Cellular delivery of doxorubicin mediated by disulfide reduction of a peptide-dendrimer bioconjugate

International Journal of Pharmaceutics 545 (2018) 64–73

Contents lists available at ScienceDirect

International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm

Cellular delivery of doxorubicin mediated by disulfide reduction of a peptide-dendrimer bioconjugate Kelly E. Burnsa,b, James B. Delehantya, a b

T

⁎

Center for Bio/Molecular Science and Engineering, U.S. Naval Research Laboratory, Code 6900, Washington DC 20375, United States National Research Council, Washington DC 20001, United States

A R T I C LE I N FO

A B S T R A C T

Keywords: pHLIP Peptide drug delivery Targeted drug delivery Tumor acidity Doxorubicin Endocytosis

In this study, we developed a peptide-dendrimer-drug conjugate system for the pH-triggered direct cytosolic delivery of the cancer chemotherapeutic doxorubicin (DOX) using the pH Low Insertion Peptide (pHLIP). We synthesized a pHLIP-dendrimer-DOX conjugate in which a single copy of pHLIP displayed a generation three dendrimer bearing multiple copies of DOX via disulfide linkages. Biophysical analysis showed that both the dendrimer and a single DOX conjugate inserted into membrane bilayers in a pH-dependent manner. Timeresolved confocal microscopy indicate the single DOX conjugate may undergo a faster rate of membrane translocation, due to greater nuclear localization of DOX at 24 h and 48 h post delivery. At 72 h, however, the levels of DOX nuclear accumulation for both constructs were identical. Cytotoxicity assays revealed that both constructs mediated ∼80% inhibition of cellular proliferation at 10 µM, the dendrimer complex exhibited a 17% greater cytotoxic effect at lower concentrations and greater than three-fold improvement in IC50 over free DOX. Our findings show proof of concept that the dendrimeric display of DOX on the pHLIP carrier (1) facilitates the pH-dependent and temporally-controlled release of DOX to the cytosol, (2) eliminates the endosomal sequestration of the drug cargo, and (3) augments DOX cytotoxicity relative to the free drug.

1. Introduction The targeted delivery of drugs to diseased tissues is fundamental to the development of effective therapies, especially tumor-specific anticancer systems. The primary challenges here have been the targeting of the therapeutic to desired sites combined with its efficient delivery across the plasma membrane to the cellular cytosol; many therapeutics are not sufficiently lipophilic to cross the membrane bilayer (Dinca et al., 2016; Lundberg and Langel, 2003; Sun et al., 2004; Tsutsumi and Neckers, 2007; Wang et al., 2000; Wijesinghe et al., 2011). To address this, various drug-carrying vehicles capable of overcoming the plasma membrane barrier while increasing the effective intracellular concentration of the therapeutic have been developed. These drug carriers include various nanoparticles (NPs), liposomal structures, and peptides (Din et al., 2017; Zaidi et al., 2017). While many of these systems have shown the ability to (1) facilitate large drug loading capacity, (2) protect against drug degradation, (3) enable controlled drug release, and (4) improve drug pharmacokinetics, the specific cellular targeting of these drug carriers and their subsequent sequestration within the endolysosomal pathway remain critical roadblocks. Peptide-mediated drug delivery represents one promising avenue

⁎

Corresponding author. E-mail address: [email protected] (J.B. Delehanty).

https://doi.org/10.1016/j.ijpharm.2018.04.027 Received 22 January 2018; Received in revised form 8 April 2018; Accepted 13 April 2018 Available online 27 April 2018 0378-5173/ Published by Elsevier B.V.

for the targeted cellular delivery of therapeutic drugs. Compared to NPbased drug vehicles, which are typically on the order of 50–100 nm, peptides are considerably smaller (∼3–5 nm) and can circulate and penetrate tissue efficiently (Acar et al., 2017). Most of the cell delivery peptides described in the literature fall under the general classification of “cell penetrating peptide” (CPP). The canonical CPP, derived from the HIV-1 transactivator of transcription (TAT), is a positively-charged (arginine- and lysine-rich) peptide that mediates the cellular uptake of attached cargos through initial electrostatic interactions with the negatively-charged cell surface followed by uptake into the endocytic pathway (Richard et al., 2005; Vives, 2003). Indeed, for the majority of CPPs described to date, the general consensus is that endocytosis is the primary means of internalization of the CPP, which necessitates strategies for the subsequent endosomal escape of the peptide-appended drug cargo (Kauffman et al., 2015). This can lead to poor bioavailability and overall decreased effective concentrations of the delivered drug (Deshayes et al., 2004; Dinca et al., 2016; Duchardt et al., 2007; Guidotti et al., 2017; Henriques et al., 2005; Richard et al., 2003). Thus, nonspecificity and endosomal sequestration remain two critical challenges in the implementation of CPPs as vehicles for therapeutic drug delivery.

International Journal of Pharmaceutics 545 (2018) 64–73

K.E. Burns, J.B. Delehanty

Fig. 1. pHLIP exists in three different states depending on its environment. At physiological pH in aqueous environments, pHLIP is an unstructured, soluble peptide (State I). In the presence of a lipid bilayer, pHLIP remains unstructured but associates with the membrane (State II). In an acidic environment pHLIP adopts an alpha helical structure and inserts across the lipid bilayer with the C-terminus in the cytosol (State III).

conjugates at neutral pH (non-triggered state). Further, the conjugates did not mediate any improvement in the IC50 of DOX compared to the free drug. To address these shortcomings, here we present a pHLIP-dendrimerdrug conjugate system for the polyvalent display and cytosolic/nuclear delivery of DOX. We chose DOX as the model drug cargo as it is a prominent anticancer drug that exerts it cytotoxic effect by intercalating between DNA base pairs and inhibiting topoisomerase II, resulting in DNA damage and the induction of apoptosis (Ai et al., 2011; Song et al., 2016; Soudy et al., 2013). The clinical application of DOX, however, is hampered by its off-target cardiotoxicity, mediated by the generation of reactive oxygen species (Danz et al., 2009). Thus, there is significant interest in improved strategies for the specific cellular delivery of DOX and the controlled augmentation of its toxicity (Maksimenko et al., 2014; Tacar et al., 2013). To this end, we synthesized a conjugate system comprised of a single copy of pHLIP covalently attached to a generation three PAMAM dendrimer (∼3.5 nm diameter) that displayed four copies of DOX per dendrimer. The DOX moieties were attached to the dendrimer surface via disulfide linkages such that insertion of the dendrimer into the cytosol resulted in release of multiple DOX molecules per pHLIP. As a control, we synthesized pHLIPDOX conjugates that displayed a single copy of DOX attached through either a disulfide linkage or a non-cleavable thioether bond. Biophysical characterization showed that all three conjugates exhibited the characteristic membrane insertion and translocation behavior of pHLIP at low pH. While both disulfide-containing conjugates efficiently delivered DOX to the nucleus of HeLa cells, we noted distinct differences between the two conjugate systems in their rate of nuclear accumulation of DOX and eventual cytotoxicity. The pHLIP construct displaying a single disulfide-linked DOX exhibited slightly faster DOX nuclear localization kinetics compared to the multiple DOX-displaying dendrimer form. Based on time-resolved confocal fluorescence microscopy, we attributed this to the slower membrane translocation and cytosolic release of DOX in the pHLIP-PAMAM-DOX conjugate system. Still, both constructs facilitated significant cytotoxicity (∼80% inhibition of cellular proliferation) at high peptide concentration (10 µM). At lower peptide concentrations (< 0.63 μM), however, the polyvalent pHLIP-PAMAM-DOX conjugate exhibited as much as a ∼17% greater anti-proliferative effect compared to the single DOX displaying conjugate and a greater than three-fold improvement in IC50 over free DOX. We attribute this to the slower membrane translocation of the dendrimer coupled with the sustained intracellular release of multiple DOX from the dendrimer surface. Cumulatively, our results demonstrate the utility of the pHLIP-dendrimer system for the facile and robust multivalent display and pH-dependent cellular delivery of drug cargos that is independent of the endocytic pathway. Further, our findings point to the exciting possibility of the use of this conjugate system for the simultaneous cellular delivery of multiple disparate drug and imaging agent cargos.

