Subcellular localization and integration activities of rous sarcoma virus reverse transcriptase

Abstract

Reverse transcriptases (RTs) alphabeta and beta from avian Rous sarcoma virus (RSV) harbor an integrase domain which is absent in nonavian retroviral RTs. RSV integrase contains a nuclear localization signal which enables the enzyme to enter the nucleus of the cell in order to perform integration of the proviral DNA into the host genome. In the present study we analyzed the subcellular localization of RSV RT, since previous results indicated that RSV finishes synthesis of the proviral DNA in the nucleus. Our results demonstrate that the heterodimeric RSV RT alphabeta and the beta subunit, when expressed independently, can be detected in the nucleus, whereas the separate alpha subunit lacking the integrase domain is prevalent in the cytoplasm. These data suggest an involvement of RSV RT in the transport of the preintegration complex into the nucleus. In addition, to analyze whether the integrase domain, located at the carboxyl terminus of beta, exhibits integration activities, we investigated the nicking and joining activities of heterodimeric RSV RT alphabeta with an oligodeoxynucleotide-based assay system and with a donor substrate containing the supF gene flanked by the viral long terminal repeats. Our data show that RSV RT alphabeta is able to perform the integration reaction in vitro; however, it does so with an estimated 30-fold lower efficiency than the free RSV integrase, indicating that RSV RT is not involved in integration in vivo. Integration with RSV RT alphabeta could be stimulated in the presence of human immunodeficiency virus type 1 nucleocapsid protein or HMG-I(Y).

FIG. 1.

Schematic representation of Pol products.

FIG. 2.

Subcellular localization of RSV RTs and of RSV IN by confocal laser scanning microscopy. Representative confocal micrographs of transiently transfected NIH 3T3 fibroblasts expressing GFP or RFP fusion proteins are shown. GFP fusion proteins were expressed with RSV IN (RSV IN-GFP) or the β subunit of RSV RT (RSV RT β-GFP). The α subunit of RSV RT was expressed as a fusion to RFP (RSV RT α-RFP). Localization of the heterodimer (RSV RT α-RFP/β) was detected by coexpression of RSV RT α-RFP and β.

FIG. 3.

(A) Integration activity of RSV RT αβ and RSV IN in the absence or presence of HIV-1 NC protein. A double-stranded 36-mer ODN substrate (100 nM) corresponding to the 3′ LTR of the viral DNA was used. The ODN harboring the 3′-terminal CATT sequence was 5′ ³²P end labeled. The concentration of the enzymes was 75 nM; HIV-1 NC was added at a concentration of 9.9 μM. Samples were incubated for 90 min at 37°C. The left panel of the figure shows a short exposure of an autoradiograph of the lower part of the gel to visualize the 3′OH processing activities of IN and RSV RT αβ. The right panel shows a long exposure of an autoradiograph of the same gel to visualize the integration products in the upper region. (B) Determination of the lowest IN concentration exhibiting integration activities. The ODN-based IN assay was performed in a 10-μl assay by using decreasing amounts of IN in order to determine the lowest IN concentration still exhibiting activity. Lane 1, 0.75 pmol of IN; lane 2, 0.5 pmol of IN; lane 3, 0.25 pmol of IN; lane 4, 0.05 pmol of IN; lane 5, 0.025 pmol of IN; lane 6, 0.005 pmol of IN; lane 7, 0.0025 pmol of IN; lane 8, 0.001 pmol of IN; lane αβ, 0.75 pmol of RT αβ. (C) Western blot with RT αβ and with different concentrations of IN to detect a possible contamination with free IN in our RT preparation. Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then blotted onto a nitrocellulose membrane. The concentration of RSV RT αβ was 48 pmol. Detection of free IN was achieved with an antiserum directed against the C terminus of RSV IN (2). The second antibody was an anti-rabbit immunoglobulin G antibody conjugated with alkaline phosphatase.

FIG. 3.

