Virus Life Cycle

Figure 2 Overview of possible genetic targets to block HIV-1 replication. The different approaches are described in detail in the text.

strain, which is resistant to 1 of such compounds and, therefore, are partially resistant to combination therapy. Such virus strains have a higher chance to further mutate and escape the inhibitory effects of the other chemical compounds. In particular, the viral enzyme reverse transcriptase, which converts the viral RNA genome into a double-stranded DNA, does not have proofreading capabilities. On average, it inserts at least 1 incorrect nucleotide into the viral genome per replication cycle. This error rate is 1 million times higher than that of the cellular DNA polymerase I, which is the main enzyme for the replication of the eukaryotic genome. This high mutation rate explains why drug-resistant virus mutants emerge rapidly in HIV-1-infected patients (15). This high mutation rate also explains why new mutant viruses continuously arise, which are ''new'' and, therefore, not recognized and inactivated by the immune system. Consequently, in the clinically latent stage of HIV-1 infection, high virus loads persist, which consistently change their genetic outfit to escape drugs that inhibit virus replication and the immune system (16-18).

In summary, the clinical application of all drugs for the treatment of AIDS has not led to a cure of the disease and even the new combination therapy may only halt the development of AIDS in infected people temporarily, as new drug-resistant variants of the HIV-1 virus start to emerge. Thus, efforts are underway in many laboratories to develop alternative therapeutics.

III. GENETIC "BULLETS" TO BLOCK HIV-1 REPLICATION

The primary target cells for HIV-1 are cells of the hematopoietic system, in particular, CD4+ T lymphocytes and macrophages. During HIV-1 infection, these cells are destroyed by the virus leading to immunodeficiency among infected individuals. To prevent the destruction of the cells of the immune system, many efforts are now underway to make such cells resistant to the HIV-1 virus. This approach has been termed

''intracellular immunization'' (19). In particular, the development of genetic agents, which attack the virus at several points simultaneously inside the cell and/or are independent from viral mutations, has gained great attention.

Such potential agents, also termed ''genetic antivirals'' should have 4 features, to overcome the shortcomings of conventional treatments. First, they should be directed against a highly conserved moiety in HIV-1, which is absolutely essential for virus replication eliminating the chance that new mutant variants arise, that can escape this attack. Second, they have to be highly effective, greatly reducing or, ideally, completely blocking the production of progeny virus. Third, they have to be nontoxic. A fourth criterion, which also should not be overlooked is that the antiviral agent has to be tolerated by the immune system. It would not make much sense to endow immune cells with an antiviral agent, that elicits an immune response against itself leading to the destruction of the HIV-1-resistant cell after a short period of time.

In the past few years, many strategies have been developed and proposed for clinical application to block HIV-1 replication inside the cell (see also Fig. 2). Such strategies use either antiviral RNAs or proteins. They include antisense oligonucleotides, ribozymes, RNA decoys, transdominant mutant proteins, products of toxic genes, and single-chain antigen-binding proteins (20-32). Antiviral strategies that employ RNAs have the advantage that they are less likely to be immunogenic than protein-based antiviral agents. However, protein-based systems have been engineered that use inducible promoters that only become active upon HIV-1 infection.

A. RNA-based Inhibitors of HIV-1 Replication

1. Antisense RNAs and Ribozymes

It is very well known that prokaryotes and bacteriophages express antisense RNAs, which provide regulatory control over gene expression by hybridizing to specific RNA sequences (33). In animal cells, artificial antisense oligonucleotides (RNAs or single-stranded DNAs) have been successfully used to selectively prevent expression of various genes (e.g., oncogenes, differentiation genes, viral genes, etc.) (33,34). Furthermore, the presence of double-stranded RNA inside the cell can induce the production of interferon and/or other cyto-kines stimulating an immune response. Indeed, it has also been reported that the expression of RNAs capable of forming a double-stranded RNA molecule with the HIV-1 RNA (antisense RNAs) can significantly reduce the expression of HIV-1 proteins, and consequently the efficiency of progeny virus production (33-38).

