The feline immunodeficiency virus (FIV) was first isolated in Petaluma, California, from cats with an immunodeficiency-like syndrome [1,2]. FIV is an enveloped virus and, similar to other lentiviruses, has a virion diameter of 105–125 nm and includes a host cell-acquired membrane with viral glycoproteins protruding as spike-like projections [1,3]. As in other complex retroviruses, the FIV genome includes the three large open reading frames (ORF) gag, pol, and env, and several genes that code for small accessory proteins such as Vif and Rev. The three large ORFs code for: (1) the structural polyprotein Gag, comprised of a myristoylated matrix (MA, p15) protein, a capsid (CA, p24) protein and a nucleocapsid (NC, p13) protein; (2) the viral polymerase (POL) protein containing reverse transcriptase (RT), integrase (IN), protease (PRO) and deoxyuridine pyrophosphatase (DU) enzymes; and (3) the Env protein comprised of the heavily glycosylated surface unit (SU) protein gp95 and the transmembrane (TM) protein gp40 [2,4].

FIV infects lymphocytes, cells of the monocyte/macrophage lineage, and cells of the central nervous system. The viral replication strategy is highly similar to that of HIV, and is initiated by interaction of the viral Env glycoprotein with CD134, a molecule up-regulated on activated CD4⁺ T cells [5]. This interaction exposes previously masked Env epitopes that bind with high affinity to the chemokine receptor CXCR4, which permits viral and cell membrane fusion and subsequent viral nucleocapsid entry into the host cell cytoplasm (Figure 1) [6]. Hence, although FIV Env does not interact with CD4, the primary receptor for HIV, CD4⁺ T cells are nevertheless targeted due to utilization of a receptor with highest expression on memory CD4⁺ T cells. The viral entry receptor, CXCR4, is utilized by both HIV and FIV, and, if present at high density, may facilitate viral entry without initial interaction with CD134 [7]. The latter mechanism might account for infection of CD134 negative cells, such as microglia. Following cell entry, viral RNA released into the cytoplasm, transcribed to complementary DNA, synthesized to viral double-stranded DNA, and transported into the nucleus for integration into the host genome by action of viral IN (Figure 1). FIV Rev assists with transport of viral mRNA, and the Orf A protein may contribute to viral release from infected cells [4,8]. Viral mRNA and genomic RNA are then transcribed, and transported to the cytoplasm for mRNA translation to viral proteins. The immature virion moves to the cell membrane, acquires the viral envelope and glycoproteins, and is then released from infected cells [9,10].

Identifying effective antiretroviral therapy (ART) has been of paramount importance since the beginning of the HIV epidemic approximately 30 years ago. The first antiretroviral agent, the RT inhibitor zidovudine, was administered to patients soon after discovery of HIV [11]. Since then, intense effort has been devoted to developing additional drugs with higher efficacy and lower toxicity, suitable for prevention of infection, and applicable in unique patient populations such as pregnant women and those with viral co-infections. In general, access to modern combination ART has turned an invariably fatal infection into a chronic but manageable condition [10,12,13]. However, ART is of limited availability to the majority of HIV-infected persons living in resource-restricted regions of the world.

At present, approximately 30 compounds are approved by the US Food and Drug Administration (FDA) for treatment of different stages of HIV infection. Types of anti-retroviral drugs include fusion or entry inhibitors, nucleoside reverse transcriptase inhibitors (NRTI), non-nucleoside RT inhibitors (NNRTI), nucleotide RT inhibitors, viral IN blockers and PRO inhibitors (Table 1) [14,15,16,17]. However, rise in resistance among circulating strains of HIV, and side-effects of available drugs, drive the persistent need to discover new therapeutics [10,12,13,14,18]. Table 2 summarizes antiretroviral compounds under development.

Antiretroviral therapeutic agents have been tested to a limited extent against animal retroviruses. However, FIV genome organization, protein chemistry and the pathogenesis of infection are very similar to HIV [3], rendering the FIV-cat model highly suitable for testing many types of antiviral compounds and strategies. Furthermore, the availability of molecular tools, reagents and in vitro and in vivo test systems is substantial and increasing, which allows detailed assessment of immune responses, viral parameters and adverse effects in this relatively large and outbred animal model [19,20,21]. Here, we review antiretroviral compounds in use or in development, their mechanism of action, their safety in cats, and their efficacy against FIV.

