key: cord-0004891-7ckqj8dw
authors: Tung, F. Y. -T.; Daniel, M. D.
title: Targeted inhibition of immunodeficiency virus replication in lymphocytes through retroviral mediated gene transfer
date: 1993
journal: Arch Virol
DOI: 10.1007/bf01313779
sha: 953ea8c5f404f4d438c7e87e1eb9ef87888ffde6
doc_id: 4891
cord_uid: 7ckqj8dw

To test the feasibility of gene therapy for AIDS patients, an animal model is needed to evaluate the efficacy and safety of this approach. Antiviral genes (encoding antisense RNA or viral protein) derived from Simian immunodeficiency virus (SIV) were efficiently targeted into CD4+ lymphocytes through retroviral-mediated gene transfer. After challenging with infectious viruses, the transduced lymphocytes that received antiviral genes were not only protected from SIV infection, but also from infection with HIV, for at least 25 days. Furthermore, little or no cytolytic effect (syncytium formation) was observed in the protected cells. These data demonstrated that SIV or HIV replication could be effectively blocked by antisense sequence(s) or negative dominant factors which were introduced into targeted cells through retroviral-mediated gene transfer.

Simian immunodeficiency virus (SIV) isolated from rhesus monkeys, has a lentivirus morphology, a CD 4 + cell tropism, and serological crossreactivity with HIV-1 and HIV-2 [3, 8] . Furthermore, SIV causes an AIDS-like disease in experimentally infected rhesus monkeys [18] . SIV has, therefore, been employed as an animal model for studying potential treatments for human AIDS [9, 21] .

Gene therapy has been proposed for treatment of AIDS [1] . The procedure would be initiated by taking bone marrow cells, including the hematopoietic stem cells, from an HIV-infected individual. The stem cells would be transduced with a recombinant virus that encoded an antiviral gene (an RNA or protein) which was capable of dominantly interfering with the intracellular growth of HIV. The modified stem cells would then be injected back into the patient where they would return to the marrow. If only a small number of specific subsets of bone marrow cells are infected with HIV in the infected individual [16, 31] , it is conceivable that, after targeting an antiviral gene, protected ceils would be selected by a virus specific cytotoxic-effect to dominate the body's hematopoietic system. This approach would probably be suitable for already infected individuals where its purpose would be therapeutic. Inhibition of HIV replication in varying degrees by antisense RNA as well as sense RNA has been shown [15, 26, 29] . We now report that using a murine retroviral vector [10, 11, 13] , an anti-SIV gene can be efficiently delivered into a cultured human CD 4 + cell line in which it can effectively block SIV and HIV replication. This report represents the first step towards the application of antisense/sense gene therapy approach in an animal model.

The SIV macaque isolate 239, HIV-2 and HIV-1 producing cell lines have been reported [22] . The porcine transmissible gastroenteritis coronavirus (TGEV) was grown on swine testicle cells (ST) as previously described [34] . The human CD 4 + cell line HUT-78 was grown in RPMI 1640 medium supplemented with 10% fetal bovine serum, 2mM L-glutamine and antibiotics (penicillin and streptomycin, 100 units/ml each, Gibco, Bethesda, U.S.A.). NIH/3T3, amphotropic packaging cell line GP + envAM12 [19] and producer lines were grown in Dulbecco's Modified Eagle Medium supplemented with 10% donor calf serum (Gibco/BRL), 2 mM L-gtutamine and antibiotics.

The retroviral vectors were derived from the Moloney routine leukemia virus [2] . SIV molecular clone, SIVmac239 (GenBank database accession no. M33262), was used for constructing retroviral vectors. For vectors SIV gp + and SIV gp-, a 5.6 kbp Nar I-Sph I viral DNA (1079 nt-6702 nt) was isolated (gag-pol region), blunt-ended and ligated to the Barn HI linearized and blunt-ended pZipNeoSV ( x ) 1 vector. For vectors SIVenv + and SIVenv-, a 3.5 kbp Sac I-Sac I DNA (6011 nt-9482 nt, env region) of SIVmac was isolated and added to a Barn HI linker, then ligated to the Barn HI linearized pZipNeoSV (x)1 vector. Producer lines were generated as described [19] . Briefly, packaging cell fine, Gp + envAM12, was electroporated with 5gg vector DNA, pSIVenv+, pSIVenv-, pSIVgp +, and pSIVgp-, respectively, and selected in the presence of 250 gg/ml active G 418 (Gibco) for 10-14 days. The individual G 418 resistant colonies were isolated with a cloning ring and expanded to a cell line. The virus titer of the producer line was determined on NIH/3T3 cells as described [20] .

