Albumin corona on nanoparticles – a strategic approach in drug delivery

Abstract

Nanomaterials have been used widely for delivery of therapeutic agents. Protein–nanoparticle (NP) complexes have gained importance as vehicles for targeted drug delivery due to increased ease of administration, stability and half-life of drug, and reduced toxic side effects. Designing of phospholipid–bovine serum albumin (BSA) complexes and stealth NPs with BSA has paved the way for drug delivery carriers with prolonged blood circulation times. Preformed albumin corona has shown to decrease non-specific association and thereby reduce the clearance rate. Albumin corona has enabled the localization of drug carriers in specific tissues such as liver and heart, thus regulating biodistribution. Tailored albumin–NP conjugates have also enabled controlled degradation of NP and drug release. However, the binding of albumin with NP is associated with conformational and functional modulations in protein as observed with silver, gold and superparamagnetic iron oxide NPs. In this review, we highlight the various potential albumin–NP hybrids as nano drug carriers.

Keywords:

• Albumin
• drug delivery
• nanoparticles

Introduction

There has been an increased exploration of nanoparticles (NP) in the field of nanomedicine due to their unique physical and chemical properties. They are known to have high surface-free energy which makes them quite unstable. They acquire stability by binding to biomolecules such as proteins that reduces the free surface energy. Coating of proteins on NPs leads to the formation of “corona” that influences biodistribution, targeting and reduces the toxicity of bare NPs (Salvati et al., 2011; Monopoli et al., 2012). In the recent years, several studies have been undertaken to understand the mechanism and stoichiometry of binding of protein–NP bioconjugates as there are potential applications of these hybrids in nanomedicine. Albumins are known to be one of the residents in the corona formed around NPs (Dobrovolskaia et al., 2009). They act as surface active agents/surfactants by binding to NPs. Albumin is a non-glycosylated, negatively charged protein with a single polypeptide chain containing three homologous α-helical domains. It has a diameter of ∼10 nm. They are also one of the multifunctional abundant proteins in plasma with crucial physiological roles such as transport of drugs, free radical scavenging and maintaining osmotic pressure (Peters, 1996).

Nanomaterials such as metallic NPs (Asadishad et al., 2010), quantum dots (Gao et al., 2014), polymeric NPs, fullerenes, carbon nanotubes, liposomes, albumin NPs (Sripriyalakshmi et al., 2014), solid lipid NPs (Gaur et al., 2014) and nanostructured lipid carriers (Koo et al., 2005) have been investigated as nanocarriers for drugs. Bionanoparticles can be considered to be safer than metal-derived NPs which are potentially toxic and immunogenic. There are several advantages of using albumin-based NPs as a drug carrier. Being an endogenous protein, they can enhance the bioavailability and distribution of the drug. They are non-immunogenic and non-toxic. They are readily available, highly soluble and their surfaces can be modified and manipulated depending on the proposed application. Albumin-based NPs are easily accumulated in tumors due to enhanced permeability retention effect. Augmented albumin uptake in tumors is attributed to the interaction of albumin with albondin, a 60-kda glycoprotein (gp60) receptor and SPARC (Secreted Protein, Acidic and Rich in Cysteine) an extracellular matrix glycoprotein which is overexpressed in cancer cells. Albumin binds to gp60 that mediates the binding to caveolin-1, thereby initiating transcytosis and facilitating the accumulation of albumin in tumors. This mechanism forms the basis of nab technology wherein 130 nm paclitaxel bound albumin (Abraxane®, Los Angeles, CA) formulations have been developed (Desai, 2007). These solvent-free albumin-based formulations reduced the risk of hypersensitivity reactions caused by toxic solvents (Vishnu & Roy, 2011). Drug release in albumin-based systems can be initiated naturally by protease digestion or based on pH responsive system (Li et al., 2012). Another attractive advantage of albumin is that the payload release can be tuned by tuning the amount of albumin on NP (Cifuentes-Rius et al., 2013). An additional role of albumin in vivo demonstrated recently by Mortimer et al. (2014) is the unfolding of albumin on the NP surface thereby revealing a cryptic epitope that promotes recognition of NPs by scavenger receptors associated with mononuclear phagocyte system. Thus, albumin alone could direct NP uptake by human macrophages and enable clearance of these materials. In this review, we have attempted to provide a snapshot of albumin bound NPs as nano drug carriers, approaches to design them, their biodistribution and fate in vivo and potential albumin–NP bioconjugates that could be explored in future as candidates for drug delivery vehicles.

