summary

Journal of Controlled Release 243 (2016) 303–322

Contents lists available at ScienceDirect

Journal of Controlled Release

journal homepage: www.elsevier.com/locate/jconrel

Review article

Hybrid protein-inorganic nanoparticles: From tumor-targeted drug delivery to cancer imaging

Ahmed O. Elzoghby a,b,⁎, Ayman L. Hemasa c, May S. Freag a,d a Cancer Nanotechnology Research Laboratory (CNRL), Faculty of Pharmacy, Alexandria University, Alexandria 21521, Egypt b Department of Industrial Pharmacy, Faculty of Pharmacy, Alexandria University, Alexandria 21521, Egypt c Department of Pharmacy, School of Medicine, Faculty of Health, University of Tasmania, Hobart, Tasmania, Australia d Department of Pharmaceutics, Faculty of Pharmacy, Alexandria University, Alexandria 21521, Egypt

⁎ Corresponding author at: Cancer Nanotechnology Res E-mail address: [email protected] (A.O. E

http://dx.doi.org/10.1016/j.jconrel.2016.10.023 0168-3659/© 2016 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history: Received 17 August 2016 Accepted 23 October 2016 Available online 26 October 2016

Recently, a great interest has been paid to the development of hybrid protein-inorganic nanoparticles (NPs) for drug delivery and cancer diagnostics in order to combine the merits of both inorganic and protein nanocarriers. This review primarily discusses the most outstanding advances in the applications of the hybrids of naturally-oc- curring proteins with iron oxide, gadolinium, gold, silica, calcium phosphate NPs, carbon nanotubes, and quan- tum dots in drug delivery and cancer imaging. Various strategies that have been utilized for the preparation of protein-functionalized inorganic NPs and the mechanisms involved in the drug loading process are discussed. How can the protein functionalization overcome the limitations of colloidal stability, poor dispersibility and tox- icity associated with inorganic NPs is also investigated. Moreover, issues relating to the influence of protein hy- bridization on the cellular uptake, tumor targeting efficiency, systemic circulation, mucosal penetration and skin permeation of inorganic NPs are highlighted. A special emphasis is devoted to the novel approaches utilizing the protein-inorganic nanohybrids in combined cancer therapy, tumor imaging, and theranostic applications as well as stimuli-responsive drug release from the nanohybrids.

© 2016 Elsevier B.V. All rights reserved.

Keywords: Nanohybrids Protein nanoparticles Inorganic nanoparticles Tumor targeting Anti-cancer drug delivery Cancer imaging Theranostics

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304 2. Hybridization strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305

2.1. Chemical conjugation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 2.2. Desolvation-chemical crosslinking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 2.3. In situ coating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 2.4. Spray-drying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306 2.5. Protein template-directed biomimetic synthesis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306 2.6. Miscellaneous . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

3. Drug loading mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 3.1. Covalent bonding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 3.2. Physical entrapment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

3.2.1. Desolvation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 3.2.2. Hydrophobic interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 3.2.3. Electrostatic attraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308 3.2.4. Soaking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308

4. Impacts of protein functionalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308 4.1. Improved biocompatibility and reduced toxicity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308

4.1.1. Reducing the release of free metal ions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308 4.1.2. Hiding the residual toxic capping agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308 4.1.3. Enhancing the water dispersibility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308 4.1.4. Reducing the immunotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308

earch Laboratory (CNRL), Faculty of Pharmacy, Alexandria University, Alexandria 21521, Egypt. lzoghby).

304 A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

4.1.5. Improved renal excretion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 4.1.6. Reducing the thrombogenic activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

4.2. Prolonged circulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 4.3. Improved colloidal stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 4.4. Altered skin permeation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310 4.5. Enhanced targeting efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310

4.5.1. Small molecules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310 4.5.2. Antibodies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310 4.5.3. Peptides. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311

5. Pharmaceutical applications of hybrid protein-inorganic nanoparticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 5.1. Combinatorial cancer therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311

5.1.1. Combined chemotherapy and magnetic targeting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 5.1.2. Combined chemotherapy and magnetic hyperthermia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 5.1.3. Combined chemotherapy and photothermal therapy (PTT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 5.1.4. Combined chemotherapy and photodynamic therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 5.1.5. Combined photothermal and photodynamic therapy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 5.1.6. Combined nanophotothermolysis and protein targeting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 5.1.7. Combined chemo- and radio-therapy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313

5.2. Cancer imaging. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 5.2.1. Improved MR contrast imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 5.2.2. Enhanced fluorescence imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 5.2.3. Dual imaging modality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 5.2.4. Multimodal imaging modality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 5.2.5. Blood-pool angiography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315

5.3. Cancer nano-theranostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 5.4. Stimuli-responsive drug release . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317

5.4.1. Magneto-responsive drug release . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 5.4.2. pH-responsive drug release . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 5.4.3. Enzyme-responsive drug release . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318 5.4.4. Thermo-responsive drug release . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318 5.4.5. Multi-stimuli-responsive drug release . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318

6. The physicochemical properties of nanohybrids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318 6.1. Particle size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 6.2. Surface charge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319

7. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320

1. Introduction

Nanocarriers provide endless opportunities in the area of anti-can- cer drug delivery. Therefore, the research in the area of Cancer nanomedicines has been progressively enlarged since the early 2000s. Nanoparticles offer the possibility to entrap poorly soluble anti-cancer drugs rendering them suitable for parenteral administration and modify their blood circulation and tissue distribution by their preferential accu- mulation at the tumor site. Furthermore, some nanosystems have been reported to overcome multi-drug resistance [1–3]. A great interest has been paid to the design of nanocarriers for therapeutic drug delivery with simultaneous imaging capability for monitoring of their biodistribution, the “nano-theranostics”. Nanoparticulate contrast agents are more advantageous to the conventional small contrast ones by exhibiting more prolonged systemic circulation in addition to its ca- pability for modification with targeting ligands to confer tissue-specific contrasting properties [4,5].

Inorganic NPs show optimal drug delivery characteristics, such as availability, biocompatibility, inertness and stability [5,6]. Compared to conventional polymer or lipid nanocarriers, the unique optical and mag- netic characteristics of inorganic NPs make them potential theranostic carriers for cancer therapy and imaging [6]. On the other hand, protein nanocarriers have been developed as drug delivery devices due to their safety, biodegradability, non-antigenicity, and significantly high drug binding potential [7]. For tumor-targeted delivery, protein nanocarriers can offer the following features: (a) Prolonged systemic circulation due to lower uptake by the reticuloendothelial system (RES) thus facilitating passive drug accumulation at tumor tissues by the enhanced permeation and retention (EPR) effect. (b) Proteins

contain multiple functional groups (e.g. NH2, COOH, and OH) available for drug conjugation or ligand-mediated tumor-targeting [8]. (c) Some proteins such as albumin and lactoferrin can enhance tumor targeting of anti-cancer drugs via interaction with specific receptors over- expressed on tumor cells. Abraxane®, albumin-bound paclitaxel (PTX) NPs, showed enhanced drug accumulation in solid tumors via binding to albondin gp60 and SPARC (secreted protein, acidic, rich in cysteine) receptors. (d) Some other proteins such as casein demonstrated ex- traordinary cancer cell penetration ability comparable to cell-penetrat- ing peptides [9,10]. (e) β-Lactoglobulin, gelatin, and elastin showed pH- , enzyme-and thermo-responsive drug release, respectively so they can be exploited in development of stimuli-responsive nanosystems. (f) Some plant proteins such as zein and gliadin can be used for controlled delivery of poorly soluble anti-cancer drugs by virtue of their hydropho- bicity [7].

Recently, a great effort has been devoted to the functionalization of inorganic NPs with proteins to combine the merits of both inorganic and protein nanocarriers and to mitigate the pitfalls of inorganic NPs. In addition to the reported toxicity of some inorganic NPs, their high chemical stability may hinder their metabolism resulting in their long- time accumulation in the body [11]. It was reported that PEGylated quantum dots (QDs) could reside in the body for two years [12]. More- over, inorganic NPs have a high tendency for aggregation due to their small size. Besides, hydrophobic capping agents such as oleic acid or oleylamine are usually coating the surface of inorganic nanocarriers thus hindering their aqueous dispersibility. These obstacles can impair the clinical utility of inorganic NPs for therapeutic or diagnostic pur- poses [11,12]. One way to overcome the pitfalls of inorganic NPs is to hybridize them with lipids, polysaccharides, and proteins. Among

305A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

them, proteins offered various advantages especially in tumor-targeted drug delivery and cancer imaging [4–6]. Therefore, this article reviews the advanced applications of protein-inorganic nanohybrids in drug de- livery to tumors and cancer cell imaging.

2. Hybridization strategies

Various techniques were successfully employed for development of hybrid protein-inorganic NPs including covalent conjugation and phys- ical entrapment via desolvation, coating or spray-drying. Other methods such as biomimetic synthesis, salting out and electrospinning may also be involved.

2.1. Chemical conjugation

By virtue of their chemical structure, proteins can be successfully conjugated to the surface of inorganic NPs via covalent bonding be- tween the protein reactive moieties (such as carboxyl and amino groups) and the functional groups conferred by the capping agent of NPs. Carbodiimide coupling is the most commonly used reaction for protein-NP conjugation via amide or ester bond formation. Arginine was employed to decorate the surface of iron oxide NPs (IONPs) to pro- vide free amine groups giving an opportunity for the amide bond forma- tion with the carboxylic groups of bovine serum albumin (BSA) [13]. Similarly, the carboxylic groups of both gadolinium diethylene triamine pentaacetic acid (Gd-DTPA) complexes and thioglycolic acid functional- ized CdTe quantum dots (TGA-QDs) were conjugated via carbodiimide coupling to the amino groups of HSA and BSA, respectively [14,15]. The use of ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochlo- ride (EDC) as a linker to conjugate TGA-QDs to BSA is found to be com- parable to bifunctional crosslinkers e.g. glutaraldehyde while avoiding their toxicity. Alternatively, to increase the protein amino groups for better conjugation efficiency, albumin could be cationized by replacing the side chain carboxylic groups with ethylenediamine. Then, amino groups of the cationized albumin were linked to the surface carboxylic groups of the citrate-capped magnetic NPs thus elaborating albumin- magnetic nanohybrids [16]. Thus, the conjugation efficiency has been improved to be 28% cationic albumin coupled to MNPs as demonstrated by Bradford assay.

A common linker; succinic anhydride could also be used to conju- gate mesoporous silica NPs (MSNs) to the proteins; gelatin, BSA, and ly- sozyme leading to fabrication of MSN-protein nanohybrids [17]. In this approach, the surface of amino-functionalized MSNs was first covalent- ly decorated with succinic anhydride molecules producing carboxylated MSNs. Then, the proteins were immobilized via their amino groups onto the surface of carboxylated MSNs by carbodiimide coupling. It was found that albumin and gelatin were coupled to MSNs in higher

Fig. 1. Schematic diagram of the desolvation technique for preparation of BSA-FeNi3 nanohybrid (modified from ref. [22]).

amounts compared to lysozyme, most likely because of their higher mo- lecular weight.

Another technique; Thiol-maleimide coupling was also utilized to link thiolated proteins to the surface of maleimide-derivatized inorganic NPs. Thiolated transferrin (Tf) was successfully attached to PEG– maleimide activated HSA-Gd-DTPA NPs with 84% Tf binding efficiency [18]. Another conjugation method involves the formation of Schiff- base bonds between residual aldehyde groups on the surface of inor- ganic NPs and amine groups of the proteins [19]. By treating amine- modified MSNs with glutaraldehyde, the aldehyde-functionalized MSNs were reacted with gelatin allowing the construction of gelatin co- rona on the surface of MSNs.

2.2. Desolvation-chemical crosslinking

The hydrophilic nature of proteins makes it possible to prepare pro- tein nanohybrids by organic solvent-induced desolvation from aqueous protein solution entrapping the inorganic NPs. In this technique, inor- ganic NPs were added as nuclei to the aqueous protein solution prior to desolvation by adding either ethanol or acetone thus producing sta- ble core-shell composite NPs. The abundant amino groups on the pro- tein surface enable further crosslinking using dialdehydes such as glutaraldehyde for enhanced stabilization of the developed nanohybrids. Fe3O4 and FeNi3 NPs were encapsulated within folate-con- jugated HSA [20], and BSA [21] nanospheres prepared via desolvation, respectively. Since albumin is negatively charged above its pI (isoelec- tric point; 4.8), BSA molecules were coated onto FeNi3 NPs with its an- ionic carboxylate group. Upon adding ethanol, the attached BSA molecules were denatured and fixedon the FeNi3 NPs exterior, which lead to the formation of stable core-shell FeNi3-BSA nanohybrids. Simi- larly, gold nanorods (AuNRs) were simultaneously encapsulated with paclitaxel (PTX) into HSA NPs by acetone desolvation procedure (Fig. 1). Albumin could strongly bind the surface of gold that is coordi- nated by amines and sulfur-containing moieties in albumin [22]. How- ever, the use of chemical crosslinkers such as glyoxal and glutaraldehyde was avoided to prevent quenching of the nanocluster lu- minescence [23]. In this technique, the protein coating acts as a protec- tive layer physically encapsulating the inorganic NPs, imparting stability and biocompatibility.

2.3. In situ coating

In this strategy, inorganic NPs were neither entrapped within pro- tein NPs via desolvation nor covalently bonded to the protein. Instead, the inorganic NPs were only coated with a protein layer via either elec- trostatic or hydrophobic interaction. Being charged macromolecules (above or below their pI), proteins can interact electrostatically with the oppositely-charged surface-capped inorganic NPs forming a protein

s (modified from ref. [21]), TEM image of AuNR encapsulated into HSA NPs by desolvation

306 A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

shell around the NPs. The cationic poly(allylamine hydrochloride)-coat- ed AuNPs bind electrostatically to the negatively charged HSA (above its pI ~4.7) [24]. HSA binding was concluded from the decrease in positive zeta potential (from +42 to +20 mV) and the increase in hydrodynam- ic diameter (from 77 to 112 nm). Similarly, the electrostatic interaction between negatively charged CdTe QDs and cationic Lysozyme (Lys) (at pH below its pI ~10) was utilized for preparation of QDs-loaded Lys- carboxymethyl cellulose coacervates. Then, the samples were heated at 80 °C for 30 min to induce thermal denaturation of Lys forming more stable nanogels [25]. It was found that the electrostatic complex- ation between QDs and Lys has significantly potentiated the fluores- cence intensity of QDs. This could be attributed to the enlarged size and light scattering of the nanogel complex. In another investigation, hybrid gelatin-anionic ammonium heptamolybdate (AHM), as model anti-cancer polyoxometalate NPs, were developed via electrostatic complexation between negatively charged AHM and the cationic frac- tion of zwitterionic gelatin [26,27]. Additional stabilization of the formed complex may be attributed to the repulsion offered by the an- ionic fraction of gelatin. i.p. injection of gelatin-AHM NPs with high HM content (70%)into hepatoma-bearing mice demonstrated a superior antitumor efficacy in comparison with free AHM [27].

Proteins may be layer-by-layer assembled onto the surface of inor- ganic NPs via electrostatic interaction. Positively charged BSA layer at pH 4 (below its pI) were adsorbed onto the negatively charged nitric oxide (NO)-releasing S-nitroso silica NPs (SNO-SiNPs) for controlling NO release. BSA binding has reduced the surface charge of bare SNO- SiNPs from −40 mV to become nearly neutralized/slightly positive with a size increase from 200 nm to 300 nm [28]. The polyanionic drug, suramin acted as a linker between cationic BSA layers where the positively charged amino acids (lysine, arginine, and histidine) of BSA molecules form salt bridges with negatively charged sulfonate groups of suramin [28].

Hydrophobic interaction between the hydrophobic amino acids within the protein sequence and the hydrophobic capping agent of the NPs could also be utilized as a hybridization mechanism. Based on the hydrophobic interaction between hydrophobic amino acids of gela- tin and the surface hydrophobic oleylamine molecules, oleylamine- coated IONPs could be transferred from chloroform to water by gelatin encapsulation (with the particle diameter increased to 178 nm) [29]. In another approach, the hydrophobicity of gelatin could be increased via grafting a hydrophobic moiety (hexanoyl anhydride) thus facilitated the hydrophobic interaction with the adsorbed oleic acid on magnetic nanocrystallites [30]. Upon self-assembly, the hydrophobized gelatin provoked aggregation of the IONPs to compose the core thus producing hybrid amphiphilic gelatin–iron oxide (AGIO) NPs.