The pH (Low) Insertion Peptide (pHLIP), a peptide that exhibits specific targeting/insertion into cancer cells mediated by the acidic microenvironment of many tumor tissues, is a promising alternative peptidyl motif for the delivery of drugs directly to the cellular cytosol. The low pH environment associated with cancer cells arises from the “Warburg effect”, as cancer cells rapidly undergo aerobic glycolysis, resulting in the acidification of the immediate extracellular environment (Gatenby and Gillies, 2004; Gogvadze et al., 2008; Kim and Dang, 2006; Ristow, 2006; Semenza et al., 2001; Seyfried and Mukherjee, 2005; Seyfried and Shelton, 2010; Warburg, 1956). The pHLIP peptide exhibits three distinct states depending on its local environment (Fig. 1). At physiological pH in aqueous (extracellular) environments, pHLIP is an unstructured, soluble peptide (State I). In the presence of a lipid bilayer (e.g., the cell’s plasma membrane), pHLIP remains unstructured but becomes preferentially associated with the membrane (State II) due to the thermodynamically favorable burying of hydrophobic residues into the lipid bilayer (Reshetnyak et al., 2008). Upon lowering of extracellular pH, the membrane-associated pHLIP adopts an alpha helical structure and inserts across the lipid bilayer (State III) with the peptide’s C-terminus presented to the cytosol (Hunt et al., 1997; Reshetnyak et al., 2006). pHLIP has been shown to translocate a myriad of cargos across the plasma membrane, including fluorescent molecules, peptide nucleic acids, and antimicrobial peptides (An et al., 2010; Burns et al., 2016; Burns and Thevenin, 2015; Cheng et al., 2015; Moshnikova et al., 2013; Onyango et al., 2015; Reshetnyak et al., 2006; Wijesinghe et al., 2011), pHLIP has also been shown to translocate clinically relevant chemotherapeutics, including doxorubicin (DOX) and a variety of NPs in a concentration- and pH-dependent manner (Antosh et al., 2015; Burns et al., 2017; Burns et al., 2015; Davies et al., 2012; Han et al., 2013; Kyrychenko, 2015; Song et al., 2016; Tian et al., 2017; Wei et al., 2017; Wijesinghe et al., 2013; Yao et al., 2013a; Yao et al., 2013b; Yu et al., 2016; Zeiderman et al., 2016; Zhao et al., 2013). These attributes of pHLIP served as motivation for us to develop a pHLIP-based NP bioconjugate system for multivalent drug display coupled with pH-dependent control over drug release to the cytosol. To date, studies demonstrating the use of pHLIP-NP conjugate systems for the multivalent display of drugs combined with improved drug efficacy remain scant. In a study aimed solely at imaging, Janic et al. reported the use of pHLIP for the translocation a generation five polyamidoamine (PAMAM) dendrimer decorated with ∼44 Gd-DOTA4AmP contrast agent chelates and showed pH-dependent delivery to breast cancer and glioblastoma cells using magnetic resonance imaging (Janic et al., 2016). More recently, Zhao et al. reported the synthesis of DOX-loaded mesoporous silica NPs (MSNs) conjugated to multiple copies of pHLIP via disulfide linkages (Zhao et al., 2013). In that study, the ensemble pHLIP-MSN conjugates were rather large (140 nm diameter) and were bulk loaded with DOX in the MSN core. Thus, they displayed minimal temporal control of DOX release as evidenced by the significant inhibition of cellular proliferation in cells exposed to the 65

International Journal of Pharmaceutics 545 (2018) 64–73

K.E. Burns, J.B. Delehanty

2. Materials & methods

2.3. Preparation of POPC liposomes for characterization of pHLIP membrane insertion

2.1. Materials POPC lipids (10 mg) in 0.5 mL chloroform (initial concentration of 20 mg/mL) were desolvated in a rotary evaporator for 2 h at 55 °C. The dried lipid film was rehydrated in 1 mL of 5 mM sodium phosphate (pH 7.0) to give a suspension of multilamellar vesicles (MLVs). The MLVs were subjected to repeated extrusion through polycarbonate membranes of iteratively smaller pore size (down to 100 nm diameter pore size) using a mini extruder (Avanti Polar Lipids) to produce large unilamellar vesicles (LUVs).

Dulbecco’s phosphate buffered saline (DPBS), Dulbecco’s Modified Eagle Medium (DMEM), fetal bovine serum (FBS), antibiotic-antimycotic, bovine fibronectin, Live Cell Imaging Solution, Alexa Fluor 488-transferrin conjugate and trypan blue stain (0.4%) were all purchased from Life Technologies (Grand Island, NY). 1-palmitoyl-2oleoyl-sn-glycero-3-phosphocholine (POPC) was purchased from Avanti Polar Lipids (Alabaster, AL). CellTiter 96® Aqueous One Solution Cell Proliferation Assay (MTS) was obtained from Promega (Madison, WI).

2.4. Biophysical characterization of pHLIP membrane insertion pHLIP-DOX conjugates were resuspended with 5 mM sodium phosphate (pH 7.0) to 20 µM, incubated for 1 h at room temperature, added to POPC liposomes at a 1:300 (7 µM peptide: 2.1 mM lipid) and incubated for 1 h. The pH was adjusted using 0.25 M HCl and the peptide/ liposome mixtures were incubated for ∼30 min prior to spectroscopic measurements. Tryptophan fluorescence spectroscopy measurements were carried out on a Shimadzu RF6000 Spectrofluorophotometer (Shimadzu, Japan). Peptide samples (7 µM) were excited at 295 nm and emission spectra were collected from 310 to 450 nm using a 5 nm spectral bandwidth. Far-UV circular dichroism (CD) spectra were collected on a Jasco J-815CD Spectrometer (Jasco, USA). All sample measurements were carried out at peptide concentrations of 7 µM in a 0.1 cm quartz cuvette. CD intensities are expressed in mean residue molar ellipticity (degrees cm2 dmol−1 [θ] = θobs/(10 × lcn)) where θobs is the observed ellipticity in millidegrees, l is the optical path length in centimeters, c is the final molar concentration of the peptides, and n is the number of amino acid residues. Wavelength scans were acquired from 260 to 190 nm at 0.5 nm intervals with a 50 nm/min acquisition rate and at least three scans were averaged for each sample. The background POPC spectrum was subtracted from all liposome containing samples.

2.2. Peptide and peptide conjugate synthesis All pHLIP and pHLIP-DOX conjugates were custom-synthesized by Bio-Synthesis Inc. (Lewisville, TX, USA). pHLIP (H2N-GGEQNPIYWAR YADWLFTTPLLLLDLALLVDADEGTCG-CONH2) was prepared by solidphase peptide synthesis using standard fluorenylmethyloxycarbonyl chemistry and purified via reverse phase high performance liquid chromatography (RP-HPLC) to > 84% purity and the identity confirmed by matrix assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) (calculated (M + H+ = 4210.76); experimental (M + H+ = 4210.24)). For each peptide-DOX construct the thiol of the cysteine residue at the C-terminus of pHLIP was utilized for conjugation to DOX or to dendrimer-DOX. pHLIP-M-DOX (DOX conjugated to pHLIP via a non-labile thioether bond) was prepared by modification of DOX with a sulfosuccinimidyl 4-(N-maleimidomethyl) cyclohexane-1-carboxylate (SMCC) linker to allow for conjugation to the cysteine of pHLIP. pHLIP-S-S-DOX (DOX conjugated to pHLIP via reducible disulfide bond) was prepared by modification of DOX with a succinimidyl 3-(2-pyridyldithio)propionate (SPDP) crosslinker followed by reaction with pHLIP. pHLIP-PAMAM-DOX (pHLIP conjugated via a non-labile linkage to a generation 3 polyamidoamine (G3 PAMAM) dendrimer displaying ∼ 4 DOX via disulfide linkages) was prepared by first reacting the primary amines on DOX and the G3 PAMAM with SPDP crosslinker. These activated species were then reacted with each other through thiol exchange to yield G3 PAMAM displaying ∼4 DOX/ dendrimer linked by disulfide bonds. The PAMAM-DOX was then modified with an azidobutyric acid NHS ester crosslinker to convert the remaining primary amines on the PAMAM to azide functions. The converted PAMAM-DOX was then reacted with pHLIP that had been modified at its C-terminal cysteine thiol with a dibenzocyclootyne (DBCO)-maleimide. Reaction of the azide-PAMAM-DOX with the cyclooctyne pHLIP resulted in the conjugation of pHLIP to the G3 PAMAM via a non-labile triazole linkage. The purities of the single DOX-containing pHLIP conjugates were determined by RP-HPLC to be > 90%. MALDI-TOF MS confirmed the MW of the pHLIP conjugates to be as follows: pHLIP-M-DOX, (calculated (M + H+ = 4973.52); experimental (M + H+ = 4973.93)) and pHLIP-S-S-DOX (calculated (M + H+ = 4840.40); experimental (M + H+ = 4842.11)). The pHLIP-PAMAM-DOX construct was purified by dialysis and the presence of the full length peptide was confirmed using amino acid analysis (experimental average 15795 Da; calculated using equation: MW = MWPAMAM-DOX + MWpeptide-DBCO + 28(MWazide), where 28 represents the remaining primary amines on the PAMAM after DOX conjugation). The conjugation ratio of DOX per PAMAM was determined based on absorption of DOX at 481 nm and extinction coefficient of 1.041 × 104 M−1 cm−1 to be ∼4 DOX per PAMAM (Tian et al., 2007). Based on amino acid analysis it was estimated that on average ∼0.6 pHLIP peptides were conjugated per PAMAM dendrimer. The peptide concentrations of all pHLIP-DOX constructs were quantified by UV/vis absorbance spectroscopy using ε280 = 13,940 M−1 cm−1 based on the tryptophan residues in pHLIP and confirmed using the Micro BCA Protein Assay Kit (ThermoFisher Scientific).

