A doxycycline-dependent human immunodeficiency virus type 1 replicates in vivo without inducing CD4+ T-cell depletion

Abstract

A novel genetic approach for the control of virus replication was used for the design of a conditionally replicating human immunodeficiency virus (HIV) variant, HIV-rtTA. HIV-rtTA gene expression and virus replication are strictly dependent on the presence of a non-toxic effector molecule, doxycycline (dox), and thus can be turned on and off at will in a graded and reversible manner. The in vivo replication capacity, pathogenicity and genetic stability of this HIV-rtTA variant were evaluated in a humanized mouse model of haematopoiesis that harbours lymphoid and myeloid components of the human immune system (HIS). Infection of dox-fed BALB Rag/γc HIS (BRG-HIS) mice with HIV-rtTA led to the establishment of a productive infection without CD4(+) T-cell depletion. The virus did not show any sign of escape from dox control for up to 10 weeks after the onset of infection. No reversion towards a functional Tat-transactivating responsive (TAR) RNA element axis was observed, confirming the genetic stability of the HIV-rtTA variant in vivo. These results demonstrate the proof of concept that HIV-rtTA replicates efficiently in vivo. HIV-rtTA is a promising tool for fundamental research to study virus-host interactions in vivo in a controlled fashion.

Fig. 1.

Schematic of the HIV-rtTA genome compared with the parental HIV-1 LAI virus and HIV-1 LAI-ΔNef variant. (a) Schematic of the HIV-1 LAI and HIV-1 LAI-ΔNef proviral DNA genomes. The out-of-frame, 106 bp deletion in the nef gene of the HIV-1 LAI-ΔNef variant is indicated. (b) Schematic of the HIV-rtTA proviral DNA genome. In HIV-rtTA, the Tat–TAR axis of transcription regulation has been inactivated by mutations in both Tat (Y26A) and TAR (five point mutations). Transcription and replication of the virus was made dox dependent by the introduction of tetO elements in the U3 promoter region and by replacing the nef gene with the rtTA gene.

Fig. 2.

HIV-rtTA does not induce CD4⁺ T-cell depletion in blood, despite active replication in the presence of dox. (a) BRG-HIS mice were bled before and regularly after HIV-1 infection. Virus replication was determined by quantification of viral RNA in the plasma of the animals and is expressed as means±sd. Significant differences in plasma RNA viral load between HIV-1 LAI- and HIV-rtTA-infected BRG-HIS mice are indicated (*P<0.05; **P<0.01). (b–e) PBMCs were stained for human T-cell markers and analysed by flow cytometry. The frequency of human CD4⁺ T-cells (among CD3⁺ T-cells) was determined for mice infected with HIV-1 LAI (b), HIV-1 LAI-ΔNef (c), HIV-rtTA (d) and the mock-infected BRG-HIS mice (e). Each symbol corresponds to an individual animal. The results were pooled from two independent experiments.

Fig. 3.

Viral capsid protein is expressed in human T-cells from the spleen of HIV-rtTA-infected BRG-HIS mice. Splenocytes of mock- and HIV-rtTA-infected BRG-HIS mice were isolated, stained and analysed by fluorescence-activated cell sorting (FACS). (a) The flow cytometry dot plots show the percentage of human T-cells expressing the viral CA-p24 protein. The values in the upper right quadrant indicate the percentage of human CA-p24⁺ CD4⁺ T-cells. (b) The percentage of human CA-p24⁺ T-cells from the spleen of mock- and HIV-rtTA-infected BRG-HIS mice are presented at 3 weeks p.i. (upper graph) and 10 weeks p.i. (lower graph). Each dot represents one animal. Horizontal bars represent the mean values. *Significantly different at P<0.05.

Fig. 4.

HIV-rtTA-infected human splenocytes isolated from HIV-rtTA-infected BRG-HIS mice can propagate and productively infect new human PBMCs in vitro in a dox-dependent manner. (a, b) Splenocytes of mock-, HIV-1 LAI- and HIV-rtTA-infected BRG-HIS mice obtained at 3 weeks p.i. (a) and 10 weeks p.i. (b) were co-cultured with activated human PBMCs in the presence of 1000 ng ml⁻¹ dox. (c) Splenocytes of HIV-rtTA-infected BRG-HIS mice obtained at 10 weeks p.i. were co-cultured with activated human PBMCs with or without 1000 ng dox ml⁻¹. Virus replication was monitored during the 14 days of co-culture by determination of the CA-p24 levels in the co-culture supernatants.

Fig. 5.

HIV-rtTA infection does not trigger the production of HIV-1-specific antibodies. Plasma from mock-, HIV-1 LAI- and HIV-rtTA-infected BRG-HIS mice at 10 weeks p.i. was tested for HIV-1-specific antibodies by ELISA. The negative (−) and positive (+) bars correspond to negative and positive human serum controls provided in the Genscreen HIV-1/HIV-2 kit, respectively. Results are presented as means±sd.