Field-actuated antineoplastic potential of smart and versatile PEO–bPEI electrospun scaffold by multi-staged targeted co-delivery of magnetite nanoparticles and niclosamide–bPEI complexes

Abstract

Surgical removal of tumor mass is often followed by its recurrence at the same or a distant location by proliferation and metastasis of remnant cancer cells at the site of surgery. In order to circumvent such complications, we have designed a biocompatible smart stimuli responsive anti-cancer nanofiber patch which can effectively eliminate cancer cells and, at the same time, acts as physical barrier for cancer metastasis. In pursuit of this, in this work we fabricated a composite bPEI–PEO nanofiber scaffold loaded with folic acid functionalized octagonal magnetite nanoparticles (FA-bPEI@Fe₃O₄ NPs) and bPEI–niclosamide complexes and further evaluated its field responsive anticancer therapeutic efficacy. Highly magnetic FA-bPEI@Fe₃O₄ NPs impregnated within nanofiber renders them susceptible to partial disintegration by an external alternating magnetic field (AMF) and thereby enables them to attain on-demand stipulated release of FA-bPEI@Fe₃O₄ NPs and bPEI–niclosamide complexes. Upon release, free FA-bPEI@Fe₃O₄ NPs and niclosamide perpetrated cell death by instigating apoptosis and oxidative stress (in KB cells (FR+) and L132 cells (FR−)) which was subsequently investigated on a qualitative and quantitative basis in a time dependent manner by fluorescence microscopy and flow cytometry, and further corroborated with semi-quantitative RT-PCR analysis. In summary, the composite nanofiber fabricated herein can manifest stipulated on-demand release of diverse bioactive molecules solely under the influence of external AMF and thereby effectively attain the therapeutic perquisites of cancer prognosis exactly where and when needed.

---

1. Introduction

As usually observed in case of cancer patients, disease prognosis tends to take different course with passage of time; thus there is an imminent need for a smart stimuli responsive drug infusing system which can act as drug reservoir and can dispense a stipulated quantity of drug proportionally in response to external field.¹ Recent progress in the field of drug delivery and material chemistry has provided the impetus for innovative nanoscale stimuli responsive hybrid nanomaterials. Such stimuli responsive nanoscale systems are categorized on the basis of stimuli as exogenous stimuli-responsive, endogenous stimuli responsive, and multi-stimuli responsive drug delivery systems.² Exogenous stimuli like temperature changes, magnetic fields, ultrasounds, light, and electric fields are exerted at will by external means so as to manipulate drug delivery whereas endogenous stimuli are characteristic conditions pre-existing in the milieu of cancer cells which includes pH, redox potential, or concentrations of enzymes or specific analytes.³

When it comes down to efficient cancer management strategies, an exogenous stimuli responsive drug delivery system provides better control and versatility as compared to its counterpart. Among all such systems, magnetite (Fe₃O₄) based nanomaterials have witnessed profound success in eradication of cancer by various means.⁴ Apart from rendering stimuli responsive nature to the drug delivery system, these magnetite nanoparticles are also capable of inflicting cell death by hyperthermia (by Neel and/or Brownian relaxation) and ROS (by Fenton and Haber–Weiss reaction) when triggered by an external magnetic field.^5,6 Such magnetite based nanoparticles have already witnessed dramatic success in targeted drug delivery systems owing to a high magnetization property and their inherent propensity for extensive surface functionalization. Moreover, owing to their nanoscale dimensions they easily tend to permeate across the fenestrations of endothelial cells and subsequently land themselves in a tumor mass with leaky vasculature and poor lymphatic drainage (EPR effect).⁷

The ability to maneuver iron oxide nanoparticles to an intended site of action by means of external magnetic field refrains the therapeutic effects to the vicinity of the cancer and overcomes toxic bystander effects to normal cells. In order to further sequester the therapeutic efficacy of such magnetite based systems solely to cancer cells, targeting moieties/ligands like peptides, FA, and other stimuli responsive polymers have been utilized.^8,9 Folic acid (FA), a non-immunogenic ligand, has emerged as an attractive targeting molecule for anticancer drug delivery because folate receptors are often over-expressed on the surface of many human cancers including ovarian, lung, breast, endometrial, renal, and colon cancers.^10–13 The simple well defined conjugation chemistry of folic acid has further made it a favorable candidate for cancer targeting. With the aforementioned perception, the present work investigates folic acid functionalized bPEI capped magnetite nanoparticles anticancer potential against folate receptor positive KB cells. In addition to the anticancer role, the magnetite nanoparticle synthesized herein also renders the system susceptible to an external magnetic field which provides the basis for AMF triggered drug release.

Although magnetite nanoparticles can certainly eliminate cancer by themselves, their toxic side effects limit administration dosage levels. Thus, in order to augment the anticancer potential of magnetite nanoparticles in the present work, an anticancer drug, niclosamide, also was utilized alongside magnetite nanoparticles. The anti-cancer efficacy of niclosamide against a wide range of cancers already has been established in pre-existing literature.¹⁴ In order to circumvent the highly hydrophobic nature of niclosamide, it was encapsulated within bPEI moieties (bPEI–niclosamide complexes) as reported in our previous work.^15,16

In order to assess the anticancer potential of niclosamide and FA-bPEI@Fe₃O₄ NPs, PEO–bPEI based differentially cross-linkable nanofibers were adapted in this work as a controlled drug/nanoparticle delivery platform. In the absence of an external magnetic field, the nanofibrous scaffold exhibits controlled release of both FA-bPEI@Fe₃O₄ NPs and bPEI–niclosamide complexes, whereas upon exertion of AMF the Fe₃O₄NPs impregnated within nanofibers develops an inherent magnetic moment within them which begins to disintegrate intact nanofibers and thereby triggers a sudden surge in release of both niclosamide and FA-bPEI@Fe₃O₄ NPs. The distinctive magnetic field susceptible nature of the system under study enabled us to realize an on-demand nanoparticle and drug delivery platform for better management of cancer prognosis.¹⁷

2. Materials and methods

2.1 Materials

Folic acid (FA), niclosamide, poly(ethylene oxide) (PEO) (MV: 900 000) and branched poly(ethylenimine) (bPEI) (M_w: 25 000) were purchased from Sigma-Aldrich (USA). Fluorescent cellular probes, monodansylcadaverine, rhodamine 123, Hoechst 33342, and 2′,7′-dichlorfluorescein diacetate (DCFH-DA) were also procured from Sigma-Aldrich (U.S.A) and stored at an appropriate temperature until used. Iron(II) sulfate heptahydrate, N-(3(dimethylaminopropyl)-N-ethylcarbodiimide) (EDC), N-hydroxy succinimide (NHS), dimethyl sulfoxide (DMSO), dichloromethane (DCM), N,N-dimethylformamide (DMF), and cellulose dialysis membrane-132 (molecular weight cutoff (MWCO), 12 kDa) were acquired from HiMedia (Mumbai, India). Lysotracker red DND-99 and fetal bovine serum (FBS) were purchased from Life Technologies (U.S.A.). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was bought from Amresco Life Science, USA. Superscript II reverse transcriptase, PCR supermix, random primers, dNTP mixture, and RNase out were acquired from Invitrogen (California, U.S.A.). PCR primers were ordered from Imperial Life Science, India.

