns are merged in the right column. Scale bars = (A, B) 100 [mu]m; (C-F) 50 [mu]m.

FIGURE 7. Expression of APOBEC3G in peripheral organs as demonstrated by single immunohistochemistry. (A) APOBEC3G is expressed in dendritic-like cells in the glomeruli and in the parenchyma in the kidney of a noninfected control monkey. (B) The antibody recognizes APOBEC3G specifically as its staining is preabsorbed by coincubation with excess recombinant APOBEC3G protein on an adjacent section. (C) APOBEC3G is upregulated in inflammatory infiltrates in the kidney of a monkey of the simian immunodeficiency virus, + acquired immunodeficiency syndrome (SIV,+AIDS) group. ([D] with inset) APOBEC3G is stained in periglandular cells, presumably mononuclear cells, in the cortex of the adrenal gland of an AIDS- diseased monkey. (F) Additionally, this staining is preabsorbed with excess recombinant APOBEC3G protein on an adjacent section. (H, E) In the mucosa of the small intestine APOBEC3G is expressed by macrophages, multinucleated giant cells, dendritic-like cells, and lymphocytes in an inflammatory area in late-stage SIV infection. In the liver, inflammatory infiltrates and hepatocytes surrounding the central veins are stained for APOBEC3G in an AIDS-diseased monkey (G, I). Scale bars = (A-F, H, I) 100 [mu]m; (G) 25 [mu]m.

DISCUSSION

The essential findings of this study are as follows: 1) constitutive expression of APOBEC3G is restricted to perivascular macrophages in the brain; 2) APOBEC3G is markedly increased in cells of monocyte/macrophage-lineage forming nodules and syncytia and in some T-lymphocytes in SIV^sub deltaB670^-induced encephalitis; 3) expression of APOBEC3G is closely related to SIV burden and is fully susceptible to 6-Cl-ddG; 4) cerebral increase of APOBEC3G during SIV infection is accompanied by increased fingerprinting G-to-A hypermutation in SIV DNA; and 5) APOBEC3G expression is mainly restricted to the subset of monocytes/macrophages, dendritic-like cells, and T-lymphocytes in the peripheral nervous system and peripheral organs, with the exception of the liver in which APOBEC3G also occurred in hepatocytes surrounding the central veins in AIDS.

Our data demonstrate that the cellular host defense comprising APOBEC3G exists constitutively in perivascular areas known to represent the first immune barrier between the brain parenchyma and the blood. Immune cells in these areas are ideally located to sense perturbations and are continuously repopulated from the circulation. The number of APOBEC3G-positive perivascular macrophages is slightly increased in asymptomatic SIV-infected monkeys compared with noninfected controls, perhaps through the early presence of SIV in the brain, although SIV is not actively replicating at this stage, as demonstrated in our previous studies (17, 18). As brain perivascular macrophages are shown to be the primary cell type productively infected early during SIV infection (31), APOBEC3G may play an important role in limiting neuroinvasion of SIV early in disease. Expression of APOBEC3G in some perivascular macrophages in control animals suggests another undescribed cellular role of APOBEC3G rather than inhibition of retrotransposition of endogenous retroviruses, as recently reported (32). In late-stage disease, when SIV is highly replicating in the brain, APOBEC3G is concomitantly upregulated. Its expression is largely restricted close to areas of SIV burden with no evidence that it participated in the global activation of microglia in the brain after SIV infection. By genotype-specific ISH it has been demonstrated that predominantly microglia/macrophage-tropic subclones of SIV^sub deltaB670^ are present in the CNS of infected rhesus monkeys (33). Not unexpectedly, we detected APOBEC3G mostly in macrophage-type cells. In addition, we showed that CD3-positive T-lymphocytes can express APOBEC3G in SIV-induced encephalitis. In contrast to the study by Hill et al (12), we found no evidence for constitutive expression of APOBEC3G in neurons in cortex, basal ganglia, thalamic, and brainstem nuclei and cerebellum and there is also no induction of APOBEC3G biosynthesis in neurons throughout the brain during SIV disease.

