6  Conclusion and future directions
The HIV/AIDS pandemic has lasted almost 3 decades, and a wide range of anti-HIV drugs have been developed in an attempt to ameliorate the infection and ultimately eliminate this highly fatal disease. To tackle HIV and prevent the formation of viral sanctuary sites, antiretroviral drugs must be able to access the brain. However the normal function of the BBB and blood–CSF barrier to shield the brain from harmful substances and provide a precisely regulated unique environment in the CNS, hinders the penetration of anti-HIV drugs into the brain, promoting viral replication, the development of drug resistance and ultimately sub-therapeutic concentrations of drugs reaching the brain, leading to therapeutic failure. Consequently, Letendre's CNS penetration-effectiveness ranking concept for quantifying antiretroviral drug penetration into the CNS is clearly worthy of serious consideration as an additional tool in designing more effective drug treatment strategies (Letendre et al., 2008a). However, it is important that the actual raw information that was used to determine each drug's individual rank becomes available, so that the ranking scheme could be further and more easily enhanced by considering drug–drug interactions at the level of blood–CNS interfaces and possibly other factors (e.g. toxicity issues) as they come to light.
The plethora of transporters expressed at the brain barriers all act as selective gatekeepers, and this remains a major obstacle for antiviral therapy. Multi-drug transporters have overlapping substrate specificity for numerous hydrophobic compounds and there seems to be considerable overlap in the efflux transporters used by certain anti-HIV drugs raising the possibility that inhibiting more than one transporter at a time may be one approach to improve CNS delivery of anti-HIV drugs. Although certain studies have demonstrated that simultaneously inhibiting multiple transporters can augment the concentration of anti-HIV drugs in the CNS, this would not appear to be a viable strategy, because it leaves the brain susceptible to damage by putative toxins. However, pharmacological modulation of individual transporters by concomitantly administering HAART with transporter-specific inhibitors may be a more successful alternative to enhance anti-HIV drug levels in the brain, without causing these general and systemic adverse effects (Eilers et al., 2008).
Importantly, the relative functional importance of specific transporters in individual tissues is slowly being revealed. In a recent study by Kamiie et al. (2008) the development of a sensitive and simultaneous quantification method using in silico peptide selection criteria and multi-channel multiple reaction monitoring was implemented. By using this novel method, they were able to determine simultaneously the absolute protein expression levels of multiple membrane transporters in mouse brain capillary endothelial cells. Consequently, a quantitative atlas of membrane transporter proteins was constructed. Of relevance to this review, the transporters quantified in mouse BBB were Mdr1a, Mrp4, Bcrp, Oat3 and Oatp2 (Kamiie et al., 2008). Their findings regarding protein expression levels were consistent with previous functional data (Schinkel et al., 1997). This advance in pharmacoproteomics and its extension to human tissues will not only provide significant insight into transport proteins at the BBB and other tissues, but will also hasten the translation of preclinical work to clinical studies and drug development (Kamiie et al., 2008).
Information regarding the interaction of anti-HIV drugs with recently discovered novel transporters such as URAT1 and RST are also needed, in order to acquire a complete picture of the impediments that restrict anti-HIV drug transport across the brain barriers. Conversely, the interaction between newer drugs, such as Enfuvirtide and Maraviroc, and transporters also should be fully investigated. Additionally, it would be valuable to explore whether those transporters that have altered expression during HIV infection affect the penetration of HAART across the brain barriers, and if so, whether this can be pharmacologically modulated. Modulating transporters is one method by which drug delivery into the brain could be improved (Miller et al., 2008). Thus rather than modifying substances in order to allow them to penetrate the BBB, a reasonable alternative strategy would be to modify the permeability of the BBB. Recent studies have explored factors responsible for the modulation of P-glycoprotein, Mrp2 and BCRP and the signalling mechanisms responsible for this (Bauer et al., 2004; Imai et al., 2005; Wang et al., 2006). Furthermore, although studies have revealed that the integrity of the BBB is affected during HIV, possible strategies to prevent this have not yet been determined.
Intranasal delivery of antiretroviral drugs has been proposed as a potential strategy to overcome the poor penetration of these drugs into the brain and to target HIV that harbours in the CNS. The intranasal administration of the viral entry inhibitor peptide T has recently been explored for the prevention of neuro-AIDS development such as cognitive impairment associated with HIV. As well as being an entry inhibitor, Peptide T reduces the initial infection of cells expressing CCR5 receptors such as monocytes and microglia and also acts as an antagonist of free gp120 and thereby reduces toxic effects (Hanson and Frey, 2007). More interestingly, peptide T has demonstrated antiviral and immunological benefits in HIV patients receiving intranasal peptide T with decreased viral load in monocyte reservoirs, increased antiviral cytotoxic T-cells with no drug-related toxicity and increased CD4 (Polianova et al., 2003).
