Porphyrin-Gold Nanomaterial for Efficient Drug Delivery to Cancerous Cells.

With an aim to overcome multidrug resistance (MDR), nontargeted delivery, and drug toxicity, we developed a new nanochemotherapeutic system with tetrasodium salt of meso-tetrakis(4-sulfonatophenyl)porphyrin (TPPS) armored on gold nanoparticles (TPPS-AuNPs). The nanocarrier is able to be selectively internalized within tumor cells than in normal cells followed by endocytosis and therefore delivers the antitumor drug doxorubicin (DOX) particularly to the nucleus of diseased cells. The embedment of TPPS on the gold nanosurface provides excellent stability and biocompatibility to the nanoparticles. Porphyrin interacts with the gold nanosurface through the coordination interaction between gold and pyrrolic nitrogen atoms of the porphyrin and forms a strong association complex. DOX-loaded nanocomposite (DOX@TPPS-AuNPs) demonstrated enhanced cellular uptake with significantly reduced drug efflux in MDR brain cancer cells, thereby increasing the retention time of the drug within tumor cells. It exhibited about 9 times greater potency for cellular apoptosis via triggered release commenced by acidic pH. DOX has been successfully loaded on the porphyrin-modified gold nanosurface noncovalently with high encapsulation efficacy (∼90%) and tightly associated under normal physiological conditions but capable of releasing ∼81% of drug in a low-pH environment. Subsequently, DOX-loaded TPPS-AuNPs exhibited higher inhibition of cellular metastasis, invasion, and angiogenesis, suggesting that TPPS-modified AuNPs could improve the therapeutic efficacy of the drug molecule. Unlike free DOX, drug-loaded TPPS-AuNPs did not show toxicity toward normal cells. Therefore, higher drug encapsulation efficacy with selective targeting potential and acidic-pH-mediated intracellular release of DOX at the nucleus make TPPS-AuNPs a "magic bullet" for implication in nanomedicine.

Introduction

Current cancer treatments are mainly based on radiation and chemotherapeutic agents, which have several side effects including severe damage to epithelial surfaces, infertility, swelling of soft tissues, and other side effects.[1] Multidrug resistance (MDR) is another primary limitation to the success of chemotherapy.[2] Furthermore, many effective drugs are hydrophobic or if soluble, it is difficult for them to reach their targets.[3] The development of highly efficient therapeutics to reverse MDR is still a challenging task. In a similar tune, several antiviral drugs have showed low potency because many of these drug molecules fail to effectively cross the blood–brain barrier. Several reports suggested that nanoparticles (NPs), particularly from gold, could be a choice that has a strong potential to deliver a drug inside cancer cells. Nanosized gold particles possess distinctive shape and size and unique chemical properties that rendered them to be used as a drug delivery cargo.[4−6] To make them a better choice as a delivery system, modifications of the nanosurface are often made with ligands that preferentially bind to a specific receptor of the target cells.[7−9] Such modification enhanced the permeability and retention of drug molecules, which were then preferably localized to tumor sites with a high concentration level and thus enhanced the drug effect. Some of the modifications also enhanced the drug loading capacity, thereby reducing the large dose and the toxic effect of the drug molecules.[10] Doxorubicin (DOX) is a common drug used in a wide range of cancers, and DOX-functionalized gold nanoparticles (AuNPs) exhibited significantly higher efficacy compared to that of free DOX in a biological environment.[11−14] DOX loaded with poly(ethylene glycol) (PEG)-AuNPs could significantly overcome MDR in MCF-7/ADR breast cancer cells compared with free doxorubicin.[15] Suarasan et al. showed that DOX-loaded gelatin-coated AuNPs can be used as a nanochemotherapeutic system for the treatment of MCF-7 cells.[16] Elbialy et al. demonstrated higher accumulation of drug with DOX-loaded magnetic AuNPs compared to that by passive targeting.[17] Lee et al. prepared DOX-loaded oligonucleotide-conjugated gold nanoparticles as a promising in vivo drug delivery system for colorectal cancer.[18] Dual activity including chemotherapy and photothermal treatment using AuNPs coated with DOX-loaded fucoidan was also observed in eye tumors.[19] In the current article, we developed tetrasodium salt of meso-tetrakis(4-sulfonatophenyl)porphyrin-modified gold nanoparticles (TPPS-AuNPs) that efficiently delivered the loaded doxorubicin (DOX) molecule within the nucleus of tumor cells and thereby improved the therapeutic efficacy of the drug.

Porphyrins are highly conjugated aromatic molecules and show interesting spectroscopic and electronic properties. They have immense applications in the field of photodynamic and photothermal therapy.[20−23] Porphyrin-stabilized gold nanoparticles using different porphyrins are reported.[24,25] Structural analysis of the gold nanocomposite of thioester derivatives of tetrakis-5,10,15,20-(2-acetylthiophenyl)porphyrin showed that all four pyrrole nitrogen atoms of the porphyrin ring may participate in coordination with the gold nanosurface.[26] The coupling of TPPS and magnetic nanoparticles possibly has wider applications in modern medicine.[27] Porphyrins are reported to form hybrid nanostructures in the presence of different nanomaterials such as gold, graphene, tin oxide, etc.[28−30] Several porphyrin analogs were also conjugated with gold nanoparticles for better use in photodynamic therapy because of high triplet states and singlet oxygen quantum yield efficiency of porphyrin molecules.[31−34]

Here, we showed that the water-soluble tetrasodium salt of meso-tetrakis(4-sulphonatophenyl)porphyrin-modified gold nanoparticles (TPPS-AuNPs) has high efficiency to bind with doxorubicin and effectively delivered the loaded DOX within the nucleus of tumor cells. Our results established that TPPS-AuNPs can significantly reduce the dose of DOX and thereby showed improved therapeutic efficacy toward killing of brain cancer cells, which is a challenging task because of the fast development and poor prognosis of this tumor.

Results and Discussion

Porphyrin and its derivatives are often used in making important hybrid materials and assembly of gold nanoparticles.[26,29,35−39] It contains a flat and planar electron-rich conjugated aromatic ring and facilitates interaction with metal nanosurfaces. Scheme represents the molecular structure of TPPS and DOX. We utilized a simple reduction method to prepare bare gold nanoparticles (AuNPs) using aqueous solution of sodium borohydride (NaBH4). Porphyrin-coated gold nanoparticles (TPPS-AuNPs) were subsequently prepared by incubating the freshly prepared AuNPs with TPPS via a continuous stirring method (Scheme , discussed in the Experimental Section). Figure A depicts the absorption behavior of a water suspension of freshly prepared TPPS-AuNPs. The surface plasmon resonance (SPR) peak of TPPS-AuNPs appeared at 523 nm, and for the bare gold nanoparticles (AuNPs) in aqueous suspension, the peak was at 515 nm.[40] Thus, an ∼8 nm shift in the peak position was observed. The presence of porphyrin also resulted in the broadening of the plasmonic band, indicating a strong association of TPPS with the gold nanosurface. The Soret absorption peak in the UV–vis spectrum of free TPPS (aqueous solution, pH ∼ 9.0) appeared at ∼412 nm (Figure A, inset).[41] The peak was quite broadened in TPPS-AuNPs, and a visible hump appeared at ∼420 nm.

Scheme 1

Molecular Structure of Tetrasodium salt of meso-Tetrakis(4-sulfonatophenyl)porphyrin (TPPS) (A) and Doxorubicin (DOX) (B) Used in the Preparation of Complexes

Scheme 2

Schematic Representation of the Possible Mechanism of Formation of TPPS-Conjugated AuNPs and Subsequent Loading of DOX on Their Surface

Figure 1

(A) UV–vis absorption spectra of TPPS-modified gold nanoparticles prepared at pH 9.0 and dispersed in high-performance liquid chromatography (HPLC) water (red trace) along with bare AuNPs (black trace). The arrows at 412 and 420 nm are the Soret absorption peaks of porphyrin and those at 515 and 523 nm are for the SPR band of AuNPs. The inset figure shows the UV–vis absorption spectrum of free TPPS in the same aqueous basic medium. (B) High-resolution transmission electron microscopy (HRTEM) images of TPPS-AuNPs with a scale bar of 20 nm. (C) Size distribution histogram and the fitted normal distribution curve of the nanoparticles. (D) Powder X-ray diffraction (XRD) pattern of TPPS-AuNPs. The sample was drop-cast on a glass slide and dried at 60 °C. The peaks are assigned based on JCPDS card no. 03-065-2870.

