Controlling Ligand Density on Nanoparticles as a Means to Enhance Biological Activity

Keywords::

• cancer therapy
• cellular uptake
• drug delivery
• functionalization of nanoparticles
• gold nanoparticle
• ligand density

Nanoparticle interactions with biological structures

Nanoparticle (NP)-based delivery systems (nanoconstructs) have gained recent attention because of their promise in enhancing delivery efficiency and therapeutic efficacy for cancer treatment [1]. By design, engineered nanomaterials offer advantages over other approaches because their sizes and surface ligand presentations are commensurate with biological markers and active regions on the plasma membrane [2]. However, there is still a lack of detailed information on interactions between the ligands grafted to the NP core and targets in physiological conditions. Since NPs enter most cells via energy-dependent processes that depend on the properties of the nanoconstructs [3], an understanding of how physicochemical parameters including NP size, NP shape, ligand density and overall surface charge affect local interactions is needed in order to optimize the cellular uptake pathway [4]. Most reports on nanoconstruct optimization have focused on in vivo applications [5]; however, the type and density of ligands on NP cores have recently emerged as critical factors that dictate how nanoconstructs interact with receptors on the cell membrane [2], which will ultimately affect their effectiveness in tumors. The next leap in nanomedicine will occur when we understand – at the nanoscopic level – how a single, engineered nanoconstruct interacts with an individual cell.

Cell stimulation and induced cellular responses are sensitive to local ligand concentrations of cytokines, chemokines, growth factors and related molecules [6]. For example, membrane–receptor clustering and lipid-raft formation can initiate signal cascades within ligand-induced cellular pathways [6]. Such processes modulate uptake by controlling when endocytic pathways are initiated [7]. Hence, controlled ligand density and presentation on nanoconstructs is emerging as a promising route to modulate cellular responses and as a potential tool to understand the mechanism of NP uptake. Recently, gold NPs (AuNPs) have been exploited to investigate how surface ligand densities affect interactions with cells at the nanoscale [8]. AuNPs offer distinct advantages, including: they are biocompatible and do not produce adverse effects in vitro or in vivo [9]; their unique optical (localized surface plasmon) properties such as absorption and scattering enable their use in bioaffinity sensors, photothermal therapy and bioimaging [9]; and unlike soft materials, AuNPs do not deform in physiological conditions, which enables long-term studies of how they impact cellular behavior [9]. These strengths suggest that AuNPs can function as a model system to study ligand–receptor interactions in gene and therapeutic agent delivery processes [10].

How ligand density & the presentation of ligands on NPs can be controlled

Spherical AuNPs with high ligand loading are believed to derive their unique properties via a multivalent effect that can result in high effective affinities to cell surface receptors [11]. A wide range of ligands (e.g., nucleic acids, peptides and proteins) [12] can be grafted to the Au surface. Because they can be easily synthesized and attached via thiol chemistry to AuNPs, nucleic acids are the most common system to study the effects of ligand density on cellular behavior [13]. Moreover, their small size (duplexes ~2 nm in diameter) enables dense packing and oriented presentation for binding targets, in contrast to proteins that can suffer from low loading, site-specific conjugation and low activity on NPs [13]. The packing density of ligands such as ssDNA is less sensitive to the diameter of NPs compared with the composition of DNA. As spherical NP diameters increase to 60 nm, the surface coverage effects are similar to those of a planar Au surface [14]. Nucleic acid sequences containing poly(thymine) (10-mer) spacers near AuNP surfaces show decreased loading as a function of NP size, while the same sequence with poly(adenine) spacers has nearly the same ligand density for different NP sizes because of the much stronger relative affinity of adenine compared with Au for thymine [15].

The method most widely used to control the loading of ligands on AuNPs is to tune the molar ratio of ligands and NPs during the conjugation process [16]; however, such approaches do not allow tight control over the efficiency of ligand packing. To address this issue, recent reports have suggested tailoring the chemical environment around AuNPs by adjusting the concentrations of salt and pH conditions in solution [15,17]. This approach can improve ssDNA ligand grafting on AuNPs by reducing repulsion between the particle surface and oligonucleotides. When we have applied salt-aging strategies at low pH to nucleic acids with secondary structures (G-quadruplexes), not only was the necessary excess of ligands reduced and the functionalization time significantly faster (24 vs 1 h), but also higher numbers of ligands (over two-times higher) could be attached to the AuNP surface [8].

An alternative approach to modulate the ligand density is to control the spatial presentation of the ligands, which can easily be accomplished by changing the shape of the AuNP core. Different NP shapes with average sizes similar to spherical ones can affect ligand loading due to differences in surface area, in that anisotropic NPs support higher surface-to-volume ratios [18]. In addition, the protruding structures of anisotropic AuNPs can result in more accessible binding sites for molecules at the surface of NPs, which can influence their reactivity [18]. Therefore, tuning the shape of NPs can provide an effective way to achieve quantitative loading as well as an understanding of ligand surface coverage on NPs as a function of curvature.