Ribozymes are very similar to antisense RNAs (e.g., they bind to specific RNA sequences), but they are also capable of cleaving their target at the binding site catalytically. Thus, they have the advantage that they may not need to be overex-pressed to fulfil their function. Certain ribozymes (e.g., hairpin and hammerhead ribozymes, require only a GUC sequence) (Fig. 3). Thus, many sites in the HIV-I genome can be targeted. However, several questions still remain to be answered: for

Figure 3 Schematic representation of 2 ribozymes to block HIV-1 replication. The structures shown are paired with actual HIV-I target sequences. (Top) A hammerhead ribozyme pairs specifically with a sequence in the gag region of the HIV-1 genome. (Below) A hairpin ribozyme designed to bind to and cleave the 5' end of the viral genome, abolishing the reverse transcription and integration of progeny virus.

Figure 3 Schematic representation of 2 ribozymes to block HIV-1 replication. The structures shown are paired with actual HIV-I target sequences. (Top) A hammerhead ribozyme pairs specifically with a sequence in the gag region of the HIV-1 genome. (Below) A hairpin ribozyme designed to bind to and cleave the 5' end of the viral genome, abolishing the reverse transcription and integration of progeny virus.

example, it is unclear, (a) whether efficient subcellular colo-calization can be obtained, in particular, in vivo, (b) whether the target RNA will be efficiently recognized due to secondary and tertiary folding of the target RNA, or (c) whether RNA-binding proteins would prevent efficient binding. Thus, more experimentation will be necessary to address these problems (39-52). Efforts are also underway to develop multimeric/ multivalent ribozymes for targeting all major clades of HIV-

1, besides optimizing expression potential of these labile moieties (53-55).

2. RNA Decoys

In contrast to ribozymes and antisense RNAs, RNA decoys do not attack the viral RNAs directly. RNA decoys are mutant RNAs that resemble authentic viral RNAs and have crucial functions in the viral life cycle. They mimic such RNA structures and decoy viral and/or cellular factors required for the propagation of the virus (39,56-64). For example, HIV-1 replication largely depends on the 2 regulatory proteins Tat and Rev. These proteins bind to specific regions in the viral RNA, the transactivation response (TAR) loop and the Rev response element (RRE), respectively. Tat binding to TAR is crucial in the initiation of RNA transcription, Rev binding to RRE is essential in controlling splicing, RNA stability, and the trans port of the viral RNA from the nucleus to the cytoplasm. These 2 complex secondary RNA structures within the HIV-1 genome appear to be unique for the HIV-1 virus and no cellular homologous structures have been identified. Thus, such structures appear to be valuable targets for the attack with genetic antivirals.

The strategy here is to endow HIV-1 target cells with genes that overexpress short RNAs containing TAR or RRE sequences. The rationale for this is to have RNA molecules within the cells in abundance, which will capture Tat or Rev proteins, preventing the binding of such proteins to their actual targets. Based on Tat nuclear localization capabilities, recently, a chimeric small nucleolar RNA-TAR decoy that localizes to the nu-cleolus and nucleoli of human cells has also been described (65,66). As such it will be possible to halt early HIV-1 gene expression. Encouraging in vitro data from combination anti-HIV-1 gene therapy approaches using multiple anti-HIV-1 gene therapy moieties (i.e., antisense, decoy and trans-dominant negative mutants) supports its possible future clinical applications (67,68). Combination gene therapy is envisioned as an alternative for HIV-1 therapy due to its labile genome (69). Consequently, HIV-1 replication is markedly impaired. This strategy has the advantage over antisense RNAs and ribozymes in that mutant Tat or Rev, which will not bind to the RNA decoys will also not bind to their actual targets. Thus, the likelihood that mutant strains would arise that would bypass the RNA decoy trap is low. However, it still remains to be elucidated, whether cellular factors also bind to Tat or Rev decoys, and whether overexpression of decoy RNAs would lead to the sequestering of the resulting protein-RNA complexes in the cell.