The SPR band is a key to envisage the size, shape, and interaction pattern of metal nanoparticles with other molecules. Several investigators observed a shift in the plasmonic band in nanocomposites because of coating with other molecules and in some cases because of agglomeration of the particles. Shaikh et al. observed a large shift (525–570 nm) in the AuNP plasmon band when treated it with porphyrin and purified it by passing through a specific column.[37] However, agglomeration of the particles in the purification process could be the possible reason of the formation of bigger particles and the red shift in the SPR band. They also observed a small shift (515–528 nm) in the SPR band of the freshly prepared gold nanoparticles in the presence of porphyrin and purified it by a different method. Kanehara et al. prepared porphyrin-stabilized gold nanoparticles from citrate-protected gold nanomaterials, and only small changes in the SPR band were observed.[26] They found that the Soret (from porphyrin ring) absorption peak shifted by 4–11 nm depending on the gap between the metal surface and the porphyrin ring. Ohyama et al. also developed a “lunar-lander-like” porphyrin ligand that can stabilize the gold nanosurface with face-to-face parallel geometry.[24] They observed a broad SPR band for the nanosurface and a bathochromic shift of ∼4 nm in the Soret absorption. The peak was broadened compared to that of free porphyrin in water. In our TPPS-armored gold nanoparticles, the Soret peak was broadened and a hump was observed at ∼420 nm compared to those of its monomer in solution (∼412 nm). The broadening and red shifting of the SPR band and the TPPS Soret absorption indicated that TPPS was involved in strong interaction with the gold nanosurface.[37]

Figure B shows the high-resolution transmission electron micrograph (HRTEM) of TPPS-AuNPs, which revealed that the particles were largely spherical in shape and were uniformly distributed. The transmission electron microscopy (TEM) image also confirmed low level of particle aggregation. Figure C shows the size distribution histogram of the nanoparticles and the fitted normal distribution curve. About 30 particles were randomly chosen from the enlarged TEM images to obtain the plot. The distribution analysis found that the average particle diameter was ∼10 nm. The hydrodynamic size and ζ potential of the nanoparticles were also measured by dynamic light scattering (DLS) analysis in aqueous suspension. The distribution plots are shown in Figures S1 and S2. The average hydrodynamic diameter was ∼10 nm. The measured surface potential of TPPS-AuNPs was ∼−41 mV, whereas for bare AuNPs, it was ∼−19 mV. The significant increase in the negative ζ-potential value was due to surface modification caused by negatively charged sulphonated groups of TPPS.

The X-ray diffraction (XRD) pattern was also used to determine the crystalline nature of the synthesized particles. Figure D shows the powder X-ray diffraction pattern of TPPS-AuNPs in the 2θ range 32–90°. It exhibited intense peaks at ∼38.10, 44.30, 64.50, 77.60, and 81.70 that correspond to diffraction from (111), (200), (220), (311), and (222) crystal planes, indicating that the synthesized gold nanoparticles were in the form of a face-centered cubic lattice in its solid state. The peaks were assigned using standard literature data based on JCPDS card no. 03-065-2870. The XRD peak patterns suggested nanocrystaline nature of the particles with Fm3m symmetry.

Isothermal titration calorimetry (ITC) is a sensitive technique that provides several thermodynmical parameters pertinent to binding of small molecules to the nanosurface and other macromolecular systems. It measures parameters such as the binding constant (Ka), change of Gibbs free energy (ΔG°), and entropy change (ΔS°) for the association processes. Figure A shows the calorimetric profile of TPPS interaction with AuNPs at 25 °C. The top panel shows the change in heat as a function of time for successive injection of TPPS. The heats of dilution obtained from control experiments of injecting identical amounts of the TPPS solution into the aqueous medium alone were subtracted to get corrected values of heat (shown at the top portion of the upper panels, curve). The lower panel shows the corrected integrated heats (black squares) plotted against the molar ratio of the association of TPPS with the gold nanoparticles, and the data were fitted to a single-site binding model. The calculated binding constant, standard molar enthalpy change, entropy contribution, and standard molar Gibbs energy change were 3.13 × 105 M–1, −11.22 kcal/mol, −12.5 cal/mol deg, and −7.5 kcal/mol, respectively. The negative enthalpy and negative entropy values suggested the presence of possible electrostatic interactions between porphyrin and the metal nanosurface. The binding constant was similar to that for the binding of nonfunctional porphyrin to the gold nanosurface, as investigated by Shaikh et al., and for the interaction of TPPS to magnetic iron oxide nanoparticles and other nanosystems.[27,37,42−44] It is likely that porphyrin got adsorbed on the AuNP surfaces through the coordination interaction between gold and pyrrolic nitrogen atoms of the porphyrin ring and formed a strong association complex.

Figure 2

ITC profile for the binding of TPPS to AuNPs. The top panel presents raw heat effects for the titration of AuNP suspension with 1.8 mM TPPS at 25 °C in aqueous basic medium (pH ∼ 9.0). The bottom panel shows the integrated heat effects after correction of heat of dilution against the molar ratio of AuNPs to TPPS.

Fourier transform infrared (FT-IR) analysis also suggested ample interaction of the metal surfaces with the porphyrin molecules. Figure S3 shows the FT-IR spectra of TPPS and TPPS-AuNPs. Some of the prominent FT-IR bands of TPPS appeared at 639, 738, 804, 1009, 1039, 1125, 1190, 1221 (shoulder), and 1395 cm–1. The bands were assigned based on the available literature.[6,45,46] TPPS contains a π-conjugated porphyrin ring and four negatively charged sulphonated groups (SO3–) connected to the phenyl moiety of the molecule. Many of the observed bands were linked to the phenyl moiety of the molecule.[46,47] The bands at 1221 and 1190 cm–1 could be assigned to the vibrations of SO3– groups of TPPS. Most of the vibration bands of TPPS also appeared in TPPS-AuNPs, however, with weak signals. The bands for the SO3– groups were also present in the TPPS-AuNPs and shifted slightly. This indicated porphyrin interaction with the metal nanosurface. The bands at 1470, 1009, and 804 cm–1 were linked to the vibrations of bonds connected to ring nitrogens, and associated bond vibrations were more affected and some changes in the frequency position were observed. The ring nitrogen atoms might be involved in coordination with the metal nanosurface and perturbed the ring vibrations to a certain extent, causing changes in some of these vibrational frequencies.[46]

We tested the stability of TPPS–AuNPs under high salt concentration conditions. Figure shows the optical absorption spectra of TPPS-AuNPs in several ionic buffers at high ionic strength (∼100 mM) after incubating for 30 min at 37 °C. The SPR band remained at 523 nm, and no significant change was observed. The porphyrin Soret absorption peak was also found to be less affected. These two observations suggest an excellent colloidal stability of the nanoparticles. A charged nanosurface is usually prepared through surface coating with specific molecules. The charge over the nanoparticle surface creates a potential difference (ζ potential) and stabilizes the colloidal nanoparticle suspension in solutions by repulsive electrostatic interactions. However, electrostatically stabilized nanoparticles are prone to aggregation in different chemical or biological media because of neutralization of the surface charge by ionic species present in the media. The prolonged colloidal stability experienced by TPPS-AuNPs in different media (Figure ) may be due to the presence of stable TPPS embedment. Highly hydrophilic TPPS coatings have been shown to resist ion adsorption and also provide steric hindrance, thus preventing aggregation and coagulation to form bigger particles.

Figure 3

UV–visible absorption spectra of the TPPS-AuNPs in different ionic buffer mediums (100 mM), namely, N-(2-hydroxyethyl)piperazine-N′-ethanesulfonic acid (HEPES) (black), KCl (red), NaCl (blue), phosphate-buffered saline (PBS) (dark cyan), phosphate (magenta), Tris–HCl (dark yellow), and water (navy) under normal physiological conditions (pH ∼ 7.4, 37 °C).