In ligand exchange approach, a compatible ligand replaces the con- ventional hydrophobic capping layer of inorganic NPs (e.g. oleic acid or oleylamine) to improve the NP hydrophilicity [4,11]. Being consid- ered as amphiphilic copolymers composed of hydrophilic and hydro- phobic amino acids, proteins form a coating layer around inorganic nanocarriers via hydrophobic interaction with the oleic acid residue

Fig. 2. Schematic diagram of ligand exchange techniq

while exposing their hydrophilic fraction to the aqueous media. Howev- er, this conventional exchange technique requires harsh reaction condi- tions (heating or sonication) which can easily denature proteins. A novel exchange-encapsulation method was recently developed to transfer the hydrophobic IONPs to the aqueous phase to be capped with casein (CN-IONPs) [31]. In this process, glucose was first oxidized and polymerized into oligosaccharides which then partially replace oleic acid followed by encapsulation of IONPs within CN micelles to ob- tain water-soluble CN-IONPs (Fig. 2). The encapsulation resulted in swelling of micelles with a size increase from 38 nm to 142 nm and made the micelles respond to external magnetic field. Similarly, the hy- drophobic trioctylphosphine oxide (TOPO)-coated CdSe QDs were con- verted into hydrophilic QDs via a ligand-exchange method [32]. TOPO was exchanged with the hydrophilic ligand 11-mercaptoundecanoic acid then hydrophilic QDs were entrapped into gelatin NPs via desolvation.

2.4. Spray-drying

In this technique, hybrid inorganic-protein nanocomposite spheres could be prepared by encapsulating inorganic NPs into protein matrix by spray-drying-ultrasonic atomization. Then, heating the spray-dried protein-inorganic NPs (150–160 °C) above their denaturation tempera- ture leads to thermal crosslinking of the hybrid nanocomposites thus enhancing their stability without need for chemical crosslinkers. Using this technique, the fluorescence of spray-dried QD-BSA nanocomposites could be controlled by changing the QD size. Thus, multi-fluorescent NPs may be obtained via incorporation of different-sized QDs into indi- vidual BSA nanospheres [33]. The same technique was utilized to embed gold selenium nanostructures together with the photosensitizer zinc phthalocyanine (ZnPc) into BSA nanocomposite spheres [34]. This tech- nique offered many advantages including rapid and low-cost large-scale production of hybrid inorganic-protein nanocomposites. Moreover, this method produces nanoscale spheres (most of them are b600 nm) with narrow size distribution. The use of ultrasonic atomizer is advantageous as it produces smaller aqueous droplets (b5 μm) than those formed by the rotary atomizer (up to several hundred micrometers) resulting in large inorganic-protein microscale spheres with a broad size distribu- tion [33,34]. Finally, this preparation technique is safe with no residual organic solvent.

2.5. Protein template-directed biomimetic synthesis

Biomimetic techniques are highly recommended for the fabrication of inorganic nanocarriers due to the mild conditions utilized in their preparation. Biomolecules, such as polysaccharides, lipids, proteins, and nucleic acids have been explored as templates for manufacturing of nanomaterials. Among them, the stabilizing properties of proteins en- able their use as templates for biomimetic synthesis of inorganic NPs under mild conditions [35]. BSA was used as a stabilizer for the biomi- metic preparation of GdNPs. Then, the tyrosine residues of BSA were

ue for preparation of casein-coated IONPs [31].

307A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

completely iodinated via chloramine-T method to form iodinated BSA- GdNPs (2.5 nm) as a CT probe [36]. Injection of I-BSA-GdNPs did not in- duce any damage to susceptible organs demonstrating a good biocom- patibility. In another study, BSA was used as template for preparation of silica NPs. The cationic polyethyleneimine covers the surface of the negatively charged BSA core by electrostatic interaction and thus forming a compact sphere inducing silica precipitation via hydrolysis of tetramethyl orthosilicate [37]. The positively charged highly porous silica particles of small size (250–380 nm) were synthesized within mi- nutes compared to those without the protein core (700–1000 nm). Without albumin during silica precipitation, agglutinated NPs were ob- tained. The high positive charge of the NPs also enabled ionic binding of anionic bacterial-derived proteins with a high loading capacity (98.7 mg/g) of particles.

2.6. Miscellaneous

Other approaches including salting out, electrospinning and avidin- biotin coupling could be exploited for development of protein-inorganic nanohybrids. A salting-out technique was adopted to develop doxorubi- cin (DOX)-loaded magnetic silk fibroin (SF) NPs. SF solution was added to potassium phosphate solution of both DOX and MNPs under low- temperature. With the assistance of β-sheet formation catalyzed by the salting-out phenomena, the surface Fe3+ of MNPs, and the ice-iso- lated microcrystals, the SF protein forms well-defined spherical SFNPs entrapping both MNPs and DOX. MNPs act as starting nuclei in the salt- ing out procedure where the addition of MNPs significantly reduced the diameter of SFNPs from 1.4 μm to 130 nm [38]. Electrospinning technique is commonly used for preparation of hybrid protein-inorganic nanofi- bers. Fluorescent hybrid zein nanofibers encapsulating CdS QDs (about 330 nm) have been successfully fabricated via electrospinning [39]. In this technique, the QDs were directly dispersed into zein solution to ob- tain a homogeneous mixture. Then, the protein–inorganic NP solution was injected via a syringe pump where an electric field was established between the collection plate which acted as a cathode and the needle tip as an anode. Then, the nanofibers were gathered about 10 cm below the needle tip. Non-covalent avidin-biotin coupling was also used to prepare hybrid transferrin-QD conjugates where streptavidin-coupled TGA- coated CdTe/CdSe QDs reacted successfully with biotinylated Tf elabo- rating hybrid Tf-QDs [40].

3. Drug loading mechanisms

Drugs were loaded into the hybrid protein-inorganic nanocarriers via different mechanisms including covalent bonding, and physical en- trapment (via desolvation, hydrophobic or electrostatic interaction and soaking).

3.1. Covalent bonding

Owing to the numerous active groups in proteins, drugs can be cova- lently attached to the protein shell of the inorganic nanohybrids either directly or via an intermediate linker. The active amine groups on gela- tin were utilized for functionalization of Fe3O4-gelatin nanohybrids with two moieties: fluorescein isothiocyanate (FITC) through direct re- action with thiocyanate group and the carboxylated low-toxic platinum (IV) prodrug via carbodiimide coupling [29]. After gelatin digestion by pancreatin, the active form Pt(II) was released from the nanohybrids upon intracellular reduction. Free radical grafting reaction was also used to covalently conjugate the flavonoid catechin (CT) to gelatin followed by non-covalent incorporation of carbon nanotubes (CNTs) [41]. The hydroxyl radicals produced by the reaction between the redox initiator pair (H2O2 and ascorbic acid) attack the susceptible res- idues in the gelatin structure. Then, the resulting free radicals formed on the gelatin side chains react covalently with CT to achieve a loading of 0.9 mg of CT/g of gelatin. Alternatively, the drug may be indirectly linked

to the protein-inorganic nanohybrids via an intermediate linker. High DOX loading of about 85% could be achieved by indirect conjugation onto Tf-coated Fe3O4@SiO2 NPs via a multi-armed linker, poly-L- glutamic acid (PLGA) [42]. PLGA has many carboxyl groups and a termi- nal amino group so that it reacted with the carboxylic groups of NPs, via carbodiimide reaction, imparting many carboxyl groups onto the NP surface. Then, DOX could be anchored via its amino group onto the PLGA-coated layer of the NPs through amide bonding [42]. In the cova- lent bonding mechanism, the drug release is mainly dependent on the cleavage of covalent bond between the protein and drug. This bond is commonly designed to be stable in systemic circulation while can be cleaved within tumor cells in order to achieve site-specific release. The protein nanohybrids could induce release of the conjugated drug in tumor cell lysosomes consisting of several enzymes capable of digesting protein.

3.2. Physical entrapment

Covalent conjugation strategy requires the presence of reactive func- tional groups in the drug molecules. In addition, some other therapeu- tics may lose their efficacy in a conjugated form. Therefore, drugs can be readily physically encapsulated within protein nanohybrids. Many mechanisms can be involved in the drug entrapment together with in- organic NPs into protein matrix including desolvation, hydrophobic in- teraction or electrostatic attraction.

3.2.1. Desolvation Cisplatin was effectively encapsulated with high drug entrapment

(89.75%) and loading (15.25%) together with magnetic NPs within HSA NPs prepared by desolvation-chemical crosslinking method [20]. About 40% of cisplatin was released from HSA-magnetic NPs in normal saline after 24 h. In this technique, since the drug is not covalently con- jugated to the protein, the rate of drug release is mainly dependent on the biodegradation rate of the encapsulating proteins by proteolytic en- zymes. Moreover; the protein crosslinking density is another important factor affecting the rate of drug release from the protein matrix. On the other hand, the drug may be physically entrapped alone into the protein NPs then the inorganic NPs were conjugated to the drug-loaded protein NPs forming the hybrid nanocarrier. Gemcitabine was encapsulated into HSA NPs by desolvation method then the NPs were conjugated to graphene quantum dots (GQD) via carbodiimide chemistry [43]. De- spite its high water solubility, HSA NPs could prolong the release of gemcitabine up to 20 h following first-order diffusion-dependent pro- cess following an initial burst release.

3.2.2. Hydrophobic interaction Hydrophobic drugs can be incorporated within the layer of hydro-

phobic capping agent of inorganic NPs via hydrophobic interaction. Hy- drophobic DOX base was efficiently loaded together with IONPs in the inner hydrophobic layer of the amphiphilic copolymer poly(maleic acid)-octadecene [44]. Casein (CN) was then assembled onto the outer hydrophilic layer of the copolymer and crosslinked with glutaraldehyde to form layer-by-layer IONPs for oral DOX delivery (Fig. 3). Presence of the outer crosslinked CN layer effectively reduced the initial burst re- lease of DOX from the amphiphilic inner polymer layer in the acidic gas- tric medium. Thus, a higher DOX loading was reserved for release in the intestine. In another research work, incorporation of CTAB-capped AuNRs in the hybrid AuNR-HSA NPs resulted in about 45% enhancement in PTX loading to reach a concentration of 91.8 μg PTX/mL AuNR-HSA NPs solution able to induce apoptosis of cancer cells [22]. The hydropho- bic PTX could be embedded within the hydrophobic pocket created via hydrocarbon interaction between hydrocarbon chains of residual CTAB molecules during the formation of the HSA coating [22]. Additionally, increasing the HSA concentration caused an improvement in PTX load- ing. The amphiphilic structure of HSA entrapping AuNRs may have con- tributed to enhance PTX loading. Another contributing factor is the

Fig. 3. Entrapment of the hydrophobic DOX base together with IONPs in the inner hydrophobic layer of the amphiphilic copolymer poly(maleic acid)-octadecene coated by casein layer [44].

308 A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

protein crosslinking degree of the NPs where 40% crosslinking efficiency resulted in maximal drug loading. Crosslinking can keep the drug mol- ecules trapped into the NP matrix until its release.

3.2.3. Electrostatic attraction For ionizable water soluble drugs, charged drug molecules can inter-

act with the oppositely charged protein macromolecules or capping agent via electrostatic interaction. Both cationic DOX molecules and CaP were simultaneously co-deposited on the surface of the amphiphilic hexanoyl-gelatin-iron oxide (AGIO) core via electrostatic interaction with the anionic carboxylic groups of gelatin [30]. Such simultaneous fabrication of CaP-DOX nanoshell avoided premature drug escape dur- ing the conventional drug loading process. In this co-deposition tech- nique, CaP nucleation started first followed by continuous growth ending with the formation of corona with 38% DOX entrapment effi- ciency. In comparison, only 10% DOX was entrapped via a two-phase de- position loading in which cationic DOX molecules were first adsorbed to the anionic core followed by removal of un-entrapped DOX and then CaP was added forming the shell around AGIO core. Similarly, the nega- tively charged citrate-stabilized MNPs effectively adsorbed the positive- ly charged DOX via electrostatic interaction during the formation of silk fibroin-MNPs (SFNs) by salting out technique [38]. Further addition of MNPs during the salting-out process resulted in 50% enhancement in DOX loading compared with the direct loading of drug onto the SFNs in absence of MNPs.

3.2.4. Soaking Some types of porous NPs (such as MSNs) offer the opportunity for

post-preparation drug loading by soaking of the protein-inorganic nanohybrids in the saturated drug solution. DOX was loaded inside the mesopores of gelatin-coated MSNs by incubation in DOX solution for 48 h to reach the loading saturation [19]. A higher loading of DOX (7.4%) into gelatin-MSNs compared to uncoated MSNs (6.5%) may be at- tributed to the additional drug entrapped within the gelatin shell. More- over, gelatin coating forms a dense matrix blocking the pore exits thus preventing premature drug release in blood before reaching the tumor. Thus, both the high porosity and gelatin corona of the MSNs were essential for high drug loading.

4. Impacts of protein functionalization

Functionalization of inorganic NPs with proteins improved their physicochemical, pharmacokinetic, biological and toxicological charac- teristics via different mechanisms (Fig. 4).

4.1. Improved biocompatibility and reduced toxicity

Functionalization of the surface of inorganic NPs with proteins has been found to improve the biocompatibility of the NPs via different ways;

4.1.1. Reducing the release of free metal ions Coating the inorganic NPs with biocompatible polymers such as pro-

teins is reported to significantly improve their cytocompatibility via hin- dering its reaction with the bioenvironment and reducing the leakage of free metal ions. The cytotoxicity of Fe3O4 NPs may result from mem- brane damage caused mainly by ROS generation due to the release of metal ions resulting in oxidative stress, tissue inflammation, cell apo- ptosis, and DNA damage [45,46]. Moreover, Fe3O4 NPs can pass through the BBB causing iron-induced oxidative stress, and neuronal degenera- tion in the brain. Coating of Fe3O4 MNPs with silk fibroin (SF) demon- strated good neural cytocompatibility via reducing the release of iron from NPs resulting in decreased ROS generation thus promoting the dif- ferentiation of SK-N-MC neuroblastoma cells instead of cell apoptosis [46]. Similarly, HSA corona successfully reduced the leakage of Gd3+

ions from HSA-Gd-DTPA NPs to negligible (b5%) resulting in lower cyto- toxicity to Huh-7 hepatocellular carcinoma cells [14]. After incubation of HSA-Gd-DTPA NPs in blood serum for 1 h, the NPs retained as much as 95.5% of Gd3+ thus confirming the decreased in vivo release of free Gd3+ ions in circulation revealing their stability and safety. Compared to TGA-CdTe QDs, gelatin-coated QDs displayed superior biocompatibil- ity and lower toxicity on THP-1 macrophage cells via reducing the leak- age of toxic Cd2+ ions from the metalloid core [47].

4.1.2. Hiding the residual toxic capping agents Wrapping of CTAB-coated AuNRs by the protective HSA shell could

reduce the harmful effect of the residual toxic CTAB to cell membranes resulting in improved biocompatibility (N91% viability on 4T1 breast cancer cells) [22]. On the other hand, the free AuNRs without albumin coating induced a significant reduction in cell viability, mainly due to the CTAB traces on the AuNR surface.

4.1.3. Enhancing the water dispersibility One of the principal hurdles of pristine CNTs is their aggregation in-

duced by hydrophobic interactions between the sp2 carbon tube shells. Well-dispersed CNTs were found to be safe for mesenchymal stem cells. Coating with proteins enhanced the water dispersibility of CNTs mainly via breaking both the hydrophobic interface with water and the tube– tube interactions in the bundles [41,48]. Incorporation of CNTs into the gelatin-catechin conjugate could enhance their water dispersibility with minimal negative cellular effects. Similarly, since most photostable QDs are commonly synthesized in hydrophobic environments, protein functionalization has enabled synthesis of QDs in aqueous media and thus facilitating its application in a biological system [49].

4.1.4. Reducing the immunotoxicity The protein shell was found to significantly reduce the auto-immune

response to the inorganic NPs. Albumin-magnetic nanohybrids were less phagocytosed by macrophages resulting in higher cell viability for higher albumin concentration (45%) on mouse macrophage cells com- pared to the lower albumin composition (15%) [50]. It was also noticed that the interactions of surface silanol groups of MSNs with normal cell membranes induce potential toxicity of uncoated MSNs. Modifying

Fig. 4. Schematic diagram illustrating various impacts of protein functionalization of inorganic nanoparticles.

309A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

MSNs with protein corona could reduce their toxicity to normal cells, inhibiting their phagocytic uptake by macrophages [51]. The macro- phage RAW264.7 cells were severely activated after uptake of un-coat- ed MSNs inducing severe inflammatory responses as revealed by the high expression levels of inflammatory cytokines secreted by macro- phages (3.8-fold TNF-α and 4.2-fold interleukin-1b compared to the control group). In contrast, the immune inflammatory responses of MSNs were significantly mitigated after coating with natural proteins; gelatin, BSA, and lysozyme. The expression levels of both TNF-α and IL-1b in RAW264.7 cells incubated with protein-coated MSNs were sig- nificantly lower than those of un-coated MSNs.

4.1.5. Improved renal excretion Ultra-small BSA conjugated-Gd NPs (b3 nm) were found to be clin-

ically important as MRI contrast agents because they could be excreted via the renal system [52]. After i.v. injection of BSA-Gd NPs, 3 T T2MR images of the mouse liver displayed significant negative contrast en- hancements which then returned to the initial contrast after 24 h main- ly owing to the excretion of BSA-Gd NPs.