2.5. Cell culture HeLa cells (human cervical adenocarcinoma, ATCC® CCL-2™) were cultured in DMEM supplemented with 10% (v/v) heat-inactivated fetal bovine serum and 1% (v/v) antibiotic/antimycotic in a humidified atmosphere of 5% CO2 at 37 °C. Cells were routinely passaged every two to three days and were typically used between passage numbers 6 and 16. 2.6. Cellular delivery and imaging 35 mm MatTek dishes with 14 mm glass bottom inserts (#1.0 cover glass, MatTek Corp., MA, USA) were coated for 2 h with bovine fibronectin (30 µg/mL) prior to seeding HeLa cells at a density of ∼100,000 cells/dish. After overnight culture, complete growth media was removed and the cells were washed with DMEM. pHLIP-DOX conjugates (10 µM pHLIP concentration in all cases) were resuspended in DMEM (pH 7.4), sonicated in a bath sonicator (Branson Ultrasonics) and vortexed until solubilized. This solution was added to cells and, after a brief equilibration period of 5 min, the pH of samples was adjusted to pH 5.0 by the addition of 25 µL of pH 3.0 DMEM. pH 5.0 was selected to allow for efficient pHLIP insertion as it is well below the pKa of 6.0 (the pH at which 50% of pHLIP is in the membrane-translocated state (State III)) as demonstrated in previous reports (Burns et al., 2017; Burns et al., 2016; Burns et al., 2015; Burns and Thevenin, 2015). The peptide solution (total volume 100 µL) was incubated on the cell monolayers for 2 h. After this incubation period, cells were washed and cultured in 3 mL complete DMEM for 24, 48 h, or 72 h. For endocytosis studies, cells were cultured for 24 h after deliver. Following this, endosomes were labeled with AlexaFluor488-transferrin for 45 min prior 66

International Journal of Pharmaceutics 545 (2018) 64–73

K.E. Burns, J.B. Delehanty

Fig. 2. Chemical structures of pHLIP-DOX conjugates and membrane insertion mechanism. (A) The sequence of pHLIP used in this study. The C-terminal Cys (in red) is the attachment point for DOX in each pHLIP-DOX conjugate. The chemical structures of each construct: (B) pHLIP-M-DOX, DOX conjugated via non-labile thioether bond; (C) pHLIP-S-S-DOX, DOX conjugated via reducible disulfide bond; and (D) pHLIP-PAMAM-DOX, DOX conjugated to PAMAM G3 dendrimer via disulfide bonds and dendrimer conjugated to pHLIP via non-labile triazole. The membrane schematic shows pHLIP (in green) inserted into the cellular membrane at low pH presenting the Cterminal DOX to the cellular cytosol. pHLIP-PAMAMDOX has ∼ 4 DOX per pHLIP carrier. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

to imaging. The cells were then washed with live cell imaging solution and imaged by DIC and confocal laser scanning microscopy using a Nikon A1RSi equipped with a 488 nm diode laser with fluorescence detection set to 570–620 nm for DOX emission. All images were collected using a Plan Apo 60× objective. Laser power, PMT gain, and threshold were held constant across different samples throughout all imaging.

percentage of trypan blue-positive cells (compromised membranes) was determined using a Countess™ automated cell counter (Thermo Scientific).

2.7. Cellular proliferation assay

In the work reported herein, our strategy was to use the unique transmembrane-spanning peptide pHLIP to target and deliver the wellcharacterized chemotherapeutic, DOX, to affect (1) the translocation of DOX across the plasma membrane in a low pH-dependent manner and (2) the subsequent release of free DOX via reduction of disulfide linkages by the strong reducing environment of the cellular cytosol. This approach has previously been shown to be effective when pHLIP was conjugated to the cell permeable monomethyl auristatin E (MMAE) and the cell impermeable monomethyl auristatin F (MMAF) via disulfide bonds. In those studies, the pHLIP-auristatin conjugates exhibited potent cytotoxic effects (up to 90% inhibition of cellular proliferation), effective targeting of triple-negative breast tumor xenografts in mice, and significant therapeutic efficacy without off-target toxicities (Burns et al., 2017; Burns et al., 2015). Building upon these previous works, we chose DOX as the model drug cargo in the current study for several reasons. First, there is wide interest in modulating/controlling the toxicity of this potent anticancer drug due to its significant off-target toxicity and its susceptibility to be efficiently effluxed from cells via P-glycoprotein multidrug-resistance pumps (Maksimenko et al., 2014; Shen et al., 2008; Tacar et al., 2013; Tsou et al., 2015). Second, DOX is amphiphilic and stable in the aqueous cytosolic environment. Lastly, DOX is inherently fluorescent which avails the tracking of its location throughout the uptake and cellular distribution process. Our working hypothesis was that pHLIP could be used to modulate, and potentially augment, the cytotoxicity of DOX depending on how the pHLIP drug vector was used to present DOX to cells. With this in mind, we designed a number of pHLIP-DOX variants in which the DOX was attached to the C-terminus of pHLIP through disulfide linkages that would be amenable to reduction and release of DOX upon peptide insertion across the plasma membrane into the

3. Results & discussion 3.1. Rationale and synthesis of pHLIP-DOX conjugates

96 well tissue culture-treated microplates were coated for 2 h with bovine fibronectin (30 µg/mL) prior to seeding HeLa cells at a density of 3000 cells/well and incubation overnight. pHLIP-DOX conjugate aliquots were resuspended in pH 7.4 DMEM without FBS, sonicated and vortexed until complete solubilization was obtained. This solution was then added to each well and incubated to allow for peptide equilibration with the cellular plasma membranes. After 5 min, pH 3.0 DMEM was added to each well to obtain a final pH of 5.0 at the desired peptide concentrations and cells were incubated for 2 h at 37 °C. After incubation, the solutions were removed, cells were washed and cultured in 100 µL complete DMEM for 72 h. After this proliferation period, cell viability was determined using the colorimetric MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)2H-tetrazolium) assay. Tetrazolium substrate (20 µL) was added to each well and incubated for 3 h at 37 °C. The formazan product was read at 590 nm and 620 nm (for subtraction of background absorbance) using a Tecan Infinite M1000 (Tecan, USA) microtiter plate reader. Cell viability was calculated and plotted as percent of control cells treated at physiological pH. 2.8. Cellular membrane integrity assay 96 well plates were coated for 2 h with fibronectin (30 µg/mL) prior to seeding HeLa cells at a density of 7000 cells/well and incubated until ∼90% confluent. pHLIP-DOX conjugates were introduced to the cells and allowed to incubate for 5 min prior to pH adjustment to pH 5.0 at the desired peptide concentration of 10 µM. After 2 h, the cells were trypsinized, incubated with 0.2% trypan blue solution, and the 67

