CCR-5 and CXCR-4 co-receptors have specific chemokine ligands/antagonists that possess the ability to block the virus infection. The molecules that bind to the co-receptors can be divided into four categories: naturally occurring chemokines and their derivatives, peptides and small molecules (< 1 kDa), and Mabs, which recognize epitopes on for instance the extracellular domains of certain receptors.

Examples of chemokine ligands (beta-chemokines) that inhibit infection of R5 isolates include RANTES, a physiological ligand for the HIV-1 co-receptors CCR3 and CCR5. RANTES is actively secreted by normal T-cells. Derivatives of this peptide have been used, including aminooxypentane (AOP)-RANTES 9, and from a recent study, N^α-(n-nonanoyl)-des-Ser¹ [L-thioproline², L-α-cyclohexyl-glycine³] RANTES (PSC-RANTES) 10. RANTES is an antagonist that besides having the ability to interact with CCR5 also has a downregulating effect on the co-receptor. RANTES can however induce chemotaxis and promote unwanted inflammatory side effects. Therefore AOP-RANTES was created by chemical modification of the amino terminus. This analogue does not promote any inflammatory side effects, and in addition it can prevent chemotaxis induced by e.g. RANTES. AOP-RANTES is a very strong antagonist that has a high affinity for CCR5, elicits rapid endocytosis of CCR5, and prevents recycling of the co-receptor back to the surface. PSC-RANTES is chemically identical to native RANTES except for the substitution of a nonanoyl group, thioproline, and cyclohexylglycine for the first three N-terminal amino acids of the native protein. This analogue acts in the same way, but has shown more potent in vitro antiviral activity than AOP-RANTES. Furthermore, it has successfully protected rhesus macaques from intravaginal exposure to a chimeric simian/human immunodeficiency virus containing an R5-tropic envelope of HIV-1 10.

In addition to RANTES and its derivatives, the chemokine ligands macrophage inflammatory proteins 1alpha/beta (MIP-1alpha and MIP-1beta) also show an inhibiting effect by mediating a receptor blockade 8111213. Examples of chemokine ligands that inhibit infection of X4 isolates include stromal cell-derived factor-1alpha (SDF-1alpha) and its derivatives that inhibit HIV-1 fusion and entry by minimizing the accessibility to the co-receptor on the cell surface and by inhibiting the SU-CXCR4 interaction 9111213.

The CCR5 amino-terminal domain is thought to play an important role in virus fusion and entry. This knowledge has been utilized in the development of anti-CCR5 Mabs whose epitopes include residues in the amino-terminal domain. Mabs of this kind strongly inhibit SU binding to CCR5 but only moderately inhibit HIV-1 fusion and entry 12. Another type of Mab, the anti-ECL2 Mab whose epitopes include residues from one of the extracellular loops on CCR5 (ECL2), potently inhibits HIV-1 fusion and entry, but only moderately inhibits SU binding 12. PRO 140, also an anti-CCR5 Mab, inhibits viral fusion with the cell membrane at concentrations that do not prevent the CCR5 chemokine receptor activity. It binds a complex epitope spanning multiple extracellular domains on CCR5, and although it acts as a weak antagonist it does not induce signaling or downregulation of CCR5. It is thought that the antiviral effect is exerted through a mechanism involving receptor blockade 14. Mab 12G5 is a monoclonal antibody that recognizes an epitope on CXCR4. This epitope is also present in ECL2, and binding inhibits HIV-1 fusion 1215. A potential disadvantage of this strategy is that binding of the antibody to a receptor may trigger unwanted signal transduction 1416.

Peptides, resembling the CCR5 transmembrane helices, inhibit HIV-1 replication and chemokine signaling by disrupting helix-helix interactions, which may influence the CCR5 structure 12. T22 is a positively charged cyclic 18-mer antimicrobial peptide, which presumably inhibits SU-CXCR4 interaction by associating with the negatively charged surface of CXCR4 812. A truncated form of SDF-1alpha, consisting of the 16 amino-terminal residues of SDF-1alpha, also seems to possess such a blocking effect 12.