Classes of NPs with an albumin “halo”

Nanomaterials are organized into various groups based on their shape, composition and dimension. Association of proteins with these diverse types of nanomaterials is an area that has been delved into extensively in the recent times. Here, we specifically highlight the NPs that have been crowned with albumin.

Inorganic NPs

Metallic NPs

Colloidal metallic NPs such as gold and silver have been studied extensively with proteins as they are stable, easy to synthesize and show interesting optical properties. Formation of human serum albumin (HSA) corona on silver (Ag) NPs was examined by transmission electron microscopy. The average diameter of bare NPs were ∼30 nm and while that of AgNP–HSA corona was ∼80 nm. The interaction of these nanohybrids was further investigated with artificial lipid vesicles. It was demonstrated that both NPs and protein corona interacted with vesicles and enhanced its fluidity. Moreover, the presence of lipid vesicles alleviated the conformational changes of the protein induced by NPs possibly due to the electrostatic repulsion between the vesicles and NPs (Chen et al., 2012). Contrastingly, ZnO NP dispersions stabilized with bovine serum albumin (BSA) triggered an increase in ordering of lipids in phospholipid dioleoyl phosphocholine (DOPC) membranes (Churchman et al., 2013). Gold NPs capped with BSA and functionalized with various amino-glycosidic antibiotics have proved to be effective drug carriers. Streptomycin, neomycin, gentamicin and kanamycin loaded on these particles exhibited enhanced antibacterial activity against Gram-negative and Gram-positive strains, compared to pure antibiotic at the same concentration. As proposed by the authors, the plausible reason for enhancement of bacterial activity is the transport of a large number of antibiotic molecules into a highly localized volume at the site of bacterium–particle contact (Rastogi et al., 2012).

Magnetic NPs

Due to their magnetic properties, this class of NPs has been researched widely for theranostic applications. In a novel approach, Quan et al. (2011) developed HSA-coated iron oxide NPs (HINPs) and loaded 0.5 mg of doxorubicin (Dox) in 10 mg of the HSA matrices (Figure 1). The resulting Dox–HINPs were of ∼50 nm and showed a sustained release profile. The HINPs were found to assist the translocation of Dox across the cell membrane with its accumulation in the nucleus. A striking tumor suppression effect was observed in the 4T1 murine breast cancer xenograft model that outperformed free doxorubicin. Biocompatible ferrofluids with potential application in drug delivery has been designed. BSA adsorbed onto lauric acid-stabilized superparamagnetic iron oxide NPs (SPIONs) were used to carry payloads of up to 800 μg/ml of the cytostatic drug mitoxantrone without losing colloidal stability. The drug-loaded system exhibited excellent therapeutic potential in vitro, exceeding that of free mitoxantrone. The hybrid-coated particles were found to be biocompatible with primary human endothelial cells (Zaloga et al., 2014). In another approach, HSA-coated Fe₃O₄ NPs prepared by microemulsion technique has been radiolabeled with ¹⁸⁸Re and evaluated for labeling efficiency and stability. Labeling efficiency of 90% was achieved with the particles being stable upto 72 h in BSA (Zhang et al., 2004). These particles were found to be suitable for magnetically targeted therapy in vivo.

Figure 1. Targeted delivery by albumin-based nanoparticle.

Silica NPs

Silica NPs have attracted attention due to their high adsorption capacity, biocompatibility and optical properties. Alkylterminated porous silicon NP surfaces encapsulated with BSA via hydrophobic interactions were engineered as stealth NPs by Xia et al. (2013). The albumin coating improved their water-dispersibility and long-term stability under physiological conditions. It also reduced non-specific cellular uptake in vitro and prolonged blood circulation in vivo upon intravenous injection into mice. Thus, this approach circumvented the problem of fast biodegradation encountered when NPs are used as drug delivery agents. The biocompatibility of silica NPs is governed by the pore size and geometry of the NPs. Rod-shaped silica NPs (aspect ratio ∼3) demonstrated higher hemolytic activity than spherical ones and larger pore-sized NPs (3 nm) exhibited enhanced hemolysis in comparison to smaller pore-sized NPs (2 nm). Interestingly, corona of HSA on these silica NPs prevented RBC hemolysis (Ma et al., 2014) implying that the albumin coating strategy is efficient in enhancing biocompatibility.