International Journal of Pharmaceutics 545 (2018) 64–73

K.E. Burns, J.B. Delehanty

the peptide with 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine large unilamellar vesicles (POPC LUVs) at neutral pH (7.4); State III in the presence of POPC LUVS at pH 5.0 to drive pHLIP membrane insertion. Fig. 3A–C shows the Trp fluorescence spectra of all three states of the pHLIP-M-DOX, pHLIP-S-S-DOX, and pHLIP-PAMAM-DOX conjugates, respectively. In aqueous buffer at neutral pH, each conjugate exhibited a modest peak centered at ∼330–332 nm, indicative of the adoption of a random State I conformation. When incubated with POPC LUVs at neutral pH, pHLIP-M-DOX exhibited Trp emission centered at 330 nm (Fig. 3A). Upon lowering the pH to 5.0, the emission spectra was modestly blueshifted to ∼327 nm and was associated with an increase in intensity, consistent with the Trp residues being in a hydrophobic environment. This observation agrees well with reports of other pHLIP peptide conjugates, where a blue shift of 3 nm was observed along with an increase in intensity due to the transition from State II to State III (Burns and Thevenin, 2015). Compared to pHLIP-M-DOX, the pHLIP-S-S-DOX and pHLIP-PAMAM-DOX conjugates both exhibited greater blueshifts upon transition from State II to State III, with spectral shifts from 338 to 325 nm and 325 to 314 nm, respectively, along with an increase in intensity. These data provided strong evidence that each of the pHLIP constructs experienced a change in environmental polarity upon transition from state II to state III, signifying the interfacing of Trp residues with the hydrophobic environment upon peptide insertion into the lipid bilayer of the POPC LUVs. In conjunction with Trp fluorescence analysis, CD was employed to determine the secondary structure of pHLIP in each of its states (Fig. 3D–F). This technique has been used previously to assess the degree of α-helix formation in pHLIP upon adoption of the State III conformation. The CD spectra for pHLIP-M-DOX shown in Fig. 3D is typical for the transition of pHLIP from a random coil in States I and II to an αhelix in state III. The spectra for State III shows the distinguishable minima centered at ∼208 and ∼227 nm, indicative of formation of αhelix. The CD spectra of pHLIP-S-S-DOX, shown in Fig. 3E also indicates unstructured peptide conformation at States I and II and a transition to an α-helix in the presence of lipid vesicles at low pH, albeit with broader minima between 208 and 230 nm. Similar α-helix CD spectra has been previously observed with other pHLIP peptide and fluorophore conjugates (Burns and Thevenin, 2015; Karabadzhak et al., 2014). The CD spectra of pHLIP-PAMAM-DOX (Fig. 3F) in state III is indicative of α-helix formation as well, due to the minima at 208 and 222 nm. The spectra corresponding to States I and II indicate random coil, although the degree of ellipticity at state I is rather low suggesting weak peptide signal. This is notable considering that in these analyses, the peptide concentration was held constant across all three conjugates (as confirmed by BCA assay). Similar quenched CD spectra for peptidedendrimer and peptide-polymer conjugates have been observed elsewhere (Nilsson et al., 2003; Polcyn et al., 2013). Cumulatively, our data provides strong evidence that each of the pHLIP-DOX conjugates displays the characteristic behavior of pHLIP insertion into lipid membranes at low pH.

cytosol. One critical facet of this hypothesis was the concept that, through the polyvalent display of multiple DOX moieties per pHLIP carrier, the efficacy of DOX could be enhanced beyond that achievable using a pHLIP construct bearing only a single DOX per peptide. Fig. 2 shows schematically the three pHLIP-DOX conjugates used in this study. The goal of each construct is to control the manner in which DOX is delivered to the cytosol by the pHLIP carrier. In each conjugate, the DOX moiety is linked to pHLIP through the Cys residue at the Cterminus (sequence shown in Fig. 2A). The conjugation chemistries used to generate each pHLIP-DOX are shown in the Supporting Information (SI). pHLIP-M-DOX bears a single DOX per peptide where the DOX is attached via a non-labile thioether linkage using a succinmidyl-4-(N-maleimidomethyl)cyclohexane-1-carboxylate) (SMCC) crosslinker (Fig. 2B, Fig. S1). The pHLIP-M-DOX was developed to serve as a control, where the non-labile pHLIP-DOX linkage is expected to facilitate negligible intracellular DOX release. For cytosolic release of DOX from pHLIP, we designed two variations of pHLIP-displayed DOX. In the first iteration (pHLIP-S-S-DOX), a single DOX moiety was appended per peptide and was attached through a disulfide bridge using a succinimidyl 3-(2-pyridyldithio) propionate (SPDP) crosslinker (Fig. 2C, Fig. S2). Alternatively, to increase the effective payload of the delivered DOX per pHLIP, we synthesized a pHLIP-dendrimer-DOX construct that appended onto a single pHLIP a generation three polyamidoamine (PAMAM) dendrimer that displayed multiple copies of DOX that were attached to the dendrimer scaffold by disulfide linkages (Fig. 2D). As shown in the synthetic scheme in Fig. S3, the PAMAM dendrimer was first decorated with multiple copies of DOX by functionalizing both the PAMAM and DOX with the same SPDP crosslinker, followed by their reaction with each other. The remaining free amines were capped with an azidobutyric acid NHS ester crosslinker which subsequently facilitated PAMAM strain-promoted click chemistry with pHLIP that was modified with a dibenzocyclootyne-maleimide at the Cys of the C-terminus. This capping step also served to neutralize the positively-charged free amines to facilitate membrane translocation of the PAMAM. The absorption of DOX at 481 nm was used to determine the number of DOX per PAMAM dendrimer and was found to be on average ∼4 and amino acid analysis determined that the average number of pHLIP peptides per PAMAM was ∼0.6. Importantly, the dithiol conjugates in the studies here were specifically designed so that upon intracellular reduction the linker-modified DOX molecules released to the cytosol were identical. 3.2. Biophysical characterization of pHLIP-DOX conjugate membrane insertion We first investigated whether the tethering of DOX or PAMAM-DOX to pHLIP had any deleterious effects on the peptide’s characteristic membrane insertion behavior. Previous studies provided evidence that the conjugation of small molecule drugs (as well as larger cargos) to the C-terminus of pHLIP does not significantly alter the pH-dependent conformational change or membrane insertion properties of the peptide (Burns et al., 2017; Burns et al., 2015; Burns and Thevenin, 2015; Janic et al., 2016; Moshnikova et al., 2013; Onyango et al., 2015). Still, we felt it was imperative to directly assess the activity of our pHLIP conjugates using biophysical techniques. Thus, we employed tryptophan fluorescence and far-UV circular dichroism (CD) spectroscopy to characterize each of the states of pHLIP for each DOX conjugate studied herein (Fig. 3). Tryptophan (Trp) fluorescence is widely used in the study of peptides and proteins as the intensity and wavelength maximum (ƛmax) report on the polarity of the local environment experienced by Trp residues in the sequence and correlate with the degree of Trp solvent exposure (Vivian and Callis, 2001). Far-UV CD is used to determine the secondary structure content of the pHLIP-DOX conjugates in each of the states. For biophysical characterization experiments, analysis of the conjugates was performed as follows: State I was assessed in aqueous sodium phosphate buffer; State II by incubation of

3.3. Cellular delivery and imaging of pHLIP-DOX conjugates Having characterized the membrane association/insertion activity of the pHLIP-DOX conjugates by biophysical analysis, we next sought to determine the efficacy of delivery of DOX to the cytosol and, ultimately, to the nucleus. To achieve this, HeLa cells were incubated with the conjugates (final pHLIP concentration, 10 µM) for an initial equilibration period of ∼5 min (pH 7.4), after which the pH was adjusted to 5.0. The cells were then incubated with the conjugates for 2 h, washed with complete growth medium and then cultured for 24, 48, or 72 h prior to imaging by confocal microscopy. As shown in the fluorescence micrographs in Fig. 4 (and Figs. S4 and S5), no deleterious morphological changes were noted for non-treated control cells exposed to the lowered pH. As anticipated, we observed no cytosolic or nuclear accumulation 68