4.1.6. Reducing the thrombogenic activity Pristine and PEGylated CNTs were found to be strongly

thrombogenic may be because their negatively charged surface facili- tates the enzymatic activity of clotting factors [53]. Some plasma pro- teins e.g. fibrinogen and von Willebrand factor may play a role in the interaction of CNTs with blood cells where the CNT-adsorbed plasma proteins can induce platelet adhesion and activation. Since albumin

resists protein adsorption and platelet adhesion, coating CNTs with HSA alleviated their prothrombotic effect and hence reduced their po- tential to cause platelet aggregation or thrombus formation thus im- proving their biocompatibility [54]. HSA adsorbed on CNTs seems to cover the regions where platelet adhesion can occur.

4.2. Prolonged circulation

At physiological pH, the uncharged IONPs tend to agglomerate and precipitate thus quickly removed by RES macrophages preventing them from reaching the target cells. Compared with DOX alone, DOX- loaded HSA-coated IONPs demonstrated more prolonged circulation half-life (about 87 min vs 3 min) [55]. The size of NPs was too small (50 nm) to be captured by RES, meanwhile they were very large so that they cannot be eliminated by the kidney. An alternative method of creating long circulating intravascular contrast agents is by conjugat- ing the small molecular Gd-chelates to large proteins, such as HSA. Al- bumin binding reduced the rapid clearance of the contrast agents and enhanced the interaction between the Gd(III) ion electrons and the nu- cleus of water protons. Based on MRI findings, Gd-albumin conjugates demonstrated longer blood clearance half-lives (40–47 min) compared to Gd-dendrimer conjugates [56].

4.3. Improved colloidal stability

One of the interesting advantages of protein coatings is their ability to improve the physical stability and reduce aggregation of inorganic

310 A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

NPs. Casein (CN)-coated IONPs (30.5 nm) remained stable for months at 4 °C without aggregation or size change after 48 h incubation in pH 7.4 phosphate buffer or medium containing serum. This high stability may be partly attributed to the high negative surface charge (−38.0 mV) provided by large number of carboxyl groups in CN imparting sufficient electrostatic stabilization [31]. On another avenue, the protein shell could enhance the stability of the NPs during lyophilization. Coating of SNO-SiNPs and AuNPs by an outer layer of BSA or SF, respectively inhibited lyophilization-induced particle aggregation without any pro- tectants mainly via steric protection rather than electrostatic repulsion [28,57]. Lyophilization induced aggregation for the un-coated SiNPs (from 200 nm to around 1800 nm), but not for albumin-coated particles.

Proteins may also enhance the surface stability of QDs, which direct- ly influences their fluorescence properties. The maximum fluorescence emission wavelength and intensity of gelatin-CdTe QDs stayed essen- tially the same after 15 days storage due to the protecting effect of gel- atin which prevented the photo-oxidation damage of QDs [23]. Gelatin could also control the nucleation process of CdTe QDs resulting in 1.5 folds enhancement in the luminescence intensity. The gelatin functional groups successfully coated the free atoms leached from the metal inte- rior through electrostatic and hydrogen bonding interactions besides decreasing the availability of nonradiative centers [47]. Once formed, gelatin by its high viscosity hindered the diffusion of nanoclusters caus- ing slower Ostwald ripening.

Protein shell-crosslinking has been successfully exploited as an effi- cient strategy to improve the stability of protein-inorganic nanohybrids via inhibiting desorption of protein coating during blood circulation. Glutaraldehyde cross-linking of BSA shell of MNPs increased their col- loidal stability with only 15% particle size increase after 10 days incuba- tion with fetal bovine serum, compared to nearly 400% change for un- crosslinked BSA-MNPs [58]. In another approach, the acid etching-in- duced quenching of QDs photoluminescence was inhibited by coating the 11-mercapto-undecanoic acid capped-CdSe QDs (MUA-QDs) via layer-by-layer assembly. First, MUA-QDs were electrostatically immobilized onto the polycationic polydiallyldimethyl ammonium chloride (PDAC)-coated gelatin NPs. Then, a polyanionic layer of poly(sodium 4-styrenesulfonate) was added as a terminal coating layer in order to reduce the effect of acidity used during gelatin NPs preparation thus showing proton-resistant properties [59]. The gela- tin-QDs were resistant to quenching upon decreasing pH from 9 to 1 with the average lifetime (τave) of gelatin-QDs (about 914 ps) remained approximately the same compared to the τave of MUA-QDs which decreased significantly from 285 to 112 ps.

4.4. Altered skin permeation

Based on the nontoxic nature of silk fibroin (SF) to human skin, as well as its water-retaining, and skin-adhesive properties, SF-MNPs com- bined with magnetic field were explored for methotrexate (MTX) deliv- ery via the skin [60]. Compared to MTX solution, SF-Fe3O4 NPs increased the MTX percutaneous permeation by 2.78 folds. Application of alter- nating magnetic field enhanced the stratum corneum (SC) lipid fluidity and skin penetration by modification of the organized phase of the SC lipids. Additionally, the combined stationary and alternating magnetic fields enhanced the follicular transport and the percutaneous drug per- meation mainly by inducing a massage-like driving force that causes the movement of MNPs in the epidermis in both horizontal and vertical di- rections. On the contrary, conjugation with proteins may reduce the skin penetration of inorganic NPs. Various types of uncoated QDs could penetrate intact porcine and human skin [61]. When zein was conjugated to CdTe QDs, the diameter of QDs was significantly increased from 5 nm to 24 nm [62]. The uncoated QDs penetrated the stratum corneum of nude mice skin in vivo by passing through the intercellular corneocyte spaces (19 nm) reaching deeper dermal layers after 8 h followed by rapid clearance from the blood stream so no QDs were

observed after 24 h. The larger size of zein-QDs hybrid particles (about 24 nm) limited their penetration through the epidermis compared with smaller QDs. Thus, protein conjugation has changed the skin pen- etration of QDs reducing the risk of QD skin toxicity.

4.5. Enhanced targeting efficiency

The surface of protein-inorganic nanohybrids could be functional- ized with targeting ligands with the aim of site-specific tumor-targeted drug delivery or imaging.

4.5.1. Small molecules Active targeting with folic acid (FA) enables the NPs to reach and

penetrate into tumor cells that over-express FA receptors. Conjugation of FA to protein-coated NPs usually occurs via carbodiimide reaction be- tween folate carboxylic group to the amino groups of proteins. FA-DOX- BSA MNPs demonstrated a two-fold higher cytotoxicity to nasopharyn- geal carcinoma KB cells compared with non-targeted NPs [63]. To de- crease the non-selective internalization of FA-targeted NPs by normal non-cancerous cells, novel switchable FA-targeted nanocarrier ap- proaches have emerged including the use of shrinkable-spreadable polymer coating or degradable PEG shell. In the first approach, pH-re- sponsive polymers (PP, PNIPAM-MAA-b-AMA) were assembled onto the surface of Fe3O4-BSA NPs to hide or expose the FA molecule on the nanocarrier surface on-demand [64]. At pH 7.4, FA was shielded in the spreading hydrophilic polymer PP thus preventing FA from binding to the non-cancerous cells. At acidic tumor pH 5.5, the hidden FA was ex- posed due to the polymer PP shrinkage, and thus become able to bind to FR on the surface of cancer cells. The second approach introduced PEGylated gelatin onto FA-MSNs to protect the targeting ligand. In sys- temic circulation, FA-MSN was shielded against non-selective internali- zation by the PEGylated gelatin layer. At tumor sites, the gelatin layer was hydrolyzed by the over-expressed MMP-2 to remove PEG thus switching on FA activity and enhancing the internalization by tumor cells [65]. In addition to FA, EGCG which specifically binds the laminin receptor (67LR) over-expressed on some cancer cells was used to deco- rate DOX-gelatin-AuNPs showing improved cellular uptake and superi- or anti-cancer effect against PC-3 cells [66]. Phenylboronic acid (PBA) is another promising ligand for targeted drug delivery to liver cancer cells with a high affinity for sialic acid (SA), a well-known indicator for tumor metastasis up-regulated on HepG2 cells [67]. PBA-conjugated HSA was grafted onto the surface of MSNs as an end-capping agent showing en- hanced uptake by HepG2 cells [51]. In vivo, PBA-targeted HSA-DOX- MSNs effectively reduced the tumor growth rate in mice by targeting SA up-regulated on tumor cells.

4.5.2. Antibodies Different types of antibodies were conjugated to the surface of pro-

tein-inorganic nanohybrids to enhance their tumor targeting. The epi- dermal growth factor receptor (EGFR) is up-regulated on pancreatic and lung cancer cells. Therefore, cetuximab, an anti-EGFR monoclonal antibody, was conjugated to albumin-MNPs containing gemcitabine [68] and plasmid-survivin/shRNA [69] to mediate co-delivery of anti- body-targeted drug and Fe3O4 NPs to pancreatic and lung cancer cells. Simultaneously, magnetic-hyperthermia induced by MNPs could also occur parallel with double-targeted drug delivery. The NPs showed in- creased uptake, antitumor efficiency and apoptosis of cancer cells signif- icantly higher than those of non-targeted thermochemotherapy and chemotherapy alone. High-grade gliomas are characterized by severe necrosis and hypoxia, thus provoked secretion of angiogenesis vascular endothelial growth factor (VEGF). Therefore, monoclonal antibodies against VEGF (mAbVEGF)-conjugated to BSA-Fe3O4 NPs successfully en- abled visualization of glioma microvessels at the early time points (up to 2 h) with the detection of the highly active neoangiogenic areas 24 h after administration [58]. The drug amifostine was exploited as a linker to conjugate the amphiphilic gelatin-iron oxide (AGIO) NPs with

Fig. 5. Schematic diagram of the formation of the HER-AGIO@CaP-CD nanoparticles [70].

311A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

Herceptin (Trastuzumab), a monoclonal antibody that target HER2-pos- itive breast cancer cells. The phosphonate and amino groups of amifostine interacted with both CaP shell of AGIONPs and the carboxyl group of Herceptin, respectively (Fig. 5) [70]. The dual (ligand and mag- netic)-targeted NPs encapsulated curcumin and DOX showed dramati- cally enhanced internalization in the highly HER2-overexpressing SKBr3 breast cancer cells. Compared to non-targeted therapy, a more re- markable tumor growth inhibitory effect of the HER-AGIO@CaP NPs entrapping both drugs was observed due to the Herceptin-enhanced cellular internalization via the antigen–antibody reaction with the HER2 receptor.

In addition to protein-inorganic nanohybrids decorated with anti- bodies, several studies have revealed the enhanced tumor targeting ef- ficiency of inorganic NPs directly conjugated to antibodies for both cancer therapy and imaging. Xiangbao et al. specifically delivered the cytotoxic As2O3 NPs conjugated to anti-VEGFR-2 scFv antibody to liver cancer xenografts [71]. The nanoconjugates remarkably reduced the growth of human hepatoma cancer cells and increased the survival time of treated mice. In another study, cetuximab was conjugated to AuNPs as radiosensitizing agents to increase the radiation inhibitory ef- fect on head and neck tumor growth [72]. Thus, it could overcome the tumor radioresistance by increasing the radiation absorbed in the tumor. For tumor imaging applications, magnetic IONPs were conjugat- ed with anti-epidermal growth factor receptor 2 (HER2) single-chain antibody (scFv-IONPs) [73]. The developed scFv-IONPs successfully demonstrated HER2-specific tumor-targeted MR imaging.

4.5.3. Peptides t-PA-ligands, peptides derived from tissue plasminogen activator

with high binding capacity to galectin-1 receptor upregulated in pan- creatic cancer, were conjugated onto the surface of HSA-IONPs causing considerable accumulation of the circulating NPs in pancreatic tumor tissue [74]. MRI-based R2 relaxometry at 1.5 T could detect an enhanced uptake for the t-PA-targeted vs. non-targeted NPs. Tat and SynB pep- tides, another family of BBB cell-penetrating cationic peptides, could en- hance the uptake of gelatin–siloxane nanohybrids (GSNPs) by brain capillary endothelial cells owing to the electrostatic attraction between the cationic peptides and the anionic phospholipid heads in the bio- membrane [75,76]. Tat-GSNPs loaded with the plasmid encoding calci- tonin peptide (pLXSN-CGRP) demonstrated enhanced CGRP expression72 h after transfection [75]. Further use of additional fusogenic peptide (HA2) to Tat-modified GSNPs significantly enhanced their uptake, endosomal escape, and nucleus internalization resulting in adequate nucleic acid transfection both in vitro to HeLa and HEK-293

cells and in vivo to HeLa tumor-bearing mice [77]. Another tandem pep- tide of RGD and octaarginine (RRGD) enabled the best glioma targeting of gelatin-AuNPs-DOX, with good colocalization within neovessels in vivo [78].

5. Pharmaceutical applications of hybrid protein-inorganic nanoparticles

Hybrid protein-inorganic NPs have provided promising opportuni- ties for drug delivery and tissue imaging especially in the field of oncology.

5.1. Combinatorial cancer therapy

Combination of various cancer treatment strategies is favored to im- prove therapeutic efficiency and overcome drug-resistance while re- ducing the side effects. Chemotherapy can be combined with different therapeutic modalities such as magnetic hyperthermia, radiation, gene, photothermal and photodynamic therapies, to achieve synergistic anti-tumor effect (Table 1).

5.1.1. Combined chemotherapy and magnetic targeting Application of magnetic field on the tumor improves the accumula-

tion of anti-cancer drugs incorporated in magnetic NPs at the tumor tis- sue [79]. Magnetic tumor targeting does not need conjugation of specific ligands on the nanocarrier or upregulated receptors on the cancer cells [80]. Hybrid protein-magnetic NPs can offer combined magnetic- and protein-mediated tumor targeting. Two tumor-targeting mechanisms were involved by encapsulating HSA co-loaded with 5-flurouracil (5- Fu), and magnetic NPs into 800 nm nanocomposite spheres: an external magnetic targeting and internal targeting via binding of albumin to its receptors [81]. After administration, a magnet pulled the NPs into the tumor tissue preventing its diffusion to the rest of the body. On the other hand, HSA coating improved the extravasation and enhanced up- take of the magnetic nano-composites into cancer cells via its binding with albondin receptor and SPARC. The magnetically targeted nano- composites demonstrated enhanced anti-cancer efficacy in mice with SCC skin cancer [81]. In another study, DOX-loaded SF-magnetic NPs were i.v. injected into animals with two subcutaneous breast tumors on both sides and the left side tumors were linked to magnets [38]. After 2 h, the NPs accumulated and markedly inhibited the magnet-at- tached tumor compared to the tumor without magnet.

Table 1 Hybrid protein-inorganic nanoparticles used in combined cancer therapy.

Protein Inorganic nanoparticles Drug Combined cancer therapy Ref.

HSA Gelatin Silk fibroin

Fe3O4 NPs Fe3O4 NPs Fe3O4 NPs

5-Fu Platinum (IV) prodrug Doxorubicin

Combined chemotherapy and magnetic targeting [81] [29] [38]

BSA BSA BSA HSA

Fe3O4 NPs Fe3O4 NPs Fe3O4 NPs Fe3O4 NPs

Doxorubicin Plasmid-survivin/shRNA Gemcitabine 20(s)-ginsenoside Rg3

Combined chemotherapy and magnetic hyperthermia [63] [69] [68] [106]

HSA BSA HSA

Au NRs Au NPs Au NRs

Paclitaxel – Sorafenib

Combined chemotherapy and photothermal therapy [22] [107] [86]

HSA Mn(II)NPs Paclitaxel & Chlorine e6 Combined chemotherapy and photodynamic therapy [89] BSA BSA-polypyrrole

Au2Se/Au NPs Gd(III) NPs

Zinc phthalocyanine (ZnPc) Chlorine e6

Combined photothermal and photodynamic therapy [34] [90]

Gelatin Tf

MWCNTs MSNs

Catechin Selenocysteine

Combined chemotherapy and radio-therapy [93] [94]

HSA HSA

MWCNTs MWCNTs – –

Combined nanophotothermolysis and protein targeting [91,92]

312 A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

5.1.2. Combined chemotherapy and magnetic hyperthermia Heating tissues at temperatures of about 42–45 °C results in impair-

ment of cell functions via preventing enzymatic activity. Magnetic- targeted hyperthermia can precisely deliver heat to the target tumor cells via MNPs [82]. As a consequence, the tissue temperature can be tuned by the applied magnetic field, thus preventing heat-induced de- fects to non-target areas. Thus, MNPs combined with chemotherapy loaded-proteins allow the use of a lower dose of chemotherapy, thus avoiding damage to normal tissue and improving therapeutic efficacy not only by enhancing the killing of tumor cells but also by preventing the development of drug resistance. Under powerful magnetic hyper- thermia, DOX-loaded FA-targeted hybrid albumin MNPs had a dramatic effect on nasopharyngeal carcinoma NPC cell viability as well as 97% in- hibition of average tumor volume and weight in vivo at 1-week post- treatment [63]. It was found that co-administration of anti-cancer gene synergistically improved the tumor growth inhibitory rate of mag- netic hyperthermia (MH). Albumin NPs were simultaneously loaded with SPIONs, and anti-cancer plasmid pDONR223-IFNG, for targeted therapy of lung carcinoma GLC-82 cells. The apoptotic index of the gene and MH combined treatment group was 67.68%, whereas the indi- ces of the gene therapy group and MH group were 16.34%, and 20.04%, respectively [69].