Polymeric NPs

Polymeric NPs have demonstrated significant therapeutic efficacy and have been investigated for smart drug delivery because in contrast to other types of NPs they are biodegradable with minimal systemic toxicity. Polymeric particles can be categorized in two groups based on their origin, i.e. natural or synthetic. Natural polymers include chitosan, starch, cellulose, albumin and gelatin while synthetic polymers include polyethylene glycol, polylactic acid, polylactic-glycolic acid and poly(methyl methacrylate). Poly(isobutylcyanoacrylate) (PICBA) is another popularly characterized synthetic polymeric NP for drug delivery. Configuration of BSA adsorbed on these NPs grafted with dextran has revealed that albumin adsorbs in a flat configuration with their largest face bound to the PICBA surface and it accommodates 64 albumin molecules per NP (Vauthier et al., 2009).

Based on the method of preparation, polymeric particles can take the form of nanospheres in which the drug is dispersed throughout the polymer matrix or nanocapsules in which the drug is confined in a cavity surrounded by a unique polymeric membrane. Jung & Anvari (2012) has synthesized and characterized BSA-coated nanocapsules loaded with indocyanine green. The nanocapsules were formed by electrostatically crosslinking poly(allylamine) hydrochloride with phosphate anions. These polymeric NPs were further conjugated to BSA via glutaraldehyde. These nanoconstructs were found to be effective for intracellular optical imaging in normal human endocervical epithelial cells. As proposed, these nanoconstructs may potentially serve as a multifunctional platform for combined optical imaging, phototherapy and drug delivery.

Polymeric NP–albumin conjugates have proved to be beneficial in achieving steady and controlled release of the drug. Complexation of HSA with cholesterol-modified pullulan NPs loaded with mitoxantrone exhibited dual sustained drug release in vitro wherein the drug released from the NP surface was initially adsorbed to HSA. Then the drug–HSA complex rapidly bound to the particle surface. Subsequently, the drug released from the NP core was still adsorbed by the complexed HSA. In addition, mitoxantrone-loaded NPs coated with the adsorbed or completely complexed HSA molecules inhibited the drug release by a steric hindrance effect and repeated binding whereas mitoxantrone loaded on NPs uncomplexed with HSA showed rapid drug release (Tao et al., 2012).

In a study by Qi et al. (2010), BSA–dextran–chitosan NPs were fabricated by heating the mixture of chitosan and BSA–dextran conjugates. The formation the particles was favored by the electrostatic attraction between BSA and chitosan as well as the gelation of BSA. Doxorubicin was effectively loaded into the NPs after changing the pH of their mixture to 7.4 by virtue of the electrostatic and hydrophobic interactions between the NPs and doxorubicin. The antitumor effects of doxorubicin-loaded NPs were investigated by the tumor inhibition and survivability of murine ascites hepatoma H22 tumor-bearing mice. The loaded NPs were found to largely decrease the toxicity of doxorubicin and significantly increase the survivability of the tumor-bearing mice.