Other chemicals used in this work were analytical grade and were used as received without further modification. KB cells (FR+) (HeLa derivative) and L132 cells (FR−) (lung epithelial cells) were received from National Centre for Cell Science, Pune, India. They were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin in a 37 °C incubator with 5% CO₂ and 95% air.

2.2 Synthesis of iron oxide nanoparticles

Octagonal hematite nanoparticles were synthesized by slight modification of an oxidation–precipitation synthesis procedure mentioned elsewhere.¹⁸ In brief, 0.6 M solution of iron sulfate heptahydrate (FeSO₄·7H₂O) was prepared and purged with N₂ in order to lustrate dissolved oxygen in it. A stipulated quantity of NaOH preheated to 80 °C was added gradually to iron sulfate heptahydrate solution until pH attained a value of 9.5. The addition of NaOH was carried out under vigorous stirring conditions (i.e. 1000 rpm) for 30 minutes. After the incubation period, excess of NaOH present in the reaction mixture was neutralized with HNO₃ to form NaNO₃. The reaction mixture was kept under stirring for 12 hours in order to allow completion of the reaction during which growth of magnetite nanoparticles takes place. The magnetite nanoparticles thus synthesized were magnetically separated and washed with ethanol and deionized water to remove remnant salts and unreacted components. The separated magnetite nanoparticles were then dried at 45 °C for 48 hours.

2.3 Synthesis of FA-bPEI capped iron oxide nanoparticles

The octagonal hematite nanoparticles synthesized herein were coated with bPEI (25 kDa) by a sonochemical method. Around 10 mg of iron oxide nanoparticles suspension was probe sonicated in the presence of 8 mg bPEI under reducing conditions of NaOH (0.5 M) for 30 minutes with pulse duration of 3 s “on” and 2 s “off”. Intermittent flash cooling of the sonicated sample was carried out in order to effectively entangle bPEI moieties onto the reactive sites created on the surface of iron oxide nanoparticles during sonication. The resultant reaction mixture was magnetically separated and then repeatedly washed in distilled water for 4–5 times in order to wash away excess unreacted bPEI moieties.

Folic acid was further conjugated to terminal amine groups on the surface of bPEI capped magnetite nanoparticles by EDC/NHS chemistry. Initially, the –COOH group of folic acid was activated by addition of 0.33 mmol NHS and 0.17 mmol EDC to 0.15 mmol FA dissolved in DMSO. The reaction mixture was then stirred for 24 hours to complete the reaction. The bPEI capped hematite nanoparticles were then gradually added to activate the FA reaction mixture and kept under stirring for another 24 hours. The resultant product was subsequently dialyzed against water (cutoff M_w – 12 kDa) until there was no further diffusion of unreacted FA into the dialysis medium. The FA-bPEI capped iron oxide nanoparticles retained in the dialysis membrane were lyophilized to obtain a viscous black product.

2.4 Fabrication of FA-bPEI@Fe₃O₄ nanoparticle and niclosamide loaded PEO–bPEI nanofibers

Pre-synthesized bPEI–niclosamide complexes were added to 3.5 wt% of PEO and 1 wt% of bPEI in DCM : DFM (3 : 2) solvent mixture and blended for 5 hours under magnetic stirring. Once the solution attained homogenous phase 10 mg of FA-bPEI@Fe₃O₄ was added to the mixture followed by stirring for 24 hours. The polymer blend was then electrospun at 14 kV operating voltage with a constant flow rate of 0.4 mL h⁻¹ to obtain random aligned composite nanofibers over a static collector positioned at a distance of 35 cm from the spinneret tip. The cabinet temperature was maintained at 32 °C throughout the electrospinning process. The FA-bPEI@Fe₃O₄ nanoparticle and niclosamide loaded PEO–bPEI nanofibers thus obtained were crosslinked in the presence of glutaraldehyde vapor for 20 seconds. The crosslinked nanofibers were dried in a hot air oven for 3–4 hours at 40 °C and then UV sterilized for 30 min before being used for cell culture studies.

2.5 Characterization of niclosamide and FA-bPEI@Fe₃O₄ nanoparticle loaded PEO–bPEI nanofibers

The surface morphology of niclosamide and FA-bPEI@Fe₃O₄ nanoparticle loaded bPEI–PEO nanofibers was observed by an Ultra plus-Carl Zeiss field emission-scanning electron microscope (FE-SEM) operating at 5 kV. The nanofibers were gold coated for 30 s in a Denton gold sputter unit before being mounted in the FE-SEM. The FE-SEM images of nanofiber were then analysed further by ImageJ software to calculate fiber diameter distribution and mean fiber diameter. An energy dispersive X-ray detector (EDX) operating at an accelerating voltage of 15–20 keV was used to perform compositional analysis of the FA-bPEI@Fe₃O₄ nanoparticle and composite nanofiber.

The monocrystalline nature and crystal phase of as-synthesized FA-bPEI@Fe₃O₄ was identified using a Bruker AXS D8 advance powder X-ray diffractometer (Cu-K_α radiation λ = 1.5406 Å in the range of 10–90° at the scan rate of 0.2° min⁻¹). Transmission electron microscope (TEM) analysis of FA-bPEI@Fe₃O₄ nanoparticle and composite nanofiber was performed using TEM (FEI TECHNAI G2) operating at 200 keV. The composite nanofibers were directly collected over carbon-coated copper TEM grids (200 mesh × 125 μm) during electrospinning whereas FA-bPEI@Fe₃O₄ nanoparticles were drop casted over carbon coated copper grids for TEM observation. The magnetite nanoparticle release from composite nanofibers and its subsequent cellular uptake was estimated by AAS (Avanta M, GBC Scientific Equipment) with pure Fe (2, 4, and 8 μg mL⁻¹) standards as references. The niclosamide entrapment efficiency was calculated by following equation:

Niclosamide entrapment efficiency = (total mass of drug released from nanofiber/mass of total drug added) × 100

Total mass of drug released was estimated by disrupting the stipulated mass of un-crosslinked nanofibers by probe sonication and then estimating the concentration of drug in the release medium from a niclosamide standard curve. The nanofiber degree of swelling and weight loss was also estimated at 24 hours and 72 hours by a procedure mentioned elsewhere. The composite nanofibers contact angle analysis was carried out in a Drop Shape Analysis System-DSA30 (Krüss) by the sessile drop method.¹⁵

