Expression of naturally occurring antisense RNA inhibits ... - CiteSeerX

Journal of General Virology (1997), 78, 2503–2511. Printed in Great Britain ...........................................................................................................................................................................................................................................................................................

Expression of naturally occurring antisense RNA inhibits human immunodeficiency virus type 1 heterologous strain replication Nailya E. Tagieva and Catherine Vaquero† Institut Cochin de Ge! ne! tique Mole! culaire, 22 rue Me! chain, Paris 75014, France

Recently, the presence of human immunodeficiency virus type 1 (HIV-1) RNA transcripts with negativestrand polarity has been shown in tissue culture models of acute and persistently infected cells. One of these transcripts encodes a 189 amino acid open reading frame. This highly conserved antisense sequence is complementary to the structured Rev-responsive element and extends through the cleavage site of the Env protein. We tested the ability of this antisense RNA to modulate HIV-1 replication and the mRNA profile when expressed stably or transiently in several cell types. Different cell lines and PBLs were transduced by retroviral

Introduction In developing approaches to anti-human immunodeficiency virus (HIV) gene therapy, the inhibitory properties of antisense RNA can be exploited to bring about selective inhibition of viral gene expression. Inhibition of specific genes with stably integrated antisense DNA templates (also called antisense genes) was first demonstrated in bacteria by Pestka et al. (1984) and Coleman et al. (1984) and in eukaryotic cells by Izant & Weintraub (1984). In these cases, the cells are genetically altered to continuously express antisense RNA, which leads to the inhibition of the target gene in the transduced cells. Genetic alteration of cells mediated via gene transfer, leading to the inhibition of virus replication, has been termed intracellular immunization (Baltimore, 1988), and the use of antisense RNA represents one approach to achieve such inhibition. Retroviral vectors are widely used for introduction of transgenes, gene therapeutic antiviral agents and expression of antisense or sense RNA in different cells. The ability of Author for correspondence : Catherine Vaquero. Fax 33 1 45 83 88 58. e-mail vaquero!icgm.cochin.inserm.fr † Present address : INSERM U313, 91 Bd de l’Ho# pital, Paris 75013, France.

0001-4759 # 1997 SGM

vectors producing antisense RNA and were then challenged by HIV infection. We have shown that the endogenously expressed antisense RNA containing the natural open reading frame inhibits HIV-1IIIB and HIV-1NDK replication in these cells. The level of inhibition varied according to the cells, but was significant in all cases. The production of HIV-1 (BRU, IIIB, NDK) mRNAs was also significantly decreased. HIV-2 replication was not inhibited by expression of the antisense RNA. Our results also suggest that this inhibitory effect is due to the antisense RNA and not to the protein which is encoded by this sequence.

retroviral vectors to integrate into the genome of target cells renders them effective gene transfer vehicles. Retrovirus vectors are useful reagents for the efficient transfer of functional HIV proteins and the establishment of cell lines constitutively expressing these genes (Eglitis & Anderson, 1988 ; Palmer et al., 1989 ; Morgenstern & Land, 1990 ; Purcell et al., 1996). The efficacy of anti-HIV gene therapy via retroviral vectors has been tested in different T-cell lines and in primary CD4+ T-lymphocytes (Rhodes & James, 1990, 1991 ; Sczakiel et al., 1992 ; Garcia & Miller, 1994 ; Chuah et al., 1994 ; Duan et al., 1994 ; Vandendriessche et al., 1995 ; Sun et al., 1995 ; Lisziewicz et al., 1995 ; Kim et al., 1996). For our study we have chosen the sequence of the HIV-1BRU open reading frame (ORF) on the plus strand of the genomic DNA, which was first described by Miller (1988). This HIV-1BRU ORF is capable of encoding a protein (ASP, in vitro, and antibodies against ASP were detected in sera from HIV-1+ individuals (Vanhe! e et al., 1995). HIV-1 antisense RNA transcripts have been detected in cell lines acutely and chronically infected with HIV-1 (Vanhe! e et al., 1995) and in peripheral blood mononuclear cell samples using RT–PCR (Michael et al., 1994). One of these transcripts encodes a previously described ORF. This ORF is located across the gp120–gp41 boundary and encodes a 189 amino acid protein

CFAD

N. E. Tagieva and C. Vaquero

Fig. 1. Schematic localization of the HIV-1BRU antisense RNA. The positive-strand env gene is shown below the position ruler, the negative-strand RNA and ORF are shown above the ruler. The positions are aligned with the viral genome. Restriction enzyme sites are indicated and the nucleotide positions are relative to the start of HIV-1BRU transcription (AC K02013).