5.1.3. Combined chemotherapy and photothermal therapy (PTT) Gold (Au) nanocarriers were successfully utilized in plasmonic

photothermal therapy (PPTT) of cancer because of their high near-infra- red radiation absorptivity, and consequent release of thermal energy [83]. This induced mild hyperthermia causes enhanced drug uptake and cellular damage whereas severe hyperthermia (N46 °C) results in cell death. Unfortunately, the clinical utility of PPTT is limited by the dis- persion of energy into the laser beam and the heat deterioration in cells far from the AuNRs [84]. Thus, design of a hybrid system containing AuNRs together with chemotherapeutic drug will allow killing and eradicating cancer cells even the resistant ones via both hyperthermia and chemotherapy [85]. The combinatory effects of PTX and AuNRs co-entrapped within HSA NPs resulted in 94% apoptosis of 4T1 mouse breast cancer cells following one irradiation cycle compared to 82% cell death without irradiation [22]. Additionally, PTX-AuNR-HSA NPs could be irradiated in vivo at tumor causing no damage to the surround- ing healthy cells. After 15 min of laser exposure, the NPs induced photothermal heating up to 46 °C resulting in dramatic necrosis. In a re- cent study, irradiation of 0.1 and 0.05 μM concentrations of HSA-AuNRs was found effective to produce high temperatures N50 °C thus inducing coagulative necrosis in vivo. In addition, the NPs could deliver sorafenib to renal cell carcinoma RCC 786–0 cells resulting in increased cytotoxic- ity compared to single HSA-AuNRs therapy [86].

As an approach to reduce the laser power required for PTT, AuNPs were coated with Tf, a protein that binds to TfR cell surface receptor up-regulated on cancer cells approximately 100-fold higher than on normal cells due to their rapid growth and increased iron requirements. Thus, the cellular uptake of Tf-AuNPs by breast cancer cells (Hs578T, ATCC) was enhanced by 6 folds compared to that of un-coated NPs [87]. Consequently, the required laser power for PTT was successfully reduced from 1600 W/cm2 to 7 W/cm2.

5.1.4. Combined chemotherapy and photodynamic therapy Photodynamic therapy (PDT) had been successfully used in the ther-

apy of some malignant tumors, where the photosensitizers (PS) were activated by visible light within tumor cells to produce reactive oxygen species (ROS) resulting in cytotoxicity [88]. With the assistance of PTX- induced albumin self-assembly via hydrophobic interactions, HSA mod- ified with the photosensitizer chlorine e6 (Ce6) and HSA-RGD co-as- semble forming NPs that could target αvβ3-integrin and enable combined photodynamic and chemotherapy [89]. Meanwhile, Ce6 acted as a chelating agent for Mn2+ to enable MR imaging. Under irra- diation by the 660-nm light for 30 min, the NPs caused synergistic kill- ing of U87MG cancer cells partly due to the accelerated endosomal drug escape by photodynamic effect-induced disruption of endo/lysosomes [89].

5.1.5. Combined photothermal and photodynamic therapy A beneficial synergistic effect for cancer phototherapy can be

achieved by combined photothermal PTT and photodynamic PDT ther- apy. After incubation with human esophageal carcinoma cells, the (Au2Se/Au and photosensitizer zinc phthalocyanine, ZnPc)-loaded BSA NPs synchronously demonstrated photothermal and photodynamic ac- tions by laser treatment (Fig. 6) [34]. Laser irradiation for 20 min in- creased the NPs temperature by over 20 °C resulting in accelerated molecular motion and faster release of ZnPc from the NPs for efficient photodynamic therapy [34]. BSA pre-conjugated with the photosensi- tizer Ce6 was used as stabilizing agent for fabrication of polypyrrolePPy-BSA-Ce6 NPs [90]. After i.v. administration of PPy- BSA-Ce6chelating Gd3+into cancer animal models, both PDT and PTT synergistically inhibited the tumor growth. Mild photothermal heating (to 43 °C) enhanced the cellular internalization of the photodynamic agent, and thus promoting the anti-tumor activity of the combined ther- apy. In addition, PDT sensitized tumor cells to thermal ablation, resulting in enhanced anti-tumor efficacy [90].

5.1.6. Combined nanophotothermolysis and protein targeting Carbon nanotubes (CNTs) have the capability to change near-infra-

red (NIR) laser beam into thermal energy. The photothermal killing of tumor cells associated with CNTs is termed as “nanophotothermolysis”

Fig. 6. Schematic representation of the synthesis of (Au2Se/Au and ZnPc)-encapsulated BSA nanospheres designed for cancer phototherapy [34].

313A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

[91,92]. To enhance their selectivity to cancer cells, HSA was non-cova- lently bonded onto the surface of oxidized MCNTs allowing selective at- tachment to albondin receptor on the cancer cells thus facilitating endocytosis. Therefore, a selective photothermal ablation of liver HepG2 [91] and pancreatic [92] cancer cells was achieved using HSA- CNTs. When HepG2 cells were incubated with HSA–CNTs and then treated by laser irradiation, the death rate of HepG2 cells ranged from 88.24% at 60 s to 92.34% at 30 min [91]. Similar extensive necrosis of ex-vivo pancreatic cancer specimens was observed upon treatment with HSA-CNTs under external laser irradiation [92].

In addition to their potential photothermal ablation of tumor cells, CNTs have a high affinity for cell membrane and a great capacity for cel- lular uptake by various mechanisms. This allows enhanced cellular ac- cumulation of the associated drug thus reducing the required drug dose and the risk of toxicity. Therefore, incorporation of CNTs into gela- tin-catechin (Gel-CT) conjugate enhanced its anti-cancer efficacy on HeLa cancer cells. At CT concentration of 150 μg/mL, free CT demonstrat- ed 20% cell killing, while CNT-CT and CNT-Gel-CT killed 38 and 72% of cells. The stabilizing effect of gelatin on CT also contributed to the supe- rior activity of CNT-Gel-CT compared with CNT-CT [41].

5.1.7. Combined chemo- and radio-therapy When combined with X-ray irradiation, catechin-loaded gelatin-

CNTs showed a reduction in the protein content of stem cell transcrip- tion factors and enhanced the cancer cell radiosensitivity [93]. A signif- icant inhibition of sphere composing ability of prostate cancer cells was observed only for X-ray irradiated DU145 cells pre-incubated with the catechin-gelatin-CNTs, revealing the powerful eradicating capacity of the combined strategies on cancer stem cells. Similarly, a synergistic chemo-/radiotherapy could be attained where selenocysteine (SeC)- loaded Tf/TAT-MSNs have remarkably increased the sensitivity of cervi- cal cancer cells to the growth inhibitory action of X-ray radiotherapy [94].Upon radiation, the NPs induced intracellular ROS overproduction, which caused apoptosis by affecting p53, AKT and MAPKs oncogenic mechanisms.

5.2. Cancer imaging

The protein-inorganic nanohybrids exhibited unique characteristics enabling their utility in various imaging applications (Table 2).

5.2.1. Improved MR contrast imaging Hybrid protein-inorganic NP-based contrast agents provide stronger

contrasting than the free contrast agent. Casein-coated Fe3O4 and Fe5C2

NPs demonstrated substantially higher r2 relaxivity and enhanced MRI contrast compared with un-coated NPs. Casein-coated Fe5C2 NPs exhib- ited a significantly high r2 of 973 mM

−1 s−1 compared to the reported T2 contrast agents [31,95,96]. This superior MRI contrast is likely caused by the soluble hairy phosphoprotein κ-casein coating the NPs that may present long, hydrophilic channels, enabling water molecules to enter and interact with the inner aqueous layer adjacent to particle surface [97,98]. Moreover, the protein coating increases the particle diameter and prevents fast water diffusion thus causing an enhanced τD, leading to increase of apparent r2 [31,96]. In addition to IONPs, protein nanohybrids with gadolinium (Gd), a promising T1-MRI agent, were de- veloped to overcome Gd3+ limitations such as non-specificity to target organs, high cytotoxicity and rapid elimination by renal filtration lead- ing to short imaging [99]. BSA-Gd@PEG-b-PLGA vesicles exhibited rapid improvement of the signal intensity after injection into HeLa tumor-bearing mice indicating the superior T1-MRI function of the ves- icles within short time, leading to fewer side effects [100].

Modification of the nanohybrids with specific protein ligands such as Tf and Lf with their receptors over-expressed on cancer cells leads to significantly improved MRI function [101,102]. Presence of TfRs on the brain capillaries endothelia enabled transcytosis of Tf-conjugated Gd- HSA-NPs resulting in significantly higher MRI signal enhancement in the brain compared with unmodified NPs [103]. In addition to brain, Tf-Gd-HSA and Tf-calcium phosphosilicate NPs could induce a higher contrast enhancement in the liver and MDA-MB-231 breast cancer cells due to additional expression of TfR2 and TfRs (CD71) on hepato- cytes and breast cancer cells, respectively [18,102]. When Lf was coupled to SPIONs, a higher uptake into glioma cells could be obtained via LfR-mediated endocytosis where LfRs (LRP1) are over-expressed in brain glioma compared to normal brain tissue. After parenteral admin- istration, the low intensity T2-MRI images was continual for 48 h with full details of tumor tissue [104,105].

5.2.2. Enhanced fluorescence imaging QDs are inorganic particles (1–10 nm) that constitute a potential

substitute to organic dyes as imaging tool because of their optical char- acteristic including strong fluorescence intensity, high quantum yield, resistance to photobleaching and size-related emission [32,49]. Hybrid- ization of QDs with proteins such as albumin, gelatin, lysozyme, and transferrin was previously reported [49]. The enhancement in the photoluminescence intensity of QDs in presence of BSA has been attrib- uted to the surface radiationless recombination [15]. Similarly, gelatin caused a substantial increase in emission intensity of QDs with quantum yield up to 1.5 times higher than those without gelatin [47].

Table 2 Hybrid protein-inorganic NPs used in cancer imaging and theranostic approaches.

Protein Inorganic NPs Imaging modality/Theranostic combination Ref.

Improved MR imaging CN Fe3O4 NPs Improved T2-MRI contrast [31,95] CN Fe5C2 NPs Improved T2-MRI contrast [96] BSA Gd2O3NPs Improved T2-MRI contrast [52] BSA Gd(III) NPs Improved T1-MRI contrast [100] HSA Gd(III) NPs Improved T1-MRI contrast [56]

Enhanced fluorescence imaging BSA CdTe QDs “Always on” fluorescent probe [15] Gelatin CdTe QDs “Always on” fluorescent probe [23,47] BSA CdTe QDs Multi-fluorescent NPs [33] Tf AuNCs/GO NCs “Activatable/turn-on” NIR fluorescence [109] Gelatin AuNFs Metal-enhanced fluorescence (MEF) [110]

Dual- & Multi-modal imaging modality BSA SPIONs T2-MRI/US contrast [111] HSA Gd-Rho123 T1-MRI and fluorescence imaging [14] BSA-polypyrrole Gd(III) NPs T1-MRI and fluorescence imaging [113] I-BSA Gd(III) NPs MRI/CT [90] BSA SPIONs-NIR 797 Superparamagnetic/NIRF imaging [36] Tf Fe3O4/NaYF4NPs Superparamagnetic and upconversion fluorescence imaging [114,124] Tf Fe3O4/MS-Cy7 MRI/NIRF imaging [125] HSA Fe3O4NPs SPECT–CT, gamma camera and MRI [74] BSA Ag2S-DTPA-Gd pQDs MR/NIRF imaging [112] Gelatin AuNPs Dark-field microscopy, FLIM, SERS [115] BSA AuNRs, IONPs and AuNCs T2-MRI, one-photon and two-photon photoluminescence (PL) [108] BSA Cypate-GdNCs NIRF/PA/MR imaging [116] HSA Gd(III) NPs Blood-pool MR angiography [18,36]

Nano-theranostics AG Fe3O4/CaPNPs DOX and T2-MRI contrast [30] BSA Fe3O4NPs DOX and T2-MRI contrast [29] Gelatin Fe3O4 NPs Pt(IV) prodrug and T2-MRI [126] BSA Gd(III) NPs DOX and T1-MRI contrast [100] HSA Gd(III) NPs siRNA and MRI [119] BSA Au NCs DOX and photoluminescence [117] HSA-HA Graphene QDs Gemcitabine and fluorescence imaging [43] Zein ZnS-Mn QDs 5-FU and fluorescence imaging [118] Lys CdTe QDs Methotrexate and fluorescence imaging [25] Lf AEC–CP–Fe-bLf NCs Lactoferrin and NIRF, MRI and CT [121–123] BSA Mn-Fe3O4NPs Photothermal therapy and T1-MRI [120] Tf Fe3O4/NaYF4 NPs Paclitaxel and MRI/NIRF imaging [125] BSA Fe3O4NPs DOX and T2-MRI [126] BSA-FITC Gd(III) NPs BCNU and T1-MRI/fluorescence [127]

314 A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

Gold nanoclusters (AuNCs) were reported as a fluorescent probe preferable to photo-bleachable organic dyes or toxic QDs because of their stable photoluminescence and inertness [108]. Furthermore, the two-photon luminescence of gold nanorods (AuNRs) falling within the NIR biological window of 650–900 nm could offer a novel in vivo imaging modality. Activatable “turn-on” rather than “always on” NIR fluorescent probe was designed where graphene oxide (GO), with its super-fluorescence quenching characteristics, was adsorbed onto Tf-AuNCs [109]. Due to the high binding affinity of Tf to TfR and the competition of TfR with GO for Tf in Tf-AuNCs/GO nanocomposite, the fluorescence of the activatable nanocomposite was efficiently

Fig. 7. Schematic diagram showing the Tf-AuNCs/GO composite as a Turn-On N

recovered after 4 h incubation with Hela cells (over-expressing TfR) (Fig. 7). No activation of the probe fluorescence occurred upon pre-blocking TfR on Hela cells with Tf for 6 h prior to incubation with the composite revealing the role of TfR in the activation of fluo- rescent signal [109].

Based on the metal-enhanced fluorescence (MEF) effect, fluorescent NPs have been fabricated by self-assembly of conjugated fluorene dicyano-1,4-phenylene-derived polymer (PFVCN), onto the gold nanoflower (AuNF)@gelatin NPs via ionic interaction [110]. Because the fluorophore emission would be quenched by direct linkage to the metallic NPs, the AuNF interior was hybridized with a gelatin layer via

IR fluorescent probe for bioimaging cancer cells over-expressing TfR [109].

315A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

sol–gel transition to separate the fluorophore and the metal core. A 10 nm thick gelatin shell resulted in 2-fold increase of the fluorescence intensity of PFVCN [110].

5.2.3. Dual imaging modality Because each imaging modality has specific merits and demerits,

combining two imaging techniques such as computed tomography (CT), magnetic resonance (MR), fluorescence or ultrasound (US) is strongly recommended. Hence, BSA-based microbubbles (MBs) were functionalized with amphiphilic carboxymethyl hexanoyl chitosan (CHC) micelles encapsulating SPIONs and camptothecin to be traced by both MR and/or diagnostic US for combined extravascular-intravas- cular imaging [111]. The nano-sized CHC/SPIO micelles were suitable for extravascular targeted imaging and delivery via wide fenestrations of the tumor leaky vasculature. As the microsize of MBs limits their ex- travasation, the MBs showing a concentration-dependent T2-MRI con- trast could enhance imaging of intravascular areas to observe the rate of angiogenesis. The B-mode contrast of the developed carriers was ob- vious in the blood vessels of rats as revealed in the in-vivo US images. Additionally, drug release was induced via 1 MHz therapeutic US for 30 min [111]. In another investigation, iodination of BSA-Gd NPs en- abled the detection of osteosarcoma via MR/CT [36]. After i.v. injection into osteosarcoma-bearing rats, the prolonged circulating I-BSA-Gd NPs accumulated in the tumor tissue through EPR effect, demonstrating both powerful X-ray attenuation, and enhanced MR contrasting. Among dual imaging modalities, MR/fluorescence dual imaging has attracted a great attention; (1) Non-invasive and non-radiative nature of both tech- niques permit their safe periodic use for body scan; (2) Infinite tissue penetration and immense 3D resolution allow successful use of MR im- aging in diagnosis; (3) Fluorescent imaging can overcome the low sen- sitivity of MRI at subcellular levels with its great detection sensitivity, rapid feedback and multiplexing [112]. Protein-Gd nanohybrids could provide a dual MRI/fluorescence imaging modality where HSA NPs were linked with both Gd-DTPA and the fluorescent dye, rhodamine [14,113]. These HSA-Gd-Rho NPs could be detected by MRI in vivo and also in tissue homogenates by their fluorescence. The i.v. injected HSA-Gd-Rho NPs were found to primarily accumulate to the liver, may be due to their internalization into liver macrophages. The NPs maintained powerful T1-MRI signal in the liver thus enhanced the de- tection of HCC tissues in a transgenic mouse model [113]. For confirma- tion, the highest fluorescence of the NPs was obtained in the liver as revealed by fluorescence activated cell sorting (FACS). In a recent study, NaYF4 Up-conversion NPs (UCNPs) were found to emit green fluorescence when treated with 980 nm wavelength NIR radiation. Therefore, Tf-labeled Fe3O4/NaYF4 nanocomposites with excellent up- conversion fluorescent and superparamagnetic characteristics were de- veloped to obtain fluorescent and magnetic imaging simultaneously [114].

