Novel multi-component nanopharmaceuticals derived from poly(ethylene) glycol, retro-inverso-Tat nonapeptide and saquinavir demonstrate combined anti-HIV effects

Background Current anti-AIDS therapeutic agents and treatment regimens can provide a dramatically improved quality of life for HIV-positive people, many of whom have no detectable viral load for prolonged periods of time. Despite this, curing AIDS remains an elusive goal, partially due to the occurrence of drug resistance. Since the development of resistance is linked to, among other things, fluctuating drug levels, our long-term goal has been to develop nanotechnology-based drug delivery systems that can improve therapy by more precisely controlling drug concentrations in target cells. The theme of the current study is to investigate the value of combining AIDS drugs and modifiers of cellular uptake into macromolecular conjugates having novel pharmacological properties. Results Bioconjugates were prepared from different combinations of the approved drug, saquinavir, the antiviral agent, R.I.CK-Tat9, the polymeric carrier, poly(ethylene) glycol and the cell uptake enhancer, biotin. Anti-HIV activities were measured in MT-2 cells, an HTLV-1-transformed human lymphoid cell line, infected with HIV-1 strain Vbu 3, while parallel studies were performed in uninfected cells to determine cellular toxicity. For example, R.I.CK-Tat9 was 60 times more potent than L-Tat9 while the addition of biotin resulted in a prodrug that was 2850 times more potent than L-Tat9. Flow cytometry and confocal microscopy studies suggest that variations in intracellular uptake and intracellular localization, as well as synergistic inhibitory effects of SQV and Tat peptides, contributed to the unexpected and substantial differences in antiviral activity. Conclusion Our results demonstrate that highly potent nanoscale multi-drug conjugates with low non-specific toxicity can be produced by combining moieties with anti-HIV agents for different targets onto macromolecules having improved delivery properties.


Background
Most current anti-Acquired Immunodeficiency Syndrome (AIDS) drugs target two key enzymes in the human immunodeficiency virus-1 (HIV-1) replication cycle, reverse transcriptase and protease. While the remarkable efficacy of protease and reverse transcriptase inhibitor combinations for the treatment of HIV-1 infection has been clearly established in vitro and in the clinic, not even a single AIDS patient has ever been cured. Accordingly, new anti-HIV drug candidates having alternate mechanisms of action are under investigation. For example, ALX40-4C [1,2] blocks viral coreceptor CXCR4 and TAK-779 [3] blocks coreceptor CCR5. T-20 [4,5] and T-1249 [6][7][8] inhibit virus-cell fusion by binding to the viral envelope glycoprotein gp-41. Tat antagonists [9,10] interrupt viral transcription. NCp7 inhibitors [11] hamper viral assembly and budding. To date, combination pharmacotherapy remains the most effective strategy for reducing viral loads in HIV-infected patients. However, given the variety of new chemical entities under development, combination therapies hold even greater future promise.
A major impediment to successful anti-HIV-1 therapy is the emergence of drug resistant strains harboring mutations in genes encoding these viral enzymes [12]. Factors that are known or expected to contribute to the failure of highly active antiretroviral therapy (HAART) include preexisting resistance [13], low and fluctuating drug concentrations due to poor drug absorption or patient non-compliance [6,14,15], and the presence of viral reservoirs and sanctuary sites [16]. Other mechanisms of resistance are becoming increasingly recognized in AIDS therapy. For example, drug-induced biopharmaceutical "resistance" (i.e., multidrug resistance), an established concept in cancer pharmacotherapy [17,18], occurs when the upregulation of cell efflux transporter activity results in lower cellular exposure and decreased drug efficacy. Therefore, the ability to control blood and cellular drug concentrations is critical for managing the emergence of classical viral and multidrug resistance.