(A) Integration activity of RSV RT αβ and RSV IN in the absence or presence of HIV-1 NC protein. A double-stranded 36-mer ODN substrate (100 nM) corresponding to the 3′ LTR of the viral DNA was used. The ODN harboring the 3′-terminal CATT sequence was 5′ ³²P end labeled. The concentration of the enzymes was 75 nM; HIV-1 NC was added at a concentration of 9.9 μM. Samples were incubated for 90 min at 37°C. The left panel of the figure shows a short exposure of an autoradiograph of the lower part of the gel to visualize the 3′OH processing activities of IN and RSV RT αβ. The right panel shows a long exposure of an autoradiograph of the same gel to visualize the integration products in the upper region. (B) Determination of the lowest IN concentration exhibiting integration activities. The ODN-based IN assay was performed in a 10-μl assay by using decreasing amounts of IN in order to determine the lowest IN concentration still exhibiting activity. Lane 1, 0.75 pmol of IN; lane 2, 0.5 pmol of IN; lane 3, 0.25 pmol of IN; lane 4, 0.05 pmol of IN; lane 5, 0.025 pmol of IN; lane 6, 0.005 pmol of IN; lane 7, 0.0025 pmol of IN; lane 8, 0.001 pmol of IN; lane αβ, 0.75 pmol of RT αβ. (C) Western blot with RT αβ and with different concentrations of IN to detect a possible contamination with free IN in our RT preparation. Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then blotted onto a nitrocellulose membrane. The concentration of RSV RT αβ was 48 pmol. Detection of free IN was achieved with an antiserum directed against the C terminus of RSV IN (2). The second antibody was an anti-rabbit immunoglobulin G antibody conjugated with alkaline phosphatase.

FIG. 3.

(A) Integration activity of RSV RT αβ and RSV IN in the absence or presence of HIV-1 NC protein. A double-stranded 36-mer ODN substrate (100 nM) corresponding to the 3′ LTR of the viral DNA was used. The ODN harboring the 3′-terminal CATT sequence was 5′ ³²P end labeled. The concentration of the enzymes was 75 nM; HIV-1 NC was added at a concentration of 9.9 μM. Samples were incubated for 90 min at 37°C. The left panel of the figure shows a short exposure of an autoradiograph of the lower part of the gel to visualize the 3′OH processing activities of IN and RSV RT αβ. The right panel shows a long exposure of an autoradiograph of the same gel to visualize the integration products in the upper region. (B) Determination of the lowest IN concentration exhibiting integration activities. The ODN-based IN assay was performed in a 10-μl assay by using decreasing amounts of IN in order to determine the lowest IN concentration still exhibiting activity. Lane 1, 0.75 pmol of IN; lane 2, 0.5 pmol of IN; lane 3, 0.25 pmol of IN; lane 4, 0.05 pmol of IN; lane 5, 0.025 pmol of IN; lane 6, 0.005 pmol of IN; lane 7, 0.0025 pmol of IN; lane 8, 0.001 pmol of IN; lane αβ, 0.75 pmol of RT αβ. (C) Western blot with RT αβ and with different concentrations of IN to detect a possible contamination with free IN in our RT preparation. Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then blotted onto a nitrocellulose membrane. The concentration of RSV RT αβ was 48 pmol. Detection of free IN was achieved with an antiserum directed against the C terminus of RSV IN (2). The second antibody was an anti-rabbit immunoglobulin G antibody conjugated with alkaline phosphatase.

FIG. 4.

Integration reaction with a supF containing substrate to investigate coupled joining. In vitro integration products with IN and RT αβ are shown in the absence or presence of HMG-I(Y) protein. The upper band (RF II) of the integrated products represents a mixture of plasmids, with one molecule of donor integrated either in a concerted (both ends) or nonconcerted (one end or two ends integrated independently) way. The lower band (RF III) of products represents plasmids where two donor molecules are integrated in a concerted way, thus leading to a linear product (29).