International Journal of Pharmaceutics 545 (2018) 64–73

K.E. Burns, J.B. Delehanty

Fig. 3. Biophysical characterization of pHLIP membrane insertion. (A-C) Tryptophan fluorescence of the three states of (A) pHLIP-M-DOX, (B) pHLIP-S-S-DOX and (C) pHLIP-PAMAM-DOX. (D-F) Circular dichroism of the three states of (D) pHLIP-M-DOX, (E) pHLIP-S-S-DOX and (F) pHLIP-PAMAM-DOX. The black lines represent pHLIP conjugates in an aqueous environment (State I), the blue lines represent pHLIP conjugates in the presence of lipid membranes at physiological pH (State II) and green lines represent peptides in the presence of POPC LUVs at pH 5.0 (State III). All sample measurements were conducted at peptide concentration of 7 μM and a peptide:lipid ratio of 1:300. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

showed none of these characteristics (Fig. 4B). Our results with the pHLIP-S-S-DOX used here showed a greater extent of nuclear localization (100% cells DOX-positive; 70–80 cells analyzed over two separate experiments) compared to a similar construct used by Song et al. who reported DOX distribution primarily in the cytosol (Song et al., 2016). A number of subtleties between the two studies likely explain these observed differences. We used a slightly higher pHLIP concentration

of DOX in HeLa cells treated with any of the pHLIP-DOX constructs at pH 7.4 across all culture timepoints (Fig. 4A). Cells treated with pHLIPM-DOX, even at the lower pH, still showed no cellular uptake of DOX, consistent with the non-labile nature of the thioether linkage. In contrast, cells treated with pHLIP-S-S-DOX at pH 5.0 showed intense DOX fluorescence staining in the nucleus, condensed nuclear morphology, and decreased cell density after 72 h while cells treated at pH 7.4

Fig. 4. Cellular uptake and distribution of DOX. Cellular internalization and nuclear localization of DOX was assessed using laser scanning confocal microscopy. HeLa cells were incubated with medium alone (non-treated), pHLIP-M-DOX, pHLIP-S-S-DOX and pHLIP-PAMAM-DOX at 10 μM for 2 h at either (A) pH 7.4 or (B) pH 5.0. After washing, the cells were cultured in complete growth medium for 72 h prior to imaging. The arrows indicate cells exhibiting condensed nuclei which is indicative of DOX-induced toxicity. Scale bar is 20 µm. 69

International Journal of Pharmaceutics 545 (2018) 64–73

K.E. Burns, J.B. Delehanty

Fig. 5. Quantification of time-resolved fluorescence intensity of DOX. Shown are the average DOX fluorescence intensities per cell measured at 24 h, 48 h and 72 h post-incubation with pHLIP-S-S-DOX and pHLIP-PAMAM-DOX at low pH. In all cases, 100% of cells were positive for DOX uptake. Results are expressed as mean ± SEM (n = ∼70 cells).

analyzed over two experiments). We did, however, note a significant difference in the rates of DOX nuclear accumulation between the two conjugate systems. Fluorescence imaging at 24 h and 48 h post-delivery (Figs. S4 and S5, respectively) revealed a significantly lower intensity of DOX fluorescence per cell for the pHLIP-PAMAM-DOX conjugate. While the overall percentage of DOX-positive cells for the pHLIP-S-S-DOX and dendrimer constructs were identical to each other at these time points (100% for both systems), the average fluorescence per cell for the pHLIP-PAMAM-DOX construct was 61% and 69% lower than that measured for the pHLIP-S-S-DOX system at 24 h and 48 h, respectively (Fig. 5). These results clearly demonstrate that while the overall efficiency of cellular interaction and labeling was comparable between the two conjugate systems, the kinetics of cellular uptake and nuclear localization were slower for the PAMAM-DOX system. We attribute this to possible differences in the rate of PAMAM translocation across the plasma membrane and the efficiency of DOX release, such as slower disulfide bond reduction. Slow disulfide reduction in pHLIP systems has been reported elsewhere. Karabadzhak et al. used fluorescence quenching to monitor the time-resolved insertion of pHLIP and intracellular release of dye reporters via disulfide reduction (Karabadzhak et al., 2014). The authors observed a slow, yet steady increase in dye reporter release (∼16-fold) over a 2 day period. This continual increase in fluorescence intensity over time is in keeping with our observations in the current study, suggesting that it is the context of the disulfide bonds that plays a critical role in determining the rate of their reduction. It was also apparent that both pHLIP-DOX conjugate systems showed negligible colocalization of DOX with endosomal compartments as evidenced by tracking with the endosomal marker, transferrin (Fig. S6). This provided confirmation of the direct delivery of DOX to the cytosol/nucleus in a manner that was not dependent on uptake through the endocytic pathway. This is a critical observation as it demonstrates that the presence of the PAMAM dendrimer does not induce endocytosis of the pHLIP-PAMAM-DOX conjugate as has been observed extensively in other peptide-based dendrimer drug delivery systems where it can significantly impact drug efficacy (Albertazzi et al., 2010; Eggimann et al., 2014; Rewatkar et al., 2016).

Fig. 6. Cytotoxicity of pHLIP-DOX constructs. The effect on HeLa cell proliferation after incubation with pHLIP-DOX conjugates was assessed by MTS assay. HeLa cells were treated with (A) pHLIP-M-DOX, (B) pHLIP-S-S-DOX and (C) pHLIP-PAMAM-DOX at the concentrations shown. Concentrations correspond to that of pHLIP peptide. In (C), the concentration of DOX in the context of pHLIP-PAMAM-DOX is four times that shown (DOX valence per dendrimer = 4). Cells were incubated with the constructs for 2 h at either pH 7.4 (blue bars) or pH 5.0 (green bars) followed by a culture period of 72 h in complete medium prior to MTS assay. Results are normalized to non-treated control cells and are expressed as mean ± SEM (n = 9–12 wells and are representative of two separate experiments). A two-tailed Student’s t test analysis at the 95% confidence level was performed to compare the toxicities of pHLIPS-S-DOX and pHLIP-PAMAM-DOX (*, p = 0.03; **, p = 0.009; ***, p = 0.004). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.4. Analysis of pHLIP-DOX conjugate cytotoxicity To ascertain the anti-proliferative effect of DOX delivered to HeLa cells by the pHLIP-DOX conjugates we performed MTS cellular proliferation assays after incubation with the various constructs. HeLa cells seeded to the wells of a microtiter plate were treated with pHLIP-DOX conjugates in identical fashion as described for the imaging studies. Briefly, conjugates were added to the cells at pH 7.4, the pH was then adjusted to 5.0, followed by incubation for 2 h and a 72 h culture period to allow for proliferation. At the end of this period, MTS tetrazolium reagent was added to the wells. As shown in Fig. 6A, pHLIP-M-DOX had a negligible effect on HeLa cell proliferation when incubated on cells at either pH at pHLIP concentrations as high as 10 µM. This is consistent with our imaging data that showed no release of DOX into the cell interior at either pH. HeLa cells treated with pHLIP-S-S-DOX at low pH showed significant toxicity

(10 µM) during pHLIP-DOX incubation for the imaging studies and we allowed for as long as a 72 h culture period after initial incubation with the conjugates as opposed to imaging the treated cells after a 3 h incubation and fixation. Also, Song et al. performed their pHLIP-DOX deliveries at the slightly elevated pH of 6.0 compared to our use of pH 5.0. Similar to the pHLIP-S-S-DOX construct, the pHLIP-PAMAM-DOX conjugate exhibited comparable labeling efficiency at all three culture time points (24, 48 and 72 h; 100% cells DOX-positive; ∼70 cells 70