Recent successes with HIV peptide fusion inhibitors such as T20 (e.g., enfuvirtide and fuzeon) [8] suggest that small anti-HIV peptides can provide clinical utility complementing the antiviral activity of reverse transcriptase or protease inhibitors. However, many of these peptide drugs are poorly absorbed or are rapidly cleared from the body. HIV-1 encodes a small non-structural protein, Tat (trans-activator of transcription), which is essential for transcriptional activation of virally encoded genes. Viruses with deletion of the Tat-function are non-viable [19]. Efficient replication and gene expression of HIV-1 requires a specific interaction of the Tat viral protein with the transactivation responsive element (TAR), a highly stable stemloop RNA structure [20]. The interaction with TAR is mediated by a nine-amino acid basic domain (RKKRR-QRRR, residues 49-57) of the Tat protein ( Figure 1). This domain is essential for TAR RNA binding in vivo and is sufficient for TAR recognition in vitro [21]. A Tat-derived basic arginine-rich peptide alone binds TAR RNA with high affinity in vitro [9]. A peptidyl compound, N-acetyl-RKKRRQRRR-(biotin)-NH2, containing the 9-amino acid sequence of Tat protein basic domain, was shown to inhibit both Tat-TAR interaction in vitro and HIV-1 replication in cell culture [9]. In addition to the TAR RNA interaction, the basic domain in Tat has at least three other functional properties. It constitutes a nuclear/nucleolar localization signal [22,23]. The basic Tat peptide is also a prototypic cell penetrating peptide that can bring a cargo molecule across the plasma membrane. This was originally discovered when it was observed that Tat protein could freely enter cells [24]. Small Tat peptides derived

Compound
Number EC 50  from the basic domain have also been shown to inhibit HIV replication in cultured T-cells by interacting with the HIV CXCR4 co-receptor present on the surface of T cells, thereby blocking infection by T-tropic HIV-1 strains [25][26][27]. These peptides may also have translational effects [28].
Therefore, we have been investigating Tat peptides as therapeutic agents [9,28]. Considering the pleiotrophic effects of Tat domain peptides, it is not clear whether the delivery of these peptides to extracellular or intracellular targets or both is important for their antiviral effect. Furthermore, since these Tat peptides have cell penetrating activity [29], they can also potentially be used to enhance the cellular uptake of an appended drug [30]. However, Tat peptides have certain disadvantages, such as high systemic clearance due to in vivo degradation, non-specific binding to other biological components, and rapid renal clearance due to their low molecular weight and positive charges [31]. A variety of different strategies have improved biopharmaceutical properties of peptide drugs. In our studies, we have utilized retro-inverso (RI) peptides and macromolecular PEG conjugates to overcome the many biopharmaceutical challenges faced by Tat peptides. R.I.CK (retro-inverso-D-cysteine-lysine)-Tat9, N-acetyl-ckrrrqrrkkr-NH 2 , consists of D-amino acids assembled in the reverse order of the natural L-amino acid Tat9 peptide, Nacetyl-RKKRRQRRR-NH2. Thus, R.I.CK-Tat9 has a similar shape and charge distribution to the natural L-amino acid peptide but is more stable to proteases and retains pharmacological activity [32]. PEGylation has been shown to be one of the most successful techniques for improving the pharmacokinetic and pharmacodynamic properties of peptide drugs by increasing stability and reducing renal clearance and protein binding [33].
The second agent used in this study, saquinavir (SQV) was the first HIV-protease inhibitor approved by the U.S. Food and Drug Administration. Its structure mimics the phenylalanine-proline cleavage sequence at positions 167 and 168 of the HIV gag-pol polyprotein [34]. Thus, SQV prevents cleavage of gag and gag-pol protein precursors by HIV protease in acutely and chronically infected cells, arresting maturation and blocking nascent virions from becoming infectious [35]. However, therapeutic use of SQV suffers from problems of low absorptive and high secretory permeability, bioconversion to inactive metabolites, and poor solubility [36,37]. The oral bioavailability of SQV in clinical formulations is low and/or variable with limited penetration into the lymphatic and central nervous systems (CNS) [38,39]. While its low and variable bioavailability is primarily attributed to metabolism by cytochrome P-450 3A, recent results published by our group [40] and others [41] suggest that multiple membrane transporters may also contribute significantly to the delivery problems of SQV.