Porphyrin molecules can form stable molecular junctions, and addition of anchoring groups to the porphyrin backbone increases the stability of the junctions.[48] It was observed that porphyrin with four amino groups binds strongly to AuNPs compared with porphyrin having less number of/no functional amino groups. With no functional amino groups, the porphyrin fluorescence intensity is quenched because of the standard face-on and face-off interactions of porphyrin with gold nanoparticles. Time-resolved fluorescence measurements indicated that the quenching of the fluorescence of porphyrin by AuNPs is of static type.[37] Furthermore, it was established that the thiol functional groups have a higher affinity to the gold surface and showed strong binding to gold nanoparticles.[49] Porphyrins with diamino groups and thiols show similar binding constants. Some groups studied the formation of Au and tetrapyridyl porphyrin complexes and suggested that Au atoms and small Au clusters can form dative bonds with the lone electron pair on a nitrogen atom; hence, the Au clusters are likely to bond to one of the pyridyl nitrogen atoms.[50−52] Bhaumik et al. developed porphyrin–gold nanoconjugates as biocompatible photosensitizers and showed that the porphyrin with a carboxyl tether was successfully attached to the surface of the bioinspired metal nanoparticles (rich in −OH group) via covalent ester bond formation.[53] Nada et al. reported that π-conjugated phthalocyanine rings can serve as stabilizing ligands for gold nanoparticles through van der Waals interaction between parallel-adsorbed phthalocyanine ligands and the gold nanoparticle surface.[54] Evdokimova et al. found that porphyrin with four bromine groups has a stronger affinity for the surfaces of gold nanoparticles than that of porphyrin without functional groups.[55] Zhang et al. suggested that porphyrin monomers are first adsorbed on the gold nanosurfaces through the coordination interaction between gold and pyrrolic nitrogen atoms, followed by the formation of linear or necklace chain assemblies of the gold nanospheres via the π–π stacking interaction between porphyrin monomers.[35] Kanehara et al. showed that all of the ring nitrogen atoms of porphyrin are involved in binding to the nanosurface and aligned the porphyrin rings quite parallel to the gold nanosurface.[26] The additional study established a very close face-on configuration that may generate a strong coupling between the π orbital of the porphyrin ring and the 6s orbital of gold.[56]

TPPS, used in our investigation, contained no sulfhydryl or −NH2 functional groups that could easily bind to the nanosurfaces. However, it contained a highly electron rich π-conjugated macrocyclic ring with two pyrrolic nitrogen (−NH) atoms. The four sulphonated groups render the porphyrin molecule water-soluble but not stably functionalizing the AuNPs. However, pyrrolic nitrogen atoms have strong propensities to be involved in coordination with the metal surface.[27] It could be a strong association complex. ITC measurements established the binding/association constant for binding of TPPS to gold nanoparticles to be about 3 × 105 M–1. The adherence of TPPS to the nanosurface was further confirmed by the ζ potential measurement of both AuNPs and TPPS-AuNPs. A significant increase in the negative ζ potential value (from −19 mV of AuNPs to −41 mV of TPPS-AuNPs) could be attributed to the presence of TPPS (SO3– groups) on the gold nanosurface. FT-IR investigation (Figure S3) also showed the presence of TPPS on the gold nanosurface. This investigation also showed that some of the vibration bands are associated particularly with the macrocyclic ring (that contains nitrogen atoms) of the molecule affected, and not much difference was observed in the vibration of sulphonated groups. We further performed dialysis of the nanocomposite for 2 days in aqueous medium (pH ∼ 7.0) and observed that the porphyrin Soret absorption peak remained at ∼412 nm; however, the peak was broadened. Steady-state fluorescence investigation (not shown) also suggested very strong binding of TPPS molecules to the nanoparticles. Upon repeated washing, some loss of porphyrin from the nanosurface was possible that could cause agglomeration of the nanostructures.[37]

The efficient delivery and release of a drug molecule to a targeted cell are a crucial challenge to improve therapies for a range of human diseases. We have selected tetrasodium salt of meso-tetrakis(4-sulfonatophenyl)porphyrin (TPPS) to modify the gold nanosurface. Its unique conductivity properties due to the conjugated π-system, structural flexibility, and several possible architectural modifications make it attractive for use in nanocomposites and the modulation of conductive properties of gold nanoparticles. TPPS is water-soluble. In vivo tumor localization potential of the porphyrin was found to be reasonably high. Furthermore, it has four −SO3– groups that may interact with −NH2 groups of DOX through hydrogen bonding. These hydrogen bonds are very sensitive to pH. In an acidic environment (cancer cells), the hydrogen bonding interactions become weak, resulting in the release of DOX from the nanosurface and thus making it available to kill the cells effectively. Here, we used a commonly used clinically approved anticancer drug (DOX) as a model system and embedded it on TPPS-AuNPs to make a ternary nanochemotherapeutic system, DOX@TPPS-AuNPs (Scheme ). Loading of DOX was monitored by UV–vis spectroscopy, ζ potential, DLS, TEM, and fluorescence measurements. Figure A shows the UV–vis spectra associated with the SPR band of DOX-loaded and free TPPS-AuNPs. The inset in the figure shows the time-dependent doping of DOX on the porphyrin-coated gold nanosurface. We found that after 1 h incubation of DOX with TPPS-AuNPs at pH 7.0, ∼60% of DOX was entrapped. It increases slowly thereafter and reached a plateau, upto ∼90% encapsulation, at 24 h. However, AuNPs entrapped only ∼25% of DOX for 1 h of incubation (Figure S4). The weight ratio of DOX to AuNPs/TPPS-AuNPs was 1:7. The reason of enhanced encapsulation by TPPS-AuNPs could be the favorable interactions between −SO3– groups of TPPS on the gold nanosurface and −NH2 groups of DOX. The binding interaction between the two groups was highly possible, and it was reported earlier.[45] Upon DOX loading, the size of the nanoparticles increased, as confirmed by the large bathochromic shift of the SPR band from 523 to 584 nm (Figure A). The TEM micrograph also confirms the increased particle size upon DOX loading (Figure B). Figure C illustrates the usual fluorescence emission spectrum of DOX (30 μM) in water. On addition of TPPS-AuNPs to the solution, mainly the bright fluorescence of DOX was substantially quenched, and it strongly suggested the encapsulation of DOX on TPPS-AuNPs. Thus, a novel ternary complex (DOX@TPPS-AUNPs) was formed. ζ potential and DLS measurements further supported the fact that DOX was sufficiently loaded in TPPS-AuNPs (Figure S5). The ζ potential value of TPPS-AuNPs was −41.1 mV, and it was increased to −34.3 mV for DOX-loaded TPPS-AuNPs. The hydrodynamic diameter also increased from 10 to 26 nm after DOX loading.

Figure 4

(A) Optical absorption spectra of TPPS-AuNPs and DOX@TPPS-AuNPs in water. The inset shows the encapsulation efficiency (EE) of DOX as a function of time on the surface of TPPS-AuNPs in aqueous medium at pH ∼ 7.0. (B) TEM micrograph of TPPS-AuNPs loaded with doxorubicin (scale bar 100 nm). (C) Fluorescence spectrum of DOX (30 μM) in water (black trace) and fluorescence spectrum of the supernatant obtained upon adding TPPS-AuNPs to the DOX solution followed by filtration (red trace with reduced intensity). (D) Fluorescence spectrum of DOX released from the TPPS-AuNPs nanosurface after incubating DOX@TPPS-AuNPs at pH = 5.0 (citrate buffer) at 37 °C for 40 h. The inset shows the corresponding UV–vis spectrum. (E) Release of DOX from DOX@TPPS-AuNPs at 37 °C incubated under three different solution conditions (pHs varies) with time. The calculations of encapsulation efficiency and release % are explained in the Materials and Methods section.

The flat porphyrin ring can possibly cradle and facilitate the incorporation of doxorubicin in the porphyrin–gold nanocomposite. Interactions (electrostatic and H-bonding) among the polar and ionic groups (SO3–) of TPPS and −NH2 groups of DOX or π–π staking between the conjugated structure of the TPPS moiety and DOX could provide ample stabilization in the binding processes (Scheme ).[57] This kind of interaction was already observed between DOX and graphene oxide (containing electron-rich π-orbital substituted with the CO2– group).[58,59] Therefore, DOX was noncovalently loaded on TPPS-AuNPs and the electrostatic/hydrogen bonding interaction was the major force in the association processes. The ζ potential of TPPS-AuNPs was −41.1 mV at neutral pH (pH = 7.4), AuNPs being covered with the anionic TPPS porphyrin. The ζ potential of DOX-loaded TPPS-AuNPs was −34.3 mV. The decrease in the ζ potential is attributed to the charge neutralization of suphonated groups. This also justified a large quenching of fluorescence of DOX (Figure C) in DOX@TPPS-AuNPs, and both the facts supported the formation of a strong and stable ternary complex.