The human CD 4 + lymphoid cell line HuT 78 was co-cultured with producer lines in the presence of 8 gg/ml polybrene at a ratio of 1 : 1. The transduced HUT-78 cells were harvested and grown in the presence of G418 (500 gg/ml, active) and passaged 1:3 every 3 days for at least two weeks. Before challenging with uncloned SIVmac, HIV-1, and HIV-2 [4, 24] , the HUT-78 cells were replenished with fresh medium in the absence or presence of G418 for at least 24 h. All transduced cells had similar growth rate prior to challenge with viruses. Approximately 10 6 viable cells in 5 ml medium were infected with 10 000 cpm reverse tran- Polymerase chain reaction analysis of producer line and transduced HUT-78 cells DNA was extracted from producer cell lines and transduced HUT-78 cells by proteinase K/phenol/chloroform method [14] . Standard PCR technique was used for amplification. In brief, 1 gg sample DNA was added to the 50 gl solution containing 10raM Tris-HC1 pH 8. 

Total RNA was extracted from transduced HUT-78 cells with a Stratagene RNA Extraction Kit. RNA was fractionated on a 1.2% formaldehyde gel and capillary blotted onto charged nylon paper (Micro Separation Inc.). The blot was probed with a [32p]labeled DNA fragment prepared by PCR.

Viral replication was monitored by RT activity in the cell-free supernatant [5] . Two milliliter aliquots of cell-free supernatant were removed at different days postinfection for RT activity assay. Two milliliters of RPMI 1640 medium with 10% FBS were added into each cell culture after each removal.

A 5.6 kbp DNA fragment of SIVmac 239 (Nat I-Sph 1) encompassing the primer binding site, gag-pol, viv, and 5' env (gag-pol region) and a 3.5 kbp DNA frag-ment of SIVmac239 (SacI-SacI) encoding vif, vpx, vpr, rev, tat, and env (env region) were separately subcloned into the Bam HI site of the expression vector pZipNeoSV(x) 1 in both sense (SIVgp+ and SIVenv+) and antisense (SIVgp-and SIVenv-) orientations (Fig. 1 ). The orientation of the inserts was determined by restriction mapping and DNA sequencing. These retroviral vectors were used to transfect the amphotropic packaging cell line, Gp + envAml2. The producer lines which produced a high titer of recombinant virus (approximately 106-107 cfu/ml) were isolated for transduction experiments.

There was no evidence of replication competent virus in the transduced cells by checking the infectivity of culture medium on NIH/3T3 cells as described [20] .

After G 418 selection (more than two weeks), total cellular DNA was extracted from the producer lines and pooled transduced cells for PCR analysis. The results showed positive for "env region" for DNA extracted from producer cells, SIVenv-and SIVenv + ( from SIVenv-transduced HUT-78 cells (Fig. 2 B) . When DNA extracted from SIVenv+ transduced HUT-78 before G418 selection was used for PCR, weak positive band for env region was also observed (data not shown). However, the PCR analysis of neo gene showed positive for both SIVenv-and SIVenv + transduced HUT-78 cells (Fig. 2 C) . This data suggested that both the full-length and spliced RNA transcripts ( Fig. 1) were transferred into the SIVenv + transduced HUT-78 cells. A possible explanation is that splicing donor (nt 6853) and acceptor (nt 9062) sites in the sense RNA of env region resulted in the spliced RNA being more favourably packaged into viral particles. The Northern RNA analysis also supports this hypothesis (Fig. 3 ).