In a previous report [42], we showed that the activity of SQV prodrug conjugates was reduced when SQV was conjugated to PEG 3.4 K , compared to the maximal achievable antiviral efficacy. But activity was restored by the addition of R.I.CK-Tat9 to the conjugate (EC 50 = 15 nM). However, the mechanism of enhancement for the SQV-PEG 3.4 K -R.I.CK-Tat9 conjugates was not clearly established since, in addition to targeting intracellular TAR, Tat possesses cell penetrating properties [29,30] that may promote conjugate uptake into the cell and/or it may exert anti-HIV-1 activity by means of cell surface binding to CXCR-4 receptors [25][26][27]. In the current study, the preclinical in vitro effectiveness of a small peptidic Tat

Synthesis of R.I.CK-Tat9 and SQV conjugates
A series of R.I.CK-Tat9 and SQV bioconjugates was synthesized and characterized. The peptides, L-Tat9, R.I.CK-Tat9, R.I.CK(biotin)-Tat9, and R.I.CK(ε-carboxyfluorescein)-Tat9 ( Figure 2) were synthesized, purified and their structures were confirmed by electrospray ionization mass spectrometry (ESI-MS). For R.I.CK-Tat9 PEG bioconjugates, the thiol group of the cysteine residue at the N-terminus of R.I.CK-Tat9 or its derivatives was linked to the maleimide group on mPEG-MAL ( Figure 3A) or amino groups of branched 8-arm PEG 10 K -(NH 2 ) 8 through a stable thioether bond using a heterobifunctional cross-linker N-maleimidobutyryloxysuccinimide ester (GMBS) (Figure 3B). As a control, the eight amino groups of 8-arm PEG amine were reacted with carboxyfluorescein-NHS and yielded PEG 10 K -(ε-carboxyfluorescein) 8 lacking Tat peptides. These bioconjugates were purified using sizeexclusion chromatography on a Sephacryl S-100 column. The formation of each bioconjugate was confirmed by MALDI-TOF mass spectrometry and the concentration of each bioconjugates was determined by quantitative amino acid analysis. The overall design for all SQV conju-Schematic representation of R.I.CK-Tat9 and its derivatives Figure 2 Schematic representation of R.I.CK-Tat9 and its derivatives.
gates was to link the various components using covalent bonds that varied in their stability properties ( Figure 4)

Stability studies
The stability of the covalent linkages attaching pharmacophores to the conjugates was determined. R.I.CK-Tat9 was linked to PEG 10 K using a relatively stable thioether bond. The stability of this bond was assessed by incubating PEG10 K-[R.I.CK(fluorescein)-Tat9] 8  SQV was linked to R.I.CK-Tat9 through a SQV-Cys ester bond. The stability of the ester was evaluated in PBS at pH 7.4 and in spiked plasma, measured at 37°C using a recently developed fluorogenic protease assay for free SQV [42]. The release of active SQV was observed with halflives of 14 h and 0.9 h in PBS at pH 7.4 and in spiked plasma, respectively [42].

Biological activity
The anti-HIV activity of each bioconjugate derived from R.I.CK-Tat9 and SQV was evaluated in vitro in HIVinfected MT-2 cells ( Figure 6) utilizing an established antiviral assay [42]. The L-form of Tat9 showed weak anti-HIV activity (EC 50 = 51.3 µM) while the retro-inverso form of Tat peptide showed much stronger anti-HIV activity (EC 50 = 0.85 µM). Biotin appended to the R.I.CK-Tat9 peptide greatly enhanced the activity of the peptide (EC 50 = 0.018 µM), possibly due to the increased cellular uptake (~ 30fold) conferred by biotin [43]. The conjugation with a 10 kDa branched PEG might be expected to enhance the stability of the peptide by protecting it from attack by peptidases, even though the unnatural D-amino acids in the R.I. peptides may already confer ample protease-resistance. We found that R.I.CK-Tat9 is released from PEG 10 k -(R.I.CK-Tat9) 8 conjugates very slowly (t 1/2 = 50.6 h) and that most of the fluorescence labeled PEG 10k -(R.I.CK-Tat9) 8  µM and 1.5 µM, respectively, which was weaker than that of the non-PEGylated forms ( Figure 6 and Table 1). This result suggests that the intracellular inhibitory effect of Tat peptide may be quantitatively more important than the extracellular blocking of HIV infection previously described for Tat-based peptides, although targeting cell surface receptors might still be an important secondary mechanism of viral inhibition by either biotinylated or non-biotinylated peptides.