Recently, a new kind of CCR5 antagonist has been discovered in a protozoan parasite, Toxoplasma gondii 17. This protein, cyclophilin-18 (C-18), has several potential antiviral properties including CCR5 binding, induction of the production of interleukin-12 (IL-12) from murine dendritic cells, inhibition of fusion and infectivity of R5 isolates by co-receptor antagonism and blocking of syncytia formation.

Small organic molecules, such as AMD3100, potently inhibit HIV-1 replication by an interaction with residues present on one of the CXCR4 extracellular loops, ECL2, and residues within a transmembrane helix, TM4. Upon binding to these residues AMD3100 spans the main ligand-binding cavity of CXCR4, which probably constrains the co-receptor in an inactive conformation 12.

Individuals with a homozygous deletion in the gene encoding CCR5 are healthy and protected against HIV-1 transmission, which suggests that down regulation may not pose any clinical side effects. This knowledge has led to the development of strategies that directly target the mRNA encoding CCR5 or CXC4, either by ribozymes 618, anti-sense RNA 61819 or RNAi 20. The latter strategy, the siRNA approach, has led to successful blocking of HIV-1 entry, protection of cells from infection and delay of virus replication 21222324. Interestingly, it is thought that single-stranded siRNAs (the anti-sense strand of a siRNA duplex) act through the same RNAi pathway, but at a later stage than double-stranded siRNA, thereby requiring less time to exert their antiviral activity 2125.

Soluble CD4 (sCD4) is an anti-HIV-1 protein, which can be expressed and secreted from genetically engineered cells. It is a truncated form of the CD4 receptor, composed of the ectodomain that inhibits laboratory-adapted strains of HIV-1. sCD4 probably prevents the binding of the virus to the cell, by binding directly to Env, or indirectly by inducing or repressing cellular factors that influence the viral gene expression 1819.

When sCD4 binds to SU it acts by extending the distance to TM, which inhibits the fusion. But when used against primary isolates, sCD4 was much less successful because of a lower affinity for sCD4. Surprisingly, some isolates became more infectious upon sCD4 treatment. An explanation for this may be that an interaction between the SU protein and sCD4 induces changes in SU, allowing it to bind the co-receptor with higher affinity or increased kinetics. In addition this interaction can eventually facilitate the fusion of HIV-1 with CD4^- cells expressing the co-receptor 13. This has led to the development of a tetrameric version of sCD4, PRO542, in which the SU-binding region of CD4 is fused to the conserved region of human immunoglobulin IgG2. This fusion protein has a high affinity for the SU protein and has shown promising results in phase I clinical trials 813.

siRNA-directed silencing of CD4 mRNA expression has been shown to specifically inhibit HIV-1 entry and thus HIV-1 replication 2324. However, CD4 silencing in vivo may interfere with its role in normal immune function. Thus an approach targeting the CD4-binding domain of the SU protein would be more relevant. This has successfully been achieved by expressing a 0.5 kb dsRNA containing the major CD4-binding domain of the SU protein, as the target of the env gene. By this approach it has been possible to significantly suppress the expression of the HIV-1 CA-p24 antigen in human peripheral blood mononuclear cells (PBMCs) and in HeLa-CD4⁺ for a relatively long period of time 26.

Strategies based on the intracellular expression of antibodies specific for the HIV-1 envelope (anti-SU) have also been shown to inhibit virus replication. This strategy is based on the usage of sFvs, containing the smallest structural domain that still possesses complete antigen and binding-site specificity of the parental antibody. They are secreted into the medium where they probably act as inhibitors by direct interaction with the viral proteins 18 to neutralize the virus 19.