Reverse transcriptase inhibitors (RTI) are divided into three categories: nucleoside, nucleotide, and non-nucleoside reverse transcriptase inhibitors (NRTIs, NtRTIs and NNRTIs, respectively). Each type of compound blocks the catalytic activity of viral RT by a slightly different mechanism. Because of its essential role in the viral life cycle, RT has been a major target of ART since the discovery of HIV (Figure 1, step 3).

The hallmark of the retroviral replication cycle is integration of viral double-stranded DNA (ds-DNA) into the host genome. For this process, the pre-integration complex (PIC) is assembled in the cytoplasm of infected cells. The PIC is comprised of two strands of viral RNA, viral IN, and accessory proteins required for nuclear translocation such as viral protein R (Vpr), MA and nucleocapsid (NC) protein, and host proteins [64]. Synthesis of viral dsDNA by RT takes place in the PIC, which is then translocated to the nucleus. Both nuclear translocation and proviral DNA integration have been targets for novel antiretroviral therapy (Figure 1, step 4). Studies targeted to these processes have focused on HIV, but may also be applicable to other retroviruses, including FIV [65]. Protein trafficking from the cytoplasm to the nucleus of eukaryotic cells occurs in either passive or active mode. In passive mode, proteins less than 40 kDa are thought to freely diffuse through nuclear pores [66]. However, translocation of larger proteins from the cytoplasm to the nucleus takes place via an active process which utilizes cellular importins. Importins are proteins that recognize a stretch of basic amino acids, called the nuclear localization signal (NLS), on the cargo protein, and then shuttle such proteins into nucleus [67,68]. HIV proteins such as IN, MA and Vpr contain NLS motifs and contribute to the nuclear import of the PIC [64,69], while inhibitors of nuclear transport aim to interfere with this process. For example, the arylene-bis compound CNI-H0294 binds to the lysine residues in the NLS of MA. By occupying the NLS, CNI-H0294 blocks the interaction of cellular importins with PIC, and prevents nuclear translocation [70]. More recently identified compounds mimic the NLS of MA and by binding to importin α, prevent its interaction with MA [65]. Another novel anti-HIV compound blocked interaction of Vpr with importin α, which prevented HIV replication in primary macrophages [71].

Retroviral IN catalyzes the integration of retroviral dsDNA into the host cell genome and interacts with PIC, making it an ideal target for two-site interference by antiretroviral compounds [65,72]. Integrase inhibitors are designed to either foil enzyme activity and therefore viral dsDNA integration, or to prevent formation of the PIC by interfering with the interaction of IN with cellular factors [65]. Transcriptional co-activator p75 (or lens epithelium-derived growth factor, LEDGF) is a cellular protein that not only plays a crucial role as a functional component of the PIC, but also facilitates IN interaction with nuclear chromatin at integration sites after nuclear transport [73]. D77 is a benzoic acid derivative that interrupted IN/LEDGF interaction and HIV IN function, which subsequently prevented viral replication [74]. The IN strand transfer inhibitor naphthyridine carboxamide (L-870810) prevented FIV DNA integration into the host cell genome and resulted in lower virus replication in the feline lymphoid cell line MBM [75,76].

Generation of progeny viral mRNA and transfer to the cytoplasm of the host cell are the next steps in the retroviral replication cycle (Figure 1, step 5). Transcription of proviral DNA is controlled by the flanking long terminal repeats (LTR). The 5’ LTR initiates transcription and the 3’ LTR is essential for polyadenylation of the transcripts. Lentiviral accessory proteins such as HIV Tat, Rev and Vpr, and FIV Vif and Vpr, as well as cellular factors such as NF-ĸB, control viral transcription. Therefore, the accessory proteins have also been of interest as targets of antiviral therapy [14]. The transfer of HIV mRNA to the cytoplasm is dependent on the regulatory protein Rev, which binds to a Rev responsive element (RRE) in the viral mRNA. Rev contains both NLS and nuclear export signal (NES) motifs that enable shuttling of the protein between cytoplasm and nucleus. NES is a leucine-rich sequence required for the nuclear export of proteins via the CRM1/exportin 1 protein [14]. The interaction of Rev with RRE can be disrupted with tetracationic diphenylfuran, which binds to specific nucleotides in the RRE and blocks binding of Rev [77]. There are also compounds that directly bind to the NES binding site in CRM1 and prevent interaction with Rev. For example, compound PKF050-638 binds to Cys-539 of CRM1 and prevents CRM1-mediated Rev nuclear export [78]. The course of action of PKF050-638 and other inhibitors of CRM1 is similar to leptomycin B, which was the first agent identified as preventing nuclear export of Rev by blocking CRM1 [79]. Leptomycin, therefore, had anti-HIV activity, but toxicity in cell culture precluded clinical application [79]. Other CRM1 blockers of Rev nuclear export, such as 5,6-dihydrovaltrate, may also be effective in reducing HIV assembly in the cytoplasm, and are promising new agents for anti-retroviral therapy [80].