International Journal of Pharmaceutics 545 (2018) 64–73

K.E. Burns, J.B. Delehanty

(∼50% viability at the lowest concentration tested (0.16 µM pHLIP)) which, in this construct, corresponds to the DOX concentration. Cellular viability tracked inversely with increasing pHLIP conjugate concentration, peaking at ∼18% viability at 10 µM. The cytotoxicity data for the pHLIP-PAMAM-DOX conjugate mirrored quite closely that of the pHLIP-S-S-DOX conjugate, particularly at the higher concentrations (> 1.25 µM). However, at lower pHLIP concentrations (0.16 µM–0.63 µM), the PAMAM conjugate exhibited higher cytotoxicity than the single DOX conjugate (∼up to 17% higher toxicity). Given the multivalent display of DOX (4 DOX per dendrimer) in this construct, this concentration range corresponds to DOX concentrations of 0.64 µM–2.5 µM. In previous studies of NP-peptide cellular DOX delivery our laboratory determined the IC50 of free DOX in HeLa cells to be ∼2 µM (Sangtani et al., 2017) which agrees well with values reported in the literature (Nguyen et al., 2015; Tomankova et al., 2015). In the current study, the measured cellular viability of ∼37% at 0.16 µM pHLIP (0.64 µM DOX) for the pHLIP-PAMAM-DOX conjugate reflects an improvement in DOX IC50 of greater than three-fold compared to the free drug. We confirmed that the observed toxicity was due to the anti-proliferative effects of DOX and not a result of undermined membrane integrity caused by membrane insertion of the pHLIP constructs. Trypan blue assays showed that across all three conjugate systems and at both pH values, membrane integrity remained largely intact (Fig. S7). The cytotoxicity data is consistent with our observations in fluorescence imaging which, for the pHLIP-PAMAM-DOX construct, showed a slower rate of nuclear accumulation over 48 h while at 72 h, the intensity of nuclear DOX staining was nearly identical between the two systems. Similar enhancement of DOX cytotoxicity mediated by the slow, sustained release of DOX from NP conjugates has been reported elsewhere. For example, Spillmann et al. showed an ∼40-fold improvement of DOX IC50 when it was delivered as a liquid crystal NPtransferrin formulation with DOX loaded in the core of the NP (Spillmann et al., 2014). Cumulatively, our data provide strong evidence of the ability of the pHLIP-PAMAM system to enhance DOX IC50 by mediating the slow, sustained release of DOX from the dendrimer surface.

by the low pH-induction of pHLIP membrane insertion/translocation. The improvement of DOX efficacy compared to free drug is attributed largely to the slow sustained release of DOX from the dendrimer surface, an observation that is consistent with that reported for other NPDOX delivery systems (Spillmann et al., 2014). Further, our findings point to important design considerations for peptide-mediated drug delivery going forward. First, the ease of synthesis of the conjugates used herein provides for the possibility of the use of non-natural amino acids or protease-resistant peptidyl motifs for enhanced conjugate stability. Second, the strategy shown here clearly demonstrates the ability to circumvent the sequestration of delivered cargos in the endolysosomal system. This has been one of the chief challenges facing not only peptide-mediated but also NP-mediated drug delivery, as the typical route of cellular entry of these materials is via the endocytic pathway (Nazarenus et al., 2014). Finally, the generation three dendrimer has a total of 32 available surface amine groups for conjugation to therapeutic drugs. Thus, the four DOX moieties conjugated to the dendrimer backbone leave available further “handles” for the decoration of the dendrimer surface with either larger DOX payload or other drug species. This presents the exciting possibility for the multivalent display/delivery of multiple disparate drug species on the same peptide-dendrimer vector. Given the diameter of the generation three dendrimer used herein (∼3.5 nm) (Lim et al., 2013) and the demonstrated ability for its insertion across the membrane bilayer shown here, these nanomaterials have now emerged as a viable scaffold for pHLIPfacilitated multidrug delivery to cells and tissues. As a wide array of commercial and custom dendrimer materials are available for conjugation to pHLIP, current studies are underway in our laboratory to explore the upper limit of both dendrimer size and the corresponding drug payload that can be efficiently translocated across the plasma membrane by the pHLIP vector.

4. Conclusion

Notes

Acknowledgments The authors acknowledge financial support from the NRL Base Funding Program. K.E.B. is supported by a Research Associateship through the National Research Council.

The authors declare no competing financial interest

The development of efficient and nontoxic peptide vehicles is crucial to the successful implementation of peptide-mediated drug delivery systems. Critical to the advancement of these systems is the need for both efficient targeting to specific cells and the ability to bypass the endolysosomal system during drug uptake. To this end, we developed multiple peptide-DOX conjugates that were designed to take advantage of the targeted intracellular insertion of pHLIP peptide in low pH environments. The pH-induced membrane spanning properties of pHLIP, coupled with the disulfide-linked DOX cargo, facilitated the efficient reduction-mediated cytosolic release and nuclear accumulation of DOX in HeLa cells, a model cancer cell line. The pHLIP-PAMAM-DOX construct displaying four DOX per pHLIP linked to the dendrimer by disulfide bonds availed the delivery of a larger payload of DOX per pHLIP compared to single-displayed DOX. We observed that the rate of nuclear delivery of DOX was faster when displayed as a monomer on the pHLIP vector compared to the multivalent form, likely due to a slower rate of cytosolic insertion of the PAMAM dendrimer and DOX release in the latter construct. Functionally, however, the pHLIP-PAMAM-DOX system mediated greater cytotoxicity compared to pHLIP-S-S-DOX, particularly at lower peptide concentrations. The pHLIP-PAMAM-DOX system we detail here exhibits several significant improvements over previously described peptide-based drug delivery systems. Its synthesis is facile and allows for the controlled stoichiometric ratios of pHLIP (1 copy) and DOX (4 copies) in the conjugate. It exhibits tight control over the pH-dependent cytosolic delivery of DOX with no observed toxicity until DOX release is triggered

Appendix A. Supplementary data Details of the chemical synthesis of pHLIP-DOX conjugates, cellular uptake of DOX, membrane integrity assay are given in Supporting Information. Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ijpharm. 2018.04.027. References Acar, H., Ting, J.M., Srivastava, S., LaBelle, J.L., Tirrell, M.V., 2017. Molecular engineering solutions for therapeutic peptide delivery. Chem. Soc. Rev. 46, 6553–6569. Ai, S., Duan, J., Liu, X., Bock, S., Tian, Y., Huang, Z., 2011. Biological evaluation of a novel doxorubicin-peptide conjugate for targeted delivery to EGF receptor-overexpressing tumor cells. Mol. Pharm. 8, 375–386. Albertazzi, L., Serresi, M., Albanese, A., Beltram, F., 2010. Dendrimer internalization and intracellular trafficking in living cells. Mol. Pharm. 7, 680–688. An, M., Wijesinghe, D., Andreev, O.A., Reshetnyak, Y.K., Engelman, D.M., 2010. pH(low)-insertion-peptide (pHLIP) translocation of membrane impermeable phalloidin toxin inhibits cancer cell proliferation. Proc. Natl. Acad. Sci. U.S.A. 107, 20246–20250. Antosh, M.P., Wijesinghe, D.D., Shrestha, S., Lanou, R., Huang, Y.H., Hasselbacher, T., Fox, D., Neretti, N., Sun, S., Katenka, N., Cooper, L.N., Andreev, O.A., Reshetnyak, Y.K., 2015. Enhancement of radiation effect on cancer cells by gold-pHLIP. Proc. Natl. Acad. Sci. U.S.A. 112, 5372–5376. Burns, K.E., Hensley, H., Robinson, M.K., Thevenin, D., 2017. Therapeutic efficacy of a family of pHLIP-MMAF conjugates in cancer cells and mouse models. Mol. Pharm. 14, 415–422.