Our analysis of the periphery reveals that APOBEC3G expression is largely restricted to a subset of immune cells in which SIV^sub deltaB670^ variants are able to replicate (33). Sites of peripheral APOBEC3G synthesis include dorsal root ganglia, adrenal gland, lymphoid tissues, gut, heart, and skin of our SIV-infected monkeys. In the liver, hepatocytes surrounding the central veins, interstitial macrophages, and T cells were found to express APOBEC3G. APOBEC3G expression in hepatocytes in the course of hepatitis B and C infection in conjunction with viral hypermutation was reported (34, 35). In contrast to Hill et al, who found constitutive APOBEC3G expression in epithelial cells of the proximal convoluted tubules in the kidney (36), our data indicate that APOBEC3G expression in the kidney is absent from epithelial cells and only present in macrophages, glomerular mesangial dendritic- like cells and lymphoid cells. Possible differences between our data and that of Hill et al (12, 36) may be attributed to the antibodies used and tissue processing. We were unable to get sufficient immunohistochemical staining using the mouse monoclonal antibody against APOBEC3G that Hill et al (12, 36) had used. Under culture conditions it was shown that APOBEC3G either exclusively localized in the cytoplasm (37, 38) or additionally in the nucleus (6). In our in vivo model we immunolocalized APOBEC3G in the nucleus and/or the cytoplasm of positive cells. Traffic into nucleus suggests that APOBEC3G may have a genomic DNA target. More studies in this direction are needed to determine whether APOBEC3G possibly targets chromatin-integrating or nonintegrating proviral DNA (39), nuclear translocated virus compounds such as vif (37) or retrotranspositioning endogenous retroviruses (32) or has an as yet unidentified cellular function.

Our observations that SIV sequences from pol and gag regions are hypermutated in all macaques with AIDS/encephalitis, whereas only 1 of 5 asymptomatically SIV-infected macaques without AIDS/ encephalitis and 4 of 8 antiretrovirally treated SIV-infected monkeys exhibited hypermutations, can be interpreted in the following way. We believe that the increased incidence of hypermutations in the brain of animals of the SIV,+AIDS group is a consequence of increased APOBEC3G levels due to increased local viral replication provoking mutationslocally locally in the brain. Hypermutations induced in peripheral immune cells before they enter the brain early during acute infection or infiltrate ongoingly during encephalitis contribute only to a minor extent to hypermutations seen in macrophages and T-lymphocytes in the brain. In contrast, hypermutations in the brains of asymptomatically infected and treated monkeys are attributed mainly to hypermutations occurring early during viral infection in macrophages and T- lymphocytes that entered the brain.

Other regions analyzed, including insular and frontal cortex and bordering white matter, exhibited the same mutation patterns as basal ganglia (data not shown), suggesting no evidence for regional variability of hypermutations. In support of this view, closely related immunodeficiency viral sequences were found in different brain regions (33, 40, 41). On the other hand, heterogeneous immunodeficiency viral populations were found to exist in different brain regions, suggestive of little gene flow among different brain regions (42, 43). The observation that APOBEC3G induction and SIV were mostly colocalized throughout the encephalitic brains of AIDS monkeys strongly suggests that hypermutations through APOBEC3G induction in the brain are region-independent, as seen in our model.

We cannot rule out additional deaminases such as APOBEC3F or APOBEC3B to contribute to some extent for G-to-A hypermutations that we observed. Nevertheless, we believe that APOBEC3G is the main deaminase acting on SIV because of following reasons: 1) based on our own immunohistochemical observations APOBEC3F is less expressed than APOBEC3G in SIV encephalitis (unpublished observations); 2) G- to-A hypermutations preferentially took place in GG/GGG sequences that are favored by APOBEC3G and less by APOBEC3F and APOBEC3B (14, 44, 45); and 3) it is shown that APOBEC3G is much more potent against retroviruses than APOBEC3F and APOBEC3B (44, 45).

One may argue that treatment with antiviral 6-Cl-ddG does not affect APOBEC3G levels directly, independent of the effect of 6-Cl- ddG on SIV. However, to our knowledge there is no evidence in the literature that antiviral agents in general and 6-Cl-ddG in particular would affect members of the APOBEC family directly. We did not address this question in this study but have reason to believe that the possibility of a direct regulatory effect of 6-Cl- ddG on APOBEC3G expression during SIV infection is rather unlikely. The viral burden levels of the monkeys differ between the 3 SIV groups, and the reduction in infectivity is not directly proportional to the mutagenic activity. Therefore, it is reasonable to assume that the antiretroviral treatment does not affect APOBEC3G directly. The lower expression of APOBEC3G in the SIV,+AIDS,+ddG group according to the SIV,+AIDS group is primarily a consequence of suppressed viral replication and downregulation of concomitant productive inflammation as a result of 6-Cl-ddG action on the SIV.