A fruitful area of future research will be to determine if there are any systems that can target drug delivery to the brain and also enable the use of recombinant protein therapeutics for this purpose. Novel developments emerging in the field of polymer science and nanotechnology provide an option by which the obstacles of limited brain entry can be surmounted (Kingsley et al., 2006). Although several different polymeric materials have been explored for localised delivery to the brain, these have been unsuccessful due to a variety of factors such as inflammatory responses to implants and their invasiveness. Despite this setback, nanomedicines such as polyethylene glycol-coated liposomes carrying chemotherapeutic drugs for systemic release have been successfully developed and allowed for clinical use, raising the possibility that similar methods for delivery to the brain can be developed in the near future. Other examples of nanomaterials include nanoparticles, polymeric micelles and nanogels (Vinogradov et al., 2004; Kabanov and Gendelman, 2007; Begley, 2004). All have potential as drug delivery systems targeting the brain.A number of recent studies involving polymeric micelles and drug nanosuspensions are of particular relevance to this review. Using brain microvessel endothelial cells Miller et al. (1997) showed that micelles of Pluronic copolymers can affect drug transport by inhibition of P-gp and redirection of vesicular transport (Miller et al., 1997). Subsequently, the same research group found that Pluronic P85 unimers increase accumulation of a P-gp dependent drug in Caco-2 and bovine brain microvessel endothelial cell monolayers by inhibiting P-gp efflux (Batrakova et al., 1998, 2003; Kabanov and Gendelman, 2007). Furthermore, Pluronic P85 inhibited substrate efflux via MRP1 and MRP2 (Miller et al., 1999). Spitzenberger et al. (2007) explored the effects of Pluronic P85, antiretroviral therapy (AZT, 3TC and nelfinavir) or both on a severe combined immunodeficiency animal model of HIVE. This model of HIVE involved inoculating mice with human monocyte-derived macrophages infected with HIV-1 (Spitzenberger et al., 2007). The authors found that Pluronic P85, as well as both the drugs and P85 combined, showed the most significant decrease in percentage of HIV-1 infected monocyte-derived macrophages (8–22% of control) which was superior to the antiretroviral drug group alone (38% of control). This study not only demonstrates that Pluronic P85 increases the penetration of antiretrovirals drugs into the brain but also that block copolymers may have antiretroviral effects, particularly in cells such as macrophages which serve as a viral reservoir in the CNS (Spitzenberger et al., 2007). A recent study has also shown that Pluronic P85 effectively inhibits the interaction of P-gp with the PIs, nelfinavir and saquinavir and that other transporters (including MRP) may also be inhibited by Pluronic P85 (Shaik et al., 2008).Drug nanosuspensions refer to drug particles that are often stabilised by non-ionic PEG-containing surfactants or with mixtures of lipids. This technology is highly advantageous, in the sense that it has a high-drug loading capacity, it is relatively simple to use and can be applied to many drugs. One application of this technology was in studies that used indinavir nanosuspensions for cell-mediated delivery to the CNS. The concept of cell-mediated delivery of nanocarriers to the CNS when loaded with a drug is based upon leukocyte recruitment during an inflammatory response to a pathogen. In particular the process of migration, phagocytosis and exocytosis of inflammatory cells such as macrophages, neutrophils and T cells (Kabanov and Gendelman, 2007). Indinavir nanosuspensions were internalised into bone marrow derived macrophage lysosymes and subsequently the drug was released into the extracellular environment. Thus, these experiments found that indinavir nanosuspension-laden bone marrow derived macrophages were able to carry and release the drug in tissue (Dou et al., 2007). Furthermore, a study in which indinavir nanosuspension bone marrow derived macrophages were administered to HIV-1-challenged mice demonstrated reduced numbers of HIV-infected cells in plasma, lymph nodes, spleen, liver and lungs. The impressive result from this study demonstrates the potential of cell-mediated delivery of nanocarriers to the brain in diseases such as HIV (Kabanov and Gendelman, 2007).It would therefore be advantageous if in vivo models could be used to validate novel methods of drug delivery to the brain and if genomic and proteomic techniques can be used to identify new blood–brain and blood–CSF barrier transporters. This could possibly be done if a greater emphasis is placed on the molecular interaction between different transport systems in the BBB, the toxicity induced by drug delivery technologies and molecular imaging of the brain and vasculature (Neuwelt et al., 2008).
The advent of HAART has had a profound effect on HIV/AIDS and its associated complications such as HAD, HIVE and MND. However, to further improve the morbidity associated with this destructive pandemic, methods to circumvent the brain barriers must be accomplished so that sufficient concentrations of anti-HIV drugs can enter the brain. As several studies have acknowledged, this is not an easy task and its difficulty, at least in part, is attributed to the powerful transporters present at both the blood–brain and blood–CSF barriers. Members of the ABC and SLC transporter superfamilies act in concert to protect the brain and consequently restrict the entry of anti-HIV drugs into the brain (Table 3). Several factors add to the complexity of the problem, such as the multi-specificity of the transporters, the diverse expression profiles of transporters in different tissues (e.g. the BBB compared with the choroid plexus), the changing expression of transporters as a consequence of HIV infection and differences between the viral populations in the CNS and the plasma, making treatment immensely difficult. A thorough understanding of each transporter, its location, level of expression and transport potency is needed to help improve the treatment of HIV infection and AIDS.