XRD patterns for FA-bPEI@Fe₃O₄ nanoparticle and composite nanofiber was acquired by a Bruker AXS D8 Advance powder X-ray diffractometer (Cu-Kα radiation, λ = 1.5406 Å) in the range of 10–90° at a scan speed of 0.05 min⁻¹. The synthesis of FA-bPEI@Fe₃O₄ nanoparticle was further confirmed by acquiring FTIR spectra (4000–400 cm⁻¹) by a Thermo Nicolet spectrometer using KBr pellets. The effect of an external magnetic field on composite nanofiber surface topology also was analyzed by AFM (NTEGRA PNL) with Si cantilever (spring constant – 21 N m⁻¹) probes operating in semi-contact mode at a resonance frequency of 160 kHz. The images were acquired with scan field of 5 μm × 5 μm and then processed further using NOVA software for size and roughness analysis. The hydrodynamic diameter of as-synthesised Fe₃O₄ nanoparticles, bPEI capped Fe₃O₄ nanoparticles, and FA-bPEI@Fe₃O₄ nanoparticles in distilled water were estimated by dynamic light-scattering (DLS) measurements using a Malvern Zetasizer Nano ZS90 instrument equipped with 4 mW He–Ne laser operating at a wavelength of 633 nm and a detection angle of 90°. The size distribution intensity was obtained from analysis of correlation functions using the algorithm based upon non-negative least square fit. The zeta potential corresponding to these nanoparticles were acquired by using a disposable folded capillary cell (DTS1070) at 25 °C. Triplicate measurements were made for each sample and they were subsequently converted into zeta potential using Smoluchowski's formula to arrive at a zeta potential distribution curve.

2.6 bPEI–niclosamide complex and FA-bPEI@Fe₃O₄ nanoparticle release study

The crosslinked niclosamide and FA-bPEI@Fe₃O₄ nanoparticle incorporated nanofibers were incised into small pieces each weighing approximately 20 mg and incubated in 50 mL PBS for 94 hours at 37 °C. The niclosamide released from the composite nanofiber was monitored at periodic intervals by sampling 1 mL of release medium at each time point and measuring its absorbance value at 410 nm in a Cytation3 multimode reader (Biotek Instruments, Inc., USA). The standard plot of the bPEI–niclosamide complex was extrapolated to deduce the concentration of niclosamide released at each time point. The volume of release medium withdrawn at each time point for release study was subsequently replaced with equivalent PBS. A similar set of experiments was performed to estimate FA-bPEI@Fe₃O₄ nanoparticle release from the composite nanofiber scaffold. In this case, the release media sampled at each time point was carried over for AAS analysis in order to estimate amount of Fe released. All experiments were carried out in independent triplicates and niclosamide release was reported in terms of cumulative percentage drug release at each time point.

Cumulative percentage release = M_t/M_∞ × 100

M_t – mass of niclosamide/FA-bPEI@Fe₃O₄ nanoparticle released at time t, M_∞ – total mass of niclosamide/FA-bPEI@Fe₃O₄ nanoparticle loaded in 20 mg of composite nanofiber.

2.7 Cell culture study

KB cells (HeLa derivative) and NIH 3T3 (mouse embryonic fibroblast) cells were purchased from the National Centre for Cell Sciences (NCCS) Pune, India and cultured in DMEM medium supplemented with 10% (v/v) fetal bovine serum (FBS, Gibco Life Technologies (U.K.)) and 1% (v/v) penicillin–streptomycin (Sigma-Aldrich, USA). The cells were maintained in a CO₂ incubator with constant temperature of 37 °C and 5% CO₂.

2.8 FA-bPEI@Fe₃O₄ nanoparticle cell uptake study

KB cells were seeded over FA-bPEI@Fe₃O₄ nanoparticle and niclosamide-loaded nanofibers in a 12 well plate and incubated at 37 °C in CO₂ incubator. After a stipulated time span i.e. 15 h, 30 h, 45 h, and 75 h the cells in the respective wells were washed with PBS in order to remove extracellular nanoparticles and then the adherent cells were recovered by trypsinization and subsequently digested with 1 mL nitric acid (16 M). The resulting solution was diluted with deionized water, and the Fe concentration was determined by AAS to arrive at percentage cellular uptake of FA-bPEI@Fe₃O₄ nanoparticle.

2.9 Cellular ROS detection assay

Around 50 000 KB cells were seeded over FA-bPEI@Fe₃O₄ nanoparticle and niclosamide loaded nanofibers and bare bPEI–PEO nanofibers (control). At predetermined time points i.e. 24 hours, 48 hours, and 72 hours growth media was removed from respective wells and replenished with PBS containing 20 μM DCFH-DA followed by incubation at 37 °C for 30 min. The 2′,7′-dichlorfluorescein diacetate (DCFH-DA) dye is intracellularly metabolized into fluorescent dichlorofluorescein (DCF) in proportion to the intracellular concentration of reactive oxygen species and thus quantifies the extent of ROS generation by KB cell on predisposal to FA-bPEI@Fe₃O₄ nanoparticles. The treated cells were initially monitored under an EVOS cell imaging system (Life Technologies, USA) to monitor the extent of ROS generation. Furthermore, the cells were later harvested by trypsinization and their respective fluorescence was quantified by a Flowsight flow cytometer (Amnis). Around 10 000 cells fluorescence were acquired for each sample and represented as normalized frequency% versus intensity in channel 2 (Ch_02). The acquired data were further analyzed by IDEAS version 6.0 software to arrive at a relative shift in population. KB cells treated with H₂O₂ were taken as positive control in order to draw appropriate gating over sample cell populations.

2.10 Cell viability assay

The niclosamide and FA-bPEI@Fe₃O₄ nanoparticle loaded composite nanofiber anticancer potential was investigated against KB(FR+) cells and folate receptor negative L-132(FR−) cells by standard 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) based cell viability assay. Briefly, cells were seeded over different versions of nanofiber in a 96 well plate (5000 cells per well) and incubated at 37 °C with 5% CO₂. The experiment included bare bPEI–PEO nanofibers as a control, two different concentration of niclosamide alone loaded nanofibers, a FA-bPEI@Fe₃O₄ nanoparticle alone loaded nanofiber, and two different composite nanofibers. At 18 and 48 hours an external alternating magnetic field was applied on specific sets of composite nanofibers after which cells were transferred back to incubator for subsequent growth. After 48 and 96 hours the respective wells were supplemented with 10 μL of MTT (stock concentration – 500 μg mL⁻¹) and incubated at 37 °C. After 3–4 hours dark-blue formazan crystals that formed in each well were solubilized in DMSO and their corresponding absorbance values at 574 nm were obtained by a Cytation3 multimode reader (Biotek Instruments, Inc., USA). The absorbance values acquired were then normalized with respect to the control to give percentage cell viability.