Lipid NPs

Being a biopolymer, lipid-based systems hold tremendous potential as drug carriers. A promising strategy explained by Peng et al. (2014) serves to prolong the circulation time in blood. It involved constructing a nanocomplex with some BSA molecules fixed on inner surface of BSA–phospholipid (PL)–NPs through hydrophobic interactions between the fatty acid chain of phospholipid and hydrophobic domain of BSA and others were located in the core area. These constructs not only exhibited antiadsorption and low phagocytosis but also showed slow zero-order drug release and enhanced nanostructure stability. Moreover, these nanocomplexes showed no cytotoxicity to mouse L929 fibroblasts but rather could stimulate the cells growth, thus proving to be biocompatible. Serum albumins have been used to surface modify lipid-based nanoparticular siRNA delivery systems in order to prevent non-specific association of serum molecules. Coating of targeted N-(1-aminoethyl)iminobis[N-(oleoylcysteinylhistinyl-1-aminoethyl)propionamide] (EHCO)/siRNA NPs with BSA at 9.4 μM prior to cell transfection improved cellular uptake and gene silencing efficacy of EHCO/siRNA-targeted NPs in serum-containing media, as compared with the uncoated NPs. At a proper concentration, albumin has the potential to minimize interactions of serum proteins with siRNA NPs for effective systemic in vivo siRNA delivery (Kummitha et al., 2012). Efficient transfection has been observed with HSA-coated liposomal formulations that have been evaluated for the delivery of antisense oligodeoxyribonucleotide (ODN) G3139 in KB human oral carcinoma cells. Treatment of the cells with HSA-coated liposome–ODN complexes resulted in efficient Bcl-2 mRNA downregulation that was ∼3-fold greater than with uncoated liposomes and 6-fold greater than with free ODN (Jung et al., 2010). Thus, these nanohybrids are efficient delivery vehicles for antisense ODN. In another approach, BSA-coated liposomes were thermally denatured (DBL) and encapsulated with the drug doxorubicin. The DBLs showed higher intracellular uptake and stability in plasma in comparison to Doxil®, Titusville, NJ (Weecharangsan et al., 2009).

Carbon nanotubes

They are 1D molecular tubes formed from sheets of graphite and having diameters of few nanometres. In a study by De Paoli et al. (2014), coronas of HSA, fibrinogen, IgG and histones were formed on carboxylated-multiwalled carbon nanotubes (CNTCOOH). Formation of HSA corona minimized the interaction of nanotubes with human blood platelets, preventing their aggregation, platelet membrane microparticle shedding and lactate dehydrogenase release. The HSA corona caused delayed production of reactive oxygen species and aided in proper dispersion of the nanotubes in contrast to Ig G, histone and fibrinogen coronas that triggered agglomeration. Thus, the albumin corona preserved the platelet membrane integrity, making this complex a potential drug carrier.

Quantum dots

They are semiconductor nanocrystals with excellent optical properties that make them suitable for diagnosis and therapy. Nigam et al. (2014) synthesized graphene quantum dots and conjugated them with hyaluronic acid functionalized HSA NPs for specific delivery to pancreatic cancer cells (Figure 1). Efficient cellular uptake of these functionalized NPs was observed in Panc-1 cell lines. Due to non-specific association, these anionic nanocarriers are preferred for targeted drug delivery over cationic counterparts.

Genesis of albumin corona on NPs

The most commonly adopted approach to form albumin corona on NPs involves incubation of the NP with albumin which results in physical adsorption of the protein on the NP surface with various interactive forces existing between them as summarized in Table 1. Separation of the albumin–NP complexes is usually achieved by centrifugation followed by washing to remove the unbound proteins (Goy-López et al., 2012; Peng et al., 2013). In some cases, albumin is added to the reaction system during the synthesis procedure of the NP itself so that a coating of albumin results (Singh et al., 2005). In the latter case, albumin plays the dual role of being a reducing agent and a stabilizer. Alternatively, albumin can be chemically conjugated to NPs as described by Lu et al. (2005). In this novel approach, cationic albumin (CBSA) was conjugated to pegylated NPs. Pegylated NPs were prepared by blending two copolymers methoxy-polyethyleneglycol–poly(lactide) (MPEG–PLA) and maleimide-PEG–PLA using emulsion/solvent evaporation technique. BSA was cationized with ethylenediamine and then thiolated using Traut's reagent. The thiolated CBSA was then conjugated with the pegylated NPs by mild stirring at room temperature with reaction of the thiol group from CBSA and the maleimide group that protruded out from the pegylated NP (Lu et al., 2005). Emulsification-solvent evaporation technique is another approach by which poly(d,l-lactic acid) (PLA₅₀) NPs coated with albumin have been readily obtained. In this method, an organic solution of PLA₅₀ was dispersed in aqueous solution of HSA that acts as a surfactant. The o/w primary emulsion was homogenized to reduce the emulsion droplet size. Evaporation of the organic solvent resulted in a colloidal suspension of PLA₅₀ NPs coated with albumin. Using microemulsion approach, HSA-coated magnetic NPs have been fabricated wherein a mixture of HSA, magnetite solution, emulsion agent and cotton oil were ultrasonicated for 15 min. The mixture was then rapidly added into cotton oil at 130 °C under stirring conditions. The particles were extracted using ether, washed with acetone and vacuum dried. In another strategy devised by Xia et al. (2013), albumin was encapsulated onto alkyl-terminated porous silicon NPs via a hydrophobic interaction. The Si NPs prepared by microwave-assisted synthesis were grafted with 1-dodecene to achieve higher hydrophobicity and were dispersed in BSA solutions by ultrasonication. The authors proposed that the amphiphilic BSA could be efficiently encapsulated onto Si NPs via hydrophobic interaction, whereas the large amount of hydrophilic carboxyl or amino groups of adsorbed BSA could enhance the aqueous solubility of Si NPs. Thus, to summarize physical adsorption, chemical conjugation, hydrophobic interactions, emulsion-solvent evaporation and microemulsion techniques are the strategies employed to fabricate albumin corona on NPs.