3. Small Inhibitory RNA

RNA interference (RNAi) is a newly recognized phenomenon in the field of HIV-1 gene therapy with potential promises (70). The inhibitory mechanism involves posttranscriptional gene silencing. RNAi exploits double-stranded ribonucleic acid (dsRNA) oligonucleotides to inhibit expression of the gene with sequence similarity to the input oligonucleotide (Fig. 4). A 21-25 mer dsRNA oligonucleotide is used to interfere with the messenger RNAs (mRNAs) of a specific gene. One of the major prerequisites for achieving RNA interference is a perfect and unique match between the dsRNA oligo with mRNA that produces protein. The dsRNA binds to the target mRNA and makes it incomompetent for translation to protein. At molecular levels, a 2-step process involves editing of the dsRNA oligo by a protein complex called DICER, an RNAse-like enzyme, followed by another complex, called the RNA inhibitory silencing complex, that digests mRNA with similar sequences and hence silences that gene of interest.

There could be 3 major potential avenues for RNAi technology in HIV-1 therapeutic development. HIV-1 infection could be controlled by reducing host cellular targets such as CD4, CXCR4, and CCR5 or by silencing early HIV-1 gene products, Tat and Rev. HIV-1 integrase and gag could also be potential targets due to their crucial role in integration and morphogenesis, respectively. Several recent in vitro studies

Figure 4 RNA interference model. The mechanism of action of dsRNA interference involves a key enzyme, RNAse III, commonly known as Dicer, that processes dsRNA to siRNA. Dicer generates siRNA duplexes having 5'-phosphate and free 3'-hydro-xyl groups. This is followed by siRNA duplexes incorporation into siRNA-containing ribonucleoprotein complexes to generate RNA-induced silencing complexes (RISC) endonuclease. The RISC further mediates sequence-specific target RNA degradation.

Figure 4 RNA interference model. The mechanism of action of dsRNA interference involves a key enzyme, RNAse III, commonly known as Dicer, that processes dsRNA to siRNA. Dicer generates siRNA duplexes having 5'-phosphate and free 3'-hydro-xyl groups. This is followed by siRNA duplexes incorporation into siRNA-containing ribonucleoprotein complexes to generate RNA-induced silencing complexes (RISC) endonuclease. The RISC further mediates sequence-specific target RNA degradation.

targeting CD4, Tat, Rev, and Gag mRNA document the strength of this technology (71-76).

The in vitro findings of HIV-1 control by RNAi serve as a proof of principle that RNAi technology can be employed to suppress multiple steps in the HIV-1 life cycle (77). Moreover, targeting of HIV-1 coreceptors, such as CXCR4, CCR5, APJ, and a number of others, might be a promising future for this technology in HIV-1 gene therapy. An advantage of such interference would be its lowered effect on normal immune functions, one of the hurdles in conventional gene therapy.

RNAi technology will be facing several challenges before it acceptance in clinical scenarios. A major issue will be the stability of small oligos used for therapeutic purposes. In spite of such challenges, strength of this technology is the availability of data through the Human Genome Project to cross-reference redundancy of siRNAs selected for silencing purposes. Development of a vector-based strategy to target a combina tion of viral and cellular genes, essential for HIV-1 infection, will further add to the strength of this therapeutic intervention.

B. Protein-based Inhibitors of HIV-1 Replication

1. Transdominant Mutant Proteins

During HIV-1 replication, several regulatory proteins are essential for viral gene expression and gene regulation. Mutant forms of such proteins greatly reduce the efficiency of viral replication (78). Transdominant (TD) mutants are genetically modified viral proteins that still bind to their targets but are unable to perform their actual function. They compete with the corresponding native, wild-type protein inside the cell. The competition of several TD proteins with the wild-type counterpart has been shown to greatly reduce virus replication, especially when such TD mutants are expressed from strong promoters (the cytomegalovirus immediate early promoter, CMV-IE) (19,26,79-88).