Scheme 3

Schematic Representation of Electrostatic Interaction and Hydrogen Bonding between Adsorbed TPPS and Adsorbed DOX Molecules on AuNP Surfaces

TPPS molecules attached on the AuNP surface through coordination interaction.

For efficient drug delivery, the biocompatibility of drug nanocarriers and controlled release of the drug at the pathological site are essential and important requirements. As we showed above, the DOX@TPPS-AuNP complex was formed by the adsorption of DOX on the surface of TPPS-AuNP and the bonding was very strong at neutral pH and caused inefficient release.[58] Once DOX was loaded, its fluorescence signal significantly quenched. However, when the DOX@TPPS-AuNPs system was dispersed in acidic buffers, the chances of pyrrolic nitrogen (−NH) protonation increased and the −NH2 group (pKa ∼ 7.2) of DOX also got protonated. This caused instability in the binding interaction of porphyrin/DOX with the metal nanosurface and resulted in the release of both DOX and porphyrin molecules from the metallic surface. The absorption spectrum (Figure D, inset) shows the Soret absorption peak of diprotonated porphyrin at 433 nm under acidic conditions (pH ∼ 5.0).[60] However, the spectrum was dominated with the DOX absorption band.

To measure the pH-triggered uncapping efficiency of DOX@TPPS-AuNPs nanoparticles in greater detail, the drug release experiments were performed by varying the pH of the solutions. A small amount of DOX@TPPS-AuNPs was incubated with different buffers of varying pHs and maintained at 37 °C under gentle shaking. The amount of DOX released at different time intervals was measured based on the absorbance of the released DOX in the medium. Absorbance at 480 nm was recorded and plotted as a function of time to generate a release profile. Figure D shows the fluorescence emission spectrum of released DOX from the TPPS-AuNP surface after incubating the solution for 40 h at pH 5.0 and at 37 °C. It showed strong fluorescence emission at 592 nm, which is a characteristic emission wavelength of DOX. This observation also suggested that the molecular structure of DOX remained intact upon binding and subsequent release from TPPS-AuNPs and thus it can be available in the tumor environment (acidic pH) to act effectively. A pH-dependent DOX release profile has been shown in Figure E. This shows much less (∼25%) DOX releases after 40 h under normal physiological conditions (pH 7.4 and 37 °C). At the end of 40 h, the amount of DOX released was ∼64% at pH 6, ∼81% at pH 5 at 37 °C. Thus, it was found that most of the DOX remained in the composite for a considerable length of time in the plasma under physiological conditions (pH 7.4). Hence, a reduction in the release of DOX to the normal tissue was observed as the pH of body fluids is around pH 7.4. The release of DOX was linked to the pH of the microenvironment surrounding the nanocomposite and the protonation of both the ring nitrogen of TPPS and the −NH2 group of DOX. Protonation of the ring nitrogens causes instability in the binding interaction of porphyrin with the metal nanosurface and protonation of the −NH2 group causes instability to the electrostatic interaction between the TPPS and DOX. Chances of protonation of the pyrrolic nitrogens/–NH2 of DOX increase with a decrease in the pH of the solution. The absorption spectrum at pH ∼ 5.0 (Figure D, inset) shows the strong Soret absorption peak of diprotonated porphyrin at 433 nm. However, the pKa of porphyrin pyrrolic nitrogen is ∼4.9 and that of the cellular environment in cancerous cells is ∼6.0. The change of pH from 7.4 to 6.0 may not result in a drastic change what we observed at pH ∼ 5.0; however, the medium is sufficiently acidic to weaken the binding interaction of TPPS with AuNPs. At pH 6.0, however, the −NH2 groups of DOX (pKa ∼ 7.2) may be protonated to a greater extent. It could make the interaction of DOX and TPPS on the nanosurface very unstable. We found that DOX was released more at acidic pH (Figure E). Taken together, the pH-dependent release may help improve the efficiency of DOX@TPPS-AuNPs as an efficient delivery system, where the normal cells are less affected.

Multidrug resistances resulting from the expression of the mdr-1 and mrp-1 gene products are the major barriers for the successful chemotherapy in cancer.[61−63] Overexpression of MDR gene-encoded P-glycoproteins and MDR-associated proteins is responsible for quick drug efflux from cancer cells, which reduces the effective drug concentrations within the cells and thus decreases its sensitivity. LN229 cells express both mdr1 and mrp1 and also higher P-glycoprotein levels.[64] We examined if TPPS-AuNPs could overcome the lower drug accumulation and retention in LN229 cells. Accordingly, LN229 cells were incubated with free DOX and DOX@TPPS-AuNPs separately at different time intervals and the quantity of DOX inside the cells was determined by flow cytometry (Figure A). The fluorescence intensity was ∼5-fold-enhanced when the cells were incubated with DOX@TPPS-AuNPs compared to that with free DOX even after 2 h incubation, suggesting higher amount of DOX internalization, which reached plateau after 6 h. In contrast, cells treated with free DOX took a longer time to reach the plateau because of its poor uptake in LN229 cells. Thus, it may be stated that DOX-loaded NPs elicit their effect more prominently by increasing the cellular uptake within a short span of time and could be beneficial to reduce the drug toxicity as well as efficient delivery.

Figure 5

(A) Accumulation of total DOX in DOX- and DOX@TPPS-AuNP-treated MDR brain cancer cells (GBM, LN229). (B) DOX- and DOX@TPPS-AuNP-pretreated (6 h) LN229 were cultured in a complete medium without any drug. The concentration of DOX inside cells was measured at different time points by flow cytometry. (C, D) U87MG cells treated with free DOX or DOX@TPPS-AuNPs for 48 h. The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay showed DOX@TPPS-AuNPs having half-maximal inhibitory concentration (IC50) ∼9-fold lower than that of free DOX (C). Inverted light microscopy showed morphological changes (D). (E) Higher annexin V/propidium iodide (PI)-positive U87MG in the upper right quadrant indicating more apoptosis when treated with DOX@TPPS-AuNPs compared to that with free DOX. (F) Western blots showing upregulation of proapoptotic (Bax/Bid) and downregulation of anti-apoptotic (Bcl2) proteins in DOX@TPPS-AuNP-treated U87MG.

We further determined the retention time of DOX in LN229 cells when the drug was delivered as doxorubicin-tethered AuNPs. Accordingly, cells were incubated with either free DOX or DOX@TPPS-AuNPs for 4 h. They were subsequently washed with PBS to remove uninternalized free drug or DOX@TPPS-AuNPs. The cells were further incubated in fresh complete medium for different periods of time (1–4 h). We maintained the initial dose (500 nM) of doxorubicin during treatments with DOX@TPPS-AuNPs. However, we intentionally increased the initial concentration of doxorubicin to 4 μM when free DOX was used for treatment to increase the initial cellular concentration of this drug. As shown in Figure B, a fast decrease in the intracellular doxorubicin level was observed in LN229 cells preincubated with free doxorubicin because of rapid drug efflux. However, the amount and the efflux rate of doxorubicin were significantly lower when the cells were preincubated with DOX@TPPS-AuNPs, indicating that the efflux of drug was decreased upon its coupling with TPPS-AuNPs. These results demonstrated that doxorubicin is protected against efflux when attached with TPPS-AuNPs and thus the nanoparticles increased the cellular retention of doxorubicin in LN229 cells. As porphyrins are essential cofactors of many human proteins including hemoglobin and myoglobin, we expect that TPPS-modified gold nanoparticles can escape their uptake by macrophages and may retain in the system for a longer time and thus could be useful in vivo.