Table 1. Biophysical parameters of albumin–nanoparticle interaction.

Type of albumin–NP complex	Mechanism of quenching	Stern Volmer quenching constant at RT/37 °C (M^−1)	Binding constant (M^−1)	Microenvironment of tryptophan/tyrosine residues	Mode of interaction	Effect on secondary structure	References
BSA–SNP	Dynamic and static	5.06 × 10^9	1.512 × 10^14	Increased hydrophobicity around Trp residues	Hydrophobic	Loss of secondary structure	(Mariam et al., 2011, 2014)
BSA–ZnO	Static	7.74 × 10^6	1.7 × 10^9	Increased hydrophobicity around both Trp and Tyr residues	Van der Waal's and hydrogen bonding	Decrease in α-helix content	(Bhogale et al., 2013)
BSA–ZnO	Static	2.5 × 10^4	7.5 × 10^3	–	Electrostatic	Decrease in α-helix content from 64.51% to 54.83%	(Bardhan et al., 2009)
BSA–Cu	Static	6.43 × 10^8	5.01 × 10^6	Increased hydrophobicity around Trp residues	Hydrophobic	Decrease in α-helix content	(Bhogale et al., 2014)
BSA–starch capped CdS	Static	0.1995 × 10^6	6.6 × 10^2	Increased hydrophobicity around Trp residues	–	Conformation affected	(Jhonsi et al., 2009)
BSA–AgTiO2	Static	1.64 × 10^7	3.71 × 10^5	Increase in hydrophilicity around Tyr residues	–	–	(Kathiravan et al., 2009b)
BSA–iron oxide	Static	2.4 × 10^8	2.37 × 10^8	–	Electrostatic	Reduced α-helical content	(Yang et al., 2009)
HSA–GNP	Static	5.22 × 10^6	7.71 × 10^6	–	Hydrophobic	–	(Gao et al., 2005)
BSA–GNP	Static	5.67 × 10^6	1.12 × 10^7	–	Hydrophobic	–	(Gao et al., 2005)
HSA–TiO2	Static	6.58 × 10^5	3.49 × 10^5	Increased hydrophobicity around Tyr residues	–	–	(Kathiravan et al., 2009a)
BSA–TiO2	Static	1.01 × 10^4	5.25 × 10^5	–	–	–	(Kathiravan & Renganathan, 2008)

Biodistribution and fate of NPs with albumin corona

Adsorption of a particular protein on NP can influence the affinity and adhesion to a particular tissue (Figure 2). Arbós et al. (2002) quantified the bioadhesive properties of poly(methyl vinyl ether-co-maleic anhydride) (PVM/MA) NPs fluorescently labeled with rhodamine B isothiocyanate, and coated with either Sambucus nigra lectin (SNA–NP) or BSA–NP. The different formulations (10 mg) were administered to animals by the oral route and the fraction of adhered particles to the mucosa was estimated by measuring the fluorescent marker after the digestion of the tissue. BSA–NP displayed a highest initial affinity for the gut mucosa than for the control or SNA–NP, however, they were eliminated more rapidly from the mucosa than SNA–NP.

Figure 2. Biodistribution of albumin-coated NPs.