For example, transcription from the HIV-1 long terminal repeat (LTR) promoter is dependent on the Tat protein. Mutant Tat proteins, which still bind to the nascent viral RNA, but which are unable to further trigger RNA elongation of transcription greatly reduce the production of HIV-I-RNAs and consequently the production of progeny virus. In a similar way, mutant Rev proteins interfere with regulated posttran-scriptional events and also greatly reduce the efficiency of virus replication in an infected cell.

Although TD mutants have been shown to be effective in vitro, it still remains unclear how long cells endowed with such proteins will survive in vivo. An exciting area of transdominant protein research is exploring inhibitors of host proteins that interact with HIV-1 proteins. One such report using a protein that interacts with HIV-1 integrase [i.e., integrase interactor 1 also known as hSNF5] showed inhibitory effects on HIV-1 replication (89) There is a significant possibility that peptides of such proteins will be displayed via HLA leading to the destruction of the HIV-I-resistant cell by the patient's own immune system. There is, however, the potential to express such proteins from the HIV-1 LTR promoter, which only becomes activated upon HIV-1 infection. However, this would rule out that a TD Tat can be used, because TD Tat may also abolish its own expression. Even if other TD proteins are expressed from inducible promoters, it still remains unclear, whether such inducible promoters are really silent enough so that no protein is made (and no immune response) as long as there is no viral infection.

2. Toxic Genes

Another approach to reduce the production of progeny virus is to endow the target cells of HIV-1 with toxic genes, which become activated immediately after virus infection. The activation of the toxic gene leads to immediate cell death; therefore, no new progeny virus particles can be produced. Theoretically, this would lead to an overall reduction of the virus load in the patient. In vitro experiments have shown that the production of HIV-1 virus particles was indeed reduced, if target cells had been endowed with genes coding for the herpes simplex virus thymidine kinase or a mutant form of the bacterial diphtheria toxin protein. Such genes were inserted downstream of the HIV-1 LTR promoter, which only becomes activated upon HIV-1 infection, when the viral Tat protein is expressed (90-94).

Besides the question regarding the ''silence'' of the HIV-1 LTR promoter without Tat (discussed above), the main problem with this approach is the actual number of cells that carry a toxic gene present in the patient. Because the HIV-1 virus will not only infect cells that carry the toxic gene, but also many other cells of its host, this approach may only ''slow down'' virus replication for a short period of time until all cells that carry the toxic genes undergo self-destruction upon infection.

3. CD4 as Decoy

The CD4 molecule is the major receptor for the HIV-1 virus for entry into T lymphocytes. Thus, in a similar way to RNA decoys, mutant CD4, which stays inside the endoplasmatic reticulum has been shown to inactivate HIV-1 envelope maturation, preventing formation of infectious particles. In another approach, soluble CD4 has been used to block the envelope of free extracellular virus particles and to prevent binding to fresh target cells (95). However, the question remains if soluble and/or mutant CD4 in the blood of the patient will also serve as a trap for natural CD4 ligands leading to the impairment of important physiological functions (96-98).

4. Single-chain Antibodies

Single-chain antibodies (scA) have originally been developed for Escherichia coli expression to bypass the costly production of monoclonal antibodies in tissue culture or mice (99,100). They comprise only the variable domains of both the heavy and the light chain of an antibody. These domains are expressed from a single gene, in which the coding region for these domains are separated by a short spacer sequence coding for a peptide bridge, which connects the 2 variable domain peptides. The resulting scA [also termed single-chain variable fragment (scFv)] can bind to its antigen with similar affinity as a Fab fragment of the authentic antibody molecule.

scFvs have been developed by our group and others to combat HIV-1 replication, when expressed intracellularly (32,97,101-107). Both pre- (e.g., integrase, reverse transcriptase, matrix protein) and post- (e.g., Rev and Tat) integration sites of the viral life cycle have been targeted, with varying success (108). An HIV-1 coreceptor CXCR4 scFv has shown potent inhibitory effects on HIV-1 replication (109), and future development of CCR5 scFvs will further assist in HIV-1 control. Further studies using constructs combining multiple scFvs for potential synergistic antiviral potency are under development and should be of importance in developing robust anti-HIV-1 molecular therapeutics.