The successful delivery of doxorubicin into the cancer cell is reflected by monitoring the cell viability. The range of concentrations selected was that which was frequently used for biomedical applications of AuNPs, where 200 μg/mL or 400 nM (100 μg/mL) was used for in vitro and 151 μg/mL for in vivo biodistribution studies.[65,66] The cytotoxic effect depends on the size, shape, surface coating, and charge of gold nanoparticles.[67−70] Regarding the size of the AuNPs, citrate-stabilized AuNPs of 5 nm have been shown to be cytotoxic at a concentration higher than 50 μM with Balb/3T3 cells.[71] However, Vijayakumar et al. found that citrate-capped AuNPs having diameters 5, 6, 10, 17, and 45 nm did not decrease the cell viability of MCF-7 and PC-3 cells.[69] Furthermore, TAT-modified AuNPs having diameters 3.8 and 22.1 nm showed very low toxicity in HepG2 cells.[70] AuNPs with 15–150 nm diameters are also biocompatible, and only certain 150 nm PEG-functionalized particles reduced the viability at high concentrations.[72] Huo et al. showed that nanoparticles smaller than 10 nm (2 and 6 nm) could preferentially enter the nucleus of the MCF-7 breast cancer cell.[73] The size-dependent gold nanoparticle interaction at the nano–micro interface showed that smaller AuNPs had a lower uptake compared to that of larger NPs at a monolayer cell level. However, the order was reversed at a tissuelike multilayer level. The smaller NPs penetrated better compared with larger NPs in tissuelike materials.[74] Cytotoxicity studies including 3, 10, 25, and 50 nm of AuNPs into the HEp-2 cells indicated that after 1 h incubation 3 and 10 nm particles entered the nucleus, whereas 25 and 50 nm particles accumulated around the nucleus.[75] Furthermore, the shape of the nanoparticles has a marked effect on AuNP toxicity. Rod-shaped AuNPs demonstrate more toxicity than their spherical counterparts. Gold nanorods are more toxic to human keratinocyte cells compared with spherical gold nanomaterials.[4] Properly functionalized AuNPs not only can serve as a drug reservoir but also can increase the circulation time of the drug in the blood stream.[76−78] Therefore, suitable modification of the AuNP surface can reduce cellular toxicity, which is associated with chemical surfactants used during the synthesis of the NPs.[79,80] The ζ potential on the surface of nanoparticles can also influence AuNP toxicity. Cationic AuNPs are moderately toxic compared with anionic AuNPs; however, toxicity of both cationic and anionic AuNPs toward cells is also reported.

Considering all of these points, first, we addressed whether the nanoparticle itself has any toxic effect or not. Accordingly, we treated the normal lung fibroblast (WI38) and nonsmall-cell lung cancer (A549) with different doses of TPPS-AuNPs for 72 h in complete medium (Figure S6). Even upto 200 μM concentrations, the empty nanoparticle showed no toxicity toward both the normal and cancer cells, as determined by the MTT assay. These results shows that TPPS-coated 10 nm diameter AuNPs do not induce any acute cytotoxic effects in the cells, which provides new opportunities for the safe application in drug delivery.

Previously, we showed an acid-responsive release of the drug from the nanoparticle surface (Figure E). The core of the tumor is acidic, and many drugs fail to reach that place; therefore, delivery of DOX by DOX@TPPS-AuNPs may rapidly increase the concentration of the free drug in cancer cells and will enhance the cytotoxicity. Accordingly, we incubated various doses of free DOX and DOX@TPPS-AuNPs with two representative cancer cells having various mutations, namely, A549 and glioblastoma multiforme (GBM) cell lines (U87MG), and determined the half-maximal inhibitory concentration (IC50) of doxorubicin by the MTT assay. Both free DOX and DOX@TPPS-AuNPs were used with the same concentration of DOX. They exhibited a significant cytotoxic effect in a dose-dependent manner (Figures C and S7). The IC50 values were 80 and 94 nM for DOX@TPPS-AuNPs and 560 and 840 nM for free DOX in A549 and U87MG cells, respectively, after 48 under similar conditions. We found that DOX@TPPS-AuNPs showed ∼7- and ∼9-fold better efficacies in A549 and U87MG cells compared with free DOX. Furthermore, drastic morphological changes in these cells treated with an equal concentration (100 nM) of DOX@TPPS-AuNPs compared to those with free DOX for 48 h were visualized under light microscopy (Figure D). The DOX@TPPS-AuNP-treated cells exhibited more cell shrinkage. These cells appeared as round or oval and smaller in size. The cytoplasm of these cells was dense, and the organelles were more tightly packed. All of these morphological changes are the sign of cell death. Only DOX at the similar dose and time point did not show such an effect, indicating that TPPS-AuNPs can even reduce the time span of treatment. Thus, the reduction of doses using TPPS-AuNPs would be beneficial. In addition, we similarly treated normal cells (WI38 and SVG) at IC50 doses at different time points. We found that these cells were not at all affected with DOX@TPPS-AuNPs even after 72 h incubation, whereas free DOX showed toxicity to normal cells at IC50 concentration (Figures S8 and S9).

We also investigated whether the nanoparticle-encapsulated DOX is showing the programed cell death (apoptosis) or necrosis.[81] The detection of apoptosis was assessed by the binding of annexin V/PI with the externalized phosphotidylserine to the outer surface of plasma membrane. Here, we observed loss of membrane integrity of DOX@TPPS-AuNP-treated U87MG cells, as indicated by both annexin V and PI (29.57%) positivity compared to that in free doxorubicin treatment (1.1%), indicating late apoptosis as shown in Figure E. This data demonstrated that DOX@TPPS-AuNPs are capable of killing cancer cells via programmed cell death. For further validation, we checked two proapoptotic (Bax and Bid) and one representative anti-apoptotic (Bcl2) molecules in their protein level (Figure F). We found enhanced levels of Bid and Bax and a decrease of Bcl2 in DOX@TPPS-AuNP-treated cells, confirming induction of apoptosis. These results proved that DOX@TPPS-AuNPs have a higher therapeutic effect than that of free doxorubicin. Consequently, TPPS-AuNPs could improve the therapeutic effect of DOX and served as a good drug carrier.

Metastasis is one of the major problems through which cancer cells travel to different organs of body and thereby generate secondary tumor. Invasion and migration of cancer cells are the two main properties for establishing metastasis. Angiogenesis is another process through which cancer cells generate new blood vessels for their emerging necessity of nutrients. Therefore, it is necessary to reduce metastasis and angiogenesis to inhibit the growth of tumor. Accordingly, we checked the ability of DOX@TPPS-AuNPs to reduce migration, invasion, and angiogenesis properties of cancer cells. Scratch wounds were made through the >80% confluent U87MG cells in a six-well plate. They were incubated with free DOX and DOX@TPPS-AuNPs separately for 36 h. We found that cells treated with DOX@TPPS-AuNPs were unable to migrate to that scratch area, whereas untreated and free DOX-treated cells were able to migrate (Figure A,B). We further determined the invasion ability of cells treated with free DOX and DOX-loaded TPPS-AuNPs using matrigel-coated insert systems. Cells that invaded the membrane and adhered to the lower surface are considered for invading tissues. Three randomly selected fields on the lower side of the insert were photographed, and the migrated cells were counted (Figure C,D). The DOX-TPPS-AuNP-treated cells showed reduced invasion in comparison to that from free DOX. In addition, we compared both DOX-TPPS-AuNP- and DOX-treated cells for their capacity to form a connective tube, which is a signature of angiogenesis. U87MG cells form threadlike connective tubes between cellular colonies. However, at similar doses of free DOX and DOX-TPPS-AuNPs, treated cells showed weak connective tube formation ability (Figure E), which helps kill the cells easily. Therefore, free DOX when conjugated with TPPS-AuNPs exhibited reduction in migration, invasion, and angiogenesis properties compared to those of free DOX alone. Thus, the nanoparticle-conjugated doxorubicin cells can halt metastasis and angiogenesis of cancer cells at a much lower dose.

Figure 6

(A, B) Scratch wound assay performed by making similar scratches on plates of confluent U87MG cells and the cells treated with free DOX and DOX@TPPS-AuNPs separately for 36 h. DOX@TPPS-AuNP-treated U87MG showed lower scratch-wound closure, indicating lower migration. (C, D) DOX@TPPS-AuNPs/DOX-treated U87MG cells kept in an upper chamber of matrigel-coated insert systems to check the invasion ability. Cells that invade the membrane were counted. DOX@TPPS-AuNP-treated cells showed lower invasion ability to the membrane compared to that of free DOX. (E) DOX@TPPS-AuNP-treated U87MG cells showing a weaker threadlike structure compared to that of free DOX (red arrow), indicating inhibition of angiogenesis.