Most importantly, the corona is said to regulate the biodistribution. This is evident from the work of Borchard & Kreuter (1993) wherein radiolabeled poly(methylmethacrylate) (PMMA) NPs were coated with rat serum albumin (RSA), serum and inactivated serum, to examine the influence of these blood components on the body distribution of a model colloidal drug carrier. Serum complement inactivation was achieved by storage at 56 °C for 30 min. The different formulations of NP suspensions were injected intravenously via the tail vein of Wistar rats. After sacrificing, the animals at defined time points measurement of radioactivity of each sample revealed that coating with RSA led to no significant change in the body distribution of the particles, whereas incubation in serum, especially with complement inactivation prior to injection, very significantly reduced the uptake of particles into the organs of the reticuloendothelial system (RES), e.g. liver, spleen and bone marrow. At the same time, much higher concentrations of NPs were observed in the serum and in non-RES organs and peripheral tissues (kidneys, muscles and intestine). This effect was most pronounced after 30 min, but was still observable after 7 days. In another study, the long-term fate of SNPs functionalized with BSA was investigated by injecting these complexes in live rats. Very significant accumulations of NPs were confirmed by inductively coupled plasma mass spectroscopy (ICPMS) and transmission electron microscopy techniques in the liver and heart. In contrast, the brain tissue did not reveal evidence of particles. The authors suggested that Ag⁺ permeated across the blood–brain barrier (BBB) and followed swift clearance from the organ (Garza-Ocañas et al., 2010). In a study by Lu et al. (2005), CBSA–PEG–PLA NPs showed higher uptake than BSA–NP in rat brain capillary endothelial cells through absorptive mediated transcytosis (Lu et al., 2007) after injection in mice caudal vein. Fluorescent microscopy of brain coronal sections showed a higher accumulation of CBSA–NP in the lateral ventricle, third ventricle and periventricular region than that of BSA–NP, thereby making these conjugates a promising brain drug delivery carrier (Lu et al., 2005). The distribution of albumin–NP complexes is dependent on NP charge. The cellular binding of BSA–NP complexes formed from cationic NPs is enhanced, whereas the cellular binding of BSA–NP complexes formed from anionic NPs is inhibited irrespective of NP diameter or cell type. Using competition assays, it was demonstrated that BSA–NP complexes formed from anionic NPs bind to albumin receptors on the cell surface, whereas BSA–NP complexes formed from cationic NPs are redirected to scavenger receptors (Fleischer & Payne, 2014).

For drugs that are administered orally, the pattern of release of the drug from NP matrices into the gastrointestinal tract will be a determining factor in subsequent drug absorption. The mechanism of degradability of NPs in the gastrointestinal tract was studied by Landry et al. (1996). Entirely biodegradable poly (d,l-lactic acid) (PLA₅₀) NPs coated with albumin were prepared by the solvent evaporation technique and their degradative properties were investigated in simulated gastric and intestinal fluids. The albumin coating was found to undergo rapid degradation in both gastric and intestinal fluids However the PLA₅₀ core was gastroresistant and susceptible to intestinal fluid implying that if a drug is encapsulated within PLA₅₀ it will be released by passive diffusion in gastric fluid and actively by matrix erosion in intestinal fluid. Controlled drug release could be obtained by use of slower or non-digestible coating agents. PLA₅₀ NPs coated with either a readily digestible protein albumin or with a non-digestible coating agent such as polyvinyl alcohol (PVA) were prepared by the solvent evaporation technique. The NPs were administered perorally to guinea pigs to evaluate the gastrointestinal degradation of their PLA₅₀ matrix. In the case of PLA₅₀, NPs coated with digestible albumin, substantial gastrointestinal degradation of the PLA₅₀ matrix occurred, leading to the passage of considerable amount (≥45%) of water-soluble products across the gastrointestinal barrier. When a non-digestible coating agent like PVA was used, the degradation of the PLA₅₀ matrix in the gastro-intestinal tract was at least 2 times lower (Landry et al., 1998). Thus, by manipulation of surface coating, one can achieve controlled degradation of NP and drug release in vivo.