(Fig. 1). This region is highly conserved among HIV-1 isolates. Moreover, this antisense sequence contains 180 nucleotides which are complementary to the Rev response element (RRE), a highly structured RNA found in all unspliced and singly spliced HIV transcripts (Olsen et al., 1990 ; Holland et al., 1990). Binding of RRE with Rev protein is required for the cytoplasmic expression of these transcripts (Malim et al., 1990). For these reasons, we were interested in studying HIV-1 replication and gene expression under conditions of overexpression of the natural antisense-RNA-containing ORF. Stable secondary structures within the complementary region may prevent formation of the hybrid with the target mRNA (Rhodes & James, 1990, 1991). On the other hand, sense–antisense interactions in vivo may not be simple ‘ zipper ’ processes but may require specific structural features of the RNA and possibly also the involvement of helper proteins to mediate RNA–RNA interactions (Nellen & Lichtenstein, 1993). At the same time, the existence of an AUG or an ORF may be deleterious for antisense RNA (Inouye, 1988). In the design of anti-HIV gene therapy strategies, it is important to specifically target essential and conserved HIV genes or cis-acting elements. In this study, we addressed the question of whether a stably expressed natural antisense RNA could inhibit virus replication. We also asked if the antisense protein encoded by this negative-strand ORF was implicated in the modulation of HIV-1 replication and gene expression. For this purpose, different cell lines expressing this RNA were established ; as a control we used cells stabilized with a retroviral vector lacking the antisense insert, or a vector with a mutated insert not producing ASP.

Methods + Expression vectors. Retroviral vector pLASBRUSN was used to stably express the antisense RNA. The antisense HIV-1BRU ORF was obtained by PCR using primers designed to amplify a 570 bp fragment

CFAE

from the vector pE∆H with the addition of BamHI sites (Ellerbrok et al., 1993). This fragment was then inserted into the BamHI site of retroviral vector pLXSN under the transcriptional control of Moloney MuSV LTR (Miller & Rosman, 1989). The antisense construct, as well as the empty vector, was transfected into retroviral ψCRIP packaging cells (Danos & Mulligan, 1988). Two vectors were obtained : pRc}CMVAS and pRc}CMVASmut, producing or not producing the antisense protein, respectively. The fragment corresponding to the antisense HIV-1BRU ORF was introduced at the XbaI and HindIII sites of the pRc}CMV vector (Invitrogen). The fragment expressing antisense RNA with a mutated AUG codon was obtained by site-directed mutagenesis (Hemsley et al., 1989). All subcloning experiments were carried out via standard recombinant DNA techniques (Sambrook et al., 1989) and oligonucleotides were synthesized by GIBCO BRL. + Cell culture and construction of antisense-RNA-expressing cell lines. The T cell lines A3.01 and chronically infected ACH-2 (NAIDS) as well as CEM and Jurkat cells were maintained in RPMI supplemented with 0±30 mg}ml -glutamine, antibiotics and 10 % heatinactivated foetal calf serum. Adherent cells were grown in Dulbecco’s modified Eagle’s (DMEM) medium with 0±30 mg}ml -glutamine, antibiotics, 10 % heat-inactivated new-born calf serum (ψCRIP) or 10 % foetal calf serum (HeLa). All media and supplements were purchased from GIBCO BRL. Cells were incubated at 37 °C in an atmosphere containing 5 % CO . Transfections of ψCRIP monolayers were performed by the # calcium phosphate precipitation method (Wigler et al., 1979) with glycerol treatment. Briefly, 3 µg of plasmid DNA was precipitated and added to subconfluent cultures in 35 mm diameter dishes. After 18–20 h the precipitate was removed and washed once with PBS. Glycerol (15 % in 1¬Hanks’ buffered saline) was added to the monolayers and incubated for 1 min at 37 °C. Then the cells were washed and supplied with fresh medium. The medium was replaced after 2 days with the selective medium for the isolation of stable transformants. For G418 (Gibco BRL) selection, the ψCRIP packaging cells were selected with 0±5 mg G418}ml. Neomycin-resistant cell populations were obtained 4–5 weeks later. Lymphocytic cells (A3.01, ACH-2, CEM, Jurkat) were infected during a 48 h cocultivation period with stable retroviral transformants. At this time, lymphoid cells were harvested and suspended at each passage at a concentration of 3¬10& viable cells}ml in fresh medium. For G418