The activity of SQV in drug conjugates compared to the maximal achievable antiviral efficacy of free SQV (EC 50 Table 1. The results (Figure 9) showed that at a relatively high concentration (7µM) R.I.CK(fluorescein)-Tat9 or PEG 3.4 k -R.I.CK(fluorescein)-Tat9 (green) were mainly co-localized with the fluid phase endocytosis marker rhodamine-dextran (red) in punctate dots (orange/yellow in the merged panels, Figure 9D &9H), suggesting predominant endo-somal location. Only in cells incubated with R.I.CK(fluorescein)-Tat9, was there some faint green fluorescence that was not co-localized with the fluid phase endocytosis marker (arrows in Figure 9D), suggesting some cytosolic location.

Flow cytometry and confocal microscopy
Overall, the confocal data are consistent with the conclusion from flow cytometry analysis that the majority of R.I.CK-Tat9 is within cells, the majority of PEG 10 k -(R.I.CK-Tat9) 8

Discussion
While recent advances in anti-AIDS therapeutics have resulted in the introduction of more potent drugs, a cure for HIV infection remains an elusive goal. Many factors contribute to the inability of current therapeutic regimens to cure HIV infection. However, central to the problem is the variability of drug concentrations in the blood and target tissues resulting from poor patient adherence to complicated regimens, the inability of these potent agents to selectively target infected tissues, and the poor penetration or retention of drugs in reservoir and sanctuary sites. It is becoming more obvious that better drug delivery and targeting technologies are required to increase total body persistence, target cell exposure and retention of these potent therapeutic agents. In the current study, we have produced and tested the first of a series of nanoscale drug delivery vehicles in order to better target HIV infected cells and possibly to provide novel modes of action.
The present report describes the design, synthesis and initial characterization of a series of PEG-based bioconjugates in order to achieve maximum therapeutic payload of Tat peptide, to explore multi-drug delivery on one bioconjugate and to determine the potential role of Tat in the cellular uptake and HIV inhibition by the conjugates. These bioconjugates were designed (i) to carry multiple copies of the R.I.CK-Tat9 drug linked by stable thioether bonds to 8-arm poly(ethylene) glycol (PEG), (ii) to carry multiple drugs such as SQV for combination therapy, (iii) to have an extended biological and chemical half-life, (iv) to selectively release appended drug molecules inside the cell because of the differential in reducing capacity between blood and the internal cell environment and finally, (v) to enhance cellular uptake of the drug and bioconjugate through the use of uptake enhancing moieties like biotin attached to the R.I.CK-Tat9. These modifications were designed to increase the therapeutic peptide's bioavailability, biodistribution and delivery into HIV sanctuary sites of both the therapeutic Tat peptide and appended drugs, exemplified here by SQV.
The current study is aimed at improving the therapeutic potential of a Tat-antagonistic compound R.I.CK-Tat9. It is an analog of the nine amino acid sequence of the TARbinding basic domain of Tat protein in which the direction (polarity) of the amino acid sequence is reversed and the chirality of each amino acid residue is inverted from L to D. Retro-inverso analog peptides are expected to have shapes and charge distributions of their side chains similar to the natural L-amino acid peptides, but are highly resistant to proteolysis [44,45]. Furthermore, Wender et. al. found that the L-, D-and R.I. forms of Tat9 showed similar cellular uptake in serum-free medium, while in the presence of serum the D-form was modestly more active and the R.I. form much more active than the L-form, indicating the likely role of proteolysis in limiting the activity of natural peptides [46]. The current results show that R.I.CK-Tat9 had a 60-fold higher anti-HIV activity than L-Tat9, consistent with the enhanced stability and/or increased intracellular availability of R.I.CK-Tat9.