2.11 Semi-quantitative RT-PCR analysis

KB and L132 cells were seeded over different versions of nanofibers as adapted in the cell viability assay in a 6 well plate. After 96 hours the cells were harvested from the wells and total RNA was extracted from KB and L132 by Tri reagent (Sigma-Aldrich, USA). The cDNA was generated by reverse transcription of 1 μg of total denatured RNA using M-MLV reverse transcriptase (Sigma, USA). The cDNA product was then used for gene specific amplification of apoptotic and oxidative stress genes. The forward and reverse primers utilized for PCR amplification are mentioned in Table S1.† Beta-actin (housekeeping gene) was adapted as an internal control. The PCR products were finally resolved in 1% agarose gel and visualized by ethidium bromide staining under UV light. The difference in gene expression was computed from the gene specific bands obtained by Image lab 4.0 software. The apoptotic genes considered for gene expression studies include bad, bax, p53, caspase-3, and C-myc; apart from these, anti-apoptotic genes bcl-2 and bcl-XL were also included in the study. The oxidative stress enzymes adapted in this work includes GAPDH, catalase, Mn-SOD, Cu/Zn SOD1, and HO 1.

2.12 Nuclear fragmentation analysis by Hoechst 33342

The cell undergoing apoptosis orchestrates its own destruction by instigating a cascade of signaling pathways which at latter stages actuates characteristic morphological changes like membrane blabbing, nuclear condensation, and DNA fragmentation. The events of nuclear condensation and fragmentation could be ascertained by fluorescent DNA intercalating dye Hoechst 33342. Apoptotic nuclei upon staining by Hoechst 33342 appear small, fragmented, and highly textured.

2.13 Monitoring autophagy

In order to examine incidence of autophagy, lysosomes and autophagic vacuoles were selectively labelled with fluorescent dyes lysotracker red DND-99 and monodansylcadaverine (MDC), respectively. In brief, treated cells were initially supplemented with 1 μM lysotracker red DND-99 and incubated at 37 °C for 30 min. After incubation, excess dye was removed using a brief PBS wash. The cells autophagic vacuoles were subsequently stained by incubating them in 50 μM MDC at 37 °C for 15 minutes. After labelling autophagic vacuoles and lysosomes, the cells were fixed with 2 wt% glutaraldehyde and mounted in a DPX for fluorescence microscopy.

2.14 Mitochondrial membrane potential analysis

The iron oxide nanoparticles generate reactive oxygen species (ROS) through the Fenton and Haber–Weiss reaction at their surface. Moreover, iron oxide nanoparticles mediated ROS generation is further augmented in a time and concentration dependent manner by applying an alternating magnetic field. ROS, generated in the process destabilizes the mitochondrial membrane potential of cells in the vicinity, which can be effectively traced by a cell permeant cationic fluorescent probe rhodamine 123 (a mitochondria specific dye). Rhodamine 123 is preferentially retained in mitochondria in an amount proportional to mitochondrial membrane potential (ΔΨ_m) which is further determined by the extent of ROS generation. After a specific time point of treatment, spent media was removed from each well and a mild PBS wash was made. The cells were subsequently incubated in PBS supplemented with rhodamine 123 (0.25 μg mL⁻¹) for 30 minutes at 37 °C. Preliminary qualitative analysis of cellular mitochondrial potential was performed by fluorescence microscopy in the presence of nuclear stain Hoechst 33342 (8 μM). Further, stained cells were harvested by mild trypsinization and carried over for quantitative analysis by flow cytometry.

3. Results and discussion

3.1 Characterization of FA-bPEI@Fe₃O₄ nanoparticles

A FE-SEM micrograph of pristine Fe₃O₄ NPs reveals octahedral crystallites of Fe₃O₄ nanoparticles with clearly distinguishable facets (Fig. 1(a)). The average edge length of Fe₃O₄ octahedron was found to be 143 ± 25 nm. Furthermore, FE-SEM analysis of FA-bPEI@Fe₃O₄ NPs illustrated uniform surface modification of Fe₃O₄ NPs by FA tagged bPEI moieties (Fig. 1(d)). Similarly, TEM analysis of FA-bPEI@Fe₃O₄ NPs also indicated blurred demarcation along the edges of Fe₃O₄ nanoparticles which depicts the presence of FA-bPEI polymer phase on its surface (Fig. 1(b) and (c)). The average size distribution of FA-bPEI@Fe₃O₄ nanoparticles was estimated to be 137 ± 45 nm. The crystalline magnetite nanoparticles gave raise to distinct diffraction spots in a SAED pattern corresponding to distinct facets of octahedral Fe₃O₄ NPs crystallites (Fig. 1(e)). Fourier transform analysis of lattice fringes obtained in HR-TEM images revealed a lattice spacing of 0.20 nm which corresponds to [400] cubic plane of magnetite. During Fe₃O₄ NPs synthesis individual crystal facets evolve at different growth rates which are critically determined by energy associated with it. Thus, preferential crystal growth along 〈001〉 direction rather than lower energy {111} governs the final octahedral morphology of resultant magnetite nanoparticle. Apart from this, at Fe₃O₄ NPs core, rate of Fe(OH)₃ consumption is equivalent to the rate of Fe(OH)₂ oxidation which leads to equal crystal growth rates along three axes directions resulting in octahedral morphology.¹⁹ The elemental analysis of a nanoparticle by TEM-Energy Dispersive X-ray spectroscopy (EDAX) indicated atomic composition of 34.49% Fe and 65.5% O which also confirms the presence of magnetite nanoparticles.

Fig. 1 (a) FE-SEM and (b) TEM image of as-synthesised Fe₃O₄ NPs; FA-bPEI@Fe₃O₄ NPs, (c) TEM and (d) FE-SEM image; (e) SAED pattern, and (f) HRTEM of Fe₃O₄ NPs.