Antisense RNA inhibits HIV-1 replication selection haematopoetic cells were grown in the presence of 1 mg of antibiotic}ml. Neomycin-resistant cell populations obtained 3–4 weeks later were initially screened for the presence of antisense RNA by Northern blot hybridization. Cell populations obtained by infection using the empty vector (mock) were kept as a negative control. Peripheral blood lymphocytes (PBLs) were prepared from leukopacks of healthy donors by Ficoll–Hypaque gradient centrifugation and were then stimulated for 48–72 h with phytohaemagglutinin (Sigma ; 5 mg}ml). The stimulated PBLs were transduced by exposure to retroviralproducer cells (for 18 h), selected with G418 (0±5 mg}ml) for 7–10 days and expanded for an additional 2–5 days in the absence of G418 as described (Sun et al., 1995 ; Vandendriessche et al., 1995). It was not possible to completely eliminate the nontransduced cell fraction by augmenting the G418 concentration without causing significant cell mortality and morbidity. Cell viability was determined by trypan blue exclusion from live cells ; the viable cells were subsequently used for HIV challenge. Throughout all experiments, the PBLs were maintained in RPMI 1640 medium supplemented with 10 % foetal bovine serum, 1 % essential amino acids and human recombinant interleukin-2 at 20 units}ml. HeLaCD4 cells (Charneau et al., 1992) were transiently transfected by the calcium phosphate precipitation method with glycerol treatment, as described above. All transient transfection experiments were performed in duplicate and repeated three times. After 24 h, the medium was replaced with fresh medium and cell populations were challenged with cell-free HIV. For the isolation of HeLaEnv stable transformant (expressing antisense RNA with or without ASP) (Dragic et al., 1992), cells were selected with 0±5 mg}ml of G418 after transfection by the calcium phosphate precipitation method with glycerol treatment. Neomycin-resistant cell populations (designated HeLaEnv as ; HeLaEnv asmut) were obtained 4–5 weeks later. The cell population obtained by transfection of the empty pRc}CMV vector was kept as a negative control (HeLaEnv mock). + Virus strains and infectivity assays. Laboratory strain HIV-1IIIB was obtained from chronically infected H9 cells. To produce cell-free HIV-1NDK, HIV-2ROD and HIV-2EHO three different CEM polyclonal populations chronically infected by these viruses were used. The infectivity of HIV isolates was tested in a syncytium formation and βgalactosidase assay. Briefly, monolayers of HeLaCD4 cells carrying a Tatinducible lacZ gene were exposed to chronically infected cell lines in coculture for 12–24 h, as described previously for similar assays (Clavel & Charneau, 1994). Wells with monolayers were then washed in PBS, fixed for 5 min in 0±5 % glutaraldehyde, and stained with a 5-bromo-4chloro-3-indolyl β--galactopyranoside (X-Gal) solution for 4 h. In the absence of viral particles, no β-galactosidase-expressing cells could be detected. After 24 h, in typical experiments, most of the HeLaCD4 cells were associated with large blue syncytia. Supernatants from culture of HIV-infected cells were prepared by centrifugation for 10 min at 2000 r.p.m. twice to remove debris. Viruscontaining supernatants were then filtered (pore size 0±45 µm) and p24 content was determined by ELISA (Innotest HIV antigen kit, Innogenetics, NV). Virus stocks were stored at ®80 °C. Innotest ELISA allows the detection of p24 core antigen of both HIV-1 and HIV-2 isolates. + HIV superinfections. HIV infection of A3.01, CEM, Jurkat, PBLs and HeLaCD4 cells was performed by incubating the cells with viral supernatants previously adjusted for identical concentrations of p24. All infections were performed in duplicate and repeated three to five times. For each infection, 10' cells were resuspended in 1 ml of supplemented

fresh medium containing 50 ng of total HIV p24. We also studied the inhibition of HIV-1IIIB replication in Jurkat and CEM cell lines with various initial doses of cell-free p24 : 25, 50 or 75 ng. The cell–virus mixtures were incubated at 37 °C for 2 h, and the cells were washed three times with PBS to remove unbound virus. The infected cells were suspended in 2 ml of medium for continued incubation. Every 3–4 days virus replication was monitored by testing p24 yield in the cellular supernatants by ELISA. + RNA and protein analyses. Total cellular RNA was extracted from cells after HIV infections, or for ACH-2 after activation with phorbol myristate acetate (PMA) (Folks et al., 1988), using TRIzol (GIBCO) or RNeasy (Qiagen) and 2–10 µg samples were denatured (glyoxal–DMSO) and electrophoresed in an agarose gel. The RNA was transferred to Hybond N filters (Amersham) by SSC blotting. pLasBRUSN RNA and HIV-1 mRNA were detected using specific (sense or antisense) riboprobes as described (Ausubel & Kingston, 1987 ; Serpente et al., 1992). Immunoblot analyses were performed as described by VanSlyke et al. (1991). HeLaEnv cells transfected with antisense expression vectors pRc}CMVAS and pRc}CMVASmut were washed in PBS pH 7±0, and lysed by boiling for 5 min in an SDS–PAGE sample buffer (50 mM Tris pH 6±8 ; 1 %, w}v, SDS ; 0±1 %, v}v, 2-β-mercaptoethanol ; 1 %, v}v, glycerol). As controls, HeLaEnv cells transfected with the empty vector were used. The proteins were separated on SDS–PAGE and electrotransferred to PVDF filters in Tris–glycine methanol buffer. Western blotting was done with HIV-1IIIB gp120 antiserum (NAIDS).