IV. GENETIC "GUNS" TO DELIVER GENETIC ANTIVIRALS

In all therapeutic approaches listed above, the therapeutic agent cannot be delivered directly to the cell. Instead, the corresponding genes have to be transduced to express the therapeutic agent of interest within the target cell. Genes can be delivered using a large variety of molecular tools. Such tools range from nonviral delivery agents (liposomes or even naked DNA) to viral vectors. Because HIV-1 remains and replicates in the body of an infected person for many years, it will be essential to stably introduce therapeutic genes into the genome of target cells for either continuous expression or for availability upon demand. Thus, gene delivery tools such as naked DNA, liposomes, or adenoviruses (AV), which are highly effective for transient expression of therapeutic genes, may not be useful for gene therapy of HIV-1 infection.

Adeno-associated virus (AAV), a nonpathogenic single-stranded DNA virus of the parvovirus family, has recently gained a great deal of attention as a vector because it is not only capable of inserting its genome specifically at one site at chromosome 19 in human cells, but it is also capable of infecting nondividing cells. However, vectors derived from AAV are much less efficient and lose their ability to target chromosome 19 (110,111). Another shortcoming of AAV is the need for it to be propagated with replication-competent AV because AAV alone is replication defective. It also remains to be shown how efficiently AAV vectors transduce genes into human hematopoietic stem cells and/or mature T lymphocytes and macrophages. Because gene therapy of HIV-1 infection may also require multiple injections of the vector (in vivo gene therapy, see below) or of ex vivo manipulated cells, it also remains to be shown whether even small amounts of contaminating AV, which is used as a helper agent to grow AAV, will cause immune problems.

The most efficient tools for stable gene delivery are ret-roviral vectors (112-118), which stably integrate into the genome of the host cell, as this is a part of the retroviral life cycle (Fig. 1). This is why first virus-based gene delivery systems have been derived from this class of viruses, and continued development toward this class of safer vectors will lead to more suitable gene therapy system (119,120). This is also why they are being used in almost all current human gene therapy trials, including ongoing clinical AIDS trials.

Retroviral vectors are basically retroviral particles that contain a genome in which all viral protein-coding sequences have been replaced with the gene(s) of interest. As a result, such viruses cannot further replicate after 1 round of infection. Furthermore, infected cells do not express any retroviral proteins, which makes cells that carry a vector provirus (the integrated DNA form of a retrovirus) invisible to the immune system (112-118).

A. Retroviral Vectors Derived from C-Type Retroviruses

All current retroviral vectors used in clinical trials have been derived from murine leukemia virus (MLV), a C-type retroviruses with a rather simple genomic organization (Fig. 5). MLV contains only 2 gene units, which code for the inner core structure proteins and the envelope protein, respectively. It does not contain regulatory genes such as HIV-1. Thus, the construction of safe gene delivery systems is rather simple and straightforward. Such delivery systems consist of 2 components: the retroviral vector, which is a genetically modified viral genome, that contains the gene of interest replacing ret-roviral protein coding sequences, and a helper cell that supplies the retroviral proteins for the encapsidation of the vector genome into retroviral particles (Fig. 5). Modern helper cells contain separate plasmid constructs, which express all ret-roviral proteins necessary for replication. After transfection of the vector genome into such helper cells, the vector genome is encapsidated into virus particles (due to the presence of specific encapsidation sequences). Virus particles are released from the helper cell carrying a genome containing only the gene(s) of interest (Fig. 5). Thus, once established, retrovirus helper cells can produce gene transfer particles for very long time periods (e.g., several years). In the last decade, several retroviral vector systems have also been derived from other C-type chicken retroviruses (112,114,118).