5.2.4. Multimodal imaging modality Multifunctional nanohybrids were successfully exploited for inte-

gration of diagnostic multi-modality of inorganic NPs in order to achieve high sensitivity, precise tumor localization and good spatial resolution. As a novel multimodal imaging approach, time-resolved fluorescence imaging was combined with confocal Raman microscopy for evaluation of DOX release from gelatin-coated AuNPs and its accumulation inside MCF-7 cells [115]. In addition, the light scattering characteristics of AuNPs enabled its visualization via dark-field microscopy. DOX fluores- cence was completely quenched upon loading into gelatin-AuNPs while recovered after its release from the NPs. At the early time of cell treat- ment, wide-field fluorescence microscopy revealed no release of free DOX in the cells as indicated by the absence of any DOX signal. After 24 h of treatment, the drug was released from the NPs resulting in sig- nificant fluorescence increase throughout the cytoplasm and also in the nuclei. The few DOX molecules entrapped between NPs (“electro- magnetic hotspots”) after drug release, act as surface-enhanced

Raman scattering (SERS) reporters for the localization of plasmonic AuNPs in MCF-7 cells [115].

For simultaneous integration of plasmonic, magnetic and lumines- cence properties, multifunctional nanocarriers (MFNCs) were devel- oped by incorporating AuNRs, IONPs and AuNCs within BSA NPs via a simple desolvation process [108]. The plasmonic and magneto-lumines- cent MFNCs could be traced in HeLa cells based on their photoluminescence with either conventional one-photon or two-pho- ton imaging. In addition, the magnetic property exhibited by MFNCs due to presence of IONPs (T2-relaxivity = 448.04 mM

−1 s−1) revealed their potential capacity as T2-MRI contrasting agent. Finally, the plas- monic nature of the MFNCs was also exploited for in vitro photothermal therapy.

Upon photoirradiation, near-infrared cyanine dyes e.g. Cypate can demonstrate multi-imaging modalities such as near-infrared fluores- cence (NIRF) of above 800 nm, and photoacoustic (PA) effect, in addi- tion to photothermal effect [116]. Therefore, trimodal NIRF/PA/MR imaging could be achieved by grafting Cypate to the paramagnetic MRI agent gadolinium oxide nanocrystals (Cy-GdNCs) complexed with BSA. Compared to free Cypate, signals of Cy-GdNCs showed 23.2- fold increase of NIRF signals at tumors 24 h post-injection in mice. BSA could enhance the photostability of Cypate by preventing the damage of its unsaturated bonds [116]. GdNCs also acted as a good contrast agent for PA imaging in the range of 680–780 nm. Moreover, a fourfold increase of the T1-relaxivity of Cy-GdNCs compared to that of Gd-DTPA suggested MR imaging potential.

5.2.5. Blood-pool angiography Owing to their high molecular weight, nanoparticle-based contrast

agents conjugated to proteins do not diffuse through capillaries and elicit a longer circulation time in the blood than small molecule contrast agents, allowing a precise contrasting of the vasculature at lower doses. Thus, protein-contrast NPs allow longer imaging times in blood so they could be considered as blood-pool contrast agents for MR angiography. HSA-Gd-DTPA-NPs showed a longer circulation time in the blood than Gd-DTPA and showed significantly higher contrast enhancement in the blood providing longer MR imaging times in rat blood vessels [18, 36].

5.3. Cancer nano-theranostics

Many types of protein-inorganic nanohybrids with their unique op- tical, magnetic and electronic characteristics have been exploited as nano-theranostic probes for imaging and therapeutic applications. AGIO@CaP-DOX nanohybrids acted as theranostic nanocarriers success- fully taken up by HeLa cells where DOX escaped from the endosome and entered the nucleus. The NPs could also be used as a promising T2-MRI probe where they improved the T2 proton relaxation with high relaxivity values (r2) of 106.5 mM

−1 s−1 resulting in MR signal decrease with increasing the content of NPs [30]. Gemcitabine-HSA-graphene QDs [43], DOX-BSA-AuNCs [117], 5-FU-zein-ZnS/Mn QDs [118] and siRNA-HSA-GdNPs [119] are among many other successfully developed nano-theranostic approaches (Fig. 8). Instead of conventional chemo- therapy, another theranostic approaches may utilize photothermal therapy and cancer imaging. For example, the use of denatured BSA- coated manganese doped IONPs (MnIO-dBSA) remarkably intensified T1-MR imaging signal by two folds at the tumor tissue in 4T1 tumor an- imal model [120]. Post 20 min of intratumoral injection of MnIO-dBSA nanohybrids, the tumor was treated with 808 nm NIR irradiation to allow significant entry of the NPs into the tumor tissue. The temperature of the tumor was increased to 70 °C resulting in tumor regression one day following laser treatment.

Moreover, the protein itself may demonstrate potential anti-cancer effect e.g. bovine lactoferrin (bLf). Saturated Fe3O4-bLf nano- theranostics were developed and loaded onto calcium phosphate nanocrystals (NCs) then coated by alginate-chitosan layers [121–123].

Fig. 8. Depiction of the fabrication of DOX-loaded Au nanocluster embedded BSA nanoparticles as a theranostic carrier [117].

316 A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

The orally administered NCs targeted with locked nucleic acid modified aptamers against epithelial cell adhesion molecule (EpCAM) caused complete tumor growth inhibition in 90% of mice bearing triple positive xenograft colon cancer. This powerful effect may be attributed to: (a) Enhanced internalization and cytotoxicity in Caco-2 cells by utilizing different receptors including LfR, TfR or LRPR. (b) Inhibited degradation of encapsulated Fe-bLf in mice plasma. In addition, multimodal imaging could be obtained including NIR, MRI and computerized tomographic modalities [123]. In our laboratory, theranostic nanocarriers were

Fig. 9. A diagram illustrating stimuli-responsive dru

developed for synergistic co-delivery of pemetrexed and resveratrol to glioma cells. First, resveratrol-phospholipid complex was physically entrapped in HSA NPs by desolvation. Then, the amino group of albumin was coupled to the carboxylic group of pemetrexed via carbodiimide coupling. TGA-CdTe QDs as fluorescent probe were then conjugated to HSA NPs via amide bonding. The nanohybrids (~200 nm) maintained high fluorescence intensity of the QDs and demonstrated controlled re- lease of both drugs with a powerful cytotoxic effect to glioma cells (data to be published soon).

g release from protein-inorganic nanohybrids.

317A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

5.4. Stimuli-responsive drug release

The implementation of stimuli-responsive hybrid protein-inorganic nanocarriers offers a great opportunity for tumor-targeted drug and gene delivery. To enable on-demand drug release, a variety of control mechanisms have been explored so that the delivery system can re- spond to biological stimuli (e.g. pH, and enzymes) or external ones (e.g. magnetic field, hyperthermia, and light) which “trigger” drug re- lease from the nanocarriers (Fig. 9, Table 3).

5.4.1. Magneto-responsive drug release The enhanced drug release from hybrid protein-magnetic NPs upon

application of external magnetic field may be attributed to the align- ment of MNPs and subsequent expansion of the protein hydrogel net- works [128]. The release of DOX from magnetic polyacrylamide- gelatin nanocomposite hydrogel was increased by 11.5% after employ- ment of external magnetic field. Presence of gelatin has promoted the expansion ability and hemocompatibility of the magnetic hydrogel nanocomposites [128]. Another approach utilized magnetic field (MF)-induced perfluorohexane (PFH) gasification for magneto-respon- sive deep tumor-penetration and enhanced drug release [129]. Upon high-frequency MF exposure, the evolved heat increased the local pres- sure via gasification of PFH embedded in Lf-capped mesoporous iron oxide NPs (MIONPs). This high pressure induced rupture of the 3D tumor spheroids in addition to improved PTX release and enhanced nanocarrier penetration leading to inhibited tumor growth in 16 days after a single MF exposure.

Table 3 Hybrid protein-inorganic nanoparticles used in stimuli-responsive drug release applications.

Protein Inorganic nanoparticles Drug Stimuli-responsive drug

Magneto-responsive dru Gelatin Lactoferrin

Fe3O4 NPs Mesoporous Fe3O4 NPs

Doxorubicin Paclitaxel

Magnetic field induced a Magnetic field-induced p

pH-responsive drug rele Gelatin Amphiphilic gelatin

Transferrin Transferrin

Gelatin

Fe3O4 NPs Fe3O4/CaP NPs Fe3O4/CaP NPs CaP NPs AuNRs MSNs

MSNs

Doxorubicin Doxorubicin Curcumin Camptothecin Doxorubicin Doxorubicin

Doxorubicin

Weak interaction betwee Rapid dissolution of CaP

Acid cleavable hydrazone pH-sensitive nanovalve c at neutral conditions Detachment of gelatin co

Enzyme-responsive dru Casein Gelatin-FITC Gelatin Gelatin Gelatin HSA

Fe3O4NPs Fe3O4NPs MSNs QDs AuNPs MSNs

Doxorubicin Platinum (IV) prodrug Doxorubicin – Doxorubicin Doxorubicin

Casein degradation by in Gelatin degradation by p Gelatin degradation by M

Intermediate polypeptid

Thermo-responsive dru Gelatin HSA BSA/poly-L-lysine BSA

Fe3O4 NPs AuNRs AuNPs

AuNPs

Cisplatin Paclitaxel Doxorubicin

Fluorescein isothiocyanate

Increased molecular mot

Electromagnetic energy a

IRA radiation of hair folli particles and drug releas

Multi-stimuli-responsiv BSA

HSA

Fe5C2 NPs

CaP NPs

Doxorubicin

Platinum (IV) prodrug

pH and NIR light-dual re

pH and redox-dual respo

5.4.2. pH-responsive drug release The pH variation between systemic circulation and tumor compart-

ment could also be utilized as a biological stimulus to trigger drug re- lease from protein-functionalized inorganic NPs. Gelatin-coated IONPs showed pH-responsive DOX release resulting in enhanced drug release at pH 4 (61%) compared to pH 7.4 (32%) in 30 h [130]. This could be ex- plained by the positive charge of gelatin-coated IONPs at pH 4 leading to weaker interaction with DOX. Another pH-responsive strategy utilized the rapid dissolution of CaP in the acidic environment while being struc- turally stable at physiological pH [30]. DOX was slowly released at pH 7.4 (b20% for 24 h) from CaP/DOX shell-coated amphiphilic hexanoyl-gelatin iron oxide (AGIO@CaP-DOX) NPs whereas 100% of DOX was released after 10 h in pH 5. The increased osmotic pressure caused by the dissolved calcium and phosphate ions in endosomes/lyso- somes resulted in membrane damage, and drug release into the cyto- plasm. This pH-sensitive nature of the AG@CaP nanocarrier could control the dual drug sequential release of DOX/CUR and DOX/CPT from the pH-sensitive CaP shell and the degradable AG core, respective- ly, especially in the acidic pH within cellular compartments [70,131]. An acid cleavable hydrazone bond was also successfully used to conjugate DOX to Tf-AuNRs targeted to the TfR up-regulated on lung cancer (A549, HCC827) cells via maleimidocaproic acid hydrazine linker [132]. The fluorescence of free DOX was quenched when conjugated to AuNR and restored again when DOX was released from the NPs by hydrolysis in the acidic intracellular endosomal compartment. A novel pH-sensitive nanovalve was also formed by the inclusion complex of an aniline stalk conjugated to the of Tf-MSN surface, within α-cyclodex- trin (α-CD) to block the drug (DOX) within the pores at neutral

release mechanism Ref.

g release lignment of magnetic NPs and hydrogel swelling erfluorohexane gasification

[128] [129]

ase n protonated gelatin and cationic DOX at pH 4 in the acidic endolysosomal environment

bond using maleimidocaproic acid hydrazine linker omposed of an aniline-cyclodextrin inclusion complex sealing the pores

ating due to its protonation at acidic pH

[130] [30] [70] [131] [132] [133]

[140]

g release testinal protease ancreatic enzyme MPs

e linker (−PVGLIGG-) degradation by MMP-2

[44] [29] [19,65] [135,141] [66,78] [51]

g release ion of NPs and swelling of protein at higher temperature

bsorbed by Au NPs released as heat causing gelatin melting

cles induced plasmonic heating of AuNPs resulting in opening of the e

[136] [21] [137]

[142]

e drug release sponsive drug delivery

nsive drug delivery

[138]

[139]

318 A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

conditions [133]. Upon exposure to acidic conditions, the phenyl amine in the nanovalve was protonated (pKa ≈ 6) and the affinity between the stalk and CD decreased, inducing unbinding of CD and drug release from the pores (Fig. 10).

5.4.3. Enzyme-responsive drug release Protein-magnetic nanohybrids with enzyme responsive drug release

were reported taking the privileges of proteins as metabolizable struc- tures. The outer casein coating of DOX-loaded amphiphilic polymer- capped IONPs (CN-DOX-IONPs) showed enhanced resistance to degra- dation in enzymatic and acidic stomach fluid [44]. The outer glutaralde- hyde-crosslinked CN layer significantly suppressed the initial rapid release of the entrapped DOX from 62% to about 40% in simulated gas- tric fluid containing pepsin. In the intestine, CN coating was gradually degraded by intestinal protease thus exposing the uncoated inner am- phiphilic layer encapsulating DOX, resulting in enhanced drug release. After 6 h in simulated intestinal fluid containing trypsin, ~30% of DOX was released from CN-DOX-IONPs compared to only ~15% DOX from DOX-IONPs. Furthermore, CN dramatically increased the permeability of CN-DOX-IONPs through the small intestinal sac [44]. The CN-DOX- IONPs could penetrate the mucus deeply into the villi pits in contrast to un-coated DOX-IONPs hence increasing the effective plasma concen- tration. However, the mechanism by which CN enhanced the mucosal penetration still unclear may be related to the previously reported membrane penetrating activity of CN via an energy-independent mech- anism [134]. Similarly, the enzymatic degradation of gelatin by matrix metalloproteinases (MMPs) normally up-regulated in tumor tissues was responsible for uncapping of the pores of gelatin-coated DOX- MSNs resulting in quick DOX release at tumor sites [19]. For enhanced retention in deep tumor sites, the design of shrinkable gelatin NPs by MMPs was utilized to give relatively small-size NPs, that have higher transvascular and interstitial penetrating efficacy compared to the orig- inal gelatin NPs with low tumor penetration capability [78,135]. A core composed of 100-nm gelatin NPs was covered with 10 nm amino-PEG QDs as a multistage system (QDGelNPs). Following passing through the leaky tumor vasculature via the EPR effect, MMP-2 and MMP-9de- graded the 100 nm gelatin NPs resulting in the release of smaller 10 nm QDs. Oppositely to larger gelatin NPs, the smaller 10 nm NPs can easily penetrate the tumor's interstitial matrix. At 6 h post- intratumoral co-injecting gelatin-QD multi-stage NPs and Silica-QDs control in HT-1080 tumor bearing mice, gelatin-QDNPs had diffused up to about 300 μm while the Silica-QDs demonstrated low penetration from its administration site [135].Another approach utilized intermedi- ate polypeptide linker consisting of polyarginine and MMP-2 cleavable substrate (PVGLIG) for conjugating phenylboronic acid-linked HSA (PBA-HSA) to MSNs as a capping agent [51]. When the MSNs reach the tumor site, MMP-2 degrades the intermediate linker causing the de- tachment of the HSA sealing and consequent drug release. After incuba- tion with MMP-2 for 24 h, about 73% of DOX was released from the NPs compared to only about 15% of DOX released without MMP-2.