There are conflicting results concerning the correlation between APOBEC3G expression, HIV viremia, and disease progression (46, 47), as might be expected for a host antiviral mechanism that might succeed in both decreasing viral burden and accelerating viral mutation, leading to enhanced rates of replication in a way that is dependent on stage of disease and on the cellular and tissue compartments in which infection is ongoing. Our results suggest that APOBEC3G expression in the brain is positively correlated with virus burden and functionally active on SIV in late-stage infection. A local breakdown of the APOBEC3G-SIV balance in favor of the latter may lead to spread of the virus in the brain and then to neuro- AIDS. Enhancing intrinsic host defense factors to which APOBEC3G belongs could constitute a novel and cell-based therapy for immunodeficiency virus infection and AIDS.

ACKNOWLEDGMENTS

For excellent technical work we are indebted to R. Vertesi from L. E. Eiden’s laboratory; and to E. Rodenberg-Frank, M. Zibuschka, and R. Weber from E. Weihe’s laboratory. We thank our colleagues Hiroaki Mitsuya, Todd Reinhart, and Dianne Rausch for their collaboration and insights into mechanisms of pathogenicity and its treatment in the SIV-infected rhesus macaque model for HIV disease.

REFERENCES

1. Kato T, Hirano A, Llena JF, et al. Neuropathology of acquired immune deficiency syndrome (AIDS) in 53 autopsy cases with particular emphasis on microglial nodules and multinucleated giant cells. Acta Neuropathol (Berl) 1987;73:287-94

2. Bieniasz PD. Intrinsic immunity: A front-line defense against viral attack. Nat Immunol 2004;5:1109-15

3. Sheehy AM, Gaddis NC, Choi JD, et al. Isolation of a human gene that inhibits HIV-1 infection and is suppressed by the viral vif protein. Nature 2002;418:646-50

4. Stremlau M, Owens CM, Perron MJ, et al. The cytoplasmic body component TRIM5alpha restricts HIV-1 infection in Old World monkeys. Nature 2004;427:848-53

5. Harris RS, Bishop KN, Sheehy AM, et al. DNA deamination mediates innate immunity to retroviral infection. Cell 2003;113:803- 9

6. Stopak K, de Noronha C, Yonemato W, et al. HIV-1 Vif blocks the antiviral activity of APOBEC3G by impairing both its translation and intracellular stability. Mol Cell 2003;12:591-601

7. Bishop KN, Holmes RK, Malim MH. Antiviral potency of APOBEC proteins does not correlate with cytidine deamination. J Virol 2006;80: 8450-58

8. Mariani R, Chen D, Schrofelbauer B, et al. Species-specific exclusion of APOBEC3G from HIV-1 virions by Vif. Cell 2003;114:21- 31

9. Simmonds P, Balfe P, Ludlam CA, et al. Analysis of sequence diversity in hypervariable regions of the external glycoprotein of human immunodeficiency virus type 1. J Virol 1990;64:5840-50

10. Monken CE, Wu B, Srinivasan A. High resolution analysis of HIV-1 quasispecies in the brain. AIDS 1995;9:345-49

11. Liu Y, Tang XP, McArthur JC, et al. Analysis of human immunodeficiency virus type 1 gp 160 sequences from a patient with HIV dementia: Evidence for monocyte trafficking into brain. J Neurovirol 2000;6: 70-81

12. Hill MS, Mulcahy ER, Gomez ML, et al. APOBEC3G expression is restricted to neurons in the brains of pigtailed macaques. AIDS Res Hum Retroviruses 2006;22:541-50

13. Jarmuz A, Chester A, Bayliss J, et al. An anthropoid- specific locus of orphan C to U RNA-editing enzymes on chromosome 22. Genomics 2002;79:285-96