Total cellular RNA was extracted from transduced HUT-78 cells which were selected with G 418 and maintained in the presence or absence of G 418 (500 gg/ ml, active). The RNA blots were probed with [32p]labeled env, pol, and neo specific probes, respectively. The full-length RNA encoding antisense of env region, as well as spliced neo RNA, was seen in the SIVenv-transduced HuT-78 cells, but only spliced RNA encoding neo gene was seen in the SIVenv + transduced HUT-78 cells (Fig. 3) . For SIVgp + and SIVgp -transduced cells, both full-length (8 kb) and spliced RNA (2.5 kb) were seen (Fig. 4) . The ratio of full-length and spliced RNA was different in the SIVgp -transduced HuT-78 in the presence or absence of G418. In the presence of G418, the spliced RNA encoding the neo gene became dominant (Fig. 4, lane 1) . A 2.7 kb RNA, which comigrated with neo spliced RNA, also hybridized to the pol probe in the SIVgp + transduced HUT-78 cells (Fig. 4 B, lane 3) . This 2.7 kb RNA could be the splicing intermediate of the full-length transcript of the SIVgp +. The slightly different size of spliced RNA transcripts could be due to the distortion by 28S ribosomal RNA during blotting, or different lengths of potyadenylation. Repeated experiments with prolonged electrophoresis showed that the size difference was minimal. This data indicated that the presence of G418 in the growth medium influenced the steady state level of primary RNA transcripts.

In the presence of G 418, the cells maintained a high steady state level of spliced transcripts.

After infecting fractions of transduced cells with uncloned SIVmac or HIV-1, RT activities (Fig. 5 ) and syncytium formation (Fig. 6 ) were monitored. The HUT-78 cells transduced with antisense constructs of gag-pol and env regions were effectively protected (no RT activity) from SIVmac infection for more than 20 days when compared to cells transduced with the corresponding sense constructs. The viral activity (RT) from cells transduced with sense construct ofgag-pol region was lower than that from cells transduced with sense construct of env region (Fig. 5 A) . The partial interference could be due to the gag and polymerase proteins expressed in SIVgp + transduced cells (data not shown). The mechanism of partial interference is unclear. The higher RT in the control, untransduced HUT-78 cells, could be due to slightly higher cell number and/or G418 selection in transduced cells. There was little or no cytopathic effect (syncytium formation) in protected cells (cells transduced with SIVenv-and SIVgp-) compared to that in unprotected cells (high degree of syncytium formation in cells transduced with SIVenv +), at 20 days postinfection. Interestingly, there was little or no inhibition of SIVmac replication and strong cytopathic effect in SIVgp-transduced HUT-78 cells when culture was maintained in G418 selection (Figs. 5 B and 6) . When the fraction of pooled transduced cells was infected with HIV-1, cells transduced with sense and antisense constructs ofgag-pol region were effectively protected from viral infection. Almost no viral replication was observed in SIVgp + transduced cells and minimal viral replication was observed in SIVgptransduced cells (Fig. 5 C) . The slightly higher background RT activity (approximately 3 000 cpm/ml) in SIVgp + transduced cells was derived from the empty SIVmac viral particles (data not shown). Cells transduced with antisense and sense constructs of the env region were not effectively protected. It should be noted that significant cytopathic effect (approximately 90%) was observed in unprotected cells at 25 days postinfection. Therefore, the RT assay was terminated after 25 days. To study the mechanism of nearly complete protection of HIV-1 infection, the cellular DNA was extracted from the cells harvested at 25 days postinfection and subjected to PCR analysis. In agreement with the RT results, no HIV-1 proviralpol sequence and little HIV-1 proviralpol sequence can be detected in the SIVgp+ and SIVgp-challenged cells, respectively (Fig. 7) . The sensitivity of this PCR analysis is less than 5 copies per sample (data not shown). The inability to detect HIV-1 proviral sequence in SIVgp + transduced cells could be due to the cell death of small number of infected cells in which complete viral replication is blocked but expression of HIV-1 env gene is normal. When HIV-2 was used as the challenging virus [4] , similar results were obtained (data not shown). All of these challenging experiments were repeated at least twice. The results shown were the mean of two initial experiments. Subsequent experiments performed several months later also demonstrated similar results. These data suggested that there might be different mechanisms by which transduced cells were protected from SIVmac and HIV-1 (or HIV-2) infections.