Cyanovirin (CV-N), an 11 kDa protein originally isolated from cyanobacteria, potently inactivates diverse strains of HIV-1. It has a high affinity for the SU protein, and when bound it inhibits the SU-CD4 interaction. CV-N possesses the advantage that even high concentrations are non-toxic and it is an extremely stable protein. CV-N has also been coupled to a cytotoxin (Pseudomonas exotoxin), thereby selectively killing HIV-1 infected SU-expressing cells 15.

As the SU protein binds to CD4, it initiates conformational changes in SU, making the interaction between the SU protein and the co-receptors more favorable. After attachment to the co-receptor further conformational changes occur in both the SU and TM proteins, thus weakening their interaction. During this process a transitory pre-hairpin intermediate of the TM protein is created, freeing the previously buried fusion peptide to interact with the host-cell membrane. This exposes the N-peptide and the C-peptide regions on the pre-hairpin intermediate that have been targets for several inhibiting strategies including synthetic C-peptides, N-peptides and sFvs.

C-peptides are based on the C terminal end of the fusion peptide, and mimics this part of the fusion peptide when it has its correct fusogenic conformation. T-20, a 36-amino acid C-peptide, is a potent inhibitor of HIV-1 infection. It acts through a dominant negative mechanism and interacts by binding to a conserved domain on the N-peptide present in the pre-hairpin intermediate. The function of this domain is to mediate a structural change, which allows the pre-hairpin intermediate to form a fusogenic hairpin state. Binding of T-20 inhibits this process and thereby impedes fusion. Disadvantages of the C-peptide strategy are the cost of C-peptide synthesis and the relatively large amounts necessary for an antiviral effect. In addition, their size makes them non-amenable to oral routes of entry and they must be injected instead 132728.

The 5-Helix is a 25 amino acid N-peptide consisting of five of the six helixes constituting the C-peptide. The peptide is presumed to inhibit fusion, through binding with high affinity to the C-peptide. However because N-peptides have a strong tendency to aggregate the inhibition could also be due to their intercalation into the TM amino-terminal coiled coil 2728.

A third kind of peptides named D-peptides have also proven effective. These peptides are small 16–18 D-amino acids residues that specifically bind to three hydrophobic pockets present at the end of the N-peptide. Since such peptides are unnatural, they are resistant to proteolytic degradation, which makes them attractive for clinical use 1327.

Recently, a non-neutralizing antibody directed against epitopes exposed on the fusion peptide has been reported to possess antiviral effect 29. This antibody does not neutralize HIV-1 entry when produced as a soluble protein. However, when expressed on the cell surface as a membrane-bound sFv, it is turned into a neutralizing antibody, which markedly inhibits HIV-1 replication and cell-cell fusion by a mechanism that is thought to involve an interaction with the exposed fusion peptide. This results in inhibition of the subsequent fusion process. In the same study, this sFv was targeted into the ER and trans-Golgi network of HIV-1 susceptible cell lines where it was found to significantly block the maturation process of the viral Env protein resulting in an impairment of viral assembly.

Inhibiting strategies against RT involve the utilization of nucleosides and non-nucleosides. The nucleoside analogues lack the 3'-hydroxyl group, prevent the continued polymerization of the DNA chain, and are usually named nucleoside reverse transcriptase inhibitors (NRTIs). Clinically approved examples include Zidovudine (AZT), Didanosine (ddI), Zalcitabine (ddC), Lamivudine (3TC), Abacavir succinate and Stavudine (d4T) 8.

The non-nucleoside analogues, often referred to as non-nucleoside reverse transcriptase inhibitors (NNRTIs), act at the same step in the viral life cycle as the nucleoside analogous, but by a significantly different mechanism. Instead of acting as false nucleosides, the NNRTIs bind directly and non-competitively to RT in a way that inhibits the enzyme's activity. Examples of clinically approved NNRTIs include Nevirapine, Delaviridine and Efavirenz 8.

NRTIs bind to the deoxynucleoside triphosphate-binding pocket, which is formed partly by the template-primer nucleic acid and partly by the protein surfaces. NNRTIs bind to a hydrophobic pocket exclusively present in the RT enzyme of M subtype HIV-1. When used in combination they have a more pronounced antiviral effect.