Results HIV-1IIIB and HIV-1NDK replication is inhibited in different cells by expression of antisense HIVBRU RNA

To determine the effect of the antisense HIV-1BRU RNA in the modulation of HIV expression, different cell populations expressing this RNA were constructed. Total RNA extracted from all antisense T-cell populations was tested by Northern blot hybridization to verify expression of the retroviral antisense RNA sequence. Molecular analyses showed that the retroviral vector promotes efficient expression of antisense RNA HIV-1BRU (see Northern blots), with no detectable sense transcript in transduced cells. These cell populations were challenged with 50 ng of cell-free p24 per 10' cells. HIV-1 replication was measured by ELISA for the HIV p24 in supernatant, as described in Methods. We compared extracellular virus production by different cell populations expressing antisense RNA with that obtained from control (mock) cells. As shown in Fig. 2 (a, b) at all time-points after infection extracellular virus production was substantially less in antisense cell populations than in control cells. Virus replication varied between 0 and 25 % of the control level in Tlymphoid cell lines stably expressing antisense RNA for both HIV-1 strains. Transient expression of antisense RNA in the HeLaCD4 cells showed 40 and 47 % of HIV-1IIIB and HIV-1NDK replication, respectively. To establish if the expression of this antisense RNA inhibits HIV-1 replication in primary haematopoetic cells, challenge experiments were done on transduced PBLs. Under the same conditions of HIV-1 infection, p24 production was markedly

CFAF


(a)

(b)

(c)

Fig. 2. Effect of antisense RNA expression on HIV-1 replication in different cells. Time-course of (a) HIV-1IIIB and (b) HIV-1NDK replication in different transduced cells. Data plotted are the means from three to five challenge experiments. Error bars denote sample standard deviations. Cells were infected with 50 ng p24 per 106 cells. Virus replication was measured by quantifying p24 in culture supernatants at various times after virus challenge. (c) Dose-dependent inhibition of HIV-1IIIB replication in CEM and Jurkat transduced cells. Solid lines, control (m) cell populations ; dashed lines, antisense populations (as). Symbols : —, A3.01 m and A3.01 as ; +, PBL m ; *, PBL as ; E, HeLaCD4 m ; D, HeLaCD4 as ; U, Jurkat m ; V, Jurkat as ; _, CEM m ; ^, CEM as.

reduced : only 8 % of virus replication was observed for HIV-1IIIB (Fig. 2 a) and 12 % for HIV-1NDK (Fig. 2 b) relative to PBLs transduced with control vector. To evaluate whether the inhibitory effect of antisense expression is dose-dependent, Jurkat and CEM transduced cell populations were challenged with different initial doses of HIV-1IIIB virus. Our results showed that the level of inhibition depended on the HIV challenge dose and was significant at 25 ng of cell-free virus per 10' cells at all times tested (Fig. 2 c). Indeed, the HIV-1IIIB replication in the Jurkat antisense CFAG

population increased from 2 % of control level to 60 % for a challenge dose of 75 ng}10' cells.

Inhibition of virus replication by the antisense RNA is sequence-specific

The transduced cell populations (HeLaCD4, A3.01, Jurkat, CEM) were challenged with HIV-2ROD (Fig. 3 a) and HIV-2EHO (Fig. 3 b) isolates (initial dose of HIV-2 isolates was

Antisense RNA inhibits HIV-1 replication

(a)

(b)

Fig. 3. Replication of (a) HIV-2ROD and (b) HIV-2EHO isolates in different cells transduced with antisense RNA HIV-1BRU. Data plotted and symbols as for Fig. 2.

50 ng of p24 per 10' cells). Compared with the control cell population, the antisense-expressing cell populations show no inhibition of HIV-2 replication.

hybridized with an antisense-RNA-specific probe. The level of antisense HIV-1BRU RNA transcript (Fig. 4 a) is decreased after HIV gene expression (Fig. 4 c). The low level of antisense RNA could be caused by the selective degradation of doublestranded RNA hybrids.

The HIV-1 mRNA profile is modulated by stable antisense RNA expression

We checked the expression of antisense RNA in transduced cell lines by Northern hybridization with an antisense RNAspecific probe (Fig. 4 a). As expected, no hybridization was detected to RNA transcribed from the vector alone (lane 3). To determine whether the observed inhibitory effect of antisense RNA expression on virus replication may be due to changes in HIV-specific mRNA levels, we first used chronically infected ACH-2 cells (Folks et al., 1985). As reported, these cells harbour a single integrated copy of the HIV-1BRU viral genome in a minimally productive state and viral production occurs via PMA activation (Folks et al., 1988). Therefore, total RNA was extracted from control and antisense transduced ACH-2 cells after activation with PMA (24 h). Secondly, uninfected transduced A3.01 and Jurkat cells were analysed after HIV-1 challenge experiments (on day 3 or 5). Total cellular RNA was probed with a riboprobe complementary to the env coding region. RNA blot-hybridization (Northern) analysis demonstrated significant changes in the mRNA profile. Fig. 4 (b) shows the inhibition of the HIV-1 genomic RNA (9±2 kb) and spliced mRNAs : 4±2 kb mRNA (env) and 2 kb mRNA (tat}rev). The same membrane was subsequently