Choudhury et al. showed that Tat9-C(biotin) with S-biotinylation of the cysteine residue was taken up 30-fold more efficiently by Jurkat cells than was unbiotinylated Tat9-C (3% versus 0.1%, respectively) [9]. This was attributed to increased hydrophobic interactions with the plasma membrane [43] and it was hypothesized that biotinylation would result in enhanced inhibiton of transactivating activity by the biotinylated compound. The current results confirm this hypothesis since the biotinylated RI-Tat9 was 47 times more potent than the RI-Tat9 and was approximately as potent as SQV in inhibiting HIV-1.
The central hypothesis of the current study is that improved delivery would result in an enhancement in the pharmacological properties of Tat inhibitors, in particular R.I.CK-Tat9. Thus, it was anticipated that PEGylation of R.I.CK-Tat9 would enhance the pharmacological properties in vivo for more effective delivery. The PEG residues were chosen to avoid in vivo binding of R.I.CK-Tat9 to plasma proteins and rapid elimination from the blood [47]. Thus, PEGylation provides a way to increase the stability and body persistence of the R.I.CK-Tat9, which could result higher in vivo activity. However, the PEG conjugates reduced the antiviral activity of R.I.CK-Tat9 or R.I.CK(biotin)-Tat9 in cell culture experiments ( Table 1). The thioether bonds, used in the linkage between R.I.CK-Tat9 or R.I.CK(biotin)-Tat9 and PEG were very stable, insuring that Tat was not released from the conjugate. These results suggest the quantitatively more important antiviral effect of R.I.CK-Tat9 depends upon its release from PEG, presumably reflecting a requirement for entry into infected cells.
An oligocationic peptide compound (ALX40-4C), designed to mimic the basic domain of the HIV-1 Tat [2], was also found to interfere with viral entry through the inhibition of the chemokine receptor CXCR4 on the host cell membrane [1]. The blocking of viral entry resulted in a more potent response than the inhibition of transactivation by that compound [2]. To delineate whether the antiviral mechanism of the R.I.CK-Tat9 conjugates is by inhibition of transactivation or by blocking of cell surface coreceptor, R.I.CK-Tat9 was labelled with the fluorescence tag carboxyfluorescein-NHS and conjugated to PEG 3.4 K and PEG 10 K for stability and uptake mechanism studies of the final conjugates. Flow cytometry showed 93.8%, 53.6%, and 19.0% of total cell-associated R.I.CK(fluorescein)- Tat9 early endosomes (to be published). Once they escape from endosomes, both conjugates and their released R.I.CK-Tat9 must enter the nucleus in order to meet their TAR target. The cytosol and the nucleus are continuous through nuclear pores, which have a functional diameter of about 40 nm. This pore size roughly equals the effective size of a linear PEG molecule of 20 kDa [48], implying that the 25 kDa PEG 10 K -(R.I.CK-Tat9) 8 might not be able to enter the nucleus while the 5 kDa PEG 3.4 K -R.I.CK-Tat9 might.
In this study, we used uninfected MT2 cells that are transformed CD4 + T cells, which are capable of supporting HIV-1 replication. Depending on a particular anti-HIV-1 agent tested and the MOI (multiplicity of infection) used, MT2 cells can serve as a useful model for HIV-1 infected or uninfected T cells in vivo. Critical to current HAART regimens is that both infected and uninfected T cells should be adequately loaded with anti-HIV-1 drugs at all times.