Subsequent steps of Fe₃O₄ NPs surface modification were further confirmed at each step by monitoring change in zeta potential and hydrodynamic size of the nanoparticles (Fig. S1(a) and (b)†). The presence of molecular oxygen on the oxidized surface of iron oxide nanoparticles renders it a marginally negative zeta potential i.e., (−3.86 mV). Subsequent capping of Fe₃O₄ NPs with bPEI moieties laden with protonophoric amine groups shifted the surface zeta potential to a positive value of +8.72 mV. As a consequence of bPEI moieties capping on Fe₃O₄ NPs, a slight increase in size of nanoparticles was discerned by DLS measurements, i.e., from 98.82 nm to 125.7 nm. Further functionalization of bPEI capped Fe₃O₄ NPs with folic acid was accompanied with decline in zeta potential from +8.72 mV to −2.32 mV owing to acidic groups of folic acid and consumption of surface primary amine groups in folic acid functionalization during the EDC/NHS reaction.²⁰ Moreover, FA functionalized bPEI-Fe₃O₄ NPs hydrodynamic size was estimated to be 142.3 nm, the increase in size also confirms effective conjugation of FA to the bPEI amine groups present on the surface of Fe₃O₄NPs. The crystalline nature and structural properties of Fe₃O₄NPs was characterized by powder X-ray diffraction measurement. The acquired XRD pattern was further processed and analyzed by PANalytical X'Pert High Score Plus software. As shown in (Fig. S1(c)†), the XRD pattern of as-synthesized octahedral nanoparticles correspond to a face centered cubic lattice with inverse spinel crystal structure of Fe₃O₄NPs. The characteristic diffraction peaks of Fe₃O₄ NPs were observed at 2θ = 18.3°, 30.1°, 35.5°, 43.2°, 53.6°, 56.9°, 62.8°, and 73.9° which corresponded to (111), (220), (311), (400), (422), (511), (440), and (533) crystal planes, respectively (JCPDS 86-1345). The average size of the nanoparticle was also determined by Debye–Scherrer equation (D = Kλ/β cos θ, with K = 0.9) as 155 nm which was in close correlation with FE-SEM and TEM observation. XRD patterns of bPEI@Fe₃O₄ NPs and FA-bPEI@Fe₃O₄ NPs indicated peak suppression and broadening as compared to that of Fe₃O₄NPs due to the presence of amorphous polymer phase on its surface and subsequent increase in Fe₃O₄ NPs size upon FA-bPEI surface modification.²¹

The field and temperature dependent magnetization of bare Fe₃O₄ NPs and FA-bPEI@Fe₃O₄ nanoparticles was monitored with a vibrating sample magnetometer. The isothermal magnetization curve of FA-bPEI@Fe₃O₄ NPs at 5 K depicts a narrower hysteresis loop with lower remnant magnetization (B_R = 10 emu g⁻¹) and coercive field (H_C = 0.02 Oe) values as compared to what was observed at 300 K (i.e., B_R = 27 emu g⁻¹ and H_C = 0.23 Oe) (Fig. 2(a) and S2†). In the case of bare Fe₃O₄ NPs, only marginal shift in remnant magnetization was observed (i.e. from B_R = 11.5 emu g⁻¹ at 5 K to 4.9 emu g⁻¹ at 300 K), whereas coercive field values remained almost constant (i.e., H_C = 0.025 Oe at 5 K to 0.04 Oe at 300 K). At room temperature the coercive field of Fe₃O₄ NPs drastically increased by 0.19 Oe upon FA-bPEI functionalization due to the presence of magnetically passive polymeric phase (i.e., FA-bPEI) over Fe₃O₄ NPs.²²

Fig. 2 (a) VSM analysis of Fe₃O₄ NPs and FA-bPEI@Fe₃O₄ NPs at 5 K and 300 K, (b) field cooled (FC) and zero field cooled (ZFC) plot of Fe₃O₄ NPs and FA-bPEI@Fe₃O₄ NPs, and (c) FTIR spectroscopy analysis of FA, Fe₃O₄ NPs, bPEI@Fe₃O₄ NPs and FA-bPEI@Fe₃O₄ NPs.

Although temperature dependent progressive increase in ZFC and FC magnetization was observed, no maxima was attained in the experimental temperature range which further excluded deduction of blocking temperature (Fig. 2(b)). The ZFC magnetization curve of both pristine Fe₃O₄ NPs and FA-bPEI capped Fe₃O₄ NPs undergo transition at 55 K which arises due to glass transition or spin reorientation of Fe magnetic moments. A marginal reversal in magnetization curve was also observed after 110 K corresponding to Verwey transition temperature (T_v) of magnetite nanoparticles which is largely determined by the stoichiometry of Fe₃O₄ NPs.²³

The thermal susceptibility of FA-bPEI capped Fe₃O₄ NPs and niclosamide incorporated composite nanofibers was assessed by DTA-TGA analysis and results are represented as % weight loss with respect to temperature (Fig. 3(a) and (b)). The composite nanofibers fabricated herein exhibited multiple phases of weight loss owing to its multicomponent composition. In order to ascribe distinct weight loss phases to respective components in the composite nanofiber, pristine Fe₃O₄NPs, FA and FA-bPEI capped iron oxide nanoparticles also were included in thermal gravimetric analysis for comparison. As evident from the weight loss profile, bare Fe₃O₄ NPs did not undergo any significant weight loss up to 650 °C and instead exhibited a marginal increase in weight at around 210 °C which corresponds to progressive oxide film formation on the surface of Fe₃O₄ NPs at higher temperature.²⁴ Folic acid, being heat labile, begins to exhibit subsequent weight loss starting from 75 °C. The organic bPEI polymer phase begins to undergo thermal oxidative decomposition at 220 °C as an outcome of which significant weight loss in bPEI was observed starting from 220 °C.²⁵ The other polymer phase PEO also deteriorates in the same temperature regime as that of bPEI (i.e., 225 °C) which further augments the weight loss observed in the case of composite nanofiber.^25,26 Upon correlation of weight loss profiles of FA, bPEI, and Fe₃O₄ NPs the composition of FA-bPEI-Fe₃O₄ NPs was effectively deduced as 7.69 wt% Fe₃O₄ NPs core capped with 53.84 wt% bPEI polymer functionalized with 38.46 wt% of folic acid.

Fig. 3 (a) TG and (b) DT analysis of FA, Fe₃O₄ NPs, bPEI@Fe₃O₄ NPs and FA-bPEI@Fe₃O₄ NPs; (c) FE-SEM image of FA-bPEI@Fe₃O₄NPs and bPEI–niclosamide complex incorporated composite nanofiber with inset indicating elemental composition of composite nanofiber, and (d) elemental map of (C, O, N and Fe) in composite nanofiber by FE-SEM EDAX.

3.2 Characterization of composite nanofibers

The FA-bPEI capped Fe₃O₄ NPs and anticancer drug niclosamide were incorporated within bPEI–PEO nanofibers by the electrospinning technique. FE-SEM analysis of the composite nanofibers provided clear insight to uniform fiber morphology with fiber diameter distribution in the range of 655 ± 76 nm (Fig. 3(c)). The elemental composition analysis of composite nanofiber by EDAX revealed the presence of 2.13 wt% Fe which represents Fe₃O₄ NPs impregnated within nanofibers (Fig. 3(d)). Furthermore, elemental mapping of Fe within composite nanofiber indicated localization of Fe₃O₄ NPs at slightly bulgier sections of nanofibers as indicated in Fig. 3(d).