HIV-1 inhibition occurs even in the absence of antisense protein expression

To determine whether expression of the antisense protein (ASP) could affect HIV-1 replication and gene expression, virus production was examined in HeLaCD4 cells transiently transfected with ASP producing (pRc}CMVAS) and not producing (pRc}CMVASmut) vectors. The p24 antigen detection shows the same decrease of HIV-1 replication for both vectors (Fig. 5 a). We analysed HIV-1 mRNA expression in the HeLaCD4 antisense population not producing ASP and in control cells by Northern blots (Fig. 5 b). Our data show that the accumulation of three major HIV-1 RNA species is reduced in both transduced cells. We next examined whether stably expressed ASP can modulate HIV-1 Env expression. For this purpose, HIV-1BRU Env-producing HeLa cells were transfected with the same vectors. Cell proteins were extracted as described in Methods. Fig. 5 (c) illustrates Western blot analysis with HIV-1 gp120 antiserum. No difference in restriction of Env expression was observed between the two antisense cell populations. Env expression was greatly diminished in both populations.

CFAH

N. E. Tagieva and C. Vaquero (a)

(b)

(c)

Fig. 4. Study of antisense RNA accumulation and its effect on HIV-1 mRNA profile. (a) Northern blot analysis of antisense RNA expression in antisense transduced and control cells. (b) Inhibition of HIV-1 mRNA accumulation in antisense transduced cells. Total RNA from ACH-2 cells was extracted after PMA activation (24 h). For other transduced cell populations, total RNA was analysed on day 3 or 5 after HIV-1 infection. 10 µg ACH-2 and Jurkat samples and 2 µg A3.01 samples were probed with a riboprobe complementary to a region contained in all HIV mRNAs. Small differences in the loading and transfer of each sample were controlled by measurement of the intensity of the 28S rRNA band. (c) Level of antisense HIV-1BRU RNA transcript after HIV gene expression in transduced cells. Hybridization of the same membrane (b) with an antisense RNA-specific riboprobe. 1, Total RNA from control cell population ; 2, total RNA from antisense-expressing cell population.

(b)

(c)

(a)

Fig. 5. Effect of antisense RNA expression not producing antisense protein on HIV-1 replication and gene expression. (a) Timecourse of HIV-1IIIB replication in HeLaCD4 transfected cells. Cells were infected with 50 ng p24 per 106 cells. Solid line, HeLaCD4 transduced with control vector. Hatched lines : E, HeLaCD4 expressing the antisense RNA and protein ; D, HeLaCD4 expressing only the antisense RNA. (b) Northern blot analysis of total RNA extracted from HeLaCD4 transfected cells on day 5 after HIV-1 challenge. 1, HeLaCD4 cells transfected with empty vector ; 2, HeLaCD4 cells expressed antisense RNA and ASP ; 3, HeLaCD4 cells expressed antisense RNA with mutated AUG codon. (c) Western blot analysis for HIV-1BRU Env protein. 1, HeLaEnv cells transfected with control vector ; 2, HeLaEnv cells stably expressing antisense RNA and protein ; 3, HeLaEnv cells stably expressing antisense RNA with mutated AUG codon.

CFAI

Antisense RNA inhibits HIV-1 replication

Discussion As outlined in the Introduction, the natural antisense HIV-1BRU RNA is expressed in acutely and chronically infected cell lines and PBLs, and is conserved among different isolates of HIV-1. We determined that the expression of the antisense HIV-1BRU ORF can regulate HIV replication and gene expression. Moreover, this regulation is not restricted to particular cell types. As previously shown, the length of RNA duplexes mediates their effects on regulation of gene expression, and long antisense RNAs expressed in cells can inhibit the replication of the heterologous strains of HIV-1 (Rhodes & James, 1991). Rhodes & James showed that the expression of antisense RNA (600 bp) which was complementary to the env gene (region between 5325 EcoRI and 5926 KpnI sites ; see Fig. 1) inhibited heterologous strain HIV-1 replication from 50 to 80 %. It is believed that the length of duplexes formed between antisense RNAs and primary RNA transcripts reflects a threshold value for their stability and efficiency. The length of the antisense ORF in HIV-1BRU is 570 bp, and we aimed to determine whether the replication and gene expression of laboratory strains of HIV-1 (IIIB, NDK) would be affected by the expression of this RNA. The results reported here indicate that the antisense HIV-1BRU RNA inhibited the replication of heterologous HIV-1 strains in different cell types. The level of inhibition varied between 57 % in transient transfected HeLaCD4 cells and 98 % in cell lines stably expressing antisense RNA. Retroviral vectors have been chosen for expression of this antisense RNA in various T-lymphocytic cell lines. Human Tlymphocytic cell lines are biologically different from primary PBLs. Primary cells are the major targets for HIV infection and replication, and one of the limitations of most anti-HIV gene therapy studies is that the efficacy of these strategies has been assessed with T-cell lines. To overcome this limitation, we have transduced PBLs to test the efficacy of antiviral constructs to inhibit HIV-1 expression in these cells. As in T-cell lines, HIV-1IIIB and HIV-1NDK p24 production was markedly reduced in the PBLs expressing antisense RNA, although it was not possible to completely eliminate the nontransduced cell fraction. These results demonstrate that retroviral vectormediated expression of an antisense HIV-1BRU RNA can inhibit heterologous strain HIV-1 replication in primary Tlymphocytes. In addition, the inhibitory effect of this natural antisense RNA is dose-dependent and sequence-specific. HIV-1 challenge experiments were performed at various doses of cell-free p24, to determine whether protection against HIV-1 persisted when the amount of input virus was increased. Our studies consistently revealed that inhibition of HIV-1 production was decreased when the transduced T-lymphocytes were challenged with higher doses of HIV-1 virus. The effectiveness of an antisense RNA inhibition strategy is dictated in part by the