Cell associated transmission also inhibited in the transduced cells expressing antisense RNA HIV spreads by cell-free, as well as cell-associated transmission. The above data suggested that HUT-78 cells transduced with antiviral genes could prevent cell- ulations (transduced with sense construct) were five times greater than those in protected cell populations (transduced with antisense construct). The viral activities in protected cell populations were mostly derived from persistently infected HUT-78 cells. Furthermore, significant syncytium formation (90-95%) was observed in unprotected cell populations compared to low levels of syncytium formation (5-10%) in protected cell populations after 9 days of cocultivation. These data suggested that celt-associated viral transmission was also inhibited in the protected cell populations.

In order to demonstrate that inhibition of viral replication is virus specific, we (Table 1 ). This data indicates that the inhibition of SIVmac and HIV replication in the transduced HUT-78 cells is virus specific. 

Biological heterogeneity of HIV or SIV is reflected by differences in host range, replicative properties, and cytopathic effect in infected cells [7, 12, 28, 32] . Restriction endonuclease mapping and sequence analyses of HIV [24] and SIV [3] isolates, as well as the patterns of neutralization of different isolates by antibodies, also demonstrate genetic diversity, particularly in the envelope glycoprotein [6] . The genetic variation is a major obstacle to the design of vaccines or therapeutic agents effective to different variants. SIV macaque isolate shares about 55% amino acid sequence identity with HIV-1 in the gag and pol genes and less than 31% in the env gene [25] . Our studies have demonstrated that an antiviral gene (antisense RNA or viral gene) can be delivered to CD 4 + cells by means of a retroviral-mediated gene transfer technique. The transduced cells were not only protected from parental viral infection (SIV, macaque and sooty mangabey isolates), but also from other related viral infections (HIV-1 and HIV-2). The mechanisms by which the retroviral vectors inhibited the replication of immunodeficiency viruses is not clear. One possible mechanism is that the antisense RNA binds to the viral RNA before reverse transcription occurs or arrests translation of targeted mRNA. In the transduced cells expressing SIVmac associated gag and poI polypeptides, the SIVmac poly~eptides may function as mutant peptides, which may interact with HIV polypeptides, and inhibit HIV replication by interfering virus assembly [33] .

Three lines of evidence indicated that the effective inhibition of SIVmac and HIV replication was unlikely due to clonal variation. First, the pooled transduced HUT-78 cells were used for challenge experiments. In fact, the fraction of HUT-78 cells transduced with the antisense construct of env region effectively inhibits SlVmac but not HIV-1 and HIV-2. Furthermore, fractions of HUT-78 transduced with sense construct of gag-pol region did not support HIV-1 replication but support a certain degree of SIVmac and SIVsm (sooty mangabey isolate, data not shown). Second, height level of gene transduction efficiency (> 60%) by co-culturing HUT-78 cells with producer lines was employed in this experiment. Third, no specific pattern of integrated provirus was shown by Southern hybridization (data not shown).

A major advantage of this approach is the potential of in vivo applicability. Antiviral genes can theoretically be introduced into bone marrow stem cells from AIDS patients through the use of retroviral vectors. The manipulated marrow cells could then be infused back to the patients where progeny derived from the bone marrow stem cells carrying the antiviral gene would be resistant to viral infection. The feasibility and safety of using retroviral gene transfer for human gene therapy has been demonstrated recently [17, 23, 27] . The approaches described in this report could be tested in SIV infected monkeys.

In previous reports [t 5, 26, 29] , partial and transient inhibition of HIV was achieved in cell culture by antisense RNA. In these three cases, only small size antisense RNA was used (less than 1 kbp) for inhibition of HIV replication. In our study, a 1.5 kbp antisense construct of SIVmac gag region (including LTR, primer binding site and the majority of gag) did not effectively inhibit SIVmac replication (data not shown). Antisense RNA of SIVmac env region (3.5kb) can effectively inhibit SIVmac replication and partially inhibit HIV replication.