Presence of the hydrophobic drug niclosamide significantly contributed to larger contact angle values observed in case of composite nanofibers (Fig. S3(a) and (b)†).¹⁵ bPEI–PEO nanofibers incorporated with niclosamide alone were observed to have contact angle value of 76° which was subsequently diminished to 61° on inclusion of FA-bPEI-capped Fe₃O₄ NPs in the composite nanofiber. The increase in fraction of hydrophilic polymer bPEI in composite nanofibers upon inculcation of FA-bPEI capped Fe₃O₄ NPs accounts for the observed decline in water contact angle values (i.e. hydrophobicity).

The presence of magnetic FA-bPEI-capped Fe₃O₄ NPs within the composite nanofibers renders the scaffold susceptible to an external magnetic field. As evident from AFM images, the composite nanofibers as such exhibit uniform morphology with no surface irregularities whereas when they were subjected to 15 minutes of AMF the composite nanofiber integrity was lost followed by release of FA-bPEI capped Fe₃O₄ NPs and bPEI–niclosamide complexes in the vicinity (Fig. 4(a) and (b)). The loss in nanofiber integrity on predisposal to AMF is instigated by Fe₃O₄ NPs impregnated within nanofibers. The Fe₃O₄ NPs present within composite nanofibers tends to dislocate from its position with respect to magnitude and direction of external magnetic field.²⁷ The inherent momentum generated within magnetic Fe₃O₄ NPs disintegrates the composite nanofiber which subsequently leads to release of encapsulated drug niclosamide. Thus, an external alternating magnetic field can instigate nanofiber disintegration at any time point so as to alter nanoparticle and drug release profile according to cancer prognosis. Apart from an external magnetic field, polymer dissolution also greatly determines release of FA-bPEI-capped Fe₃O₄ NPs and bPEI–niclosamide complexes from composite nanofibers. The presence of hydrophilic polymer PEO and bPEI in the composite nanofibers inherently renders nanofibers the property of swelling and weight loss upon incubation in water. However, glutaraldehyde mediated crosslinking drastically minimizes polymer dissolution and increases the degree of swelling in a hydrophilic environment. Composite nanofibers crosslinked in the presence of glutaraldehyde vapor for 30 s attained 38% swelling whereas a 10 s crosslinked counterpart attained just 25% swelling indicating a clear correlation between degree of swelling and extent of crosslinking. Similarly, polymer dissolution rate i.e., nanofiber weight loss in hydrophilic environment was observed to decline from 19% for 10 s crosslinked nanofiber to 12% in case of 30 s glutaraldehyde crosslinked nanofiber. Thus, although glutaraldehyde mediated crosslinking renders stability to nanofiber in hydrophilic environment, it doesn't refrain solvent penetration within composite nanofiber which is a pre-requisite for controlled drug dissolution from nanofibers.²⁸

Fig. 4 AFM imaging of composite nanofiber morphology (a) before and (b) after exertion of AMF (130 Gauss field with 1 Hertz frequency) for 15 minutes in PBS. Time dependent release profile of (c) niclosamide and (d) FA-bPEI@Fe₃O₄ NPs from the composite nanofiber under three different conditions (i.e. no AMF, 15 minutes AMF at 16 hours and 15 minutes AMF at 48 hours).

3.3 Glutaraldehyde mediated crosslinking reaction

Completion of the glutaraldehyde-mediated bPEI crosslinking reaction was monitored by FTIR analysis and was observed to be same as reported in our previous work.¹⁵ The two aldehyde groups at the ends of the glutaraldehyde moiety generate Schiff's base by interacting with two amine groups of a bPEI molecule in the proximity (Fig. S4†).²⁹ Upon glutaraldehyde treatment, strong decline in absorbance for free –NH₂ (stretching vibration) in bPEI was observed as compared to untreated counterpart. This difference in free NH₂ absorption band arises due to decline in free amine groups of bPEI moieties as a certain fraction of them are consumed in amide bond formation during the glutaraldehyde-mediated crosslinking reaction. This crosslinking reaction renders the fiber better stability in a hydrophilic environment which can be further fine-tuned to our requirement by altering the time of crosslinking and concentration of crosslinking agent.

3.4 Surface functionalization of Fe₃O₄ NPs

The targeting moiety FA conjugation to bPEI by EDC/NHS chemistry was assessed by FTIR spectroscopy. Branched PEI (25 kDa) adapted in the present work is laden with umpteen numbers of primary (25%) and secondary amine (50%) groups which make them an appropriate candidate for efficient FA functionalization.³⁰ Although primary amines are more susceptible to EDC/NHS reaction, secondary amines also participate to certain extent.³¹ In order to efficiently interpret synthesis of FA conjugated bPEI@Fe₃O₄ NPs by EDC/NHS chemistry, FTIR spectra of bare Fe₃O₄ NPs, FA and bPEI@FA was also acquired apart from FA-bPEI@Fe₃O₄ NPs (Fig. 2(c)). Bare Fe₃O₄ NPs exhibits a strong absorption peak at 580 cm⁻¹ which corresponds to Fe–O–Fe bond vibration at Fe₃O₄ NPs core.³² In addition to this, Fe₃O₄ NPs also yields a broad and strong absorption peak at 3413 cm⁻¹ which arises due to the presence of –OH and/or H₂O on the surface of Fe₃O₄ NPs. The surface localized –OH groups also undergoes stretching and bending vibration under the influence of IR radiation which gives rise to broad absorption peaks at 1637 cm⁻¹ and 1108 cm⁻¹, respectively. As shown in Fig. 2(c), although FTIR spectra of FA-bPEI@Fe₃O₄ NPs included Fe₃O₄ NPs associated peaks (580 cm⁻¹ and 3413 cm⁻¹), marginal peak shift and damping was observed in case of Fe–O–Fe absorption band (i.e. from 584 cm⁻¹ to 570 cm⁻¹) as a result of surface modification.³³ Similar changes were observed in case of peaks ascribed to surface localized –OH groups present in Fe₃O₄ NPs. Similarly, FA associated peaks (i.e., 1411 cm⁻¹ (amide III band), 1270 cm⁻¹ (aromatic amine C–N bond stretching vibration)) were also present in FA-bPEI@Fe₃O₄ NPs IR spectra indicating intact FA functional groups which critically determine subsequent FA docking onto folate receptors. Apart from such Fe₃O₄ NPs, FA and bPEI associated peaks, a new peak also appeared at 1574 cm⁻¹ (in case of FA-bPEI@Fe₃O₄ NPs) which corresponds to NH₂ deformation observed in primary alkyl amide.³⁴ Thus, appearance of a new amide peak, along with concerted absence of FA associated C=O peak (i.e. –COOH (1691 cm⁻¹)) in IR spectra of FA-bPEI@Fe₃O₄ NPs, clearly illustrates an amide bond formation between the –COOH group of FA and NH₃ and the NH₂ group of bPEI.