We also investigated whether doxorubicin could be released from DOX@TPPS-AuNPs in response to the intracellular acidic microenvironment. We mixed DOX-tethered AuNPs with the culture medium at pH 5.0 for 1 h and adjusted the pH back to pH 7.4. This medium was then added to MDR brain cancer cells (LN229), and the cells were cultured for another 2 h. For comparison, we treated these cells with DOX@TPPS-AuNPs in regular complete medium. Pretreated (at acidic pH) DOX@TPPS-AuNPs exhibited a significantly lower uptake of DOX compared to that for DOX@TPPS-AuNPs at physiological pH, as measured by flow cytometry (Figure A). This is because the extracellular pretreatment at pH 5.0 led to partial release of doxorubicin. This extracellular doxorubicin could not be effectively retained in LN229 cells without carrier. In addition, confocal microscopy provided visible evidence of acid-responsive intracellular release of doxorubicin from DOX@TPPS-AuNPs (Figure B). Cells incubated with free DOX for 12 h showed only faint red fluorescence signals, whereas more fluorescence was observed in DOX@TPPS-AuNP-treated cells, indicating enhanced entry of doxorubicin into the nucleus. Doxorubicin, being a DNA intercalator, should accumulate in the nucleus. The interaction of doxorubicin with DNA by intercalation is one of the major modes of action. Therefore, the accumulation of doxorubicin in the nucleus is a necessary prerequisite. Close scrutiny of the images revealed that the fluorescence signal from cells treated with DOX@TPPS-AuNPs was significantly stronger than that from the control groups which were more dispersed in the perinuclear region. In cancer cells, the nucleus is more acidic, whereas in normal cells, the nucleus is more alkaline than the cytoplasm. That is why in cancer cells, there is release of more doxorubicin from the conjugates, playing a crucial role in the growth inhibition of cancer cells. The lower fluorescence signal within the nucleus was observed when the cells were incubated with free DOX, indicating that the release of doxorubicin is lower.

Figure 7

(A) Uptake of DOX in LN229 with/without preincubation against DOX@TPPS-AuNPs at pH 5.0 (1 h). (B) Confocal microscopy of U87MG counterstained with DAPI (blue) and CellMask Deep Red (green) showing higher accumulation of DOX (red) in the nucleus (indicated in merged orange) of DOX@TPPS-AuNP-treated cells (12 h). (C) Sodium azide-pretreated LN229 cells incubated with DOX@TPPS-AuNPs at 4 and 37 °C. (D) Cells pretreated with 0.45 M sucrose and K+-depleted medium separately and cultured with DOX@TPPS-AuNPs in complete medium. In all of the experiments, the concentration of DOX was determined by flow cytometry.

In addition, we addressed the internalization mechanism of DOX@TPPS-AuNPs into LN229 cells. Endocytosis is known to be one of the important entry mechanisms for various extracellular materials. This is an energy-dependent mechanism and can be hampered at low temperatures (e.g., 4 °C instead of 37 °C) or under adenosine 5′-triphosphate (ATP)-depleted conditions. Treatment with NaN3 is known to perturb ATP production in cells, thus affecting the endocytotic pathway. As shown in Figure C, incubation of DOX@TPPS-AuNPs at 4 °C for 2 h and with NaN3 (ATP-depleted)-pretreated cells showed significantly reduced internalization compared to that under regular culture conditions. This result suggested that DOX@TPPS-AuNPs entered into the cells via energy-dependent endocytosis. Receptor-mediated endocytosis occurs through clathrin-dependent invagination of the plasma membrane.[79] To prove further, we either pretreated the cells with sucrose (hypertonic treatment) or a K+-depleted medium prior to exposure to the DOX@TPPS-AuNPs. These conditions are known to disrupt the formation of clathrin-coated vesicles on the cell membrane. These pretreatments drastically reduced the level of cellular uptake of DOX@TPPS-AuNPs, as deduced by flow cytometry (Figures D and S10). These observations suggested the clathrin pathway for endocytotic cellular uptake of DOX@TPPS-AuNPs. As the endocytic compartments are acidic, an efficient release of DOX occurs when DOX@TPPS-AuNPs are taken up by the cancer cells via endocytosis. Therefore, a sufficiently high concentration of DOX can be released from the NP surface when it is taken up by the tumor cell through an endocytosis mechanism, thereby greatly improving the efficacy of targeted cancer therapy and enhancing the cell cytotoxicity.

Conclusions

Except a few reports, the ability of porphyrin-conjugated AuNPs to deliver the drug to the tumor cells has not been investigated thoroughly. Majority of the studies focused on the optical properties of porphyrin nanoclusters and their photosensitizing behavior for photodynamic therapy. We have developed a porphyrin-based gold nanocarrier, which can efficiently bind to doxorubicin and effectively release the drug molecules inside cancer cells. The embedment of TPPS on the gold nanosurface through coordination interaction via porphyrin pyrrolic nitrogens resulted in an increase in the stability and solubility of the nanosystem. We further validated its in vitro internalization using different cancer cells, which showed enhanced efficacy of known cancer drugs by enhanced drug delivery. DOX@TPPS-AuNPs exhibited strong tumor growth inhibition and more effective apoptosis of cancer cells compared to those of free DOX. DOX-loaded nanoparticles reduced angiogenesis and metastasis of cancer cells to a higher degree than free DOX. TPPS-AuNPs reduced the effective drug concentration to a certain level, at which normal cells were less affected. Thus, TPPS-AuNPs served dual roles: first, they reduced the concentration/dose level and second, they help in the accumulation of the loaded drug molecule preferentially in the nucleus of the diseased cell and therefore became capable of overcoming the drug resistance. Therefore, TPPS-AuNPs with a higher drug encapsulation efficacy and release of drug at the nucleus of tumor cells act as a “magic bullet” and may have an immense implication in cancer therapy.

Experimental Section

Materials and Methods

All chemicals used were of analytical grade or of highest purity available and used as received. Hydrogen tetrachloroaurate(III) hydrate (HAuCl4·3H2O), tetrasodium salt of meso-tetrakis(4-sulfonatophenyl)porphyrin (C44H26N4Na4O12S4·xH2O), HEPES buffer, antibiotic–antimycotic mixture, annexin V, and propidium iodide (PI) matrigel were from Sigma-Aldrich. Sodium hydroxide (NaOH) and sodium borohydride (NaBH4) were from Merck Millipore. The glassware was carefully cleaned with aqua regia (3:1 HCl/HNO3) and then rinsed several times with HPLC water prior to use under sonication. HPLC water was used to prepare all of the solutions. Iscove’s modified Dulbecco’s medium (IMDM), fetal bovine serum (FBS), Super Signal West Pico imaging system, and CellMask Deep Red Plasma membrane Stain were from Thermo Fisher Scientific. All of the antibodies were from Cell Signaling Technology. Protease and phosphatase inhibitor cocktails were from Calbiochem. Cell cultures insert and flasks were from BD Bioscience.

Synthesis of TPPS-AuNPs

Preparation of simple gold nanoparticles of different sizes and shapes is reported earlier in different contexts.[80,82−85] A few studies showed formation of gold nanoparticles and nanoclusters using functionalized porphyrin molecules.[26,30,39,86,87] We utilized a simple protocol to prepare gold nanoparticles armored with an electron-rich porphyrin macrocyclic ring. Bare gold nanoparticles (AuNPs) were prepared using the standard sodium borohydride (NaBH4) reduction method. Gold chloride solution (10 μL; 1.47 M) was added to 30 mL of water in a round bottom flask, and the pH of the solution was adjusted to ∼9.0 by dropwise addition of 0.5 M NaOH solution. The final concentration of Au3+ was 0.5 mM. The solution was stirred for 15 min. Subsequently, the reaction mixture was placed on an ice bath and 50 mM NaBH4 solution in water was slowly added to the reaction mixture. The molar ratio of Au3+ to BH4– was 1:2. The reaction mixture turned pinkish red, indicating the formation of gold nanoparticles (AuNPs). To avoid coagulation of particles, the reaction mixture was placed on an ice bath. The reaction mixture was continuously stirred for additional 4 h to complete the reaction. The resulting mixture was centrifuged for 10 min at 15 000 rpm and gold nanoparticles were collected in a pellet form, as precipitated.