The retroviral protease (PRO) has a pivotal role in processing Gag and Gag-Pro-Pol polyproteins to functional proteins necessary for maturation of virus particles [81]. Since PRO plays a key role in the viral replication cycle, it has been of great interest to develop inhibitors for this enzyme (Figure 1, step 6). The retroviral PRO is a homodimer with the active site located at the dimer interface. Protease inhibitors (PI), together with RT inhibitors, are two key components of combination anti-retroviral therapy in HIV patients. The initial PIs were designed to competitively bind to the active site of the enzyme, and therefore to inhibit access to viral protein substrates. The ten PIs that were first FDA approved all were competitive inhibitors (Table 1) [82]. Since the emergence of viral strains resistant to PIs, new approaches focused on designing compounds to bind at sites other than the active site of the enzyme. Through a fragment-based screening approach, several compounds were selected for their strong binding affinity to different sites of the enzyme and destabilization of enzyme conformation [83]. A specific compound, screened out of a library of inhibitors of protein-protein interactions, was able to inhibit the activity of wild-type PRO and that of six PI-resistant mutants [83,84].

In common with other retroviruses, the FIV PRO also cleaves viral Gag and Gag-Pol, yielding nine structural and non-structural proteins of FIV including MA, CA, NC, PRO, RT and IN [85]. There are a few studies on the effect of PIs on FIV replication and disease control. TL-3, a PI that binds to the active site of the FIV PRO, completely prevented virus production in cell culture [86]. In an in vivo study, TL-3 prevented FIV-induced neurological degeneration when administered at the early stages of infection in cats, but did not prevent FIV viremia in challenged cats [87]. In an in vitro study, the effects of three HIV PIs on FIV_Pet replication in MBM cells were investigated. Results indicated that all three PIs (tipranavir, atazanavir, and lopinavir) inhibited FIV replication, but only tipranavir was comparably effective against FIV and HIV [88].

The maturation of virions occurs as the result of conformational changes in CA protein molecules that have been assembled (Figure 1, steps 7 and 8). Compounds and peptides that target the C-terminal or N- terminal domains of the CA can block the required interaction for assembly of virions. Lack of CA assembly, in turn, inhibits viral replication by blocking the final cleavage of Gag protein, which precludes virus maturation. Beviramat has preventative effects on HIV replication in vitro via this mechanism. Beviramat is the only HIV assembly inhibitor that has been used in clinical trials, but has yielded poor responses [89,90].

Several host cell proteins and innate immune cytokines have antiretroviral effects. TRIM5 (a member of the tri-partite motif family of proteins) is a host protein that in some species binds to the retroviral CA in the cytoplasm, generating CA-TRIM5 complexes, which prevents retrovirus uncoating and enhanced proteasomic degredation. Cyclophilin A (a member of the peptidyl proline isomerase superfamily) contributes to retroviral restriction by interacting with TRIM5, and a fusion protein of feline TRIM5 and cyclophilin A in vitro prevented FIV and HIV infection [91,92]. The TRIM5/cyclophilin fusion protein bound the FIV capsid, and enhanced targeting to the proteasome, thereby preventing reverse transcription [93]. Therapy with such constructs may comprise a genetic approach to treatment of lentiviral infection. Additionally naturally occurring restriction factors and antiviral proteins have been described in cats and other species, but have not yet been employed for antiviral therapy.

Interferon omega had inhibitory effects against FIV in vitro, and is licensed to use in FIV-infected cats in some European countries and in Japan. Interferon alpha has also been suggested to improve health and survival of FIV positive cats [52].

Current antiretroviral agents are mostly studied for their effects on HIV. However, because of persistent emergence of drug resistant HIV strains, toxicity of many agents, and the requirement for lifetime therapy, the search for new less toxic and highly efficacious agents continues. Animal models may be very helpful for assessing antiviral properties, and among lentiviruses, FIV infection of cats is a well-established model. However, structural differences in viral proteins, and distinct differences between primates and cats in xenobiotic disposition, restrict the utility of the model to certain agents (Table 3).