3.5 Fe₃O₄ NPs and bPEI–niclosamide complexes release study

The release profiles of FA functionalized Fe₃O₄ NPs and bPEI–niclosamide complexes from composite nanofibers were monitored in a time dependent manner by AAS and UV-Vis absorbance studies. The FA-bPEI@Fe₃O₄ NPs and bPEI–niclosamide complex release study was carried out under three different conditions: i.e., without AMF throughout, with AMF for 10 minutes at 16 hours, and another batch with AMF for 10 min at 48 hours and the results were represented as % niclosamide release and % FA-bPEI capped Fe₃O₄ NPs release with respect to time (Fig. 4(c) and (d)). As clearly evident from the release profiles, when composite nanofibers were not subjected to AMF they exhibited controlled and sustained release of both bPEI–niclosamide complexes and FA-bPEI capped Fe₃O₄ NPs. In contrast, when these composite nanofibers were subjected to AMF (130 Gauss field with frequency of 1 Hertz) for 15 minutes at 16 hours, a sudden surge in the release profile of niclosamide and Fe₃O₄ NPs was observed. In response to applied AMF (at 16 hours), a certain fraction of FA-bPEI@Fe₃O₄ NPs incorporated within composite nanofibers began to extrude from the polymer nanofiber matrix and, as a result of their migration, the composite nanofiber began to disrupt. The loss in nanofiber integrity further augments polymer dissolution and solvent penetration in the composite nanofiber as an outcome of which a concerted spike in the release profile of niclosamide and FA-bPEI-capped Fe₃O₄ NPs was observed. At the end of 30 hours, around 30% higher Fe₃O₄ NPs release was observed in the case of composite nanofibers treated with AMF for 15 minutes (at 16 hours) as compared to that of untreated composite nanofibers. In concert with this, a 20% increase in bPEI–niclosamide complex release profile also was attained between these two cases at 30 hours. Although a drastic difference in release profile was observed between untreated and AMF treated (at 16 hours) composite nanofibers, their release profiles were observed to follow a similar trend after 30 hours. Similarly, when composite nanofibers were subjected to AMF at 48 hours, the Fe₃O₄ NPs and bPEI–niclosamide complexes release profile subsequently begins to deviate from the release profile attained in case of untreated composite nanofibers. The composite nanofibers subjected to AMF (at 16 hours and 48 hours) were disintegrated to higher extent and thus attained 87% bPEI–niclosamide complex release and around 90% Fe₃O₄ NPs release at 96 hours which is significantly higher than the values attained in case of untreated composite nanofibers, which was 66% bPEI–niclosamide complex release and 72% Fe₃O₄ NPs release at 96 hours. It was observed that, as compared to AMF treated composite nanofibers, untreated nanofibers attained lower Fe₃O₄ NPs and bPEI–niclosamide complex release throughout the release study.

3.6 Cellular uptake of FA-bPEI capped magnetite nanoparticles

The targeted cell specific uptake of FA functionalized Fe₃O₄ NPs by KB(FR+) was monitored by FE-SEM and estimated quantitatively by elemental analysis (AAS). The events of composite nanofibers disintegration under the influence of AMF, followed by release of FA-bPEI capped Fe₃O₄ NPs and its subsequent localization over KB cells, was effectively captured by FE-SEM. FE-SEM EDX elemental mapping at the region of interest provided concentration dependent intensity profiles of C(Red), O(green), and Fe(blue) as the major elements (Fig. 5(a) and (b)). Co-localization of the Fe elemental map onto the particles attached over KB cells confirmed Fe₃O₄ NPs localization and cellular uptake. Furthermore, in the presence of Fe₃O₄ NPs KB cells undergo morphological transformations characteristic of apoptosis which was also captured by FE-SEM (Fig. S5(a) and (b)†). The fraction of Fe₃O₄ NPs internalized and localized on cellular membrane were further estimated by AAS studies after cell lysis at different time points (Fig. 6(a) and (b)). The cellular uptake of Fe₃O₄ NPs was then normalized with respect to cell number and represented as % cellular uptake at different time points under three different conditions (i.e. without AMF, AMF applied after 16 hours, and AMF applied after 48 hours) for KB cells and L132 cells. The expression of folate receptor on KB cell's membrane facilitated distinctly higher cellular uptake of FA-bPEI functionalized Fe₃O₄ NPs as compared to L132 cells at every time point of the study which is, for instance, 53% in the case of KB cells and 27% in the case of L132 cells at 75 hours. A similar trend was observed even when AMF was applied at 16 and 48 hours which clearly indicates folate receptor mediated cellular uptake of Fe₃O₄ NPs by KB cells which is absent in case of L132 cells. Although L132 cells lack folate receptors, they exhibit non-specific Fe₃O₄ NPs uptake to a relatively small extent as compared to KB cells.

Fig. 5 (a) FE-SEM image of composite nanofiber treated KB cells depicting localization of FA-bPEI@Fe₃O₄ NPs on its surface; (b) FE-SEM EDAX frequency map of C, O, and Fe elements at FA-bPEI@Fe₃O₄ NPs localized position.

Fig. 6 Released iron oxide nanoparticle cellular uptake in (a) L132 and (b) KB cells; (c) L132, and (d) KB cell viability (MTT) assay for bare nanofiber (bare NF), bPEI–niclosamide alone loaded nanofiber (Niclo NF), Fe₃O₄ NPs alone loaded nanofiber (Fe₃O₄ NF) and composite nanofiber (C NF) with AMF applied at 48 hours (C NF(AMF@48)) and 16 hours (C NF(AMF@16)). Statistical significant values are denoted by *p < 0.05, **p < 0.005, and ***p < 0.001.

In the case of composite nanofibers subjected to AMF at 16 hours, cellular uptake of Fe₃O₄ NPs was observed to be higher than other two cases up to 45 hours whereas, at later stages (i.e. around 75 hours), it was equalized by composite nanofibers subjected to AMF at 48 hours. Thus, cellular uptake of Fe₃O₄ NPs followed a similar trend as observed in its release profile in all three cases with respect to time.