ratio of antisense RNA to its target sequence in the cell. We used cell populations rather than individual cell clones in order to have a wide range of cell variability in all cell populations and minimize the intrapopulation difference. It might be suggested that some of the transduced cells in the population express lower levels of the antisense RNA and may not be protected with a higher dose of HIV-1. Our results are consistent with the observations of others that expression of HIV-1 genes could not suppress virus replication in T-cell lines and primary CD4+ T-lymphocytes that were challenged with a higher dose of HIV-1 virus (Sczakiel et al., 1992 ; Chuah et al., 1994 ; Vandendriessche et al., 1995). Comparison of HIVBRU and HIVIIIB antisense sequences shows 98 % identity, but HIVBRU and HIVNDK show only 86 % identity. The antisense region of HIV-2ROD and HIV-2EHO shows 58 % identity with HIV-1BRU. HIV-2 replication was not suppressed by expression of HIV-1BRU antisense RNA. Thus we can conclude that the inhibitory effect of natural antisense RNA is sequencespecific. The natural antisense HIVBRU RNA contains a sequence (180 bp) that is complementary to RRE. RRE is found in fulllength and singly spliced HIV transcripts, and HIV-1 structural gene expression requires the specific interaction of HIV-1 Rev protein with RRE (Olsen et al., 1990 ; Holland et al., 1990 ; Malim et al., 1990). A recent study on the inhibition of HIV-1 by expression of the Rev response element has shown that both sense and antisense RREs inhibit HIV replication in COS and SupT1 cells. It was therefore supposed that the expression of sense and antisense RRE might inhibit Rev function (Kim et al., 1996 ; Churchill et al., 1996). Our results also demonstrate the ability of the naturally occurring antisense RNA containing the antisense RRE to suppress HIV gene expression and mRNA accumulation in susceptible cells. We can thus suggest that once RNA duplexes are formed they become targets for nucleases that degrade the dsRNA region and consequently HIV mRNA accumulation is reduced. The abundance of antisense transcript is also diminished in the transduced cells. Taken together, these data suggest that inhibition of Rev function and degradation of HIV mRNA can result in lower levels of all classes of HIV-specific mRNA. The molecular details of this presumed dsRNA degradation will require further research. This report also provides data on inhibition of de novo infectivity (the protection of uninfected cells : A3.01, CEM, Jurkat, PBLs) and data related to inhibition of viral expression in chronically infected cells : ACH-2. ACH-2 is a Tcell clone, derived from A3.01 cells infected with HIV-1BRU. ACH-2 cells are a good in vitro model for post-integration HIV latency. Expression of HIV-1 genes per se is affected by stably expressed antisense RNA. It was important to determine whether the antisense protein can modulate the HIV-1 expression. For this purpose, virus replication was studied in cell populations transfected with a plasmid carrying the antisense RNA with a mutated AUG codon. To investigate whether inhibition of HIV-1

CFAJ


replication was due to the antisense protein or only to antisense RNA, virus replication was studied in two antisense populations : one expressing this protein and another expressing only the antisense RNA. Results suggested that ASP protein is not implicated in HIV-1 inhibition. In addition, we showed that suppression of Env expression in HeLaEnv cells depends on antisense RNA, not on protein. In summary, the protein encoded by the HIV-1BRU antisense RNA appears not to be involved in inhibition of HIV-1 expression. As previously shown, the natural antisense HIV-1 sequence is conserved among different HIV-1 isolates and expressed in different cells (Miller, 1988 ; Michael et al., 1994 ; Vanhe! e et al., 1995). Moreover, a negative-strand HIV-1 promoter has recently been found (Michael et al., 1994). Based on these data, we tested the function of this antisense RNA in HIV-1 expression by its overexpression in different cells. Our results may allow further advances in understanding how this natural negative-strand transcript works. These studies will not only increase our basic knowledge of gene regulation but will also improve the design of antisense RNAs for use in antiviral gene therapy. We thank E. Fouque and L. Che# ne for technical assistance ; Drs M. Alizon and O. Pleskoff for providing HeLaCD4 and HeLaEnv cells and chronically infected T-cells producing HIV-1 and HIV-2 isolates ; and L. Galio, S. Briquet and C. Vanhe! e-Brossollet for preparation of the PBLs from leukopacks of healthy donors. We are thankful to Dr B. Shacklett for critical review of the manuscript. This work was supported by the Institut National de la Sante! et de la Recherche Me! dicale, by a SIDaction fellowship (N. E. T.) and by an ANRS grant (C. V.).

