Vpu-Mediated Counteraction of Tetherin Is a Major Determinant of HIV-1 Interferon Resistance

Abstract

Human immunodeficiency virus type 1 (HIV-1) groups M, N, O, and P are the result of independent zoonotic transmissions of simian immunodeficiency viruses (SIVs) infecting great apes in Africa. Among these, only Vpu proteins of pandemic HIV-1 group M strains evolved potent activity against the restriction factor tetherin, which inhibits virus release from infected cells. Thus, effective Vpu-mediated tetherin antagonism may have been a prerequisite for the global spread of HIV-1. To determine whether this particular function enhances primary HIV-1 replication and interferon resistance, we introduced mutations into the vpu genes of HIV-1 group M and N strains to specifically disrupt their ability to antagonize tetherin, but not other Vpu functions, such as degradation of CD4, down-modulation of CD1d and NTB-A, and suppression of NF-κB activity. Lack of particular human-specific adaptations reduced the ability of HIV-1 group M Vpu proteins to enhance virus production and release from primary CD4(+) T cells at high levels of type I interferon (IFN) from about 5-fold to 2-fold. Interestingly, transmitted founder HIV-1 strains exhibited higher virion release capacity than chronic control HIV-1 strains irrespective of Vpu function, and group M viruses produced higher levels of cell-free virions than an N group HIV-1 strain. Thus, efficient virus release from infected cells seems to play an important role in the spread of HIV-1 in the human population and requires a fully functional Vpu protein that counteracts human tetherin.

Importance: Understanding which human-specific adaptations allowed HIV-1 to cause the AIDS pandemic is of great importance. One feature that distinguishes pandemic HIV-1 group M strains from nonpandemic or rare group O, N, and P viruses is the acquisition of mutations in the accessory Vpu protein that confer potent activity against human tetherin. Adaptation was required because human tetherin has a deletion that renders it resistant to the Nef protein used by the SIV precursor of HIV-1 to antagonize this antiviral factor. It has been suggested that these adaptations in Vpu were critical for the effective spread of HIV-1 M strains, but direct evidence has been lacking. Here, we show that these changes in Vpu significantly enhance virus replication and release in human CD4(+) T cells, particularly in the presence of IFN, thus supporting an important role in the spread of pandemic HIV-1.

FIG 1

Mutant Vpus selectively defective in tetherin antagonism. (A) Alignment of Vpu amino acid sequences analyzed. The NL4-3 Vpu sequence is shown in the top row for comparison. Important functional domains are indicated above the sequences, and the mutated Ala residues are highlighted in yellow. Dots specify amino acid identity, and dashes represent gaps introduced to optimize the alignment. bdg, binding; P, phosphate. (B) Down-modulation of human tetherin by wild-type (wt) and mutant Vpu proteins in HEK293T cells cotransfected with vectors coexpressing eGFP and Vpu and a construct expressing human tetherin. Shown are the levels of tetherin cell surface expression relative to those measured in cells transfected with the control vector containing only eGFP (100%). Values are mean values (plus standard errors of the means [SEM] [error bars]) derived from three experiments. Wild-type Vpu alleles are color coded in dark colors, and mutant Vpus are shown in light colors. Values that are significantly different are indicated by asterisks as follows: *, P < 0.05; **, P < 0.01; ***, P < 0.001. (C) Virus release from HEK293T cells following transfection with vpu-defective HIV-1 NL4-3, expression constructs for the indicated Vpu proteins or eGFP only, and various amounts of plasmid expressing human tetherin. Infectious virus was determined by infection of TZM-bl indicator cells and is shown as a percentage of that detected in the absence of tetherin (100%). Infections were performed in triplicate, and the results were confirmed in an independent experiment.

FIG 2

TMD mutations in Vpu disrupt tetherin down-modulation in HIV-1-infected primary T cells. PHA-activated PBMCs were infected with HIV-1 constructs containing wt, TMD mutated, or grossly defective vpu alleles and examined for tetherin surface expression 3 days later. (A) Examples of primary data. The numbers in the graphs give the mean fluorescence intensity (MFI) of tetherin expression in the HIV-1-infected (p24-positive) cell population. CTRL, control. (B) Levels of tetherin surface expression in cells infected with the wt and Vpu mutant constructs relative to those infected with the vpu-defective HIV-1 constructs (100%). Each symbol represents the result obtained for one individual PBMC donor investigated. Values that are significantly different are indicated by asterisks as follows: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