To prepare porphyrin-embedded gold nanoparticles (TPPS-AuNPs), 600 μM aqueous solution of the tetrasodium salt of meso-tetrakis(4-sulfonatophenyl)porphyrin (TPPS) (Scheme A) was prepared in a falcon tube and covered with aluminum foil. This solution (5 mL) was added to AuNP suspension (5 mg in 5 mL H2O) in a round bottom flask in one addition, and the pH of the solution was adjusted to ∼9.0 by adding 0.01 M NaOH solution. The solution was stirred continuously (1100 rpm) for 24 h under dark conditions (covered with aluminum foil). The weight ratio of AuNPs to TPPS was 1.7:1. The TPPS-conjugated gold nanoparticles (TPPS-AuNPs) were collected by centrifugation (rpm of 15 000, 15 min). The sample was washed once/twice to remove most of the unbound TPPS molecules. Multiple washing was avoided to prevent coagulation. The embedment of the porphyrin into the nanosurface was reflected in the broadening of the SPR band (discussed in the Results and Discussion section). The collected nanocomposite remained pinkish red in aqueous suspension. The purified nanoparticles were re-dispersed in water and used for further characterization, and their ability to deliver doxorubicin into the cancer cells was tested.

Transmission Electron Microscopy (TEM)

The size distribution and morphology of TPPS-AuNPs were characterized by high-resolution transmission electron micrograph (HRTEM). For HRTEM imaging, the drop-casting method was used, in which an aqueous suspension of the prepared gold nanoparticles (TPPS-AuNPs) and the nanochemotherapeutic system (DOX@TPPS-AuNPs) were placed on a carbon-coated 300-mesh copper grid (Allied Scientific Product), and it was dried in a dust-free atmosphere. The electron microscopy images of the sample were taken using a JEOL JEM 2100 HR with electron energy loss spectroscopy instrument followed by an acceleration voltage of 200 kV.

DLS and ζ Potential Study

The mean hydrodynamic diameter and charge of the nanoparticles were determined using a Nano-ZS (Malvern Instruments, Worcestershire, U.K.) at 25 °C (5 mW, He–Ne laser, λ = 632 nm). The operating procedure was programmed using DTS software supplied with the instrument to record the average of 20 runs. Every run was collected for 30 s, and an equilibration time of 5 min was used.

Isothermal Calorimetric Titration

The binding of AuNPs to TPPS was studied by isothermal titration calorimetry (ITC) using a MicroCal VP-ITC unit (MicroCal, Inc., Northampton, MA, Now Malvern Instruments, Malvern, U.K.). To avoid air bubble formation during the course of titration, TPPS and nanoparticles solution were degassed on MicroCal’s Thermovac unit before loading. The calorimeter syringe was filled with the TPPS solution (1.8 mM), and the calorimeter cell, with 1.42 mL of the AuNP suspension (100 μg/mL). Successive injections (10 μL aliquots) of the TPPS solution into the AuNP suspension were performed by the rotating syringe (416 rpm). Control experiments were performed to determine the heat of dilution of the TPPS solution, and these values were subtracted from the integrated data before curve fitting. The area under each heat burst curve was determined by integration using Origin software to provide the heat associated with the injections. The resulting corrected injection heats were plotted as a function of the molar ratio, fit with a “single-site binding model”, and analyzed to give the binding affinity (Ka) and the standard molar enthalpy change (ΔH°) of the binding. The standard molar Gibbs free energy (ΔG°) and the entropy contribution to the binding (TΔS°) were subsequently calculated by the following equationwhere T is the temperature in kelvin. The calorimeter was periodically calibrated electrically; the mean energy per injection was ≤1.30 μcal, and the standard deviation was ≤0.015 μcal. The experimental data were analyzed by dedicated Origin 7.0 software.

FT-IR Investigation

A Bruker TENSOR27 spectrometer was used to record the FT-IR spectra of the samples using the KBr pellet technique. The solid sample (TPPS/TPPS-AuNPs) was mixed with KBr under pressure, and a solid and thin pellet was obtained. The spectra were recorded in the frequency range of 400–4000 cm–1. For each sample, background spectra were obtained with only KBr pellet. The experimental data were processed using Bruker software.

Synthesis of DOX-Loaded TPPS-AuNPs

A doxorubicin-loaded nanochemotherapeutic system (DOX@TPPS-AuNPs) was prepared by mixing TPPS-AuNPs with doxorubicin (Scheme B) in aqueous medium. Freshly prepared TPPS-AuNPs (∼2 mg) were dispersed in 15 mL of water taken in a round bottom flask. The DOX solution (150 μL; 3 mM) at pH 7.4 was added to the above solution so that the final DOX concentration was ∼30 μM. The weight ratio of DOX to TPPS-AuNPs was 1:7. The color of the solution changed from pinkish red to violet, confirming the association of DOX with the TPPS-AuNPs, forming the DOX@TPPS-AuNPs nanocomposite. The color change also indicated some increase in the size of the nanocomposite. The mixed solution was allowed to stir for 30 h at room temperature and centrifuged once at 10 000 rpm for 10 min to remove the unbound drug. The DOX@TPPS-AuNPs were collected in a pellet form. To calculate the DOX encapsulation efficiency (EE), the unloaded DOX remaining in the supernatant was quantified using a calibration curve for DOX as obtained by measuring the absorption of the free drug molecules of known concentration at 480 nm.

The encapsulation efficiency (EE) of the process was measured as a function of time using the following equation.[88,89]where Ctotal DOX is the total concentration of DOX (∼30 μM) measured from a standard calibration curve for DOX at 480 nm and Cfree DOX is the concentration of unloaded DOX measured from the same calibration curve. The DOX@TPPS-AuNPs were again redispersed in 1 mL of HPLC water for further investigation.

DOX Release Study

To quantify the drug release, small aliquots (30 μL) of DOX@TPPS-AuNPs from the stock suspension (1 mL) were rapidly added to equal volumes (1 mL) of different buffer solutions with varying the pHs and thermo stating at 37 °C. For each pH buffer solution, we made six sets, and total 18 sets were prepared for three different pH buffers, each containing 1 mL of buffer solution and 30 μL of stock DOX@TPPS-AuNPs. The solutions were gently shaken at 350 rpm. The different pH buffers (10 mM) chosen for the investigation were pH 5.0, 6.0, and 7.4. Sodium citrate was used to make buffer at pH 5.0, and sodium phosphate was used to make buffers at pH 6.0 and 7.4.

At defined time intervals, each buffer solution was centrifuged with an rpm of 10 000 for 10 min, and the absorbance of the supernatant part was measured at 480 nm to determine the amount of DOX released. The percent of DOX released was calculated from absorbance of the solution and fitted them to the following equationwhere At is the absorbance of released DOX at 480 nm, measured at specific time intervals (between 0 and 40 h), and A0 is the absorbance of total DOX loaded onto TPPS-AuNPs at the same wavelength.

Absorption Spectroscopy

The optical absorption spectra of the synthesized TPPS-AuNPs and other solutions were recorded using a JASCO V-630 Spectrophotometer (JASCO International Co. Ltd., Japan) within the wavelength range of 300–700 nm, and a high-quality quartz cuvette was used as a sample holder.

Fluorescence Spectroscopy

The steady-state fluorescence measurement of doxorubicin was performed by a Cary Eclipse fluorescence Spectrophotometer from Agilent Technologies. The machine was equipped with a Xenon lamp. DOX fluorescence measurement for the encapsulation/release study was carried out at a fixed excitation wavelength of 480 nm, and the emission maximum was at 592 nm. Both the excitation and emission slit widths were kept at 5 nm, and a quartz cuvette of path length 1 cm was used for the fluorescence measurement. All of the experiments were carried out at room temperature (25 °C). Fluorescence quenching of DOX in the presence TPPS-AuNPs indicated interaction and loading of the drug molecules to the nanoparticle. Fluorescence spectra of 30 μM DOX (total) were recorded, and it was allowed to load on the TPPS-AuNP surface. After centrifugation, the fluorescence spectrum of unloaded DOX as present in the supernatant part was recorded. The fluorescence of DOX was found to be almost completely quenched in the presence of the nanocomposite. For DOX release measurements, DOX@TPPS-AuNPs are dispersed in citrate buffer, pH ∼ 5.0 for 40 h, and centrifuged (10 000 rpm, 10 min) and the fluorescence spectrum of the supernatant part was recorded upon excitation at 480 nm and the emission intensity was measured at 592 nm.