Fig. 10. Schematic representation of the pH-respo

5.4.4. Thermo-responsive drug release In response to magnetic-induced hyperthermia, enhanced thermo-

responsive drug release was reported from protein-functionalized mag- netic NPs. Yilmaz and Sanlier noticed a 1.8 fold increase in the hyper- thermic cisplatin release rate at 42 °C from gelatin-magnetic nanohybrids at pH 7.4in comparison with the non-hyperthermic condi- tions [136]. The higher hyperthermic drug release could be attributed to the enhanced molecular movement of MNPs as well as the gelatin swelling and relaxation ability at higher temperatures. The photothermal cancer therapy combines the advantages of both hyper- thermic ablation of tumor cells as well as enhanced drug release from the NPs only at the tumor region. By virtue of the photothermal-induced heating of the PTX-AuNR-HSA NPs at 46 °C, the NPs were capable of re- leasing PTX up to 188 ng after 15 min irradiation for additive treatment besides PTT [22]. In another investigation, DOX was fastly released in re- sponse to NIR irradiation via the photothermal effect of the AuNPs in- corporated together with DOX and catalase into heat-sensitive gelatin hydrogel within self-propelled BSA/poly-L-lysine multilayer rocket [137]. The AuNPs absorbed NIR irradiation, releasing it in the form of heat resulting in melting of the gelatin hydrogel and fast DOX release. The propulsion of the gelatin-based rockets was attributed to the ejec- tion of oxygen bubbles produced by catalase-induced decomposition of H2O2.

5.4.5. Multi-stimuli-responsive drug release Fe5C2-BSA-DOX NPs, good candidates for MR imaging and light-ab-

sorbing agents, showed a double controlling of DOX release by both pH value and NIR light [138]. Upon internalization of MNPs into endosomes/lysosomes, DOX release was activated by acidic pH due to weakened interaction between DOX and Fe5C2–BSA at low pH. Within 48 h, about 88% of DOX was released at pH 5.4 compared with about 22% released at pH 7.4. Meanwhile, a triggered DOX escape from these organelles to nuclei could be modulated by 808 nm NIR irradiation where mild photothermal heating could destroy the cell membrane and increase their permeability with significantly enhanced drug re- lease up to about 49% [138]. In another work, a pH and redox bi-respon- sive NPs were developed where the Pt(IV) prodrug of cisplatin was coupled to HSA-CaP NPs via carbodiimide reaction [139]. This conjugate was very stable in systemic circulation whereas upon cellular internali- zation into tumor cells, CaP was dissolved by the endosomal acidity. Thus, Pt prodrug was released from HSA in the active form (cisplatin) upon reduction in cells by the elevated GSH concentration (Fig. 11).

6. The physicochemical properties of nanohybrids

The physicochemical properties of protein-inorganic nanohybrids particularly hydrodynamic size and surface charge can significantly af- fect their cellular interactions and in vivo behavior thus influencing their efficacy in tumor-targeted drug delivery [4]. Furthermore, the unique optical, magnetic and imaging characteristics of nanohybrids

nsive property of Tf-nanovalve-MSNs [133].

Fig. 11. A diagram illustrating the preparation of Pt-HSA-CaP nanohybrids and the stimuli-responsive release of cisplatin [139].

319A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

were reported to be influenced by their physicochemical properties mainly size and charge.

6.1. Particle size

The hydrodynamic size of protein-inorganic nanohybrids varies ac- cording to different formulation parameters. Particularly, the hybridiza- tion technique can greatly influence the nanohybrid size. For example, covalent conjugation of protein shell to the surface of inorganic NPs usu- ally results in small-sized nanohybrids. BSA-conjugated IONPs exhibited a size of 50 nm compared to 28 nm for the original un-conjugated IONPs [13]. On the other hand, nanohybrids prepared by desolvation demon- strated a larger size mainly controlled by the desolvated protein NPs rather than inorganic ones. Encapsulation of AuNRs and PTX in HSA NPs by desolvation process resulted in nanohybrids with a diameter of 299 nm [22]. Spray-drying technique produced large-sized nanohybrids ranging from 450 to 600 nm [34].

Particle size is a major influencing parameter that can significantly affect the cellular internalization, in vivo biodistribution and tumor tis- sue penetration of the nanohybrids.

a) Protein coating usually increase the size of nanohybrids above the size limit for renal barrier (5.5 nm) thus preventing their fast clearance via urinary excretion resulting in prolonged circulation of nanohybrids. HSA-coating of DOX-IONPs greatly prolonged its plasma circulation by increasing their size to 50 nm as well as inhibiting their interaction with serum proteins [55]; b) Inorganic NPs with size b10 nm usually demonstrate broad distribution into several organs resulting in proba- ble toxicity. Therefore, NPs are preferred to be large enough to inhibit rapid diffusion [4]. Thus, the increased size of protein-inorganic nanohybrids reduces the risk of non-specific biodistribution leading to enhanced tumor accumulation and reduced toxicity; c) The small size of inorganic NPs such as QDs together with the metabolizable structure of proteins enabled the design of size-shrinkable protein-inorganic nanohybrids. The original 100 nm-sized gelatin-QD nanohybrids could extravasate via the EPR effect through tumor vasculature as they are smaller than the typical cutoff size of angiogenic fenestrations (100– 200 nm). Then, nanohybrids were degraded by MMP to elaborate 10 nm QDs which could efficiently penetrate the tumor interstitium by virtue of their small size [78,135].

6.2. Surface charge

The surface charge of the nanohybrids is commonly determined by the protein shell. a) Surface charge greatly influences the colloidal sta- bility of the nanohybrids by electrostatic stabilization. Protein functionalization provides high charge magnitude to the nanohybrids' surface thus imparting colloidal stability. The maintained aqueous dispersibility of CN-IONPs without agglomeration could be attributed to the high negative charge provided by CN coating [31].

b) Surface charge of nanohybrids also has a major impact on their in vivo behavior. Nanohybrids with positively charged surface were re- ported to be more bound on the negatively charged cell membranes by electrostatic interaction resulting in higher uptake. The cationic charge of Lf (below its pI 8.65) has a significant contribution to the en- hanced uptake of Lf-functionalized SPIONs into brain glioma cells in ad- dition to LfR-mediated endocytosis [104]. The cationic nature of Lf facilitates adsorptive-mediated transcytosis (AMT) of the nanohybrids via binding with negatively charged membrane surface proteoglycans. This resulted in preferential accumulation in brain tumor and conse- quently remarkable MR contrast of brain tumor tissues with Lf-SPIONs compared to SPIONs [104,105]. Compared to neutral nanohybrids, neg- atively charged nanohybrids were more significantly bound to mem- brane despite of the electrostatic repulsion. This may be due to their binding with the much less abundant positively charged domains of cell surface [143].

7. Conclusions

Proteins can be utilized as natural alternatives to synthetic polymers for fabrication of hybrid inorganic NPs for tumor-targeted drug delivery and/or imaging applications. Proteins offered numerous advantages mitigating the limitations of inorganic NPs such as reduced toxicity, en- hanced tumor targeting and improved colloidal stability. Protein coating has also been shown to modify the biological characteristics of inorganic NPs such as prolongation of their systemic circulation via reducing their interactions with serum components and increasing their overall size preventing their rapid excretion. In addition, some proteins such as HSA, Lf and Tf have a major role in the enhanced cellular internalization of the nanohybrids through interaction with specific receptors over- expressed on target cells (e.g. albondin, LfR and TfR, respectively). In ad- dition, the functional group-rich protein structure facilitated conjuga- tion of site-specific targeting ligands and/or covalent drug bonding to the nanohybrids.

The protein-inorganic nanohybrids showed many fruitful applica- tions in the field of oncology. First, the nanohybrids have been used in combined cancer therapy to overcome the limitations of chemotherapy such as limited efficacy, severe toxicity, and development of drug resis- tance. Besides, the nanohybrids also gained great benefits from the unique characteristics of inorganic NPs such as magnetic hyperthermia, photothermal cancer ablation, and high cell internalization capacity which enabled the use of those nanohybrids in various cancer therapeu- tic strategies. In addition, the versatile imaging modalities including MR and fluorescence imaging provided by inorganic NPs such as MNPs, GdNPs, QDs and AuNPs facilitated evaluating the in vivo biodistribution of the nanohybrids and their accumulation in tumor tissues. More im- portantly, the protein-inorganic nanohybrids demonstrated controlled drug release at the tumor sites in response to biological or external stim- uli including the endosomal low pH and the enzymatic digestibility of proteins by tumor microenvironment enzymes. In addition, MNPs and

320 A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

the plasmonically-active AuNPs exhibited responsive drug release at tumor sites upon application of external stimuli such as magnetic field or NIR irradiation, respectively.

References

[1] N. Bertrand, J. Wu, X. Xu, N. Kamaly, O.C. Farokhzad, Cancer nanotechnology: the impact of passive and active targeting in the era of modern cancer biology, Adv. Drug Deliv. Rev. 66 (2014) 2–25.

[2] A.O. Elzoghby, M.W. Helmy, W.M. Samy, N.A. Elgindy, Micellar delivery of flutamide via milk protein nanovehicles enhances its anti-tumor efficacy in andro- gen-dependent prostate cancer rat model, Pharm. Res. 30 (2013) 2654–2663.

[3] A.O. Elzoghby, N.I. Saad, M.W. Helmy, W.M. Samy, N.A. Elgindy, Ionically- crosslinked milk protein nanoparticles as flutamide carriers for effective anticancer activity in prostate cancer-bearing rats, Eur. J. Pharm. Biopharm. 85 (2013) 444–451.

[4] J. Nam, N. Won, J. Bang, H. Jin, J. Park, S. Jung, S. Jung, Y. Park, S. Kim, Surface engi- neering of inorganic nanoparticles for imaging and therapy, Adv. Drug Deliv. Rev. 65 (2013) 622–648.

[5] E.C. Cho, C. Glaus, J. Chen, M.J. Welch, Y. Xia, Inorganic nanoparticle-based contrast agents for molecular imaging, Trends Mol. Med. 16 (2010) 561–573.

[6] S. Bhattacharyya, R.A. Kudgus, R. Bhattacharya, P. Mukherjee, Inorganic nanoparti- cles in cancer therapy, Pharm. Res. 28 (2011) 237–259.

[7] A.O. Elzoghby, W.M. Samy, N.A. Elgindy, Protein-based nanocarriers as promising drug and gene delivery systems, J. Control. Release 161 (2012) 38–49.

[8] A.O. Elzoghby, M.M. Elgohary, N.M. Kamel, Chapter six—implications of protein- and peptide-based nanoparticles as potential vehicles for anticancer drugs, Adv. Protein Chem. Struct. Biol. 98 (2015) 169–221.

[9] A.O. Elzoghby, B.Z. Vranic, W.M. Samy, N.A. Elgindy, Swellable floating tablet based on spray-dried casein nanoparticles: near-infrared spectral characterization and floating matrix evaluation, Int. J. Pharm. 491 (2015) 113–122.

[10] A.O. Elzoghby, M.W. Helmy, W.M. Samy, N.A. Elgindy, Novel ionically crosslinked casein nanoparticles for flutamide delivery: formulation, characterization, and in vivo pharmacokinetics, Int. J. Nanomedicine 8 (2013) 1721.

[11] S. Kango, S. Kalia, A. Celli, J. Njuguna, Y. Habibi, R. Kumar, Surface modification of inorganic nanoparticles for development of organic–inorganic nanocomposites—a review, Prog. Polym. Sci. 38 (2013) 1232–1261.

[12] B. Fadeel, A.E. Garcia-Bennett, Better safe than sorry: understanding the toxicolog- ical properties of inorganic nanoparticles manufactured for biomedical applica- tions, Adv. Drug Deliv. Rev. 62 (2010) 362–374.

[13] T.T.-D. Tran, T. Van Vo, P.H.-L. Tran, Design of iron oxide nanoparticles decorated oleic acid and bovine serum albumin for drug delivery, Chem. Eng. Res. Des. 94 (2015) 112–118.

[14] W. Watcharin, C. Schmithals, T. Pleli, V. Köberle, H. Korkusuz, F. Huebner, S. Zeuzem, H.W. Korf, T.J. Vogl, C. Rittmeyer, Biodegradable human serum albumin nanoparticles as contrast agents for the detection of hepatocellular carcinoma by magnetic resonance imaging, Eur. J. Pharm. Biopharm. 87 (2014) 132–141.

[15] M. Idowu, E. Lamprecht, T. Nyokong, Interaction of water-soluble thiol capped CdTe quantum dots and bovine serum albumin, J. Photochem. Photobiol. A Chem. 198 (2008) 7–12.

[16] E. Cheraghipour, S. Javadpour, Cationic albumin-conjugated magnetite nanoparti- cles, novel candidate for hyperthermia cancer therapy, Int. J. Hyperth. 29 (2013) 511–519.

[17] Z. Luo, Y. Hu, R. Xin, B. Zhang, J. Li, X. Ding, Y. Hou, L. Yang, K. Cai, Surface function- alized mesoporous silica nanoparticles with natural proteins for reduced immunotoxicity, J. Biomed. Mater. Res. A 102 (2014) 3781–3794.

[18] H. Korkusuz, K. Ulbrich, K. Welzel, V. Koeberle, W. Watcharin, U. Bahr, V. Chernikov, T. Knobloch, S. Petersen, F. Huebner, Transferrin-coated gadolinium nanoparticles as MRI contrast agent, Mol. Imaging Biol. 15 (2013) 148–154.

[19] J.-H. Xu, F.-P. Gao, L.-L. Li, H.L. Ma, Y.-S. Fan, W. Liu, S.-S. Guo, X.-Z. Zhao, H. Wang, Gelatin–mesoporous silica nanoparticles as matrix metalloproteinases-degradable drug delivery systems in vivo, Microporous Mesoporous Mater. 182 (2013) 165–172.

[20] D. Chen, Q. Tang, W. Xue, J. Xiang, L. Zhang, X. Wang, The preparation and charac- terization of folate-conjugated human serum albumin magnetic cisplatin nanopar- ticles, J. Biomed. Res. 24 (2010) 26–32.

[21] X. Lu, J. Wu, G. Huo, Q. Sun, Y. Huang, Z. Han, G. Liang, Protein-passivated FeNi 3 particles with low toxicity and high inductive heating efficiency for thermal ther- apy, Colloids Surf. A Physicochem. Eng. Asp. 414 (2012) 168–173.

[22] D.V. Peralta, Z. Heidari, S. Dash, M.A. Tarr, Hybrid paclitaxel and gold nanorod-load- ed human serum albumin nanoparticles for simultaneous chemotherapeutic and photothermal therapy on 4T1 breast cancer cells, ACS Appl. Mater. Interfaces 7 (2015) 7101–7111.

[23] Y. Wang, H. Chen, C. Ye, Y. Hu, Synthesis and characterization of CdTe quantum dots embedded gelatin nanoparticles via a two-step desolvation method, Mater. Lett. 62 (2008) 3382–3384.

[24] M. Schäffler, F. Sousa, A. Wenk, L. Sitia, S. Hirn, C. Schleh, N. Haberl, M. Violatto, M. Canovi, P. Andreozzi, Blood protein coating of gold nanoparticles as potential tool for organ targeting, Biomaterials 35 (2014) 3455–3466.

[25] Z. Li, W. Xu, Y. Wang, B.R. Shah, C. Zhang, Y. Chen, Y. Li, B. Li, Quantum dots loaded nanogels for low cytotoxicity, pH-sensitive fluorescence, cell imaging and drug de- livery, Carbohydr. Polym. 121 (2015) 477–485.

[26] H. Yanagie, A. Ogata, S. Mitsui, T. Hisa, T. Yamase, M. Eriguchi, Anticancer activity of polyoxomolybdate, Biomed. Pharmacother. 60 (2006) 349–352.

[27] R. Guo, Y. Cheng, D. Ding, X. Li, L. Zhang, X. Jiang, B. Liu, Synthesis and antitumoral activity of gelatin/polyoxometalate hybrid nanoparticles, Macromol. Biosci. 11 (2011) 839–847.

[28] H.-C. Chou, S.-J. Chiu, T.-M. Hu, LbL assembly of albumin on nitric oxide-releasing silica nanoparticles using suramin, a polyanion drug, as an interlayer linker, Biomacromolecules 16 (2015) 2288–2295.

[29] Z. Cheng, Y. Dai, X. Kang, C. Li, S. Huang, H. Lian, Z. Hou, P. Ma, J. Lin, Gelatin-encap- sulated iron oxide nanoparticles for platinum (IV) prodrug delivery, enzyme-stim- ulated release and MRI, Biomaterials 35 (2014) 6359–6368.

[30] W.-M. Li, S.-Y. Chen, D.-M. Liu, In situ doxorubicin–CaP shell formation on amphi- philic gelatin–iron oxide core as a multifunctional drug delivery system with im- proved cytocompatibility, pH-responsive drug release and MR imaging, Acta Biomater. 9 (2013) 5360–5368.

[31] J. Huang, L. Wang, R. Lin, A.Y. Wang, L. Yang, M. Kuang, W. Qian, H. Mao, Casein- coated iron oxide nanoparticles for high MRI contrast enhancement and efficient cell targeting, ACS Appl. Mater. Interfaces 5 (2013) 4632–4639.

[32] L. Chen, A. Willoughby, J. Zhang, Luminescent gelatin nanospheres by encapsulat- ing CdSe quantum dots, Luminescence 29 (2014) 74–78.