The RNA decoy strategy aimed at RT involves the expression of RNAs lacking the PBS region, thus preventing it from acting as template for reverse transcription. The RNA competes with HIV-1 RNA when RT makes the first jump during the first strand transfer 6. Another decoy was designed to be co-packaged together with genomic RNA into new virions where it competes subsequently with genomic RNA for RT binding 6. Also, a designed tRNA₃^Lys mutant containing an 11 nucleotide 3'-end complementary to the HIV-1 TAR region, shows an inhibiting effect. This mutant competes with cellular tRNA₃^Lys for the binding to RT and primes reverse transcription from the TAR region instead of the PBS region 6.

Other strategies against the RT enzyme involve the usage of small peptides, about 15–19 amino acids long, that inhibit RT activity by interfering with the dimerization process of the RT enzyme. The amino acid sequence corresponds to the so-called connection domain of RT, in particular a tryptophan-rich 19-mer sequence corresponding to residues 389–407, which efficiently inhibits viral replication 30. Likewise, strategies based on intracellular expression of sFvs 718 and RNA aptamers 313233 targeted against the RT enzyme are potent inhibitors of HIV-1 replication. The aptamers all recognize the same template-primer-binding cleft on RT. Some of these RNA aptamers have the potential to form pseudoknot-like secondary structures, which mimic the conformation of the template-primer when associated with the RT enzyme. Thus, these aptamers are termed template analogue RT inhibitors (TRTIs). Selectivity of the RNA aptamers is directly related to their three-dimensional structure. Utilization of the TRTI aptamers has the following benefits: 1) Aptamers have a unique specificity and a strong binding affinity for the RT enzyme. 2) Aptamers inhibit the RT enzyme competitively and will unlikely inhibit other viral or cellular proteins, thus minimizing the risk for any appreciable toxic side effects. 3) Since aptamers are expressed in the infected cell, the aptamers will be co-packaged into new virions and inhibit the next round of replication. 4) Because of the large interface of the aptamer-binding pocket, the risk of escape mutants is significantly reduced. Furthermore, mutations in essential binding domains, such as the template-primer-binding pocket, will likely impair the binding of the RT enzyme to the viral genome 31.

Anti-sense RNAs designed to be complementary to the Psi-gag and the U3-5'UTR-gag-env regions have been shown to inhibit RT in new virion particles. They are co-packaged together with the genomic RNA into the virus progeny, and inhibit reverse transcription by hybridizing to the genomic RNA 6.

siRNAs directed against several regions of the HIV-1 genome, including the viral long terminal repeat (LTR) and the accessory genes vif and nef have provided evidence that the viral genomic RNA, as it exists within the virion as a nucleoprotein reverse transcription complex, is amenable to siRNA-mediated degradation 34. In addition, siRNAs targeted against the RT gene alone have shown potent inhibition of HIV-1 replication in MAGI cells 35.

Hammerhead ribozymes targeted against the HIV-1 gag region will cleave the viral RNA before reverse transcription is completed 6. Hairpin ribozymes, designed to cleave a conserved site in the U5 region of the HIV-1 RNA can likewise inhibit replication 6. Especially the tRNA^Val-U5-ribozyme has shown promising results and is currently being tested in clinical trials. Moreover, hairpin and hammerhead ribozymes targeted against the HIV-1 pol region also show promising results 636. In the latter strategy a hammerhead ribozyme has successfully been packaged into virions by linking it to the portion of the HIV-1 genome that provides the packaging sequence 36. This intravirion targeting ribozyme has in the same study shown higher virus-suppressing activity than a nonpackageable counterpart.

Since host tRNA₃^Lys is being packaged into new virus particles, this molecule is often used when ribozymes have to be co-packaged. An example is the tRNA₃^Lys-hammerhead ribozymes targeted against the PBS region. Besides cleaving the HIV-1 RNA, the tRNA₃^Lys-ribozyme inhibits reverse transcription by competing with host tRNA₃^Lys for RT binding and/or for the binding to the PBS sequence. Also, when bound to the PBS, the tRNA₃^Lys-ribozyme is unable to prime reverse transcription 6.