The antisense RNA of SIVmac gag-pol region (5.6 kb) can effectively inhibit both SIVmac and HIV replication. From our Northern anNysis, there is no significant difference in RNA levels in those transduced cells expressing antisense and sense RNA. These data suggest that the effectiveness of the antisense RNA inhibition could be related to the size of antisense sequences. Since the viral titers used for the challenge in the previous experiments were not clearly defined, a correlation can not be drawn between these experiments. Recently, Sczakiel etal. [30] have shown effective long-term inhibition of HIV-1 by antisense RNA when a low M.O.I. of challenging virus was used. The effective inhibition of SIV and HIV replication in this report may also be due to low M.O.I. used.

Intracellular immunization

Construction and application of a highly transmissible murine retrovirus shuttle vector

Sequence of simian immunodeficiency virus from macaque and its relationship to other human and simian retroviruses

Isolation of a new human retrovirus from West African patients with AIDS

Isolation ofT-cell tropic HTLV-III-like retrovirus from macques

Prevalence of antibodies to three retroviruses in a capative colony of macaque antibodies to three retroviruses in a capative colony of macaque monkeys

Simian immunodeficiency virus from African Green monkeys

Simian immunodeficiency viruses

Vaccine protection against simian immunodeficiency virus infection

Gene expression in mice after high efficiency retroviral-mediated gene transfer

Gene transfer into hematopoietic progenitor ceils from normal and cyclic hematopoietic dogs using retroviral vectors

Differential ability of human immunodeficiency virus isolates to productively infect human cells

Development of retrovirus vectors useful for expressing genes in cultured routine embryonal cells and hematopoietic cells in vivo

Isolation of high-molecular weight DNA from mammalian cells

Inhibition of human immunodeficiency virus type I multiplication by antisense and sense RNA expression

Simian immunodeficiency virus infection of macaque bone marrow macrophages correlates with disease progression in vivo

High-efficiency gene transfer and expression in normal human hematopoietic cells with retrovirus vectors

Induction of AIDS-like disease in macaque monkeys with T-cell tropic retrovirus STLV-III

Construction and use of a safe and efficient amphotropic packaging cell fine

Improved retroviral vectors for gene transfer and expression

A formalin-inactivated whole SIV vaccine confers protection in macaques

Characterization of infectious molecular clones of Simian immunodeficiency virus (STVmac) and human immunodeficiency virus type 2: persistent infection of rhesus monkeys with molecular cloned SIVmac

Efficient retrovirus-mediated transfer and expression of a human adenosine deaminase gene in diploid skin fibroblasts from an adenosine deaminase-deficient human

Complete nucleotide sequence of the AIDS virus, HTLV-III

The complete nucleotide sequence of a pathogenic molecular clone of simian immunodeficiency virus

Inhibition of human immunodeficiency virus replication in cell culture by endogenously synthesized antisense RNA

Gene transfer into humans-immunotherapy of patients with advanced melanoma, using tumour-infiltrating lymphocytes modified by retroviral gene transduction. New infiltrating lymphocytes modified by retroviral gene transduction

Extensive variation of human immunodeficiency virus type 1 in vivo

Inhibition of human immunodeficiency virus type 1 replication in human T cells stably expressing antisense RNA

Tat-and rev-directed antisense RNA expression inhibits and abolishes replication of human immunodeficiency virus type 1: a temporal analysis

Human immunodeficiency virus-infected individuals contain provirus in small numbers of peripheral mononuclear cells and at low copy numbers

Differential syncytium-inducing capacity of hmaaan immunodeficiency virus isolate: frequent detection of syncytium-inducing isolates in patients with acquired immunodeficiency syndrome (AIDS) and AIDS-related complex

HIV-1 gag mutant can dominantly interfere with the replication of the wild-type virus

The 9-kDa hydrobphobic protein encoded at the 3' end of the porcine transmissible gastroenteritis coronavirus genome is membrane-associated

A 3.5 kbp cDNA encoding the peplomer gene of TGEV was subcloned into MNC vector under the control of cytomegalovirus early promoter

We thank Drs. A. Bank and R. C. Mulligan for the packaging cell line and retroviral vector, Dr. Y. Li for his help in this study, and Drs. K. Rand and J. Lau for critical reading of this manuscript. The HIV-1 was obtained from Reference Reagent Program, Division of AIDS, NIAID (Dr. R. Gallo). This work was supported in part by NIH grant (A131918).

Received April 19, 1993