References Ausubel, F. M. & Kingston, R. E. (1987). Current Protocols in Molecular Biology. New York : John Wiley. Baltimore, D. (1988). Gene therapy. Intracellular immunization. Nature 335, 395–396. Charneau, P., Alizon, M. & Clavel, F. (1992). A second origin of DNA plus-strand synthesis is required for optimal human immunodeficiency virus replication. Journal of Virology 66, 2814–2820. Chuah, M. K., Vandendriessche, T., Chang, H. K., Ensoli, B. & Morgan, R. A. (1994). Inhibition of human immunodeficiency virus type-1 by

retroviral vectors expressing antisense-TAR. Human Gene Therapy 5, 1467–1475. Churchill, M. J., Moore, J. L., Rosenberg, M. & Brighty, D. W. (1996).

The rev-responsive element negatively regulates human immunodeficiency virus type 1 env mRNA expression in primate cells. Journal of Virology 70, 5786–5790. Clavel, F. & Charneau, P. (1994). Fusion from without directed by human immunodeficiency virus particles. Journal of Virology 68, 1179– 1185. Coleman, J., Green, P. J. & Inouye, M. (1984). The use of RNAs complementary to specific mRNAs to regulate the expression of individual bacterial genes. Cell 37, 429–436. Danos, O. & Mulligan, R. C. (1988). Safe and efficient generation of recombinant retroviruses with amphotropic and ecotropic host ranges. Proceedings of the National Academy of Sciences, USA 85, 6460–6464.

CFBA

Dragic, T., Charneau, P., Clavel, F. & Alizon, M. (1992). Complemen-

tation of murine cells for human immunodeficiency virus envelop}CD4mediated fusion in human}murine heterokaryons. Journal of Virology 66, 4794–4802. Duan, L., Oakes, J. W., Ferraro, A., Bagasra, O. & Pomerantz, R. J. (1994). Tat and Rev differentially affect restricted replication of human

immunodeficiency virus type 1 in various cells. Virology 199, 474–478. Eglitis, M. A. & Anderson, W. F. (1988). Retroviral vectors for

introduction of genes into mammalian cells. Biotechniques 6, 608–614. Ellerbrok, H., Serpente, N., Pancino, G., Vanhe! e, C., D’Auriol, L., Sitbon, M. & Vaquero, C. (1993). Sequences in the REV-responsive

element responsible for premature translational arrest in the HIV-1 envelope. European Journal of Biochemistry 216, 459–467. Folks, T., Benn, S., Rabson, A., Theodore, T., Hoggan, M. D., Martin, M., Lightfoote, M. & Sell, K. (1985). Characterization of a

continuous T-cell line susceptible to the cytopathic effects of the acquired immunodeficiency syndrome (AIDS)-associated retrovirus. Proceedings of the National Academy of Sciences, USA 82, 4539–4543. Folks, T. M., Justement, J., Kinter, A., Schnittman, S., Orenstein, J., Poli, G. & Fauci, A. S. (1988). Characterization of a promonocyte clone

chronically infected with HIV and inducible by 13-phorbol-12-myristate acetate. Journal of Immunology 140, 1117–1122. Garcia, J. V. & Miller, A. D. (1994). Retrovirus vector-mediated transfer of functional HIV-1 regulatory genes. Aids Research and Human Retroviruses 10, 47–52. Hemsley, A., Arnheim, N., Toney, M. D., Cortopassi, G. & Galas, D. J. (1989). A simple method for site-directed mutagenesis using the

polymerase chain reaction. Nucleic Acids Research 17, 6545–6551. Holland, S. M., Ahmad, N., Maitra, R. K., Wingfield, P. & Venkatesan, S. (1990). Human immunodeficiency virus Rev protein recognizes a

target sequence in Rev-responsive element RNA within the context of RNA secondary structure. Journal of Virology 64, 5966–5975. Inouye, M. (1988). Antisense RNA : its functions and applications in gene regulation – a review. Gene 72, 25–34. Izant, J. G. & Weintraub, H. (1984). Inhibition of thymidine kinase gene expression by anti-sense RNA : a molecular approach to genetic analysis. Cell 36, 1007–1015. Kim, J. H., McLinden, R. J., Mosca, J. D., Vahey, M. T., Greene, W. C. & Redfield, R. R. (1996). Inhibition of HIV replication by sense and