B. Retroviral Vectors Derived from HIV-1

Retroviral vectors derived from MLV have been shown to be very useful to transfer genes into a large variety of human cells. However, they poorly infect human hematopoietic cells, because such cells lack the receptor, which is recognized by the MLV envelope protein. Furthermore, retroviral vectors derived from C-type retroviruses are unable to infect quiescent cells: such viruses (and their vectors) can only establish a provirus after 1 cell division, during which the nuclear membrane is temporarily dissolved. Thus, efforts are underway in many laboratories to develop retroviral vectors from lentivi-ruses [e.g., HIV-1 or the simian immunodeficiency virus (SIV), which are able to establish a provirus in nondividing cells (although the mechanism by which these viruses penetrate the nucleus is not fully understood)] (121). However, the fact that lentiviruses contain several regulatory proteins, which are essential for virus replication, makes the construction of lentiviral packaging cells more complicated. Furthermore, the fact that the lentiviral envelope proteins (e.g., that of HIV-1) can cause syncytia and/or that some viral regulatory proteins are toxic to the cells further hampers the development of stable packaging lines.

The ''envelope problem'' has been solved by generating packaging cells, which express the envelope protein of MLV or the envelope of vesicular stomatitis virus (VSV). Such envelope proteins are efficiently incorporated into lentiviral particles. The second and major problem for generating stable packaging lines is the toxicity of some retroviral regulatory proteins to the cell. Thus, retroviral vectors can only be generated in transient systems: 293T cells (human embryonic kidney cell line, highly susceptible for transfecting DNAs) are simultaneously transfected with all plasmids constructs to express the particle proteins and the vector genome. Figure 6 shows plasmid constructs used to make HIV-1-derived packaging cells. Vector virus can be harvested from the transfected cells for a limited time period and can be used to infect fresh target cells. Although this gene transfer system has been

Figure 5 Retroviral helper cells derived from C-type retroviruses. A C-type retroviral provirus (the DNA intermediate of a retrovirus is shown on the top). The protein coding genes (gag-pol and env) are flanked by cis-acting or controlling sequences, which play essential roles during replication. (Below) In a retroviral helper cell, the retroviral protein coding genes, which code for all virion proteins, are expressed (ideally) from heterologous promoters (pro) and polyadenylated via a heterologous polyadenylation signal sequence [poly(A)]. To minimize reconstitution of a full-length provirus by recombination, the gag-pol and env genes are split to different gene expression vectors. In the retroviral vector, the viral protein coding sequences are completely replaced by the gene(s) of interest. Because the vector contains specific encapsidation sequences (E), the vector genome is encapsidated into retroviral vector particles, which bud from the helper cell. The virion contains all proteins necessary to reverse transcribe and integrate the vector genome into that of a newly infected target cell. However, because there are no retroviral protein coding sequences in the target cell, vector replication is limited to 1 round of infection. LTR, long terminal repeat.

Figure 5 Retroviral helper cells derived from C-type retroviruses. A C-type retroviral provirus (the DNA intermediate of a retrovirus is shown on the top). The protein coding genes (gag-pol and env) are flanked by cis-acting or controlling sequences, which play essential roles during replication. (Below) In a retroviral helper cell, the retroviral protein coding genes, which code for all virion proteins, are expressed (ideally) from heterologous promoters (pro) and polyadenylated via a heterologous polyadenylation signal sequence [poly(A)]. To minimize reconstitution of a full-length provirus by recombination, the gag-pol and env genes are split to different gene expression vectors. In the retroviral vector, the viral protein coding sequences are completely replaced by the gene(s) of interest. Because the vector contains specific encapsidation sequences (E), the vector genome is encapsidated into retroviral vector particles, which bud from the helper cell. The virion contains all proteins necessary to reverse transcribe and integrate the vector genome into that of a newly infected target cell. However, because there are no retroviral protein coding sequences in the target cell, vector replication is limited to 1 round of infection. LTR, long terminal repeat.

shown to be functional, it is not highly efficient and better packaging cells still need to be developed. In addition, many questions regarding the safety of such vectors still need to be addressed. It is known that plasmid DNAs can recombine with each other very efficiently immediately after transfection. Thus, the question needs to be addressed, whether there is a chance that replication-competent viruses arise by recombination, which may cause a disease in gene-transduced patients.