In the infected cells, the drugs prevent the production of mature, infectious viruses. In the uninfected cells, the drugs play a prophylactic role, ensuring abortive infection or replication whenever infection occurs. Constant new HIV-1 infection must take place as both HIV-1 and virusreplicating T cells have short in vivo half-life yet the virus keeps evolving and cessation of HAART inevitably leads to viral load rebounding. We now know that during all stages of HIV disease CD4 + T cell depletion occurs predominantly in the gut [49].
Certain aspects of the performance of PEG conjugates (e.g., plasma persistence and protein binding) can only be studied in vivo and are not addressed in the current study. The known in vivo advantages of PEGylation of various pharmacophores indicates the potential that PEG-based bioconjugates could display useful therapeutic properties of increased plasma half-life [50], lower cytotoxicity [51], and reduced protein binding, which could outweigh the observed reduction in efficacy seen in this study for some PEGylated drugs. This will require further future study in in vivo models and is beyond the scope of the current studies.
Multiple-drug cocktail regimens have been associated with the recent successes in improving the quality of life of patients with AIDS. However, delivery of drugs to the bloodstream rather than to many of the known sites of high viral replication may not insure that the maximal therapeutic effect will be obtained. A potential first step in achieving this maximal effect is to better control the exposure of infected cells to multiple therapeutic agents. In this proof-of-principle study, the prototypical protease inhibitor, SQV, was conjugated to PEG 3.4 K and PEG 3.4 K -R.I.CK-Tat9. The conjugation of PEG to SQV yielded a much less active prodrug conjugate SQV-Cys-PEG 3.4 K with EC 50 at 900 nM. The 60-fold lower activity of the conjugate compared to the parent drug could be due to the slow cleavage of the ester bond and/or low cell uptake of the conjugate. However, the addition of Tat to SQV-Cys-PEG 3.4 K resulted in a conjugate with an EC 50 of 0.015µM, the same in vitro potency as the free SQV. The increased activity of SQV-PEG 3.4 K -R.I.CK-Tat9 may be attributable to a variety of factors including the enhanced intracellular uptake by the bifunctional conjugates containing both Tat peptide and SQV, the synergistic effects of SQV on HIV protease and Tat peptides on both HIV Tat-TAR binding and/or on cell surface receptor CXCR4. All of these results suggest that higher anti-HIV activity was attained mainly due to the enhanced intracellular uptake of SQV-PEG 3.4 K -R.I.CK-Tat9 resulting in improved delivery of SQV. However, the synergistic effects of SQV on HIV protease and Tat peptides on both HIV TAR and/or CXCR4 are likely to contribute.

Conclusion
In the current study, the preclinical in vitro effectiveness of a small peptidic Tat antagonist, R.I.CK-Tat9, alone or with saquinavir using PEG-based bioconjugates was demonstrated. While the PEG-linkage alone did not potentiate the activity of R.I.CK-Tat9, the addition of SQV to R.I.CK-Tat9-PEG bioconjugates significantly enhanced the antiretroviral activity to the level comparable to free SQV. These bifunctional conjugates were more potent than the PEG conjugates with R.I.CK-Tat9 or SQV alone. These results demonstrate that the macromolecular bioconju-  Figure 9D), suggesting some cytosolic localization.
gates could deliver drugs to multiple targets, bringing drug delivery systems down to the molecular level. Since in vitro studies of anti-HIV activity do not account for factors that would enhance in vivo potency (e.g., increased body persistence and decreased binding), it is quite possible that the advantages of SQV-PEG 3.4 K R.I.CK-Tat9 conjugates over the parent drugs could become evident by in vivo testing. In addition, all of the bioconjugates demonstrated low cytotoxicity.
The modular approach for producing a targeted nanopharmaceutical delivery system that we have taken allows for appending different drug combinations or combinations of drugs and cellular uptake enhancing agents in order to maximize the therapeutic effect. Therefore PEG bioconjugates could constitute a powerful delivery system for either single drug administration or combination therapy. However further mechanistic studies are needed to optimize the structure (e.g. size, shape and other features) of the PEG bioconjugates. Additional studies are needed to assess the potential in vivo benefits (long half-life, lower clearance rate) of these conjugates.