In a study closely related to the earlier mentioned CCR5 antagonist, C-18, human cyclophilin A (CyPA) has been shown to be incorporated into HIV-1 during virion assembly through interaction with an exposed proline-rich loop within the capsid domain of Gag 3738. CyPA is required for efficient viral replication but not for cell viability meaning that its cellular function is probably being compensated for by other factors. It has been proposed that CyPA enhances HIV-1 infectivity during early post-entry events, but may also be required for viral entry. The proposed molecular interaction that underlies this enhancement is the CyPA proteins ability to mask the binding site for the human host restriction factor Ref1 and thereby counteracting its inhibitory activity, allowing reverse transcription to be completed. In an attempt to reduce CyPA biosynthesis, two different anti-sense strategies were used 37. In one approach internal CyPA exons are skipped by means of modified derivatives of U7 small nuclear RNA (snRNA). U7 snRNA is the RNA component of the U7 small nuclear ribonucleoprotein (snRNP) involved in histone RNA 3'-end processing. By inserting appropriate anti-sense sequences into U7 snRNA it has successfully been converted from a mediator of histone RNA processing to an effector of alternative splicing. The other strategy involves the use of hairpin siRNA constructs targeting two different parts of the CyPA coding region. Both strategies greatly reduced the levels of CyPA, creating CEM-SS T-cells that sustain HIV-1 replication.

The next step is the translocation of the cDNA containing capsid into the nucleus. This process is mediated by independent pathways involving either the Vpr accessory protein, the matrix protein (MA) or the integrase (IN) protein. Vpr is thought to mediate the nuclear import of the preintegration complex through the nuclear pore complex (NPC) in non-dividing cells by interacting directly with proteins in the NPC. This transfer of viral DNA is mediated by a nuclear localization signal present in the Vpr protein. Furthermore, it has been shown that Vpr is involved in arresting HIV-infected cells in the G2 phase of the cell cycle, where the virus production has been shown to reach a maximum level 4567. Integration of proviral DNA is mediated by the viral IN enzyme by a process that requires the host protein integrase interactor 1 (INI1 / hSNF5). IN consists of an N-terminal zinc finger domain, a catalytic core domain, and a C-terminal domain that is important for binding HIV-1 LTR DNA 3940. It has two enzymatic functions; DNA cleavage and insertion of the provirus into the genome of the host 47. IN recognizes short inverted repeats (att sites) at both ends of the proviral DNA and cleaves an AT overhang at the 5' end. Then it catalyzes the non-specific cleavage of the host genome and the subsequently ligation of the 5' overhang to the cellular genome 4.

Several strategies aiming at the IN function have been reported:

IN has no known functional analogue in human cells and is therefore an appealing target for inhibiting strategies, which generally involves the usage of oligonucleotides, dinucleotides and different kinds of chemical agents, such as dicaffeoylquinic acids (DCQAs) 40 and 2,4-dioxobutanoic acid analogous 41. The integrase binding site in the U3 LTR region of the viral DNA contains a purine motif, 5'-GGAAGGG-3'. This motif has selectively been targeted by oligonucleotide-intercalator conjugates that interact with the viral DNA through triplex formation, thus blocking the catalytic functions of the IN enzyme 41. Disadvantages of these compounds include the low intracellular permeability and the high mutation rate of HIV-1 that may result in nucleotide substitutions in the LTR.

The inhibiting effect of a dinucleotide, named pdCpIsodU, is due to its ability to interact with the catalytic core domain 41. This molecule consists of a natural D-deoxynucleoside and an isomeric L-related deoxynucleoside joined together through a stereochemically unusual internucleotide phosphate bond, which makes the molecule resistant to 5'- and 3'-exonucleases. Through binding the molecule inhibits both the 3'-processing and the DNA strand transfer step.

