Lack of endogenous TRIM5α-mediated restriction in rhesus macaque dendritic cells

Primate cells can resist certain retroviral infections by expressing a dominant restriction factor named TRIM5α.^1,2  This member of the tripartite motif (TRIM) protein family interferes specifically with the replication of various retroviruses in a species-specific manner. In Old World monkey (such as rhesus macaque) cells, TRIM5α is responsible for the postentry block of human immunodeficiency virus type 1 (HIV-1). This endogenous restriction, originally termed Lv1, was observed in several rhesus macaque cell lines, including LL-Cmk2, FRhK4, and FrHL-2 cells, or in primary cells, such as primary lung fibroblasts.^3,,,–7  Thus, although no systematic study has been conducted so far, endogenous TRIM5α-mediated HIV-1 restriction is thought to occur identically in all rhesus macaque cells

In this work, we set out to compare TRIM5α-mediated restriction of HIV-1 infection in primary cell types that are important in vivo replication sites for HIV-1, namely, T lymphocytes, dendritic cells (DCs), and macrophages. Results revealed that, unlike primary lymphocytes, macaque DCs do not mediate TRIM5α restriction against HIV-1. We propose that the role of DCs as professional antigen-presenting cells (APCs) and their importance in stimulating host adaptive immunity preclude them from maintaining cellular restriction.

Citrate human blood was obtained from healthy donors (Etablissement Français du Sang, Hôpital Necker, Paris, France) and heparin blood from rhesus macaques (Macaca mulatta) of Chinese origin. Animals were demonstrated to be seronegative for STLV-1 (Simian T Leukemia Virus type-1), SRV-1 (type D retrovirus), herpes-B viruses, and SIVmac. Animals were housed and cared for in compliance with existing French regulations, and ethical approval for their use was granted by the Institut Pasteur. Monocytes and peripheral blood lymphocytes (PBLs) were separated using adherence on plastic (45 minutes) after Ficoll gradient. PBLs were activated for 3 days with either phytohemagglutinin or concanavalin A, 10 and 5 μg/mL, respectively, unless otherwise stated, and then transduced or infected. Monocytes were differentiated to immature DCs by incubation for 7 days with 50 ng/mL of recombinant human granulocyte-macrophage colony-stimulating factor and 20 ng/mL of interleukin-4 (R&D Systems, Minneapolis, MN). Alternatively, monocytes were differentiated to macrophages by incubation for 7 days with 50 ng/mL of human recombinant granulocyte-macrophage colony-stimulating factor alone. Phenotypic characterization before infection was carried out by flow cytometry (BD Biosciences, San Jose, CA) using antibodies (from BD Biosciences) against CD3 (SP34), CD4 (SK3), CD20 (L27), HLA-DR (L243), CD14 (MφP9), CD1a (SK9), CD80 (L307.4), and CD86 (FUN-1). DCs were CD14⁻CD1a⁺CD80^loCD86^lo, and macrophages CD14⁺CD1a^lo.

Total RNA was extracted from human or rhesus macaque CD4⁺ T lymphocytes or DCs using RNeasy Mini Kit (QIAGEN, Valencia, CA) and cDNA was prepared using SuperScript III Reverse Transcriptase (Invitrogen, Carlsbad, CA). Quantitative polymerase chain reaction (PCR) was performed in duplicate on 1 μL of cDNA on a Roche LightCycler, using the LightCycler Fast Start DNA Master SYBR Green 1 kit (Roche Diagnostics, Indianapolis, IN). The TRIM5α primer pair was designed to amplify both human and rhesus macaque TRIM5α cDNA: TTGGATCCTGGGGGTATGTGCTGG (forward) and TGATATTGAAGAATGAGACAGTGCAAG (reverse). TRIM5γ primers were: CATTATCATCAGCCACCCTGTGG (forward) andGGAGAATCATAAATCTTAAAACACGAG (reverse). For TRIM5 quantifications, a plasmid encoding either TRIM5α or TRIM5γ was used for setting standard values. Cyclophilin A (CypA) was used as an internal normalizing standard with the primer pair AGTGGTTGGATGGCAAGC (forward) and GATTCTAGGATACTGCGAGCAAA (reverse).

DCs are the focus of much interest in the field of vaccination, where their role as professional APCs can be harnessed to elicit a strong and long-lasting cellular immune response. Our laboratory is implicated in a vaccination trial against SIV infection in macaque monkeys using lentiviral vectors. Unexpected in light of the described TRIM5α-mediated restriction in macaque cells⁶  was our in vivo observation that we obtain in macaques a potent immune response to antigens expressed on a HIV backbone (A. Beignon et al, manuscript submitted). Based on these results, we set out to analyze in greater depth this apparently attenuated cellular restriction in macaques comparing T cells and DCs, which are the main professional APCs.

Activated PBLs and monocyte-derived DCs from macaque and human blood were transduced with an HIV-1–derived vector expressing the green fluorescent protein (GFP) under the control of the cytomegalovirus (CMV) promoter (ΔU3 CMV GFP) by incubation with vector supernatant from producer 293T cells followed by complementation with culture medium after 2 hours. Percentages of transduced cells were determined by flow cytometry 4 days after transduction. As expected, transduction of human PBLs was efficient (Figure 1A,B), giving rise to 40% to 60% transduced cells with 120 ng p24/100 μL, whereas macaque PBLs remained very poorly transduced (< 3 % even at 1200 ng p24/100 μL, Figure 1A; and data not shown) despite equivalent T-cell proliferation and activation status (Figure 1C). These data confirmed that there is a strong restriction of HIV-1 infection in macaque PBLs. This restriction is related to rhTRIM5α protein expression because nucleofection of 3 μg of an siRNA against TRIM5α (GCCUUACGAAGUCUGAAACdTdT) in macaque primary PBLs led to approximately 6-fold increase in infection with an envelope-defective VSV-G pseudotyped HIV virus ([VSV-G] NL43 GFP Δenv), whereas little or no effect was observed in human PBLs (Figure 1D).

