Protein Oriented Ligation on Nanoparticles Exploiting O6-Alkylguanine-DNA Transferase (SNAP) Genetically Encoded Fusion

The use of multifunctional nanoparticles (MFN) combining the emission of detectable optical and magnetic signals and a focused targeting action is attracting broad interest in cancer diagnostics [1–3]. The selectivity in targeting cancer cells is of primary importance and is usually achieved exploiting the modification of MFN with biomolecules endowed with high affinity for specific cell membrane receptors [4–6]. One of the greatest challenges in designing MFN functionalized with homing peptides and proteins to optimize molecular recognition resides in the possibility to finely control the ligand orientation on the nanoparticle surface [7, 8]. So far, three main approaches have been followed to reach this goal: (1) native proteins having domains with high affinity for small peptides can be captured via ligand immobilization on MFN [8]; (2) proteins containing affinity tags recognized by small molecules or complexes commonly utilized for protein purification can be genetically modified in order to introduce a recognition sequence specific for protein immobilization [9, 10]; and (3) sitespecific conjugation can occur via chemoselective ligation [11]. The covalent immobilization of proteins on flat surfaces has a longer story, mainly for biosensing purposes [12]. An elegant strategy involves the use of fusion proteins containing a small (typically 20-30 kDa) enzyme capable of irreversibly cross-reacting with a suicide inhibitor anchored to the solid surface. Examples include mutants of human 06-ahkylguanine-DNA alkyltransferase (SNAP-tag) [13], haloalkane dehalogenase (Halo-tag) [14], and a serine esterase [15]. This approach presents several advantages: (1) the ligand to be immobilized on the surface is a small molecule; (2) its binding to the enzyme occurs quickly under physiological conditions; (3) it is highly specific and essentially irreversible; and (4) all these binding systems involve monovalent recognition partners, which overcomes the crosslinking effects that usually occur with other conventional ligand pairs, such as biotin/avidin [16]. In principle, SNAP may overcome a crucial problem that invariably occurs when using the popular EDC/NHS method, i.e., the formation of nonspecific linkages, which results in a randomly oriented ligation of the protein (EDC = 1-ethyl-3[3-dimethylaminopropyl]carbodiimide hydrochloride; NHS = N-hydroxysuccinimide).