DCQAs are non-competitive inhibitors that act by irreversible binding to the catalytic core domain. The exact chemical mechanism for this anti-IN activity is unknown, but it is thought to be caused by a simple redox-process. Two examples are 1-Methoxy-3,5-dicaffeoylquinic acid and 3,4-Dicaffeoylquinic acid. Both are relative non-toxic 40. Finally, 2,4-dioxobutanoic acid analogous have been reported to possess potent anti-IN activity through inhibition of the DNA strand transfer step 41.

sFvs interacting with different domains on IN have been isolated, and by fusion with a nuclease, a fusion protein is created that can interact with IN in the pre-integration complex, leading to cleavage of proviral DNA. Likewise IN-specific sFvs have been shown to be inhibitory to HIV-1 replication 718.

Finally, siRNAs targeted against the capsid protein, p24-siRNA, is thought to interact with the gag gene in the unspliced viral RNA when present in the cytoplasm. Thereby, the viral RNA genome is cleaved before integration occurs 232442.

The packaging signal Psi is a highly conserved RNA sequence and is therefore an obvious target for inhibition strategies. The Psi region contains four stem-loops located near the 5' major splice donor and the start of the gag open reading frame and is essential and sufficient for RNA packaging 45.

RNA containing a 1.43 kb region anti-sense to the Psi-gag region has been reported to inhibit packaging of genomic RNA and thus HIV-1 replication with a higher efficiency than the RevM10 mutant 6745. An antiviral effect has also been observed by targeting anti-sense RNAs against the Psi sequence, which act by hindering the NC domain of the Gag protein in recognizing the packaging signal. Likewise, ribozymes directed against the packaging signal have also been shown to have a significant effect on viral replication 76.

Since the Psi element is recognized by the NC domain, one strategy is to design a NC-nuclease fusion protein, which will recognize and cleave all unspliced HIV-1 RNAs. The cleavage process occurs in the cytoplasm and this will also inhibit the production of the Gag and Gag-Pol fusion proteins 5. Expression of the gag-gene has also been reduced by using anti-sense RNAs complementary to the 5'-leader-gag region or by using ribozymes directed against the gag transcripts 718.

The ability of the NC domains to recognize the genomic RNA can be blocked with RNA decoys containing the Psi sequence. These decoys may form dimers with HIV-1 RNA and thereby compete with HIV-1 RNA dimers for packaging into new virus particles 645.

The structural proteins, Gag and Env, form multimeric complexes during viral assembly. By means of different kinds of Gag TNPs it has been possible to inhibit this step. Besides inhibiting viral assembly, these Gag TNPs also interfere with the viral release process, the uncoating of the viral genome and the reverse transcription 18. A limitation of using Gag TNPs is that the gag-gene contains an inhibiting sequence, the CRS, which hinders the expression of the gag-gene if Rev is missing 4618. The potential effect of Env TNPs has also been tested, but this strategy shows relatively low antiviral activity when compared to Gag TNPs 18.

Given that one function of the Vif protein is to block early processing of the Gag protein, it has inspired the development of a Vif TNP that is missing the blocking function. Applying this protein means that the virus assembly is impaired, but a clear disadvantage is that HIV-1 has the potential to evolve escape mutants 66. Similar inhibition has been observed by directing anti-sense RNAs against the Vif encoding region 70. Another strategy relies on the INI1 protein that is the only known host protein directly interacting with HIV-1 integrase. A minimal integrase-binding fragment of INI1, S6, comprising amino acids 183–294, potently inhibits HIV-1 assembly, particle production and replication in a transdominant manner. This inhibiting effect results from direct interaction of the S6 protein with the integrase domain within the Gag-Pol polyprotein. When the S6 protein binds to integrase it is thought to interfere with proper multimerization of Gag and Gag-Pol by steric hindrance. In addition it affects maturation, blocks an interaction of the cellular assembly machinery with Gag-Pol and mediates the mislocalization of viral proteins into different sub-cytoplasmic compartments of the cell. Besides being non-toxic, another favorable feature of S6 is that it is unlikely to be immunogenic, because it is a truncated form of a host protein. In addition, virions with mutations in the S6 binding site will also be defective for interaction with INI1. This minimizes the risk for the development of HIV-1 escape mutants 39.