How ligand density affects cellular uptake & therapeutic activity in vitro

Recent work has demonstrated that ligand density and the presentation of ligands on NPs can affect cell targeting efficacy as well as cellular uptake [4,8]. Linear nucleic acids with six-times higher loading on spherical AuNPs enhanced cellular uptake by over threefold in representative cell types that were selected to compare different species (mouse and human) and the inherent differences between cell and tissue types (yolk sac, cervix and lung) [19]. In our own work, we have shown that oligonucleotides with secondary structures grafted to AuNPs also show enhanced uptake in cancer cells depending on ligand density. We found that a 2.5-times increased loading of the DNA aptamer AS1411 (G-quadruplex) on gold nanostars (AuNSs) showed two-times higher uptake in different cancer cell lines that overexpress the surface marker nucleolin. Moreover, the highly loaded nanoconstructs were taken up by cancer cells at faster rates compared with constructs with lower densities of AS1411 [8].

So, the question remains: why and how does ligand loading density affect cellular uptake? One hypothesis is that high ligand loading can increase extracellular protein absorption on the NPs. ζ-potential results of spherical AuNPs with DNA (28-mer) indicated that nanoconstructs become more positively charged in phosphate-buffered saline as the packing density of ssDNA increased from approximately ten to 80 strands/NP [19]. This reduction in overall charge (DNA–AuNPs were still negatively charged) was due to the higher adsorption of extracellular proteins (over two-times higher), which then facilitated higher cellular uptake. Although the identification of key proteins that contribute to the internalization of NPs is still unanswered, interactions between nanoconstructs and proteins are involved in the endocytosis process [19]. A second hypothesis is that high ligand densities affect the binding capability of ligands to receptors, in that multiple ligands on AuNPs can interact with numerous target receptors simultaneously, resulting in increased affinity [8].

The densities of both nontargeting and targeting ligands on AuNPs appear to affect uptake [11]; however, downstream results – the therapeutic effects – have been largely unexplored. In our recent work, we showed that the therapeutic efficacy of AS1411 in vitro was improved through high loading on AuNSs in pancreatic cancer and fibrosarcoma cells [8]. AS1411–AuNSs with high loading densities (126 ± 6 dimers/AuNS) showed an average 42% increase in cancer cell death compared with AS1411–AuNSs with a lower loading density (55 ± 3 dimers/AuNS). These results strongly suggest that AuNS nanoconstructs with increased multivalency from higher local concentrations of aptamer drug can improve therapeutic effects.

Future perspective

Ligand density and the presentation of ligands on NPs are emerging as simple, powerful approaches for modulating extracellular and intracellular interactions. Although this article has highlighted AuNPs with nucleic acid ligands, other nanoconstructs have shown similar effects. For example, semiconductor quantum dots functionalized with peptides or proteins have shown that ligand density affects the levels of cellular internalization as well as endocytosis [4]. We have described numerous advantages of densely loaded ligands on NPs for nanomedicine, but there are also some potential drawbacks. First, while dense loading can stabilize ligands and enhance targeting efficacy on nanoconstructs, it can also be a disadvantage if the ligands are packed too tightly [11]. In addition, as the number of ligands per NP increases, the possibility of competition among multiple ligands for single receptors can increase [11]. Such issues can be overcome by optimizing densities, where ligands are designed with spacer molecules or by using anisotropic NP cores such as AuNSs in order to reduce steric hindrance. Second, detailed studies of NPs and the components in biological media are needed in order to tease out the mechanism of cellular uptake. For example, NPs exposed to physiological fluids become coated with biomolecules that may induce unwanted changes of the surface properties of the nanoconstructs, such as average diameter and charge, alterations of function and affinity reductions of NPs to the target [2]. Surface passivation of NPs with protein-resistant molecules such as PEG can help minimize the adsorption of unwanted proteins on NPs, but care needs to be taken with the length and density of PEG so as not to obscure the ligand-targeting effects [20].

Fundamentals of how to engineer NPs so that they can achieve the desired cellular and local biological interactions as well as downstream in vivo effects are critical for the future of nanomedicine. We have described a simple yet promising way to achieve such a result in vitro by optimizing ligand density and the presentation of ligands. Tuning the ligand type (e.g., oligonucleotides with spacer sequences) or NP shape (e.g., stars instead of spheres) provides a straightforward way to tailor multivalency effects between nanoconstructs and cells. Moreover, we anticipate that designer nanoconstructs with nanoscopic control over both soft (ligands) and hard (core) structures could provide insights that will guide the next-generation development of NP-based delivery systems for therapeutic applications.

Financial & competing interests disclosure

Funding is provided by the Northwestern University Center of Cancer Nanotechnology Excellence (U54 CA151880). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.

Additional information

Funding

Funding is provided by the Northwestern University Center of Cancer Nanotechnology Excellence (U54 CA151880). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. No writing assistance was utilized in the production of this manuscript.