FIG 3

Effects of alterations in vpu on HIV-1 yield and release in CD4⁺ T cells in the presence or absence of IFN-α. (A) Cell-free (CF) p24 antigen levels in the supernatant of CD4⁺ T cells at day 7 postinfection with HIV-1 IMCs expressing wt (+), Tmut (m), or no (−) Vpu proteins. Virus yield was determined after triplicate HIV-1 infection in the presence of 500 U/ml IFN-α (+) and absence of IFN-α (−). (B) Reduction of cell-free p24 antigen yield by IFN-α treatment. For calculation of n-fold reduction, the levels of p24 antigen obtained in the absence of IFN were divided by those obtained in the presence of IFN-α. (C) Enhancement of p24 release by wt and Tmut Vpu proteins in the presence (shaded) or absence of exogenous IFN-α. Data were derived from the experiment shown in panel A. The levels of cell-free p24 antigen relative to the cultures infected with the respective vpu-defective HIV-1 IMCs (100%, indicated by the dashed line) are shown. (D) Cell-free, cell-associated (CA), and total p24 yield in CD4⁺ T cells infected with HIV-1 NL4-3, CH058-TF, CH077-TF, and CH167 IMCs containing wt, mutant, or grossly defective vpu genes. The average values obtained for the respective wt IMCs were set at 100%. (E) Efficiency of p24 release in CD4⁺ T cells infected with the indicated HIV-1 IMCs. Values present percentages of cell-free p24 antigen out of the total p24 detected in the presence (shaded) and absence of IFN-α. Cell-free and cell-associated p24 antigen were quantified by an enzyme-linked immunosorbent assay (ELISA) at day 7 postinfection. (F) Effect of TMD mutations in Vpu or entire lack of Vpu function on the efficiency of virion release. Values obtained for all four IMCs analyzed are shown relative to the respective wt viruses (100%). Values that are significantly different are indicated by asterisks as follows: *, P < 0.05; **, P < 0.01; ***, P < 0.001.

FIG 4

Replication of wt and vpu mutant HIV-1 constructs in CD4⁺ T cells in the presence (+) and absence (−) of IFN-α. (A) Replication kinetics of HIV-1 IMCs expressing wt, TMD mutant, or no Vpu proteins in CD4⁺ T cells in the presence of 500 U/ml IFN-α (blue lines) or absence of IFN-α (black lines). Results show median values of p24 antigen production (n = 3) from two different donors. (B) Cumulative p24 antigen levels in the presence and absence of IFN-α measured at 1, 3, 5, 7, and 9 days postinfection. Panels B, C, and D show the results obtained from two different blood donors. (C) Reduction of cumulative cell-free p24 antigen yield by IFN-α treatment. (D) Enhancement of cumulative p24 yield by wt and Tmut Vpu proteins in the presence (shaded) or absence of exogenous IFN-α. Data were derived from the experiment shown in panel A. Values present total cell-free virus yield relative to the respective vpu-defective HIV-1 IMC (100%). (E) Ranking of wt and vpu mutant or defective HIV-1 IMCs according to their efficiency in cell-free p24 production. The levels achieved for the most potent IMC were set at 100%. Values are median values of p24 antigen production (plus SEM [error bars]; n = 3).

FIG 5

Release of wt and vpu mutant HIV-1 constructs in CD4⁺ T cells in the presence and absence of IFN-α. (A) Values present percentages of cell-free p24 antigen out of the total p24 detected in the presence (shaded) and absence of IFN-α. Results from triplicate infections of T cells derived from three PBMC donors are shown. Cell-free and cell-associated p24 antigen was quantified by ELISA at day 5 postinfection. (B) Efficiency of TF and CC virus release in CD4⁺ T cells infected with the indicated HIV-1 IMCs. Values present percentages of cell-free p24 antigen out of the total p24 detected in the presence (shaded) and absence of IFN-α. (C) Ranking of wt and vpu mutant or defective HIV-1 IMCs according to their release efficiency. The levels achieved by the most potent IMCs were set at 100%. Values are median values of release efficacy (plus SEM [error bars]; n = 3). (D) Correlation between the virus release efficiencies measured in the presence and absence (without [wo]) of IFN-α. (E and F) Correlation between the virus release efficiencies (values derived from panel C) and p24 antigen yield (values derived from Fig. 4E) in the absence (E) and presence (F) of IFN-α.