14. Liddament MT, Brown WL, Schumacher AJ, et al. APOBEC3F properties and hypermutation preferences indicate activity against HIV-1 in vivo. Curr Biol 2004;14:1385-91

15. Wiegand HL, Doehle BP, Bogerd HP, et al. A second human antiretroviral factor, APOBEC3F, is suppressed by the HIV-1 and HIV- 2 Vif proteins. EMBO J 2004;23:2451-58

16. Murray EA, Rausch DM, Lendvay J, et al. Cognitive and motor impairments associated with SIV infection in rhesus monkeys. Science 1992;255:1246-49

17. Depboylu C, Reinhart TA, Takikawa O, et al. Brain virus burden and indoleamine-2,3-dioxygenase expression during lentiviral infection of rhesus monkey are concomitantly lowered by 6-chloro- 2′, 3′-dideoxyguanosine. Eur J Neurosci 2004;19:2997-3005

18. Depboylu C, Schafer MK, Schwaeble WJ, et al. Increase of Clq biosynthesis in brain microglia and macrophages during lentivirus infection in the rhesus macaque is sensitive to antiretroviral treatment with 6-chloro-2′,3′-dideoxyguanosine. Neurobiol Dis 2005;20: 12-26

19. Fujii Y, Mukai R, Akari H, et al. Antiviral effects of 6- chloro-2′,3′-dideoxyguanosine in rhesus monkeys acutely infected with simian immunodeficiency virus. Antivir Chem Chemother 1998;9: 85-92

20. Kao S, Khan MA, Miyagi E, et al. The human immunodeficiency virus type 1 Vif protein reduces intracellular expression and inhibits packaging of APOBEC3G (CEM15), a cellular inhibitor of virus infectivity. J Virol 2003;77:11398-407

21. Braak E, Drenckhahn D, Unsicker K, et al. Distribution of myosin and the glial fibrillary acidic protein (GFA protein) in rat spinal cord and in the human frontal cortex as revealed by immunofluorescence microscopy. Cell Tissue Res 1978;191:493-99 22. Campana D, Thompson JS, Amlot P, et al. The cytoplasmic expression of CD3 antigens in normal and malignant cells of the T lymphoid lineage. J Immunol 1987;138:648-55

23. Mason DY, Comans-Bitter WM, Cordell JL, et al. Antibody L26 recognizes an intracellular epitope on the B-cell-associated CD20 antigen. Am J Pathol 1990;136:1215-22

24. Pulford KA, Rigney EM, Micklem KJ, et al. KP1: A new monoclonal antibody that detects a monocyte/macrophage associated antigen in routinely processed tissue sections. J Clin Pathol 1989;42:414-21

25. Sehested M, Hou-Jensen K. Factor VII related antigen as an endothelial cell marker in benign and malignant diseases. Virchows Arch A Pathol Anat Histol 1981;391:217-25

26. Mullen RJ, Buck CR, Smith AM. NeuN, a neuronal specific nuclear protein in vertebrates. Development 1992;116:201-11

27. Sprinkle TJ. 2′,3′-Cyclic nucleotide 3′-phosphodiesterase, an oligodendrocyte-Schwann cell and myelin-associated enzyme of the nervous system. Crit Rev Neurobiol 1989;4:235-301

28. Kent KA, Rud E, Corcoran T, et al. Identification of two neutralizing and 8 non-neutralizing epitopes on simian immunodeficiency virus envelope using monoclonal antibodies. AIDS Res Hum Retroviruses 1992;8:1147-51

29. Mannoji H, Yeger H, Becker LE. A specific histochemical marker (lectin Ricinus communis agglutinin-1) for normal human microglia, and application to routine histopathology. Acta Neuropathol (Basel) 1986;71:341-43

30. Rose PP, Korber BT. Detecting hypermutations in viral sequences with an emphasis on G[arrow right]A hypermutation. Bioinformatics 2000;16: 400-401

31. Williams KC, Corey S, Westmoreland SV, et al. Perivascular macrophages are the primary cell type productively infected by simian immunodeficiency virus in the brains of macaques: Implications for the neuropathogenesis of AIDS. J Exp Med 2001;193:905-15

32. Esnault C, Heidmann O, Delebecque F, et al. APOBEC3G cytidine deaminase inhibits retrotransposition of endogenous retroviruses. Nature 2005;433:430-33