71

International Journal of Pharmaceutics 545 (2018) 64–73

K.E. Burns, J.B. Delehanty

Reshetnyak, Y.K., Andreev, O.A., Segala, M., Markin, V.S., Engelman, D.M., 2008. Energetics of peptide (pHLIP) binding to and folding across a lipid bilayer membrane. Proc. Natl. Acad. Sci. U.S.A. 105, 15340–15345. Reshetnyak, Y.K., Andreev, Oleg A., Lehnert, Ursula, Engelman, Donald M., 2006. Translocation of molecules into cells by pH-dependent insertion of a transmembrane helix. Proc. Natl. Acad. Sci. U.S.A. 103, 6460–6465. Rewatkar, P.V., Parekh, H.S., Parat, M.O., 2016. Molecular determinants of the cellular entry of asymmetric peptide dendrimers and role of caveolae. PLoS One 11, e0147491. Richard, J.P., Melikov, K., Brooks, H., Prevot, P., Lebleu, B., Chernomordik, L.V., 2005. Cellular uptake of unconjugated TAT peptide involves clathrin-dependent endocytosis and heparan sulfate receptors. J. Biol. Chem. 280, 15300–15306. Richard, J.P., Melikov, K., Vives, E., Ramos, C., Verbeure, B., Gait, M.J., Chernomordik, L.V., Lebleu, B., 2003. Cell-penetrating peptides. A reevaluation of the mechanism of cellular uptake. J. Biol. Chem. 278, 585–590. Ristow, M., 2006. Oxidative metabolism in cancer growth. Curr. Opin. Clin. Nutr. 9, 339–345. Sangtani, A., Petryayeva, E., Wu, M., Susumu, K., Oh, E., Huston, A.L., Lasarte-Aragones, G., Medintz, I.L., Algar, W.R., Delehanty, J.B., 2017. Intracellularly actuated quantum dot-peptide-doxorubicin nanobioconjugates for controlled drug delivery via the endocytic pathway. Bioconjug. Chem. Semenza, G.L., Artemov, D., Bedi, A., Bhujwalla, Z., Chiles, K., Feldser, D., Laughner, E., Ravi, R., Simons, J., Taghavi, P., Zhong, H., 2001. 'The metabolism of tumours': 70 years later. Novart Fdn Symp 240, 251-260; discussion 260–254. Seyfried, T.N., Mukherjee, P., 2005. Targeting energy metabolism in brain cancer: review and hypothesis. Nutr. Metab. 2, 30. Seyfried, T.N., Shelton, L.M., 2010. Cancer as a metabolic disease. Nutr. Metab. 7, 7. Shen, F., Chu, S., Bence, A.K., Bailey, B., Xue, X., Erickson, P.A., Montrose, M.H., Beck, W.T., Erickson, L.C., 2008. Quantitation of doxorubicin uptake, efflux, and modulation of multidrug resistance (MDR) in MDR human cancer cells. J. Pharmacol. Exp. Ther. 324, 95–102. Song, Q., Chuan, X., Chen, B., He, B., Zhang, H., Dai, W., Wang, X., Zhang, Q., 2016. A smart tumor targeting peptide-drug conjugate, pHLIP-SS-DOX: synthesis and cellular uptake on MCF-7 and MCF-7/Adr cells. Drug Deliv. 23, 1734–1746. Soudy, R., Chen, C., Kaur, K., 2013. Novel peptide-doxorubicin conjugates for targeting breast cancer cells including the multidrug resistant cells. J. Med. Chem. 56, 7564–7573. Spillmann, C.M., Naciri, J., Algar, W.R., Medintz, I.L., Delehanty, J.B., 2014. Multifunctional liquid crystal nanoparticles for intracellular fluorescent imaging and drug delivery. ACS Nano 8, 6986–6997. Sun, J., Blaskovich, M.A., Jain, R.K., Delarue, F., Paris, D., Brem, S., Wotoczek-Obadia, M., Lin, Q., Coppola, D., Choi, K., Mullan, M., Hamilton, A.D., Sebti, S.M., 2004. Blocking angiogenesis and tumorigenesis with GFA-116, a synthetic molecule that inhibits binding of vascular endothelial growth factor to its receptor. Cancer Res. 64, 3586–3592. Tacar, O., Sriamornsak, P., Dass, C.R., 2013. Doxorubicin: an update on anticancer molecular action, toxicity and novel drug delivery systems. J. Pharm. Pharmacol. 65, 157–170. Tian, Y., Bromberg, L., Lin, S.N., Hatton, T.A., Tam, K.C., 2007. Complexation and release of doxorubicin from its complexes with pluronic P85-b-poly(acrylic acid) block copolymers. J. Control. Release 121, 137–145. Tian, Y., Zhang, Y., Teng, Z., Tian, W., Luo, S., Kong, X., Su, X., Tang, Y., Wang, S., Lu, G., 2017. pH-dependent transmembrane activity of peptide-functionalized gold nanostars for computed tomography/photoacoustic imaging and photothermal therapy. ACS Appl. Mater. Interfaces 9, 2114–2122. Tomankova, K., Polakova, K., Pizova, K., Binder, S., Havrdova, M., Kolarova, M., Kriegova, E., Zapletalova, J., Malina, L., Horakova, J., Malohlava, J., KolokithasNtoukas, A., Bakandritsos, A., Kolarova, H., Zboril, R., 2015. In vitro cytotoxicity analysis of doxorubicin-loaded/superparamagnetic iron oxide colloidal nanoassemblies on MCF7 and NIH3T3 cell lines. Int. J. Nanomed. 10, 949–961. Tsou, S.H., Chen, T.M., Hsiao, H.T., Chen, Y.H., 2015. A critical dose of doxorubicin is required to alter the gene expression profiles in MCF-7 cells acquiring multidrug resistance. PLOS One 10, e0116747. Tsutsumi, S., Neckers, L., 2007. Extracellular heat shock protein 90: a role for a molecular chaperone in cell motility and cancer metastasis. Cancer Sci. 98, 1536–1539. Vives, E., 2003. Cellular uptake [correction of utake] of the Tat peptide: an endocytosis mechanism following ionic interactions. J. Mol. Recognit. 16, 265–271. Vivian, J.T., Callis, P.R., 2001. Mechanisms of tryptophan fluorescence shifts in proteins. Biophys. J. 80, 2093–2109. Wang, J.L., Zhang, Z.J., Choksi, S., Shan, S., Lu, Z., Croce, C.M., Alnemri, E.S., Korngold, R., Huang, Z., 2000. Cell permeable Bcl-2 binding peptides: a chemical approach to apoptosis induction in tumor cells. Cancer Res. 60, 1498–1502. Warburg, O., 1956. On the origin of cancer cells. Science 123, 309–314. Wei, Y., Liao, R., Mahmood, A.A., Xu, H., Zhou, Q., 2017. pH-responsive pHLIP (pH low insertion peptide) nanoclusters of superparamagnetic iron oxide nanoparticles as a tumor-selective MRI contrast agent. Acta Biomater. 55, 194–203. Wijesinghe, D., Arachchige, M.C., Lu, A., Reshetnyak, Y.K., Andreev, O.A., 2013. pH dependent transfer of nano-pores into membrane of cancer cells to induce apoptosis. Sci. Rep. 3, 3560. Wijesinghe, D., Engelman, D.M., Andreev, O.A., Reshetnyak, Y.K., 2011. Tuning a polar molecule for selective cytoplasmic delivery by a pH (Low) insertion peptide. Biochemistry 50, 10215–10222. Yao, L., Daniels, J., Moshnikova, A., Kuznetsov, S., Ahmed, A., Engelman, D.M., Reshetnyak, Y.K., Andreev, O.A., 2013a. pHLIP peptide targets nanogold particles to tumors. Proc. Natl. Acad. Sci. U.S.A. 110, 465–470. Yao, L., Daniels, J., Wijesinghe, D., Andreev, O.A., Reshetnyak, Y.K., 2013b. pHLIP(R)-

