Inhibition of Human Immunodeficiency Virus‐1 Integrase by β‐Diketo Acid Coated Gold Nanoparticles

Gold nanoparticles (GNPs) have been proposed as carriers for drugs to improve their intrinsic therapeutic activities and to overcome pharmacokinetic problems. In this study, novel nanosystems constituted by a model β-diketo acid (DKA) grafted to the surface of GNPs were designed and synthesized following the “multivalent highaffinity” binding strategy. These first nanoscale DKA prototypes showed improved inhibition of HIV-1 integrase (HIV-1 IN) catalytic activities as compared with free DKA ligands.

A target that has been validated for developing novel antiretroviral agents is HIV-1 integrase (HIV-1 IN). 1−3 It catalyzes the insertion of retro-trascribed viral cDNA into the host genome (Chart 1A) to form a stable provirus through two coordinated reactions, 3′-processing (3′-proc) and strand transfer (ST). 4−6 Because of its vital role in the viral replication cycle, with no human counterpart of the enzyme known, the addition of an IN inhibitor to existing components of antiretroviral therapy has improved the outcome of therapy by potential synergism, without exacerbating toxicity. 1,2,7 Among a plethora of compounds with diverse structural features reported as IN inhibitors, 7 a class of compounds bearing a β-diketo acid (DKA, Chart 1B) functionality has emerged and been validated as the most effective chemical space in anti-IN drug discovery, and some compounds belonging to this family have had clinical success. 8−13 Several DKA congeners selectively inhibit the strand transfer reaction, and in cell-based assays they inhibit integration without affecting earlier phases of the HIV-1 replication cycle. 8 Mechanistically, the DKA pharmacophoric motif is involved in a functional sequestration of divalent metal ions (Chart 1C), which are critical cofactors at the enzyme catalytic site. 7,12−15 This would subsequently block the transition state of the IN− DNA complex. 2,7,13,14 Despite the outstanding advances in the HIV-IN drug discovery, it was of paramount importance to explore further possibilities to optimize the action of DKAbased compounds as IN inhibitors by the investigation of newer chemical entities. 16,17 Starting from the complexity of biological systems to exploit widespread advantage of multivalency, i.e. the combination of several entities involved in individual interactions or binding events, this characteristic has been followed as a suitable strategy in the generation of high-affinity ligands, which were able to produce simultaneous and reversible molecular recognition interactions. 18−20 In the past decade, several prototypes as multimeric ligands targeting tumor/viral receptors have been exploited, and multivalency can be envisaged as a way to improve the binding performance of a small molecule−drug construct, thus expecting to reach a similar efficiency of the antibody−drug conjugates. 18,21−23 On the other hand, the application of nanotechnology in drug design is having an impact on diagnostics and drug delivery and, more recently, in drug discovery. 24−26 Therefore, the concept of multivalency, by nanoparticle (NP)-based platforms with multiple ligands coated on their surface, has emerged as a suitable strategy to enhance the binding affinity of simple monovalent ligands. Nanoparticles (NPs) can be synthetically engineered to present multiple high-affinity molecules on their surface to tune binding affinity over several orders of magnitude. Moreover, such nanosystems may be effective in disrupting key protein/protein interactions involved in several pathogenic events. NP-based prototypes possessing multivalent ligands are also expected to produce a high local concentration of binding molecules which, consequently, can statistically promote formation of more ligand−receptor interactions, thus enhancing affinity and selectivity at the biological target.
In this context, gold nanoparticles (GNPs) are extensively investigated for various biomedical applications due to their large surface area to volume ratio and thermal stability, as well as their lower toxicity, synthetic accessibility, and amenability of functionalization. 27−29 Some gold nanosystems have been proposed as carriers for drugs, to overcome pharmacokinetic problems and to improve their intrinsic therapeutic activities. 30 The demonstration that GNPs convert a weak interaction of active small ligands into multivalent more effective and biologically active therapeutics has been offered by several seminal reports, which include, for example, inhibition of HIV fusion and antiretroviral delivery, 31,32 activation/inhibition of carbonic anhydrase, 33,34 and tumor targeting. 35,36 In this scenario, we sought to explore the development of novel GNPs that could effectively inhibit viral enzymes such as HIV-1 IN (Figure 1). Therefore, a nanoparticulate carrier was designed simultaneously with a model DKA inhibitor.