In the synthesis of R.I.CK-Tat9-PEG bioconjugates ( Figure  3), R.I.CK(fluorescein)-Tat9 was conjugated to mPEG-MAL by the reaction of the thiol group of R.I.CK(fluorescein)-Tat9 with the maleimide group on mPEG-MAL (4), which formed a stable thioether bond between PEG and R.I.CK(fluorescein)-Tat9. The eight amino groups of PEG 10 K -(NH 2 ) 8 (MW 10 kDa, 8-arm branched, Nektar Therapeutics, Huntsville, AL) (6) were first activated with three-fold molar excess of the heterobifunctional crosslinker, N-maleimidobutyryloxysuccinimide ester (GMBS) (Pierce Biotechnology, Rockford, IL), in DMF to form a maleimide activated PEG (7). The reaction was stirred overnight at room temperature. The product was precipitated with cold ether and dried under vacuum to yield the solid PEGylated product. The GMBS linker essentially converts a primary amino group to a maleimide group that can react with a thiol group to form a stable thioether bond. This activated intermediate 7 was reacted with a three-fold molar excess of 1, 2 and 3, respectively, with coupling reagents HOBt (4-fold molar excess) and BOP (3-fold molar excess) in DMF. DIEA (diisopropylethylamine, 1 % v/v) was added to adjust to neutral pH. For control, the eight amino groups of 8-arm PEG amine were reacted with carboxyfluorescein-NHS in the presence of 1%DIEA in DMF to yield 11 PEG 10 K -(fluorescein) 8 lacking Tat peptides. The products were recrystallized from cold ether and dried under vacuum overnight. These bioconjugates 8-11 were purified using size-exclusion chromatography using a Sephacryl S-100 column (Amersham, Piscataway, NJ) in 0.1 M PBS, pH 7.4, with detection at 220 nm (for 8 and 9). For 10, fluorescence was detected at 535 nm (emission) and 485 nm (excitation). The formation of the bioconjugates (8-11) was confirmed by MALDI-TOF mass spectrometry and the concentrations of bioconjugates were determined by amino acid analysis.

Release kinetics of R.I.CK-Tat9 by cleavage of thioether bond
The stability of fluorescein-labeled PEG 10 k -(R.I.CK-Tat9) 8 bioconjugates was investigated in PBS (pH 7.4) and rabbit plasma, at 37°C. The stability of the bioconjugates was also investigated in PBS (pH 7.4) at 37°C by treating them with 5 µM glutathione (GSH) (Sigma-Aldrich, St. Louis, MO), a physiologically relevant reducing reagent that is responsible for intracellular reductive environment inside cells. Initially, different concentrations (0.01 -10.0 µM) of R.I.CK(fluorescein)-Tat9 were dissolved separately in PBS (pH 7.4) and in plasma and their fluorescence measured by Tecan fluorescence microplate reader (excitation at 485 nm, emission at 535 nm) to obtain calibration curves in PBS (pH 7.4) and in plasma. Thereafter, the bioconjugate solutions were incubated separately in PBS (pH 7.4) and in spiked rabbit plasma at 37°C, along with GSH treated bioconjugates in PBS (pH 7.4) at 37°C. Aliquots were withdrawn at different time points and centrifuged at 14,000 × g for 90 min with a Microcon™ filter (molecular weight cut-off = 10,000 Da) (Amicon Inc., Beverly, MA). The drug moiety cleaved from the bioconjugate during the incubation passes through the filter whereas the drug moiety that remains linked to the PEG carrier is retained.