[33] M. Chu, F. Wu, Q. Zhang, T. Liu, Y. Yu, A. Ji, K. Xu, Z. Feng, J. Zhu, A novel method for preparing quantum dot nanospheres with narrow size distribution, Nanoscale 2 (2010) 542–547.

[34] C. Yu, F. Wo, Y. Shao, X. Dai, M. Chu, Bovine serum albumin nanospheres synchro- nously encapsulating “gold selenium/gold” nanoparticles and photosensitizer for high-efficiency cancer phototherapy, Appl. Biochem. Biotechnol. 169 (2013) 1566–1578.

[35] S.-K. Sun, L.-X. Dong, Y. Cao, H.-R. Sun, X.-P. Yan, Fabrication of multifunctional Gd2O3/Au hybrid nanoprobe via a one-step approach for near-infrared fluores- cence and magnetic resonance multimodal imaging in vivo, Anal. Chem. 85 (2013) 8436–8441.

[36] Q. Wang, L. Lv, Z. Ling, Y. Wang, Y. Liu, L. Li, G. Liu, L. Shen, J. Yan, Y. Wang, Long- circulating iodinated albumin–gadolinium nanoparticles as enhanced magnetic resonance and computed tomography imaging probes for osteosarcoma visualiza- tion, Anal. Chem. 87 (2015) 4299–4304.

[37] E. Jackson, M. Ferrari, C. Cuestas-Ayllon, R. Fernández-Pacheco, J. Perez-Carvajal, J.M. de la Fuente, V. Grazú, L. Betancor, Protein-templated biomimetic silica nano- particles, Langmuir 31 (2015) 3687–3695.

[38] Y. Tian, X. Jiang, X. Chen, Z. Shao, W. Yang, Doxorubicin-loaded magnetic silk fibro- in nanoparticles for targeted therapy of multidrug-resistant cancer, Adv. Mater. 26 (2014) 7393–7398.

[39] B. Dhandayuthapani, A.C. Poulose, Y. Nagaoka, T. Hasumura, Y. Yoshida, T. Maekawa, D.S. Kumar, Biomimetic smart nanocomposite: in vitro biological evalu- ation of zein electrospun fluorescent nanofiber encapsulated CdS quantum dots, Biofabrication 4 (2012) 025008.

[40] C. Tekle, B.v. Deurs, K. Sandvig, T.-G. Iversen, Cellular trafficking of quantum dot-li- gand bioconjugates and their induction of changes in normal routing of unconju- gated ligands, Nano Lett. 8 (2008) 1858–1865.

[41] G. Cirillo, O. Vittorio, S. Hampel, U.G. Spizzirri, N. Picci, F. Iemma, Incorpora- tion of carbon nanotubes into a gelatin–catechin conjugate: innovative ap- proach for the preparation of anticancer materials, Int. J. Pharm. 446 (2013) 176–182.

[42] W. Ding, L. Guo, Immobilized transferrin Fe3O4@ SiO2 nanoparticle with high doxo- rubicin loading for dual-targeted tumor drug delivery, Int. J. Nanomedicine 8 (2013) 4631.

[43] P. Nigam, S. Waghmode, M. Louis, S. Wangnoo, P. Chavan, D. Sarkar, Graphene quantum dots conjugated albumin nanoparticles for targeted drug delivery and imaging of pancreatic cancer, J. Mater. Chem. B 2 (2014) 3190–3195.

[44] J. Huang, Q. Shu, L. Wang, H. Wu, A.Y. Wang, H. Mao, Layer-by-layer assembled milk protein coated magnetic nanoparticle enabled oral drug delivery with high stability in stomach and enzyme-responsive release in small intestine, Biomate- rials 39 (2015) 105–113.

[45] N. Singh, G.J. Jenkins, R. Asadi, S.H. Doak, Potential toxicity of superparamagnetic iron oxide nanoparticles (SPION), Nano Rev. 1 (2010).

[46] M. Deng, Z. Huang, Y. Zou, G. Yin, J. Liu, J. Gu, Fabrication and neuron cytocompatibility of iron oxide nanoparticles coated with silk-fibroin peptides, Colloids Surf. B: Biointerfaces 116 (2014) 465–471.

[47] S.J. Byrne, Y. Williams, A. Davies, S.A. Corr, A. Rakovich, Y.K. Gun'ko, Y.P. Rakovich, J.F. Donegan, Y. Volkov, “Jelly dots”: synthesis and cytotoxicity studies of CdTe quantum dot–gelatin nanocomposites, Small 3 (2007) 1152–1156.

[48] X. Xu, Z. Wang, Non-covalent dispersed carbon nanotube–benzocyclobutene com- posites as a bonding interface material for three-dimensional integration, Microelectron. Eng. 91 (2012) 33–38.

[49] J.Y. Lee, J.S. Kim, J.C. Park, Y.S. Nam, Protein–quantum dot nanohybrids for bioanalytical applications, Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. (2015).

[50] H.E. Misak, R. Asmatulu, J.S. Gopu, K.P. Man, N.M. Zacharias, P.H. Wooley, S.Y. Yang, Albumin-based nanocomposite spheres for advanced drug delivery systems, Biotechnol. J. 9 (2014) 163–170.

[51] J. Liu, B. Zhang, Z. Luo, X. Ding, J. Li, L. Dai, J. Zhou, X. Zhao, J. Ye, K. Cai, Enzyme re- sponsive mesoporous silica nanoparticles for targeted tumor therapy in vitro and in vivo, Nanoscale 7 (2015) 3614–3626.

[52] M.W. Ahmad, C.R. Kim, J.S. Baeck, Y. Chang, T.J. Kim, J.E. Bae, K.S. Chae, G.H. Lee, Bo- vine serum albumin (BSA) and cleaved-BSA conjugated ultrasmall Gd2O3 nanopar- ticles: synthesis, characterization, and application to MRI contrast agents, Colloids Surf. A Physicochem. Eng. Asp. 450 (2014) 67–75.

[53] A.R. Burke, R.N. Singh, D.L. Carroll, J.D. Owen, N.D. Kock, R. D'Agostino, F.M. Torti, S.V. Torti, Determinants of the thrombogenic potential of multiwalled carbon nanotubes, Biomaterials 32 (2011) 5970–5978.

321A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

[54] T.V. Vakhrusheva, A.A. Gusev, S.A. Gusev, I.I. Vlasova, Albumin reduces thrombogenic potential of single-walled carbon nanotubes, Toxicol. Lett. 221 (2013) 137–145.

[55] Q. Quan, J. Xie, H. Gao, M. Yang, F. Zhang, G. Liu, X. Lin, A. Wang, H.S. Eden, S. Lee, HSA coated iron oxide nanoparticles as drug delivery vehicles for cancer therapy, Mol. Pharm. 8 (2011) 1669–1676.

[56] K. Nwe, D. Milenic, L.H. Bryant, C.A. Regino, M.W. Brechbiel, Preparation, character- ization and in vivo assessment of Gd-albumin and Gd-dendrimer conjugates as in- travascular contrast-enhancing agents for MRI, J. Inorg. Biochem. 105 (2011) 722–727.

[57] L. Jia, L. Guo, J. Zhu, Y. Ma, Stability and cytocompatibility of silk fibroin-capped gold nanoparticles, Mater. Sci. Eng. C 43 (2014) 231–236.

[58] M.A. Abakumov, N.V. Nukolova, M. Sokolsky-Papkov, S.A. Shein, T.O. Sandalova, H.M. Vishwasrao, N.F. Grinenko, I.L. Gubsky, A.M. Abakumov, A.V. Kabanov, VEGF-targeted magnetic nanoparticles for MRI visualization of brain tumor, Nanomed.: Nanotechnol., Biol. Med. 11 (2015) 825–833.

[59] L. Chen, A. Siemiarczuk, H. Hai, Y. Chen, G. Huang, J. Zhang, Development of bio- compatible and proton-resistant quantum dots assembled on gelatin nanospheres, Langmuir 30 (2014) 1893–1899.

[60] A.-Z. Chen, L.-Q. Chen, S.-B. Wang, Y.-Q. Wang, J.-Z. Zha, Study of magnetic silk fi- broin nanoparticles for massage-like transdermal drug delivery, Int. J. Nanomedicine 10 (2015) 4639.

[61] S.H. Jeong, J.H. Kim, S.M. Yi, J.P. Lee, J.H. Kim, K.H. Sohn, K.L. Park, M.-K. Kim, S.W. Son, Assessment of penetration of quantum dots through in vitro and in vivo human skin using the human skin equivalent model and the tape stripping meth- od, Biochem. Biophys. Res. Commun. 394 (2010) 612–615.

[62] L. Wang, G. Qin, S. Geng, Y. Dai, J.-Y. Wang, Preparation of zein conjugated quan- tum dots and their in vivo transdermal delivery capacity through nude mouse skin, J. Biomed. Nanotechnol. 9 (2013) 367–376.

[63] F. Miao, Q. Tang, Preparation of folic acid-conjugated, doxorubicin-loaded, magnet- ic bovine serum albumin nanospheres and their antitumor effects in vitro and in vivo, Int. J. Nanomedicine 9 (2014) 4231–4243.

[64] Z. Shen, H. Wu, S. Yang, X. Ma, Z. Li, M. Tan, A. Wu, A novel Trojan-horse targeting strategy to reduce the non-specific uptake of nanocarriers by non-cancerous cells, Biomaterials 70 (2015) 1–11.

[65] Z. Zou, X. He, D. He, K. Wang, Z. Qing, X. Yang, L. Wen, J. Xiong, L. Li, L. Cai, Pro- grammed packaging of mesoporous silica nanocarriers for matrix metalloprotease 2-triggered tumor targeting and release, Biomaterials 58 (2015) 35–45.

[66] L.-C. Tsai, H.-Y. Hsieh, K.-Y. Lu, S.-Y. Wang, F.-L. Mi, EGCG/gelatin-doxorubicin gold nanoparticles enhance therapeutic efficacy of doxorubicin for prostate cancer treatment, Nanomedicine 11 (2016) 9–30.

[67] S. Deshayes, H. Cabral, T. Ishii, Y. Miura, S. Kobayashi, T. Yamashita, A. Matsumoto, Y. Miyahara, N. Nishiyama, K. Kataoka, Phenylboronic acid-installed polymeric mi- celles for targeting sialylated epitopes in solid tumors, J. Am. Chem. Soc. 135 (2013) 15501–15507.

[68] L. Wang, Y. An, C. Yuan, H. Zhang, C. Liang, F. Ding, Q. Gao, D. Zhang, GEM-loaded magnetic albumin nanospheres modified with cetuximab for simultaneous targeting, magnetic resonance imaging, and double-targeted thermochemotherapy of pancreatic cancer cells, Int. J. Nanomedicine 10 (2015) 2507.

[69] X. Hou, H. Zhang, H. Li, D. Zhang, Magnetic albumin immuno-nanospheres as an ef- ficient gene delivery system for a potential use in lung cancer: preparation, in vitro targeting and biological effect analysis, J. Drug Target. (2015) 1–10.

[70] W.-M. Li, C.-S. Chiang, W.-C. Huang, C.-W. Su, M.-Y. Chiang, J.-Y. Chen, S.-Y. Chen, Amifostine-conjugated pH-sensitive calcium phosphate-covered magnetic-amphi- philic gelatin nanoparticles for controlled intracellular dual drug release for dual- targeting in HER-2-overexpressing breast cancer, J. Control. Release 220 (2015) 107–118.

[71] Y. Xiangbao, W. Linquan, H. Mingwen, Z. Fan, W. Kai, Y. Xin, W. Kaiyang, F. Huaqun, Humanized anti-VEGFR-2 ScFv-As2O3-stealth nanoparticles, an antibody conju- gate with potent and selective anti-hepatocellular carcinoma activity, Biomed. Pharmacother. 68 (2014) 597–602.

[72] A. Popovtzer, A. Mizrachi, M. Motiei, D. Bragilovski, L. Lubimov, M. Levi, O. Hilly, I. Ben-Aharon, R. Popovtzer, Actively targeted gold nanoparticles as novel radiosensitizer agents: an in vivo head and neck cancer model, Nanoscale 8 (2016) 2678–2685.

[73] N. Ding, K. Sano, K. Kanazaki, M. Ohashi, J. Deguchi, Y. Kanada, M. Ono, H. Saji, In vivo HER2-targeted magnetic resonance tumor imaging using iron oxide nanoparticles conjugated with anti-HER2 fragment antibody, Mol. Imaging Biol. (2016) 1–7.

[74] I. Rosenberger, A. Strauss, S. Dobiasch, C. Weis, S. Szanyi, L. Gil-Iceta, E. Alonso, M.G. Esparza, V. Gómez-Vallejo, B. Szczupak, Targeted diagnostic magnetic nanoparti- cles for medical imaging of pancreatic cancer, J. Control. Release 214 (2015) 76–84.

[75] X.-H. Tian, Z.-G. Wang, H. Meng, Y.-H. Wang, W. Feng, F. Wei, Z.-C. Huang, X.-N. Lin, L. Ren, Tat peptide-decorated gelatin-siloxane nanoparticles for delivery of CGRP transgene in treatment of cerebral vasospasm, Int. J. Nanomedicine 8 (2013) 865.

[76] X. Tian, F. Wei, T. Wang, P. Wang, X. Lin, J. Wang, D. Wang, L. Ren, In vitro and in vivo studies on gelatin-siloxane nanoparticles conjugated with SynB peptide to in- crease drug delivery to the brain, Int. J. Nanomedicine 7 (2012) 1031–1041.

[77] S.-f. Ye, M.-m. Tian, T.-x. Wang, L. Ren, D. Wang, L.-h. Shen, T. Shang, Synergistic ef- fects of cell-penetrating peptide Tat and fusogenic peptide HA2-enhanced cellular internalization and gene transduction of organosilica nanoparticles, Nanomed.: Nanotechnol., Biol. Med. 8 (2012) 833–841.

[78] S. Ruan, Q. He, H. Gao, Matrix metalloproteinase triggered size-shrinkable gelatin- gold fabricated nanoparticles for tumor microenvironment sensitive penetration and diagnosis of glioma, Nanoscale 7 (2015) 9487–9496.

[79] G. Mikhaylov, U. Mikac, A.A. Magaeva, V.I. Itin, E.P. Naiden, I. Psakhye, L. Babes, T. Reinheckel, C. Peters, R. Zeiser, Ferri-liposomes as an MRI-visible drug-delivery

system for targeting tumours and their microenvironment, Nat. Nanotechnol. 6 (2011) 594–602.

[80] C. Wang, X. Sun, L. Cheng, S. Yin, G. Yang, Y. Li, Z. Liu, Multifunctional theranostic red blood cells for magnetic-field-enhanced in vivo combination therapy of cancer, Adv. Mater. 26 (2014) 4794–4802.

[81] H. Misak, N. Zacharias, Z. Song, S. Hwang, K.-P. Man, R. Asmatulu, S.-Y. Yang, Skin cancer treatment by albumin/5-Fu loaded magnetic nanocomposite spheres in a mouse model, J. Biotechnol. 164 (2013) 130–136.

[82] R. Gilchrist, R. Medal, W.D. Shorey, R.C. Hanselman, J.C. Parrott, C.B. Taylor, Selec- tive inductive heating of lymph nodes, Ann. Surg. 146 (1957) 596.

[83] W.I. Choi, J.-Y. Kim, C. Kang, C.C. Byeon, Y.H. Kim, G. Tae, Tumor regression in vivo by photothermal therapy based on gold-nanorod-loaded, functional nanocarriers, ACS Nano 5 (2011) 1995–2003.

[84] G.S. Terentyuk, A.V. Ivanov, N.I. Polyanskaya, I.L. Maksimova, A. Skaptsov, D.S. Chumakov, B.N. Khlebtsov, N.G.e. Khlebtsov, Photothermal effects induced by laser heating of gold nanorods in suspensions and inoculated tumours during in vivo experiments, Quantum Electron. 42 (2012) 380–389.

[85] R. Khandelia, A. Jaiswal, S.S. Ghosh, A. Chattopadhyay, Gold nanoparticle–protein agglomerates as versatile nanocarriers for drug delivery, Small 9 (2013) 3494–3505.

[86] C. Callaghan, D. Peralta, J. Liu, S.H. Mandava, M. Maddox, S. Dash, M.A. Tarr, B.R. Lee, Combined treatment of tyrosine kinase inhibitor–labeled gold nanorod encapsu- lated albumin with laser thermal ablation in a renal cell carcinoma model, J. Pharm. Sci. 105 (2016) 284–292.

[87] J.-L. Li, L. Wang, X.-Y. Liu, Z.-P. Zhang, H.-C. Guo, W.-M. Liu, S.-H. Tang, In vitro can- cer cell imaging and therapy using transferrin-conjugated gold nanoparticles, Can- cer Lett. 274 (2009) 319–326.