Cell Lines and Culture Conditions

WI 38 (normal lung fibroblast), SVG (normal astrocytes), A549 (nonsmall-cell lung carcinoma), U87MG and LN229 (glioblastoma multiforme) cell lines were from ATCC, cultured in IMDM/minimum essential medium supplemented with 10% FBS and 1% antibiotic–antimycotic (complete medium), and maintained at 37 °C with 5% CO2 in a carbon dioxide incubator.

Cell Viability Assay

WI 38 and A549 cells were treated with different doses of TPPS-AuNPs for 48 h. We also tested the effect on U87MG, A549, SVG, and WI 38 cell lines when exposed to different doses of DOX and DOX@TPPS-AuNPs separately for 48 h. Cell viability was determined by the MTT assay.[90,91] The culture media were discarded, and MTT (1 mg/mL, 100 μL/well) was added to the wells, followed by incubation at 37 °C for 4 h. The supernatant was removed. Dimethyl sulfoxide (150 μL/well) was added to dissolve the formazan crystals produced by the viable cells, the plates were shaken for an additional 5 min, and the absorbance of the purple color was recorded on a microplate reader (Thermo Scientific) at a wavelength of 550 nm. The intensity of color indicates the number of viable cells.

Drug Uptake

LN229 cells (1 × 106/2 mL/well) in complete medium were seeded in six-well plates and incubated overnight. The medium was then replaced with fresh medium with the absence and presence of DOX and DOX@TPPS-AuNPs separately followed by incubation for 4 h. The medium was removed. Cells were then washed two times with phosphate-buffered saline (PBS) with pH ∼ 7.2. The amount of DOX inside the cells was immediately analyzed by a FACS Calibur flow cytometer (Becton Dickinson, San Jose, CA). Data were collected and analyzed using CellQuest software (Becton Dickinson) as we did earlier.[92,93]

Cellular uptake of DOX and DOX@TPPS-AuNPs was determined by growing the cells (5 × 105) on coverslips in a six-well tissue culture plate for 24 h. They were cultured in the presence of doxorubicin and doxorubicin-loaded TPPS-AuNPs at a concentration of 100 nM separately for 12 h. They were washed with PBS and fixed with 5% paraformaldehyde in PBS. Canadian Balsam was dropped on the slides to seal the cell samples. Cells were then washed with PBS, and the coverslips stained with the CellMask Deep Red Plasma Membrane Stain were imaged using confocal laser scanning microscopy (Olympus FV1000).

Apoptosis Assay

Externalizations of phosphatidylserine were verified as described earlier.[94,95] Briefly, U87MG cells (1 × 106) were treated with either DOX or DOX@TPPS-AuNPs for 48 h. Washed cells were resuspended in the annexin V binding buffer and kept for 45 min in the dark at 25 °C followed by incubation with fluorescein isothiocyanate–annexin V and PI (5 μg/mL) for 20 min at 4 °C in the dark according to the manufacturer’s instruction. Cells were acquired and analyzed by flow cytometry, as described above.

Immunoblotting

U87MG cells (1 × 106) were cultured to >80% confluency in six-well plate. Untreated and treated cells (1 × 106) with DOX and DOX@TPPs-AuNPs (for 24 h) were detached using a trypsin–ethylenediaminetetraacetic acid (EDTA) solution and centrifuged at 1500 rpm for 5 min. They were washed and lysed in ice-cold PBS with sonication (Qsonica-LLC, XL-2000 series). Cell lysates were centrifuged by cold centrifugation at 10 000 rpm. The supernatants were used for western blotting.[96−98]

The proteins were quantified with the Bio-Rad protein assay kit and loaded equally onto sodium dodecyl sulfate-polyacrylamide gels (10%). They were electrophoretically transferred to poly(vinylidene difluoride) membranes, as described earlier.[99] After blocking with bovine serum albumin, the membranes were incubated with respective primary antibodies overnight under cold conditions. This was followed by incubation with appropriate secondary antibodies labeled with horseradish peroxidase. The signal was visualized using chemiluminescence detection. Developed bands were detected by a X-ray plate.

Migration Assay

The migration assay was done as described earlier.[100] In brief, U87MG cells (1 × 106) were cultured to >80% confluency in a six-well plate. Three separate scratch wounds were made through the confluent cells and washed thrice to remove cell debris. They were incubated with DOX and DOX@TPPS-AuNPs for 36 h. The cell-free scratched area was calculated and represented as percent closure of area compared to that of untreated cells. Images were taken using inverted light microscopy.

Invasion Assay

The invasion assay was carried out as described earlier.[100] In brief, U87MG cells (1 × 106) were cultured to >80% confluency in a 12-well plate. Cells were detached using the trypsin–EDTA solution and centrifuged at 1500g for 10 min. Washed cells (1 × 106) were treated with DOX and DOX@TPPs-AuNPs for overnight. Treated or untreated U87MG or LN229 cells (5 × 104) were suspended in a medium without FBS (100 μL) and added to the upper chamber of an insert (6.5 mm diameter, 8 μm pore size). The insert was placed in a 24-well plate containing the medium (700 μL) with or without 10% FBS. DOX and DOX@TPPS-AuNPs were added to both the upper and the lower chambers. The invasion was monitored after 36 h, and cells were fixed with 3.7% formaldehyde. They were stained with crystal violet solution. Cells on the upper side of the insert were removed with a cotton swab. Three randomly selected fields (10 objectives) on the lower side of the insert were photographed, and the migrated cells were counted. The invasion was expressed as an average number of invaded cells in a field.

Angiogenesis Assay

The angiogenesis assay was carried out as described earlier.[101] In brief, a thin layer of matrigel in IMDM (1:3) was formed in a 12-well plate. U87MG cells (4 × 104) were layered over matrigel in a serum-free medium in a six-well plate. These cells have potentiality to form connective tissues in between cells. Cells were cultured for 48 h to form connective tissues. These cells were treated with DOX and DOX@TPPS-AuNPs separately and kept for another 48 h. Images of connective tubes were recorded by inverted light microscopy.

Endocytosis of DOX@TPPS-AuNPs

Experiments for cellular endocytosis study were carried out as described earlier.[102−104] Cells were incubated with DOX@TPPS-AuNPs (100 nM) under different conditions to inhibit the endocytosis mechanism as described below using representative drug-resistant GBM cells (LN229) followed by monitoring the entry of DOX by FACS.

Low-Temperature Incubation

LN229 cells were incubated with DOX@TPPS-AuNPs (100 nM) in complete medium at 4 °C, instead of at the physiological 37 °C temperature, to keep them in a metabolically less active condition, and the uptake was determined.

ATP Depletion

Cells were preincubated with 10 mM NaN3 and 50 mM 2-deoxy-d-glucose in PBS buffer for 30 min at 37 °C followed by incubation in a solution of DOX@TPPS-AuNPs (100 nM).

Hypertonic Incubation

Cells were preincubated with 0.45 M sucrose in PBS buffer for 30 min at 37 °C before exposure to the DOX@TPPS-AuNPs (100 nM).

Potassium Depletion

The K+ depletion was achieved as described earlier.[104] Briefly, cells were washed once with a potassium-free buffer containing 140 mM NaCl, 20 mM HEPES (pH 7.4), 1 mM CaCl2, 1 mM MgCl2, and 1 mg/mL d-glucose. Subsequently, the cells were alternatively washed three times with the potassium-free buffer diluted with water (1:1; hypotonic buffer) and the potassium-free buffer. Then, the cells were incubated with DOX@TPPS-AuNPs (100 nM) in a potassium-free buffer or complete medium for 90 min at 37 °C. Control cells were treated similarly.

Statistical Analysis

All of the data were from at least three independent experiments, and statistical analysis was performed using Graph Pad Prism 5. The differences between the groups were analyzed by the two-tail student t-test or Mann–Whitney U-test. Standard error bars represent the standard deviation (SD) of the mean (±SD), and *p < 0.05 denoted the significant differences between the means of the untreated and treated cells or two test groups.