[Note that a PEG polymer of 3,400 Da behaves as a >10,000 Da peptide on ultrafiltration.] The retentates resulting from the different incubation time points were withdrawn and subjected to fluorescence detection. Each measurement was done in duplicate. The concentrations of the bioconjugates were determined from a fluorescence calibration curve that was established in the same media. The rate constant (k) was obtained from the linear plot of ln(bioconjugate) t versus incubation time, t (h), where [bioconjugate] t = concentration of bioconjugate at different incubation time, t, ( Figure 5). The half-life (t 1/2 ) of the cleavage of thioether bond from the bioconjugates was calculated from the relation t 1/2 = 0.693/k where k is the slope of the linear plot.

Release kinetics of SQV by cleavage of ester bond
A fluorogenic protease inhibition assay was used to measure the hydrolysis kinetics of SQV-Cys ester which was the common intermediate for all the synthesized R.I.CK-Tat9-SQV bioconjugates. The chemical stability of this ester was determined in PBS at pH 7.4 and in spiked plasma both measured at 37°C as reported earlier [42].

Antiviral assays
The in vitro anti-HIV activity of PEG conjugates was determined by a MTT-based HIV-1 susceptibility assay reported previously [42] using MT-2 cells, an HTLV-1-transformed human T-cell leukemia cell line, infected with the HIV-1 strain LAV-Vbu3. The cytotoxicity of the conjugates was evaluated in parallel. The MT-2 cells were grown in RPMI 1640 DM (Dutch modification) medium supplemented with 20% FBS (fetal bovine serum), 1% w/v pen-strep, and 1% w/v L-glutamine and maintained at 37°C in 5% CO 2 in an incubator. The cultured MT-2 cells were diluted to 2 × 10 5 cells/mL and infected at a multiplicity of infec-tion (MOI) of 0.01 (1 viral particle per 100 cells), causing the death of 90% of the cells 5 days later. The tested conjugates were diluted with RPMI 1640 medium and were added to the cultured MT-2 cells after viral infection. Each conjugate was tested in triplicate for its antiviral activity. Cell viability was measured by the colorimetric MTT test at 540 nm, which is directly proportional to the number of living cells. The conjugate EC 50 and LC 50 values were determined from the curves of the percentage of viral cell killing and cytotoxicity against compound concentration (Table 1, Figure 6).

Confocal microscopy
MT2 cells were treated with fluorescein-labeled R.I.CK-Tat9 or its conjugates for 24 hours. Where indicated, a fluid phase endocytosis marker, tetramethylrhodamine-Dextran/10 kDa (Invitrogen/Molecular Probe), was used at 0.25 mg/ml in co-incubation with a fluorescein-labeled compound. All images were taken of live cells on a Leica TCS SP2 Spectral Confocal Microscope using the XYZ mode and 0.25 micrometer per section. The fluorescence wavelength windows for different dyes were well separated to ensure no breach-through from one dye to another.

Flow cytometry
MT-2 cells were grown to 2 days post-confluency and aliquots of 2 × 10 6 cells were washed briefly and incubated in 96-well microplates with 1 µM PEG 10 K , R.I.CK(fluorescein)-Tat9, PEG 3.4 K -R.I.CK(fluorescein)-Tat9 and PEG 10 K -[R.I.CK(fluorescein)-Tat9] 8 bioconjugate for 24 hours. Trypan blue staining was used for quenching of cell surface bound fluorochrome emission. After the 24 hours incubation with tested conjugates, the medium was immediately removed from the wells. Part of the cells were washed and resuspended in 0.02 M sodium acetate buffer (pH 5.8). The remaining cells were suspended in sodium acetate buffer containing 0.2 mg/ml trypan blue. After 20s, the cells were washed twice and resuspended in the sodium acetate buffer. The total cell associated fluorescence was then analyzed by flow cytometry using a Coulter EPICS PROFILE equipped with a 25 mW argon laser. For each analysis, 10,000 to 20,000 events were accumulated. The total cell associated fluorescence was the cell associated fluorescence of cells without quenching by trypan blue. The intracellular fluorescence was the cell associated fluorescence of cells quenched by trypan blue. The cell surface bound fluorescence is the difference between the total cell associated fluorescence and the intracellular fluorescence. http://www.aidsrestherapy.com/content/3/1/12