Cyclophilin A is required for TRIM5{alpha}-mediated resistance to HIV-1 in Old World monkey cells

Abstract

The peptidyl-prolyl isomerase cyclophilin A (CypA) embraces an exposed, proline-rich loop on HIV-1 capsid (CA) and renders reverse transcription complexes resistant to an antiviral activity in human cells. A CypA fusion with TRIM5 that is unique to New World owl monkeys also targets HIV-1 CA, but this interaction potently inhibits infection. A similar block to HIV-1 infection in Old World monkeys is attributable to the alpha isoform of the TRIM5 orthologue in these species. To determine whether HIV-1 restriction by Old World monkey TRIM5alpha is modulated by the CA-CypA interaction, RNA interference was used to disrupt CypA in cells from African green monkeys and rhesus macaques. HIV-1 infectivity increased in response to CypA knock-down to the same extent that it increased in response to TRIM5 knock-down. CypA knock-down eliminated the HIV-1 stimulatory effect of cyclosporin A (CsA), a competitive inhibitor of the CypA-CA interaction, or of CA mutants that block binding to CypA but caused no change in titer of retroviruses that don't interact with CypA. Simultaneous knock-down of both CypA and TRIM5 caused minimal additional increase in titer, suggesting that CypA inhibits HIV-1 replication in these cells because it is required for CA recognition by TRIM5alpha. Finally, CsA increased HIV-1 titer in otherwise nonrestrictive feline cells but only after these cells were transduced with Old World monkey TRIM5alpha. Thus, CypA is required for HIV-1 restriction by Old World monkey orthologues of TRIM5alpha.

Fig. 1.

Reduction of CypA protein counteracts HIV-1 restriction in AGM cells. (A) Vero cells were transduced with a retroviral vector expressing shRNA targeting CypA or luciferase as a control and assessed for CypA and β-actin by Western blot. (B and C) Cells were then challenged with VSV G-pseudotyped HIV-1_NL-GFP or SIV_MAC239 (B) or with N-tropic or B-tropic MLV-derived vectors expressing GFP (C); the percentage of infected cells was determined by flow cytometry and plotted against the relative amount of virus used (in reverse transcriptase units).

Fig. 2.

CypA binding to CA is critical to HIV-1 restriction in AGM cells. (A) Vero cells expressing shRNA targeting luciferase or CypA were challenged with HIV-1_NL-GFP in the presence or absence of 5 μM CsA. (B) Infections with wild-type HIV-1_NL-GFP or with HIV-1_NL-GFP bearing a CA mutation (G89V) that disrupts interaction with CypA. The percentage of infected cells was determined by flow cytometry and plotted against the relative amount of virus used (in reverse transcriptase units).

Fig. 3.

CypA protein levels determine the degree of HIV-1 restriction in rhesus macaque cells. FRhK4 cells expressing shRNA specific for CypA or luciferase were transduced with a retroviral vector expressing nontargetable CypA cDNA or with the empty vector. Cells were assessed for CypA and β-actin by Western blot (A) and challenged with HIV-1_NL-GFP in the presence or absence of CsA at 5 μM (B).

Fig. 4.

TRIM5α and CypA act together to restrict HIV-1 in rhesus macaque cells. FRhK4 cells expressing shRNA specific for CypA or luciferase were transduced a second time with vector expressing shRNA specific for TRIM5α or with the empty control vector. The resulting cell lines were challenged with HIV-1_NL-GFP (A) or EIAV_GFP (B), and the infectivity was assessed as in Fig. 1.

Fig. 5.

CsA stimulates HIV-1 infection of feline cells that express TRIM5α from Old World monkeys. HIV-1-permissive CRFK cells were transduced with TRIM5α cDNA cloned from rhesus macaque (FRhK4) or AGM (CV1) cells or with empty vector. Cells were then challenged with HIV-1_NL-GFP or SIV_MAC239-GFP in the presence of the indicated concentrations of CsA. The quantity of each virus used was such that ≈1% of the cells would be transduced in the absence of CsA. Data are plotted as the fold increase in percentage of infected cells compared with the no-drug control.