In this work, we developed ∼3.5 nm diameter DKA-coated GNPs as a nanoscale platform to construct novel HIV-1 IN multivalent therapeutics. These particles were designed starting from a model derivative of the simple 2-hydroxy-4-oxo-4phenylbut-2-enoic acid (I, Chart 1); it was coupled with lipoic acid to have both (a) a flexible linker between the ligand and the gold surface and (b) the appropriate thiolic terminal groups needed for anchoring such ligands to the NP surface ( Figure 1).
The obtained gold nanosystems resulted in precise and monodisperse nanoscale constructs, which were tested for anti-HIV-1 IN activities.
The preparation of DKA-coated GNPs (DKA-GNPs, 1) was achieved in a two-step reaction by reduction of chloroaurate with NaBH 4 in the presence of the lipoic acid-tailed diketoester (DKE-GNPs, 2), followed by an alkaline hydrolysis (Scheme 1).
The key synthone 9 was synthesized by coupling lipoic acid with the 4-amino-DKE 6 in the presence of EDCI/DMAP, as outlined in Scheme 2.
Synthetic approaches for the preparation of the key intermediate 6, and for the other free ligands 7 and 8 and their precursors are depicted in Scheme 3.
Briefly, 6 was obtained by following the Boc protection of the commercially available 4-amino-acetophenone 3 to give the 4-Boc-amino-acetophenone 4, which was converted to the diketoester derivative 5 by Claisen condensation with dimethyl oxalate and sodium methoxide, using a procedure previoulsy reported by us, with slight modification. 11 Subsequent Boc deprotection of 5 by formic acid in mild conditions gave 6 in  All small molecules have been characterized by means of NMR, IR, mass spectrometry (ESI and FAB), and elemental analysis.
GNPs 1 and 2 were characterized by transmission electron microscopy (TEM) (Figure 2). These particles are monodispersed, roughly spherical in shape, and have an average particle size of ∼3.7 nm; the number of gold atoms is calculated as ∼1430 (for DKEs-GNPs 1 and DKAs-GNPs 2). 27 Elemental analysis and ICP-OES measurements indicated that the number of ligands grafted on the NP′ surface was 193 for both systems (considering the Au/S w/w ratio). 27,33,34,37 Therefore, the obtained GNPs were consistent with a proposed empirical formula of [Au 1434 (C 18  IR spectrum of GNPs-DKE 2 is given in Figure 3A. The C O stretching bands of the COOCH 3 /COOH functionalities were found at approximately 1700 cm −1 in the spectrum of lipoic acid-tailed DKE 9, used for comparison.
Moreover, Figure 3B shows the UV−vis spectrum of DKEand DKA-capped GNPs 1 and 2. A strong absorption at about 540 nm, a resonance corresponding to excitation of surface plasmon vibrations in the GNPs, 27 was observed for both gold nanosystems.
GNPs 1 and 2 and ligands 5−8 were tested for their ability to inhibit IN catalytic activity in in vitro assays employing purified enzyme, using the raltegravir-based derivative (HL) 38 as reference compound, and lipoic acid-coated GNPs (LA GNPs), used as a control (Supporting Information). Inhibition of IN catalytic activities, 3′-proc and ST, was evaluated using oligonucleotide-based assays, and the results are reported in Table 1 (see below for schematic of IN activity, and in Supporting Information for experimental details).
With the exception of the amino-DKEs 6 and 9, all tested compounds showed anti-IN activity in low concentration ranges (IC 50s from 0.96 ± 0.34 to 44 ± 6 μM), with more potency exhibited toward the catalyzed ST process for free ligands 5 and 7 versus 3′-proc, as evidenced by their selectivity indexes. The ligands differ in activities as previously reported, thus confirming that the nature of the substituents on the aromatic ring significantly influences the potency. Surprisingly, with respect to that shown from compounds 5 and 7, as well as from the reference compound, the Bocprotected DKA 8 shared a similar inhibition profile for both catalytic activities (IC 50s = 15 ± 1 and 14 ± 1 μM for 3′-proc and ST, respectively).
This behavior was translated to the GNPs, which exhibited inhibition profiles against both catalytic processes. Specifically, GNP-DKA 1 was the most potent compound (IC 50 values of the 1.2 ± 0.85 μM and 0.96 ± 0.34 μM for ST and 3′-proc, respectively). While the LA GNPs were not active (at >5 μM) as expected, GNPs-DKE 2 was approximately 2-fold less potent than the corresponding diketoacid 1, thus demonstrating that the acid-and ester-terminal functionalities of the nanosystems similarly inhibited HIV-IN functions. Interestingly, these GNPs (i.e., 1 and 2) did not display any selectivity for IN reactions and did not show cytotoxicity in human MT-4 cells at 10 μM (Table 1).
In this study, we proposed the first application of small molecule    The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.