[88] B.W. Henderson, S.O. Gollnick, J.W. Snyder, T.M. Busch, P.C. Kousis, R.T. Cheney, J. Morgan, Choice of oxygen-conserving treatment regimen determines the inflam- matory response and outcome of photodynamic therapy of tumors, Cancer Res. 64 (2004) 2120–2126.

[89] Q. Chen, X. Wang, C. Wang, L. Feng, Y. Li, Z. Liu, Drug-induced self-assembly of modified albumins as nano-theranostics for tumor-targeted combination therapy, ACS Nano 9 (2015) 5223–5233.

[90] X. Song, C. Liang, H. Gong, Q. Chen, C. Wang, Z. Liu, Photosensitizer-conjugated al- bumin–polypyrrole nanoparticles for imaging-guided in vivo photodynamic/ photothermal therapy, Small 11 (2015) 3932–3941.

[91] C. Iancu, L. Mocan, C. Bele, A.I. Orza, F.A. Tabaran, C. Catoi, R. Stiufiuc, A. Stir, C. Matea, D. Iancu, Enhanced laser thermal ablation for the in vitro treatment of liver cancer by specific delivery of multiwalled carbon nanotubes functionalized with human serum albumin, Int. J. Nanomedicine 6 (2011) 129.

[92] L. Mocan, F.A. Tabaran, T. Mocan, C. Bele, A.I. Orza, C. Lucan, R. Stiufiuc, I. Manaila, F. Iulia, I. Dana, Selective ex-vivo photothermal ablation of human pancreatic cancer with albumin functionalized multiwalled carbon nanotubes, Int. J. Nanomedicine 6 (2011) 915–928.

[93] A. Castro Nava, M. Cojoc, C. Peitzsch, G. Cirillo, I. Kurth, S. Fuessel, K. Erdmann, D. Kunhardt, O. Vittorio, S. Hampel, Development of novel radiochemotherapy ap- proaches targeting prostate tumor progenitor cells using nanohybrids, Int. J. Cancer 137 (2015) 2492–2503.

[94] L. He, H. Lai, T. Chen, Dual-function nanosystem for synergetic cancer chemo-/ra- diotherapy through ROS-mediated signaling pathways, Biomaterials 51 (2015) 30–42.

[95] J. Sangeetha, J. Philip, The interaction, stability and response to an external stimu- lus of iron oxide nanoparticle–casein nanocomplexes, Colloids Surf. A Physicochem. Eng. Asp. 406 (2012) 52–60.

[96] T.A. Cowger, W. Tang, Z. Zhen, K. Hu, D.E. Rink, T.J. Todd, G.D. Wang, W. Zhang, H. Chen, J. Xie, Casein-coated Fe5C2 nanoparticles with superior r2 relaxivity for liver- specific magnetic resonance imaging, Theranostics 5 (2015) 1225.

[97] S.H. Koenig, K.E. Kellar, Theory of 1/T1 and 1/T2 NMRD profiles of solutions of mag- netic nanoparticles, Magn. Reson. Med. 34 (1995) 227–233.

[98] A.O. Elzoghby, W.S.A. El-Fotoh, N.A. Elgindy, Casein-based formulations as promis- ing controlled release drug delivery systems, J. Control. Release 153 (2011) 206–216.

[99] M.D. Rowe, D.H. Thamm, S.L. Kraft, S.G. Boyes, Polymer-modified gadolinium metal-organic framework nanoparticles used as multifunctional nanomedicines for the targeted imaging and treatment of cancer, Biomacromolecules 10 (2009) 983–993.

[100] Q. Liu, H. Zhu, J. Qin, H. Dong, J. Du, Theranostic vesicles based on bovine serum al- bumin and poly (ethylene glycol)-block-poly (L-lactic-co-glycolic acid) for mag- netic resonance imaging and anticancer drug delivery, Biomacromolecules 15 (2014) 1586–1592.

[101] D.P. Ferris, J. Lu, C. Gothard, R. Yanes, C.R. Thomas, J.C. Olsen, J.F. Stoddart, F. Tamanoi, J.I. Zink, Synthesis of biomolecule-modified mesoporous silica nanoparti- cles for targeted hydrophobic drug delivery to cancer cells, Small 7 (2011) 1816–1826.

[102] R. Sharma, B.M. Barth, E.İ. Altinoğlu, T.T. Morgan, S.S. Shanmugavelandy, J.M. Kaiser, C. McGovern, G.L. Matters, J.P. Smith, M. Kester, Bioconjugation of calcium phosphate nanoparticles for selective targeting of human breast and pancreatic cancers in vivo, ACS Nano 4 (2010) 1279.

[103] A. Chopra, Transferrin-coated gadolinium-labeled human serum albumin nanopar- ticles, (2013).

[104] A. Tomitaka, H. Arami, S. Gandhi, K.M. Krishnan, Lactoferrin conjugated iron oxide nanoparticles for targeting brain glioma cells in magnetic particle imaging, Nano- scale 7 (2015) 16890–16898.

[105] H. Xie, Y. Zhu, W. Jiang, Q. Zhou, H. Yang, N. Gu, Y. Zhang, H. Xu, H. Xu, X. Yang, Lactoferrin-conjugated superparamagnetic iron oxide nanoparticles as a specific

322 A.O. Elzoghby et al. / Journal of Controlled Release 243 (2016) 303–322

MRI contrast agent for detection of brain glioma in vivo, Biomaterials 32 (2011) 495–502.

[106] R. Yang, D. Chen, M. Li, F. Miao, P. Liu, Q. Tang, 20 (s)-ginsenoside Rg3-loaded mag- netic human serum albumin nanospheres applied to HeLa cervical cancer cells in vitro, Biomed. Mater. Eng. 24 (2014) 1991–1998.

[107] L. Mocan, C. Matea, F.A. Tabaran, O. Mosteanu, T. Pop, T. Mocan, C. Iancu, Photothermal treatment of liver cancer with albumin-conjugated gold nanoparti- cles initiates Golgi apparatus–ER dysfunction and caspase-3 apoptotic pathway ac- tivation by selective targeting of Gp60 receptor, Int. J. Nanomedicine 10 (2015) 5435.

[108] U.N. Pan, R. Khandelia, P. Sanpui, S. Das, A. Paul, A. Chattopadhyay, Protein-based multifunctional nanocarriers for imaging, photothermal therapy, and anticancer drug delivery, ACS Appl. Mater. Interfaces (2016).

[109] Y. Wang, J.-T. Chen, X.-P. Yan, Fabrication of transferrin functionalized gold nanoclusters/graphene oxide nanocomposite for turn-on near-infrared fluorescent bioimaging of cancer cells and small animals, Anal. Chem. 85 (2013) 2529–2535.

[110] Q. Cui, F. He, X. Wang, B. Xia, L. Li, Gold nanoflower@ gelatin core–shell nanoparti- cles loaded with conjugated polymer applied for cellular imaging, ACS Appl. Mater. Interfaces 5 (2012) 213–219.

[111] T.-Y. Liu, M.-Y. Wu, M.-H. Lin, F.-Y. Yang, A novel ultrasound-triggered drug vehicle with multimodal imaging functionality, Acta Biomater. 9 (2013) 5453–5463.

[112] J. Zhang, G. Hao, C. Yao, J. Yu, J. Wang, W. Yang, C. Hu, B. Zhang, Albumin-mediated biomineralization of paramagnetic NIR Ag2S QDs for tiny tumor bimodal targeted imaging in vivo, ACS Appl. Mater. Interfaces 8 (2016) 16612–16621.

[113] W. Watcharin, C. Schmithals, T. Pleli, V. Köberle, H. Korkusuz, F. Hübner, O. Waidmann, S. Zeuzem, H.-W. Korf, A. Terfort, Detection of hepatocellular carcino- ma in transgenic mice by Gd-DTPA-and rhodamine 123-conjugated human serum albumin nanoparticles in T1 magnetic resonance imaging, J. Control. Release 199 (2015) 63–71.

[114] C. Mi, J. Zhang, H. Gao, X. Wu, M. Wang, Y. Wu, Y. Di, Z. Xu, C. Mao, S. Xu, Multifunc- tional nanocomposites of superparamagnetic (Fe3O4) and NIR-responsive rare earth-doped up-conversion fluorescent (NaYF 4: Yb, Er) nanoparticles and their applications in biolabeling and fluorescent imaging of cancer cells, Nanoscale 2 (2010) 1141–1148.

[115] S. Suarasan, M. Focsan, M. Potara, O. Soritau, A. Florea, D. Maniu, S. Astilean, A doxorubicin-incorporated nanotherapeutic delivery system based on gelatin-coat- ed gold nanoparticles: formulation, drug release and multimodal imaging of cellu- lar internalization, ACS Appl. Mater. Interfaces (2016).

[116] Y. Wang, T. Yang, H. Ke, A. Zhu, Y. Wang, J. Wang, J. Shen, G. Liu, C. Chen, Y. Zhao, Smart albumin-biomineralized nanocomposites for multimodal imaging and photothermal tumor ablation, Adv. Mater. 27 (2015) 3874–3882.

[117] R. Khandelia, S. Bhandari, U.N. Pan, S.S. Ghosh, A. Chattopadhyay, Gold nanocluster embedded albumin nanoparticles for two-photon imaging of cancer cells accom- panying drug delivery, Small 11 (2015) 4075–4081.

[118] R.G. Aswathy, B. Sivakumar, D. Brahatheeswaran, T. Fukuda, Y. Yoshida, T. Maekawa, D.S. Kumar, Biocompatible fluorescent zein nanoparticles for simulta- neous bioimaging and drug delivery application, Adv. Nat. Sci.: Nanosci. Nanotechnol. 3 (2012) 025006.

[119] D. Mertz, C. Affolter-Zbaraszczuk, J. Barthès, J. Cui, F. Caruso, T.F. Baumert, J.-C. Voegel, J. Ogier, F. Meyer, Templated assembly of albumin-based nanoparticles for simultaneous gene silencing and magnetic resonance imaging, Nanoscale 6 (2014) 11676–11680.

[120] M. Zhang, Y. Cao, L. Wang, Y. Ma, X. Tu, Z. Zhang, Manganese doped iron oxide theranostic nanoparticles for combined T 1 magnetic resonance imaging and photothermal therapy, ACS Appl. Mater. Interfaces 7 (2015) 4650–4658.

[121] J.R. Kanwar, G. Mahidhara, R.K. Kanwar, Novel alginate-enclosed chitosan-calcium phosphate-loaded iron-saturated bovine lactoferrin nanocarriers for oral delivery in colon cancer therapy, Nanomedicine 7 (2012) 1521–1550.

[122] J.R. Kanwar, G. Mahidhara, K. Roy, S. Sasidharan, S. Krishnakumar, N. Prasad, R. Sehgal, R.K. Kanwar, Fe-bLf nanoformulation targets survivin to kill colon cancer stem cells and maintains absorption of iron, calcium and zinc, Nanomedicine 10 (2015) 35–55.

[123] K. Roy, R.K. Kanwar, J.R. Kanwar, LNA aptamer based multi-modal, Fe3O4-saturated lactoferrin (Fe3O4-bLf) nanocarriers for triple positive (EpCAM, CD133, CD44) colon tumor targeting and NIR, MRI and CT imaging, Biomaterials 71 (2015) 84–99.

[124] Q. Tang, Y. An, D. Liu, P. Liu, D. Zhang, Folate/NIR 797-conjugated albumin magnetic nanospheres: synthesis, characterisation, and in vitro and in vivo targeting evalu- ation, PLoS One 9 (2014), e106483.

[125] Y. Jiao, Y. Sun, X. Tang, Q. Ren, W. Yang, Tumor-targeting multifunctional rattle- type theranostic nanoparticles for MRI/NIRF bimodal imaging and delivery of hy- drophobic drugs, Small 11 (2015) 1962–1974.

[126] A. Semkina, M. Abakumov, N. Grinenko, A. Abakumov, A. Skorikov, E. Mironova, G. Davydova, A. Majouga, N. Nukolova, A. Kabanov, Core–shell–corona doxorubicin- loaded superparamagnetic Fe3O4 nanoparticles for cancer theranostics, Colloids Surf. B: Biointerfaces 136 (2015) 1073–1080.

[127] K.-C. Wei, F.-W. Lin, C.-Y. Huang, C.-C.M. Ma, J.-Y. Chen, L.-Y. Feng, H.-W. Yang, 1, 3- Bis (2-chloroethyl)-1-nitrosourea-loaded bovine serum albumin nanoparticles with dual magnetic resonance–fluorescence imaging for tracking of chemothera- peutic agents, Int. J. Nanomedicine 11 (2016) 4065.

[128] N.N. Reddy, K. Varaprasad, S. Ravindra, G.S. Reddy, K. Reddy, K.M. Reddy, K.M. Raju, Evaluation of blood compatibility and drug release studies of gelatin based mag- netic hydrogel nanocomposites, Colloids Surf. A Physicochem. Eng. Asp. 385 (2011) 20–27.

[129] Y.-L. Su, J.-H. Fang, C.-Y. Liao, C.-T. Lin, Y.-T. Li, S.-H. Hu, Targeted mesoporous iron oxide nanoparticles-encapsulated perfluorohexane and a hydrophobic drug for deep tumor penetration and therapy, Theranostics 5 (2015) 1233.

[130] B. Gaihre, M.S. Khil, D.R. Lee, H.Y. Kim, Gelatin-coated magnetic iron oxide nano- particles as carrier system: drug loading and in vitro drug release study, Int. J. Pharm. 365 (2009) 180–189.

[131] W.-M. Li, C.-W. Su, Y.-W. Chen, S.-Y. Chen, In situ DOX-calcium phosphate miner- alized CPT-amphiphilic gelatin nanoparticle for intracellular controlled sequential release of multiple drugs, Acta Biomater. 15 (2015) 191–199.

[132] N. Amreddy, R. Muralidharan, A. Babu, M. Mehta, E.V. Johnson, Y.D. Zhao, A. Munshi, R. Ramesh, Tumor-targeted and pH-controlled delivery of doxorubicin using gold nanorods for lung cancer therapy, Int. J. Nanomedicine 10 (2015) 6773.

[133] A.A. Hwang, J. Lu, F. Tamanoi, J.I. Zink, Functional Nanovalves on protein-coated nanoparticles for in vitro and in vivo controlled drug delivery, Small 11 (2015) 319–328.

[134] X. Zhen, X. Wang, C. Xie, W. Wu, X. Jiang, Cellular uptake, antitumor response and tumor penetration of cisplatin-loaded milk protein nanoparticles, Biomaterials 34 (2013) 1372–1382.

[135] C. Wong, T. Stylianopoulos, J. Cui, J. Martin, V.P. Chauhan, W. Jiang, Z. Popović, R.K. Jain, M.G. Bawendi, D. Fukumura, Multistage nanoparticle delivery system for deep penetration into tumor tissue, Proc. Natl. Acad. Sci. 108 (2011) 2426–2431.

[136] H. Yılmaz, S.H. Sanlıer, Preparation of magnetic gelatin nanoparticles and investi- gating the possible use as chemotherapeutic agent, Artif. Cells, Nanomed. Biotechnol. 41 (2013) 69–77.

[137] Z. Wu, X. Lin, X. Zou, J. Sun, Q. He, Biodegradable protein-based rockets for drug transportation and light-triggered release, ACS Appl. Mater. Interfaces 7 (2014) 250–255.

[138] J. Yu, Y. Ju, L. Zhao, X. Chu, W. Yang, Y. Tian, F. Sheng, J. Lin, F. Liu, Y. Dong, Multistimuli-regulated photochemothermal cancer therapy remotely controlled via Fe5C2 nanoparticles, ACS Nano (2015).

[139] H. Shi, Q. Cheng, S. Yuan, X. Ding, Y. Liu, Human serum albumin conjugated nano- particles for pH and redox-responsive delivery of a prodrug of cisplatin, Chem. – Eur. J. 21 (2015) 16547–16554.

[140] Z. Zou, D. He, X. He, K. Wang, X. Yang, Z. Qing, Q. Zhou, Natural gelatin capped mesoporous silica nanoparticles for intracellular acid-triggered drug delivery, Langmuir 29 (2013) 12804–12810.

[141] S. Ruan, X. Cao, X. Cun, G. Hu, Y. Zhou, Y. Zhang, L. Lu, Q. He, H. Gao, Matrix metal- loproteinase-sensitive size-shrinkable nanoparticles for deep tumor penetration and pH triggered doxorubicin release, Biomaterials 60 (2015) 100–110.

[142] J. Lademann, H. Richter, F. Knorr, A. Patzelt, M. Darvin, E. Rühl, K. Cheung, K. Lai, R. Renneberg, W. Mak, Triggered release of model drug from AuNP-doped BSA nanocarriers in hair follicles using IRA radiation, Acta Biomater. 30 (2016) 388–396.

[143] E.C. Cho, J. Xie, P.A. Wurm, Y. Xia, Understanding the role of surface charges in cel- lular adsorption versus internalization by selectively removing gold nanoparticles on the cell surface with a I2/KI etchant, Nano Lett. 9 (2009) 1080–1084.