antisense rev response elements in HIV-based retroviral vectors. Journal of Acquired Immune Deficiency Syndrome and Human Retrovirology 12, 343–351. Lisziewicz, J., Sun, D., Lisziewicz, A. & Gallo, R. C. (1995). Antitat gene therapy : a candidate for late-stage AIDS patients. Gene Therapy 2, 218–222. Malim, M. H., Tiley, L. S., McCarn, D. F., Rusche, J. R., Hauber, J. & Cullen, B. R. (1990). HIV-1 structural gene expression requires binding

of the Rev trans-activator to its RNA target sequence. Cell 60, 675–683. Michael, N. L., Vahey, M. T., d’Arcy, L., Ehrenberg, P. K., Mosca, J. D., Rappaport, J. & Redfield, R. R. (1994). Negative-strand RNA tran-

scripts are produced in human immunodeficiency virus type 1-infected cells and patients by a novel promoter downregulated by Tat. Journal of Virology 68, 979–987. Miller, R. H. (1988). Human immunodeficiency virus may encode a novel protein on the genomic DNA plus strand. Science 239, 1420–1422. Miller, A. D. & Rosman, G. J. (1989). Improved retroviral vectors for gene transfer and expression. Biotechniques 7, 980–990. Morgenstern, J. P. & Land, H. (1990). Advanced mammalian gene transfer : high titre retroviral vectors with multiple drug selection markers

Antisense RNA inhibits HIV-1 replication and a complementary helper-free packaging cell line. Nucleic Acids Research 18, 3587–3596. Nellen, W. & Lichtenstein, C. (1993). What makes an mRNA anti-senseitive ? Trends in Biochemical Science 18, 419–423.

Serpente, N., Sitbon, M. & Vaquero, C. (1992). Suboptimal and optimal

Olsen, H. S., Nelbock, P., Cochrane, A. W. & Rosen, C. A. (1990).

Sun, L. Q., Pyati, J., Smythe, J., Wang, L., Macpherson, J., Gerlach, W. & Symonds, G. (1995). Resistance to human immunodeficiency virus

Secondary structure is the major determinant for interaction of HIV rev protein with RNA. Science 247, 845–848. Palmer, T. D., Thompson, A. R. & Miller, D. (1989). Production of human factor IX in animals by genetically modified skin fibroblasts : potential therapy for hemophilia B. Blood 73, 438–445. Pestka, S., Daugherty, B. L., Jung, V., Hotta, K. & Pestka, R. K. (1984).

Anti-mRNA : specific inhibition of translation of single mRNA molecules. Proceedings of the National Academy of Sciences, USA 81, 7525–7528. Purcell, D. F., Broscius, C. M., Vanin, E. F., Buckler, C. E., Nienhuis, A. W. & Martin, M. A. (1996). An array of murine leukemia virus-related

elements is transmitted and expressed in a primate recipient of retroviral gene transfer. Journal of Virology 70, 887–897. Rhodes, A. & James, W. (1990). Inhibition of human immunodeficiency virus replication in cell culture by endogenously synthesized antisense RNA. Journal of General Virology 71, 1965–1974. Rhodes, A. & James, W. (1991). Inhibition of heterologous strains of HIV by antisense RNA. AIDS 5, 145–151. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning : A Laboratory Manual. Cold Spring Harbor, NY : Cold Spring Harbor Laboratory. Sczakiel, G., Oppenlander, M., Rittner, K. & Pawlita, M. (1992). Tatand Rev-directed antisense RNA expression inhibits and abolishes replication of human immunodeficiency virus type 1 : a temporal analysis. Journal of Virology 66, 5576–5581.

activation signals modulate differently the expression of HIV-1 and cytokine genes. Biochemical and Biophysical Research Communications 182, 1172–1179.

type 1 infection conferred by transduction of human peripheral blood lymphocytes with ribozyme, antisense, or polymeric trans-activation response element constructs. Proceedings of the National Academy of Sciences, USA 92, 7272–7276. Vandendriessche, T., Chuah, M. K., Chiang, L., Chang, H. K., Ensoli, B. & Morgan, R. A. (1995). Inhibition of clinical human immunodeficiency

virus (HIV) type 1 isolates in primary CD4 T lymphocytes by retroviral vectors expressing anti-HIV genes. Journal of Virology 69, 4045–4052. Vanhe! e, B. C., Thoreau, H., Serpente, N., d’Auriol, L., Levy, J. P. & Vaquero, C. (1995). A natural antisense RNA derived from the HIV-1

env gene encodes a protein which is recognized by circulating antibodies of HIV individuals. Virology 206, 196–202. VanSlyke, J. K., Franke, C. A. & Hruby, D. E. (1991). Proteolytic maturation of vaccinia virus core proteins : identification of a conserved motif at the N termini of the 4 b and 25 K virion proteins. Journal of General Virology 72, 411–416. Wigler, M., Sweet, R., Sim, G. K., Wold, B., Pellicer, A., Lacy, E., Maniatis, T., Silverstein, S. & Axel, R. (1979). Transformation of

mammalian cells with genes from procaryotes and eucaryotes. Cell 16, 777–785.

Received 3 March 1997 ; Accepted 9 June 1977

CFBB