Non-specific interaction of NPs with plasma proteins leads to rapid clearance of NPs from blood after intravenous injection. This has been overcome by a preformed albumin corona on NP that forms a protective coating, thereby inhibiting the plasma proteins adsorption and decreasing the complement activation. Incubation of NPs with preformed albumin corona with rat serum inhibited Ig G adsorption consequently decreasing complement activation and resulting in lower opsonisation. The impact of BSA corona on NP uptake by macrophages was investigated on cultured mouse macrophage cells. A significantly higher phagocytosis ratio was observed with NPs than with BSA–NPs indicating reduced uptake of NPs with corona. Analysis of pharmacokinetic profiles of NPs and NP–BSA revealed that the half-life and mean residence time are extended by 6.0- and 6.7-fold, respectively, while the clearance rate (CL) is decreased by 2.5-fold, implying that the blood circulation time of NPs is significantly prolonged in the presence of the BSA corona (Peng et al., 2013). Thus, preforming albumin corona on NPs could be a very promising strategy for drug delivery.

Biophysical aspects of albumin bound NPs

Interaction of albumins with metallic, semiconductor and inorganic NPs has been explored. The forces and the thermodynamics involved in these interactions have been studied. Fundamental understanding of these parameters would enable one to deduce the pharmacokinetics and distribution of these materials in vivo. In Table 1, we have summarized the interactions of albumin with various NPs and the influence on microenvironments of tryptophan and tyrosine residues and their effect on secondary structure of albumin. As we can see from Table 1 that quenching in most of the cases is occurring by static mechanism and the binding is associated with alteration of secondary structure of albumin. The highest binding constants were obtained with silver–albumin complexes.

The mode of interaction with albumins is not only affected by the type of NP but also by its shape. Gold nanospheres and gold nanorods showed differential interaction with BSA. Nanorods resulted in fluorescence enhancement, whereas nanospheres caused quenching. Higher binding constants were observed as the aspect ratio of the nanorods increased (Pan et al., 2007). Chakraborty et al. (2011) has also reported differential response of gold nanorods and NPs toward BSA. Complexation of gold nanorods to BSA which was entropically driven lead to substantial loss of protein secondary and tertiary structures, whereas the native structure of BSA was retained on complexation to gold NPs which was enthalpy driven. The effect of shape of NP on albumin has also been revealed by molecular dynamic simulation studies wherein cubic gold NPs showed stronger unfolding effects on albumin than spherical gold NPs (Ramezani et al., 2014). The interaction of albumin with NPs is size dependent. Smaller sized gold NPs (8 nm) were found to quench the fluorescence of BSA more efficiently due to the increased surface area available with decrease in particle size. Binding constants were found to be higher with particles in lower size regime reflecting the adsorption of larger number of BSA molecules per unit area of NP surface (Pramanik et al., 2008). Apart from type, shape and size of NP, pH of the medium is another parameter that can result in varied protein–NP interactions. Studies of albumin–gold NP bioconjugates at pH of 3.8, 7 and 9 revealed that at higher pH the decrease in α-helical content was larger and mobility of Trp was restricted (Shang et al., 2007). Highest flocculation parameters were obtained for albumin–GNP complexes at pH 4, whereas at pH ≥5, it almost remained constant implying that aggregation becomes difficult due to repulsion of GNPs and serum albumins that bear negative charge at pH ≥5 (Gao et al., 2006). Similarly, adsorption of BSA on Ag NPs prevented their aggregation in solutions of pH >5. Thus, flocculation effects can be minimized by setting the medium at appropriate pH, which could have otherwise limited the use of NPs as drug carriers (Ravindran et al., 2010).

Dynamic light scattering technique has been employed to study the interaction of diamond NPs with albumin. It was found that the carboxylation of the surface of diamond NPs and their pre-coating by albumin leads to a decrease in the adsorption of protein molecules on the surface of NPs and to a decrease in their average hydrodynamic radius (Samsonova et al., 2012).

Dendrimer-coated magnetite NPs (MNPs) with amino surface groups were found to interact very strongly with BSA (higher binding constants) in comparison to dendrimer-MNPs with surface ester groups and MNPs modified with 3-aminopropyltrimethoxysilane (APTS). Thus, surface modification with amine groups would make these albumin–dendrimer–MNP bioconjugates efficient drug carriers (Pan et al., 2005). Interaction of titanium NPs of varying crystalline phases (rutile and anatase) with BSA was studied in order to elucidate the possible influence of the crystal phase on the behavior of proteins at the NP surface. However, interaction of BSA with the rutile and anatase phase of titanium showed similar adsorption curves although surface driven conformational changes were observed (Marucco et al., 2013).