Recently, a specific inhibition of virus budding was demonstrated by overexpression of an amino-terminal fragment of tumour susceptibility gene 101 (Tsg101) 71. The role of Tsg101 is to participate in the endocytic trafficking pathway. It is presumed to bind to the Gag polyprotein and subsequently mediate the transport into multivesicular bodies (MVBs), which then carry their cargo towards the cell surface.

By targeting an intracellular sFv specifically against the CD4 binding region of the SU protein, it has been possible to make cells temporarily resistant to HIV-1 infection. This sFv, named sFv105, acts by binding to the Env protein, and traps Env in the ER. This prevents the maturation process in which Env is cleaved into the SU-TM proteins. As a result, Env is prevented from reaching the cell surface 718. Another way to inhibit the maturation of Env is by inhibiting the cellular protease furin 72. Furin is a member of the mammalian subtilisin-related proprotein convertases that mediate cleavage of the Env protein at a conserved Arg-Glu-Lys-Arg sequence. Peptides that mimic this sequence have been reported to block furin activity. Especially, a polyarginine peptide has shown promising results without showing any toxic side effects on cultured cells ex vivo or in mice in vivo.

Furthermore, anti-sense RNA approaches directed against the Env message also belong to the applied strategies 70.

Multitarget ribozymes that target different sites in the SU sequence also exert antiviral effects. Some of these designed ribozymes bind and cleave up to nine different conserved regions in the SU sequence 7.

A Gag-nuclease fusion protein can be packed into new virions and thereafter, in the proceeding rounds of infection, efficiently cleave the viral genomic RNA. Since the Gag protein has many essential functions it is unlikely that HIV-1 will develop escape mutants when using the Gag-nuclease. Vpr- and Nef-nuclease fusion proteins also seem to cleave viral RNAs, either during or after viral assembly 57.

A Nef TNP has shown an antiviral effect by inhibiting down-regulation of e.g. the CD4 cell surface protein. It is thought that CD4 may interact with the Env protein present on new viral particles, thus hampering viral release from the cell surface 66. Likewise, anti-sense RNAs and ribozymes directed against a conserved 14 nucleotide region in the nef-gene also possess an antiviral effect 7.

By overexpressing different kinds of CD4 variants in the infected cell, it is possible to inhibit virus budding. The reason is most likely that the CD4 variants possess the ability to trap and restrain new virions in the cell 19.

The HIV-1 protease plays an important role in virus maturation. As mentioned earlier, the HIV-1 protease cleaves the Gag and Gag-Pol polyproteins to form the structural and enzymatic proteins. Consequently, the protease is a potent target for inhibiting strategies. The current strategies involve protease inhibitors that bind to the active site of the HIV-1 protease and thereby inhibit processing of the Gag and Gag-Pol polyprotein precursors. This results in immature and noninfectious viral particles. The HIV-1 protease is an aspartyl protease and inhibitors have been designed that optimally bind to the catalytic aspartate residues and additionally to the water molecule that is critical for enzymatic action. The inhibitors are transition state analogues that bind the enzyme much more tightly than the natural substrate, making them competitive enzyme inhibitors. Examples of approved protease inhibitors include Saquinavir, Indinavir, Ritonavir, Nelfinavir and Amprenavir 81173.

Other strategies that target Pro involve the application of Pro TNP 7, and the overexpression of chimeric Vpr protein in which the C-terminal region is fused to several cleavage sites recognized by the protease. This will overwhelm the protease activity by a competitive mechanism and impair protease function 774. Finally, anti-sense RNAs targeting the Pol coding region inhibit this last step in the viral life cycle 70.