C. Cell Type-specific Retroviral Vectors

All retroviral vectors currently used in human gene therapy trials contain the envelope protein of amphotropic (ampho) MLV or VSV. Ampho-MLV and VSV have a broad host range and can infect various tissues of many species, including humans. Thus, the use of vectors containing such envelope proteins enables the transduction into many different human tissues. However, due to this broad host range, gene transfer has to be performed ex vivo. If injected directly into the bloodstream, the chances that the vector particles would infect their actual target cells are very low. Furthermore, such vector particles may infect germ line cells (which are continuously dividing). Thus, the target cells have to be isolated and the gene transfer is being performed in tissue culture. Gene transduced cells are then selected and reintroduced into the patient.

However, this protocol has major shortcomings in regard (not only) to gene therapy of HIV-1 infection. First, it is very expensive and requires highly trained personal. Second, human cells, which are kept in tissue culture change their physiological behavior and/or take up fetal bovine proteins (a component of the tissue culture medium) and display bovine peptides via histocompatibility antigen (HLA) on the cell sur-

Figure 6 Retroviral packaging system derived from HIV-1. (a) A provirus of HIV-1 is shown on the top. (b-d) These show plasmid constructs to express pseudotyped HIV-1 retroviral particles, and (E) shows a plasmid construct to encapsidate and transduce genes with an HIV-1 vector (the plasmid sequences to propagate such constructs in bacteria are not shown). Besides the genes encoding for HIV-1 proteins, which form the core of the virus (e.g., Gag, structural core proteins; P, protease; Pol, reverse transcriptase and integrase) and the envelope (e.g., env, envelope protein), the HIV-1 genome also codes for several regulatory proteins, which are expressed from spliced mRNAs and have important functions in the viral life cycle. (b) plasmid construct to express the core and regulatory proteins. To avoid encapsidation and transduction of genes coding for such proteins, the following modifications have been made: the 5' LTR promoter of the HIV-1 provirus has been replaced with the promoter of cytomegalovirus (CMV) to enable constitutive gene expression, the 3' LTR has been partially replaced with the polyadenylation signal sequence of simian virus 40 (polyA), the encapsidation signal has been deleted (A^), and the reading frame for the envelope and vpu genes have been blocked. (c) and (d) show plasmid constructs to express the envelope proteins of the vesicular stomatitis virus (VSV-G) or the envelope protein of murine leukemia virus (MLV), respectively. In the absence of HIV-1 envelope proteins, which are rather toxic to the cell, HIV-1 efficiently incorporates the envelope proteins of VSV or MLV into virions. The use of such envelopes also further reduces the risk of the reconstitution of a replication-competent HIV-1 by homologous recombination between the plasmid constructs. (E) shows a retroviral vector used to package and transduce a gene of interest (T gene) with HIV-1-derived vectors. Because the encapsidation sequence extends into the Gag region, part of the gag-gene (G) has been conserved in the vector. However, the ATG start codon has been mutated. The gene of interest is expressed from an internal promoter, because the HIV-1 LTR promoter is silent without Tat. sd, splice donor site.

face. Consequently, such cells become immunogenic and are eliminated by the immune system of the patient. To bypass such ex vivo protocols, efforts are now underway in many laboratories to develop cell type-specific gene delivery systems, which would involve injecting the gene delivery vehicle directly into the patient's bloodstream or tissue of interest. In the past few years, several attempts have been made to develop cell type-specific gene delivery tools, again with retrovirus-derived vectors leading the field.

The cell type specificity of a virus particle is determined by the nature of the retroviral envelope protein, which mediates the binding of the virus to a receptor of the target cell (122). Thus, experiments have been initiated in several laboratories to modify the envelope protein of retroviruses in order to alter the host range of the vector. One of the first attempts to specifically deliver genes into distinct target cells has been performed in the laboratory of Dr. H. Varmus. Using retroviral vectors derived from avian leukosis virus, these investigators incorporated the human CD4 molecule into virions to specifically transduce genes into HIV-1-infected cells (123). However, such particles were not infectious for unknown reasons. Recent reports indicate that MLV particles that carry CD4 can infect HIV-1-infected cells, although at very low efficiencies (124).