Since we have a fundamental understanding underlying the binding of various albumin–NP conjugates, these complexes must be explored further for delivery of pharmaceuticals.

Structural and functional modulations in albumin–NP conjugates

Although albumin-coated NPs seems to be a promising carrier for drugs, it is important to evaluate the structural and functional properties of albumin upon adsorption to NP surfaces. As shown by Liu et al. (2013), doxorubicin-loaded SPION was found to unfold the framework conformation of BSA, leading to changes in the microenvironment of amide moieties. Similarly, chloroquine conjugated to gold NPs was found to interact efficiently with BSA, however the binding was associated with minor alterations in its secondary structure (Joshi et al., 2011). Interaction of ifosfamide-loaded SPION NPs with HSA has been studied at physiological pH. The complex was found to be stabilized by electrostatic interaction however CD spectra showed a loss in secondary structure of albumin (Kong et al., 2015). Adsorption of HSA on mesoporous silica NPs has reflected a major reduction in α helix content and increase of random coil, thereby suggesting a degree of protein unfolding. However, Khullar et al. (2012) showed that unfolding of BSA on gold NPs was an essential reaction component as it passivated the NP surface and these complexes showed little hemolysis and cytotoxic responses making them suitable for drug delivery applications. Perturbations in esterase activity, binding of warfarin and ibuprofen, antioxidant activity and copper binding capacity of albumin were observed on interaction with silver NPs (Mariam et al., 2014). Similarly, spherical gold NPs were found to impede the binding of ligands such as 8-anilino-1-napthalenesulfonic acid, oleic acid and butanoic acid in albumin (Dieni et al., 2013). Thus, binding of albumin to NPs leads to alterations in its structure and function which is a crucial point to be considered while designing these complexes for drug delivery.

Issues and challenges in the strategy

Though NPs with albumin corona are propitious drug carriers, there are many questions that remain to be answered and areas that have to be explored. When protein corona forms around a NP it can form a monolayer or multilayer. Is the formation and thickness of this corona controlled and is the system monodisperse and homogeneous? The orientation of albumin on NPs is known to take up either two conformations, i.e. “end on” or “side on” binding, with the end on binding resulting in higher surface coverage of albumin (Ravindran et al., 2010). In another study by Maffre et al. (2014), the orientation of HSA on different NPs has been discussed. HSA was found to adhere on most of the NPs with one of its two large triangular faces. The authors suggested the absence of denaturation in HSA capable of forming reversible corona on NP surfaces such as dihydrolipoic acid functionalized quantum dots, whereas on polymer-coated Fe-Pt NPs, HSA could possibly denature as evidenced by the incomplete desorption of protein from NP surface. Thus, it is important to evaluate and understand how this directional binding can influence the drug loading capacity and release. In vivo there could be a displacement of the proteins in the preformed corona with the proteins in vivo in accordance to the formation of hard and soft corona (Vroman et al., 1980). So what would be the stability and residence time of the proteins in the fabricated corona in vivo? Since albumin undergoes conformational and functional perturbation on adsorption to NPs, does this elicit inappropriate cellular responses in vivo? And finally, are these nanohybrids bioresorbable? How are they eliminated from the body once their job is complete? Thus a thorough understanding of the above aspects could open up new exciting avenues in the area of drug delivery.

Conclusion

Thus, versatile strategies for drug delivery can be developed with NPs constituting an albumin halo. Coating these nanohybrids with specific ligands can enable targeting of tumor cells and specific organs. Stability of albumin over a wide range of pH, its abundance in plasma, easy solubility, predictable biodistribution, biocompatibility, capacity to bind to diverse type of drugs, sustained release of drugs and reduced toxicity of bare NPs are the characteristics of albumin–NP conjugates that give them an edge over other nano drug carriers. However, one must also evaluate the changes in structural and functional properties of albumins adsorbed on NPs before establishing these conjugates as potential drug delivery vectors.

Declaration of interest

The authors declare no conflict of interest. The authors gratefully acknowledge DST INSPIRE [DST/INSPIRE fellowship/2010/206] for the financial support.