The same experimental procedure was carried out in monocyte-derived immature DCs. Contrary to PBLs, both human and macaque DCs were efficiently transduced with TRIP ΔU3 CMV GFP, with up to 30% transduced cells with 120 ng p24/100 μL for both (Figure 1E-G). Equivalent transduction efficiencies were observed for human and macaque DCs, even at low multiplicity of infection, indicating that the observed transduction is not the result of a saturation of endogenous restriction in these cells. These experiments were carried out using blood from 3 different human donors and 5 different macaques to take into account possible interindividual sample variations. Macaque DC permissivity to HIV-1 was observed both with HIV-1–derived vectors and wild-type HIV-1 viruses, regardless of the method used to isolate and differentiate precursor monocytes into DCs (CD14⁺ magnetic bead-positive selection, macaque or human cytokines). Macaque DC permissivity to HIV-1 was also observed using blood from Indian macaques rather than Chinese (Figure 1H), indicating that this phenomenon is true regardless of the geographic origin of the rhesus monkeys.

To determine whether lack of TRIM5α-mediated restriction in macaque DC is a specific feature of professional APCs such as DCs, we undertook to assess the ability of HIV-1 to infect macaque macrophages. Human and macaque monocyte-derived macrophages were transduced with TRIP ΔU3 CMV GFP, and the percentage of transduced cells was assessed by flow cytometry and fluorescence microscopy at 7 days after transduction. Results revealed a strong restriction in macaque macrophages compared with human macrophages, particularly at low vector doses (Figure 1I). The observed restriction in macaque macrophages may be more easily saturable than that of macaque PBLs because the highest dose (120 ng/100 μL) could achieve 20% to 25% transduction in macrophages, and never more than 5% in PBLs.

We next asked whether TRIM5α levels are lower in macaque DCs compared with PBLs. Because no completely specific TRIM5α antibody is available, we set out to measure TRIM5α mRNA levels using reverse transcription PCR (RT-PCR). Results showed equivalent levels in both PBLs and DCs from humans and macaques, indicating that TRIM5α is not transcriptionally silenced in macaque DCs (Figure 2A). We next set out to determine whether TRIM5γ, previously described as a transdominant negative of TRIM5α,⁶  might be overexpressed in macaque DCs. Whereas macaque DCs and PBLs express comparable levels of TRIM5α mRNA, DCs were found to express a relatively lower amount of TRIM5γ mRNA, leading to a higher TRIM5α/TRIM5γ mRNA ratio compared with PBLs (Figure 2B). This indicates that, if anything, TRIM5γ is marginally lower in macaque DCs and therefore that TRIM5α is not competed out by TRIM5γ overexpression in these cells. Nonetheless, overexpression of the rhTRIM5α protein in rhesus macaque DCs led to a decrease in HIV-1 infection (Figure 2C), indicating that, although TRIM5α mRNA levels are normal, endogenous proteins are not fully functional in macaque DCs. TRIM5α mRNAs may be inefficiently translated in macaque DCs, or translated proteins may undergo posttranslational modifications that affect their function or half-life or again be routed to an inappropriate subcellular compartment.

To our knowledge, this is the first report of the absence of TRIM5α-mediated restriction in macaque DCs. This absence of restriction consolidates the use of macaque monkeys as vaccination models using HIV-1–derived vectors to deliver antigens to DCs. The uniqueness of macaque DCs in terms of HIV-1 permissivity suggests the necessity for macaque monkeys to not compromise the antigen presentation capacity of DCs by maintaining restriction against a pathogen in these cells. Although macrophages are also APCs, they are, first and foremost, the main target and producer cells for HIV-1 infection in vivo, which would account for the maintenance of restriction in macaque macrophages. Moreover, in contrast to DCs, which are essential for the initiation of primary immune responses, macrophages are less efficient for priming naive T cells and more adapted for inducing recall antigenic responses. We speculate that the pivotal role of DCs in adaptive immunity precludes the presence of any form of cellular restriction in DCs. Infection must be tolerated to ensure endogenous presentation of neo-synthesized antigens. One can conceive how evolution would stipulate that innate immunity, of which TRIM5α proteins are part, should not hamper adaptive immunity.

This work was supported by the Agence Nationale de la Recherche sur le SIDA, the Institut Pasteur, and SIDACTION (all Paris, France).

Contribution: N.J.A., S.N., and J.E. wrote the paper and performed and designed research; L.C., F.C., P.S., and A.B. performed research; and P.C. wrote the paper, designed research, and established the concept of the study.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Pierre Charneau, Molecular Virology and Vectorology, Institut Pasteur, 25-28 rue du Dr Roux, 75015, Paris, France; e-mail: charneau@pasteur.fr; or Nathalie J. Arhel, Molecular Virology and Vectorology, Institute of Virology, University Clinic of Ulm, Albert-Einstein-Allee 11, 89081 Ulm, Germany; e-mail: Nathalie.arhel@uniklinik-ulm.de.