In another attempt to target retroviral particles to specific cells, Roux et al. have shown that human cells could be infected with eco-MLV, if they added 2 different antibodies to the virus particle solution (125,126). The antibodies were connected at their carboxy termini by streptavidine. One antibody was directed against a cell surface protein, the other antibody was directed against the retroviral envelope protein (125,126). Although this approach was not practical (infectiv-ity was very inefficient and was performed at 4°C), these experiments showed that cells that do not have an appropriate receptor for a particular virus can be infected with that virus, if binding to the cell surface had been facilitated. These data also indicated that antibody-mediated cell targeting with ret-roviral vectors was possible.

To overcome the technical problems of creating an antibody bridge, it was logical to incorporate the antibody directly into the virus particle. However, complete antibodies are very bulky and are not suitable for this approach. The problem has been solved using single-chain antibody technology (127-130). Using hapten model systems, it has been shown that retroviral vectors that contained scAs fused to the envelope are competent for infection (128,129). Retroviral vector particles derived from spleen necrosis virus (SNV, an avian retrovirus) that display various scAs against human cell surface proteins are competent for infection on human cells that express the antigen recognized by the antibody (127,128,130).

Most recently, it became possible to use scA-displaying SNV to introduce genes into human T cells with the same high efficiency obtained with vectors containing wild-type envelope. In all such experiments, the wild-type envelope of SNV had to be copresent in the virus particle to enable efficient infection of human cells. However, because SNV vector particles with wild-type SNV envelope do not infect human cells at all, this requirement is not a drawback for using such vector particles for human gene therapy (131,132).

Although successful gene transfer using scA-displaying MLV vector particles has been reported from 1 laboratory (133), further experimentation in the laboratories of several other investigators revealed that MLV-derived vector particles that display various scAs are not competent for infection in human cells (125,126,134-137). The difference between MLV and SNV cell-targeting vectors is certainly based on the different features of the wild-type envelope and the mode of virus entry (138). Moreover, wild-type ampho-MLV infects human hematopoietic cells extremely poorly, most likely due to the absence of an ampho-MLV receptor on such cells (139). Thus, MLV-derived vectors are certainly not the best candidates for human gene therapy of AIDS and alternative vectors need to be developed.

D. Other New Potential Vector Systems

Most recently, other interesting attempts have been made to combat HIV-1-infected cells. Recombinant VSV has been engineered, which lacks its own glycoprotein gene. Instead, genes coding for the HIV-1 receptor CD4 and a chemokine coreceptor, CXCR4, have been inserted. The corresponding virus was able to efficiently infect HIV-1-infected cells, which display the HIV-1 glycoprotein on the cell surface. Because VSV is a virus that normally kills infected cells, the engineered virus only infects and kills HIV-1-infected cells. It has been reported that this virus indeed reduced HIV-1 replication in tissue culture cells up to 10,000-fold (140). This novel approach to combat one virus with another will certainly gain a great deal of further attention. However, it remains to be shown how effective this approach will be in vivo. Will the ''antivirus'' succeed in eliminating a large load of HIV-infected cells before it will be cleared by the immune system? On the other hand, because HIV-1 preferentially kills activated immune cells, will it destroy the immune cells that are attempting to clear the body from its own ''enemy''? How will the body tolerate a virus that does not look like one because it carries human cell surface proteins on the viral surface? The answer to these and other questions are eagerly awaited (141).

Was this article helpful?

0 0
How To Bolster Your Immune System

How To Bolster Your Immune System

All Natural Immune Boosters Proven To Fight Infection, Disease And More. Discover A Natural, Safe Effective Way To Boost Your Immune System Using Ingredients From Your Kitchen Cupboard. The only common sense, no holds barred guide to hit the market today no gimmicks, no pills, just old fashioned common sense remedies to cure colds, influenza, viral infections and more.

Get My Free Audio Book


Post a comment