33. Reinhart TA, Rogan MJ, Amedee AM, et al. Tracking members of the simian immunodeficiency virus deltaB670 quasispecies population in vivo at single-cell resolution. J Virol 1998;72:113-20

34. Komohara Y, Yano H, Shichijo S, et al. High expression of APOBEC3G in patients with hepatitis C virus. J Mol Histol 2006;37:327-32

35. Noguchi C, Ishino H, Tsuge M, et al. G to A hypermutation of hepatitis B virus. Hepatology 2005;41:626-33

36. Hill MS, Ruiz A, Gomez LM, et al. APOBEC3G expression is restricted to epithelial cells of the proximal convoluted tubules and is not expressed in the glomeruli of macaques. J Histochem Cytochem 2007; 55:63-70

37. Wichroski MJ, Ichiyama K, Rana TM. Analysis of HIV-1 viral infectivity factor-mediated proteasome-dependent depletion of APOBEC3G: Correlating function and subcellular localization. J Biol Chem 2005;280:8387-96

38. Bennett RP, Diner E, Sowden MP, et al. APOBEC-1 and AID are nucleo-cytoplasmic trafficking proteins but APOBEC3G cannot traffic. Biochem Biophys Res Commun 2006;350:214-19

39. Bell P, Montaner LJ, Maul GG. Accumulation and intranuclear distribution of unintegrated human immunodeficiency virus type 1 DNA. J Virol 2001;75:7683-91

40. Cunningham AL, Naif H, Saksena N, et al. HIV infection of macrophages and pathogenesis of AIDS dementia complex: Interaction of the host cell and viral genotype. J Leukoc Biol 1997;62:117-25

41. Chang J, Jozwiak R, Wang B, et al. Unique HFV type 1 V3 region sequences derived from six different regions of brain: Region- specific evolution within host-determined quasispecies. AIDS Res Hum Retroviruses 1998;14:25-30

42. Shapshak P, Segal DM, Crandall KA, et al. Independent evolution of HIV type 1 in different brain regions. AIDS Res Hum Retroviruses 1999;15:811-20

43. Chen MF, Westmoreland S, Ryzhova EV, et al. Simian immunodeficiency virus envelope compartmentalizes in brain regions independent of neuropathology. J Neurovirol 2006;12:73-89

44. Bishop KN, Holmes RK, Sheehy AM, et al. Cytidine deamination of retroviral DNA by diverse APOBEC proteins. Curr Biol 2004;14: 1392-96

45. Zennou V, Bieniasz PD. Comparative analysis of the antiretroviral activity of APOBEC3G and APOBEC3F from primates. Virology 2006; 349:31-40

46. Jin X, Brooks A, Chen H, et al. APOBEC3G/CEM15 (hA3G) mRNA levels associate inversely with human immunodeficiency virus viremia. J Virol 2005;79:11513-16

47. Cho S-J, Drechsler H, Burke RC, et al. APOBEC3F and APOBEC3G mRNA levels do not correlate with human immunodeficiency virus viremia or CD4^sup +^ T-cell count. J Virol 2006;80:2069-72

Candan Depboylu, MD, Lee E. Eiden, PhD, and Eberhard Weihe, MD

From the Department of Molecular Neuroscience (CD, EW), Institute of Anatomy and Cell Biology, Philipps University, Marburg, Germany; Department of Neurology (CD), Center for Nervous Diseases, Philipps University, Marburg, Germany; and Section on Molecular Neuroscience (LEE), Laboratory of Cellular and Molecular Regulation, National Institute of Mental Health, National Institutes of Health, Bethesda, Maryland.

Send correspondence and reprint requests to: Eberhard Weihe, MD, Institute of Anatomy and Cell Biology, Department of Molecular Neuroscience, Philipps University, Robert-Koch-Strasse 8, 35033 Marburg, Germany; E-mail: weihe@staff.uni-marburg.de

This study was supported by the Volkswagen Foundation to L.E.E. and E.W.

Copyright Lippincott Williams & Wilkins Oct 2007

(c) 2007 Journal of Neuropathology and Experimental Neurology. Provided by ProQuest Information and Learning. All rights Reserved.