Burns, K.E., McCleerey, T.P., Thevenin, D., 2016. pH-selective cytotoxicity of pHLIP-antimicrobial peptide conjugates. Sci. Rep. 6, 28465. Burns, K.E., Robinson, M.K., Thevenin, D., 2015. Inhibition of cancer cell proliferation and breast tumor targeting of pHLIP-monomethyl auristatin E conjugates. Mol. Pharm. 12, 1250–1258. Burns, K.E., Thevenin, D., 2015. Down-regulation of PAR1 activity with a pHLIP-based allosteric antagonist induces cancer cell death. Biochem. J. 472, 287–295. Cheng, C.J., Bahal, R., Babar, I.A., Pincus, Z., Barrera, F., Liu, C., Svoronos, A., Braddock, D.T., Glazer, P.M., Engelman, D.M., Saltzman, W.M., Slack, F.J., 2015. MicroRNA silencing for cancer therapy targeted to the tumour microenvironment. Nature 518, 107–110. Danz, E.D.B., Skramsted, J., Henry, N., Bennett, J.A., Keller, R.S., 2009. Resveratrol prevents doxorubicin cardiotoxicity through mitochondrial stabilization and the Sirt1 pathway. Free Radical Biol. Med. 46, 1589–1597. Davies, A., Lewis, D.J., Watson, S.P., Thomas, S.G., Pikramenou, Z., 2012. pH-controlled delivery of luminescent europium coated nanoparticles into platelets. Proc. Natl. Acad. Sci. U.S.A. 109, 1862–1867. Deshayes, S., Heitz, A., Morris, M.C., Charnet, P., Divita, G., Heitz, F., 2004. Insight into the mechanism of internalization of the cell-penetrating carrier peptide Pep-1 through conformational analysis. Biochemistry 43, 1449–1457. Din, F.U., Aman, W., Ullah, I., Qureshi, O.S., Mustapha, O., Shafique, S., Zeb, A., 2017. Effective use of nanocarriers as drug delivery systems for the treatment of selected tumors. Int. J. Nanomed. 12, 7291–7309. Dinca, A., Chien, W.M., Chin, M.T., 2016. Intracellular delivery of proteins with cellpenetrating peptides for therapeutic uses in human disease. Int. J. Mol. Sci. 17, 263. Duchardt, F., Fotin-Mleczek, M., Schwarz, H., Fischer, R., Brock, R., 2007. A comprehensive model for the cellular uptake of cationic cell-penetrating peptides. Traffic 8, 848–866. Eggimann, G.A., Blattes, E., Buschor, S., Biswas, R., Kammer, S.M., Darbre, T., Reymond, J.L., 2014. Designed cell penetrating peptide dendrimers efficiently internalize cargo into cells. Chem. Commun. 50, 7254–7257. Gatenby, R.A., Gillies, R.J., 2004. Why do cancers have high aerobic glycolysis? Nat. Rev. Cancer 4, 891–899. Gogvadze, V., Orrenius, S., Zhivotovsky, B., 2008. Mitochondria in cancer cells: what is so special about them? Trends Cell Biol. 18, 165–173. Guidotti, G., Brambilla, L., Rossi, D., 2017. Cell-penetrating peptides: from basic research to clinics. Trends Pharmacol. Sci. 38, 406–424. Han, L., Ma, H., Guo, Y., Kuang, Y., He, X., Jiang, C., 2013. pH-controlled delivery of nanoparticles into tumor cells. Adv. Healthcare Mater. 2, 1435–1439. Henriques, S.T., Costa, J., Castanho, M.A., 2005. Translocation of beta-galactosidase mediated by the cell-penetrating peptide pep-1 into lipid vesicles and human HeLa cells is driven by membrane electrostatic potential. Biochemistry 44, 10189–10198. Hunt, J.F., Earnest, T.N., Bousche, O., Kalghatgi, K., Reilly, K., Horvath, C., Rothschild, K.J., Engelman, D.M., 1997. A biophysical study of integral membrane protein folding. Biochemistry 36, 15156–15176. Janic, B., Bhuiyan, M.P., Ewing, J.R., Ali, M.M., 2016. pH-dependent cellular internalization of paramagnetic nanoparticle. ACS Sens. 1, 975–978. Karabadzhak, A.G., An, M., Yao, L., Langenbacher, R., Moshnikova, A., Adochite, R.C., Andreev, O.A., Reshetnyak, Y.K., Engelman, D.M., 2014. pHLIP-FIRE, a cell insertiontriggered fluorescent probe for imaging tumors demonstrates targeted cargo delivery in vivo. ACS Chem. Biol. 9, 2545–2553. Kauffman, W.B., Fuselier, T., He, J., Wimley, W.C., 2015. Mechanism matters: a taxonomy of cell penetrating peptides. Trends Biochem. Sci. 40, 749–764. Kim, J.W., Dang, C.V., 2006. Cancer's molecular sweet tooth and the Warburg effect. Cancer Res. 66, 8927–8930. Kyrychenko, A., 2015. NANOGOLD decorated by pHLIP peptide: comparative force field study. Phys. Chem. Chem. Phys. 17, 12648–12660. Lim, J., Kostiainen, M., Maly, J., da Costa, V.C., Annunziata, O., Pavan, G.M., Simanek, E.E., 2013. Synthesis of large dendrimers with the dimensions of small viruses. J. Am. Chem. Soc. 135, 4660–4663. Lundberg, P., Langel, U., 2003. A brief introduction to cell-penetrating peptides. J. Mol. Recognit. 16, 227–233. Maksimenko, A., Dosio, F., Mougin, J., Ferrero, A., Wack, S., Reddy, L.H., Weyn, A.A., Lepeltier, E., Bourgaux, C., Stella, B., Cattel, L., Couvreur, P., 2014. A unique squalenoylated and nonpegylated doxorubicin nanomedicine with systemic long-circulating properties and anticancer activity. Proc. Natl. Acad. Sci. U.S.A. 111, E217–226. Moshnikova, A., Moshnikova, V., Andreev, O.A., Reshetnyak, Y.K., 2013. Antiproliferative effect of pHLIP-amanitin. Biochemistry 52, 1171–1178. Nazarenus, M., Zhang, Q., Soliman, M.G., Del Pino, P., Pelaz, B., Carregal-Romero, S., Rejman, J., Rothen-Rutishauser, B., Clift, M.J., Zellner, R., Nienhaus, G.U., Delehanty, J.B., Medintz, I.L., Parak, W.J., 2014. In vitro interaction of colloidal nanoparticles with mammalian cells: what have we learned thus far? Beilstein J. Nanotechnol. 5, 1477–1490. Nguyen, H.N., Hoang, T.M.N., Mai, T.T.T., Nguyen, T.Q.T., Do, H.D., Pham, T.H., Nguyen, T.L., Ha, P.T., 2015. Enhanced cellular uptake and cytotoxicity of folate decorated doxorubicin loaded PLA-TPGS nanoparticles. Adv. Nat. Sci.-Nanosci. 6. Nilsson, K.P., Rydberg, J., Baltzer, L., Inganas, O., 2003. Self-assembly of synthetic peptides control conformation and optical properties of a zwitterionic polythiophene derivative. Proc. Natl. Acad. Sci. U.S.A. 100, 10170–10174. Onyango, J.O., Chung, M.S., Eng, C.H., Klees, L.M., Langenbacher, R., Yao, L., An, M., 2015. Noncanonical amino acids to improve the pH response of pHLIP insertion at tumor acidity. Angew. Chem. Int. Ed. 54, 3658–3663. Polcyn, P., Zielinska, P., Zimnicka, M., Troc, A., Kalicki, P., Solecka, J., Laskowska, A., Urbanczyk-Lipkowska, Z., 2013. Novel antimicrobial peptide dendrimers with amphiphilic surface and their interactions with phospholipids–insights from mass spectrometry. Molecules 18, 7120–7144.

72

International Journal of Pharmaceutics 545 (2018) 64–73

K.E. Burns, J.B. Delehanty

Zeiderman, M.R., Morgan, D.E., Christein, J.D., Grizzle, W.E., McMasters, K.M., McNally, L.R., 2016. Acidic pH-targeted chitosan capped mesoporous silica coated gold nanorods facilitate detection of pancreatic tumors via multispectral optoacoustic tomography. ACS Biomater. Sci. Eng. 2, 1108–1120. Zhao, Z., Meng, H., Wang, N., Donovan, M.J., Fu, T., You, M., Chen, Z., Zhang, X., Tan, W., 2013. A controlled-release nanocarrier with extracellular pH value driven tumor targeting and translocation for drug delivery. Angew. Chem. Int. Ed. 52, 7487–7491.

mediated delivery of PEGylated liposomes to cancer cells. J. Control. Release 167, 228–237. Yu, M., Guo, F., Wang, J., Tan, F., Li, N., 2016. A pH-Driven and photoresponsive nanocarrier: remotely-controlled by near-infrared light for stepwise antitumor treatment. Biomaterials 79, 25–35. Zaidi, S., Misba, L., Khan, A.U., 2017. Nano-therapeutics: a revolution in infection control in post antibiotic era. Nanomedicine 13, 2281–2301.

73