3.7 Cell viability assay

The antiproliferative efficacy of Fe₃O₄ NPs and niclosamide-loaded composite nanofibers was investigated against L132 and KB cells at different time points in presence and absence of AMF and the results were represented as % cell viability (Fig. 6(c) and (d)). As evident from cell viability results, composite nanofibers successfully fostered synergistic antiproliferative potential of Fe₃O₄ NPs and niclosamide against both cells (L132 and KB cells). The increased expression of folate receptors on KB cells rendered them as susceptible targets for FA functionalized Fe₃O₄ NPs constantly released from composite nanofiber.³⁵ In concert with this notion, cellular uptake studies also pointed out the fact that folate functionalized Fe₃O₄ NPs were preferentially taken up by KB cells (FR+) as compared to L132 cells. As an outcome of such receptor facilitated uptake of Fe₃O₄ NPs by KB cells, a drastic decline in cell viability of KB cells was observed as compared to L132 cells (FR−) at both 48 and 96 hours. In the case of niclosamide alone loaded nanofibers no significant difference in cell viability was observed as a result of external AMF (48 hours (71.5% and 68.9%) and 96 hours (62% and 59%)) whereas, in the case of in case of Fe₃O₄ NPs and niclosamide loaded nanofibers, a substantial decline in cell viability was witnessed under the influence of external AMF (i.e. 55–38% at 48 hours and 41–26% at 96 hours). The decline in cell viability solely manifested by AMF clearly illustrates AMF stimulated a burst release of Fe₃O₄ NPs from composite nanofibers which subsequently localized on KB cells and augmented cell death further by apoptosis. A similar trend in cell viability was observed in L132 cells too although in this case it was exclusively due to concentration dependent nonspecific basal uptake of Fe₃O₄ NPs by L132 cells. Apart from this, nanofibers loaded with different wt% niclosamide exhibited concentration dependent declines in cell viability in cases of both L132 and KB cells at 48 and 96 hours. In cases of both L132 and KB cells, niclosamide alone loaded nanofibers 48 and 96 hours MTT values remained consistent in the presence or absence of AMF. At the end of 48 hours no viability differences were observed in the case of composite nanofiber (C NF) (AMF@48) (i.e., 53% (no AMF) and 51% (AMF at 48 hours) with of KB cells) whereas, with C NF (AMF@16) a drastic difference in viability was attained (i.e., 52–31% with KB cells) owing to AMF-triggered Fe₃O₄ NPs release at an earlier stage (i.e., AMF at 16 hours). In the case of C NF (AMF@48), although at 48 hours cell viability remained constant irrespective of AMF, the difference in cell viability widens greatly by the end of 96 hours (i.e., 42–25% = 17%) as a result of AMF triggered release of Fe₃O₄ NPs after 48 hours. Thus, cell viability assay clearly ascertains AMF stimulated release of Fe₃O₄ NPs and niclosamide from composite nanofibers which subsequently mediates cell death apoptosis.

3.8 Induction of apoptosis and intracellular changes

Among other pathways, Fe₃O₄ NPs predominantly executes apoptosis by instigating mitochondrial membrane permeabilization and cytochrome c release which further promotes caspase activation upon cleavage of specific caspase substrates.³⁶ The activated caspase leads to production of ROS which inflicts mitochondrial damage by lipid peroxidation, disruption of electron transport chain, loss of mitochondrial transmembrane potential (ΔΨ_m), and loss in mitochondria structural integrity.³⁷ In the present study, activation of caspase and other downstream apoptotic genes was monitored by semi-quantitative RT-PCR and subsequent events including loss in mitochondrial membrane potential and ROS generation were analyzed by fluorescent dyes, rhodamine 123, and DCFH-DA (dichloro-dihydro-fluorescein diacetate), respectively. Rhodamine 123 is a lipophilic cationic dye which is accumulated within mitochondria in proportion to its membrane potential (ΔΨ_m).³⁸ The loss in mitochondrial membrane potential leads to proportional decline in Rho 123 fluorescence intensity as it permeates from mitochondria. Upon intracellular localization of DCFH-DA, it is actively converted to the polar derivative DCFH by cellular esterase. The resultant product, DCFH, is then actively oxidized to highly fluorescent DCF by intracellular ROS and peroxidase enzymes.

3.9 Mitochondrial dysfunction analysis by rhodamine 123

The mitochondrial membrane potential (ΔΨ_m) of KB cells seeded over composite nanofibers at the end of 72 hours was monitored by rhodamine 123 fluorescence on a quantitative basis by flow cytometry analysis (Fig. 7). The untreated KB cells labelled with Rho 123 were used to gate healthy cells population with respect to which fluorescence of treated KB cells population was further classified (Fig. 7(a)). In the case of KB cells seeded over composite nanofibers, a new subset of cell population (18.8%) with lower mean fluorescence (Rho 123) appeared in the histogram which clearly depicts the loss in mitochondrial membrane potential of KB cells on predisposal to the anticancer drug niclosamide and Fe₃O₄ NPs (Fig. 7(b)). At 76 hours, percentage cell population with low fluorescence increased further to 22.7% with composite nanofibers treated with AMF at 48 hours and 29.3% with composite nanofibers treated with AMF at 16 hours (Fig. 7(c) and (d)). At 76 hours of a release study, composite nanofibers subjected to AMF at 16 and 48 hours were observed to have attained same extent of niclosamide release whereas with Fe₃O₄ NPs a significant lag in release profile was observed between them which was 64.85% in the case of composite nanofibers treated with AMF at 48 hours and 84.12% for composite nanofibers treated with AMF at 16 hours (Fig. 4(c) and (d)). This significant lag in Fe₃O₄ NPs release profile in AMF-treated composite nanofibers at 76 hours lead to a corresponding difference in low fluorescence subset population which is 22.7% and 29.3% for composite nanofibers treated with AMF at 48 hours and 18 hours, respectively. Thus, at 76 hours, higher % Fe₃O₄ NPs release (in 18 hour AMF-treated composite nanofiber) clearly correlated with increased population of low fluorescence cells signifying the role of Fe₃O₄ NPs in induction of apoptosis by mitochondrial membrane destabilization.

Fig. 7 Flow cytometry analysis of mitochondrial membrane potential (ΔΨ_m) in (a) control KB cells, (b) KB cells treated with C NF without AMF, (c) KB cells treated with CNF with AMF at 48 hours, and (d) KB cells treated with AMF at 16 hours by rhodamine 123 based fluorescence assay.

Apart from flow cytometry analysis, morphology of control and treated KB cells stained with Rho 123 and Hoechst 33342 also was examined by fluorescence microscope. The control cells were observed to be intact with Rho 123 labelled highly fluorescent mitochondria scattered throughout the cytoplasm. The bright fluorescence of Rho 123 localized within the mitochondria of control KB cells clearly indicates stable negative mitochondrial membrane potential (ΔΨ_m). As a result of mitochondrial membrane permeabilization in Fe₃O₄ NPs and niclosamide treated KB cells, a sharp decline in mitochondrial membrane potential (ΔΨ_m) was observed which led to subsequent decline in Rho 123 fluorescence.

3.10 ROS estimation by DCFH-DA

The inherent property of iron oxide nanoparticles to generate extensive free radicals by Fenton and Haber–Weiss chemistry arises due to spin trapping EPR on its surface and dissolution of free Fe ions.³⁹ The FA functionalized Fe₃O₄ NPs released from the composite nanofibers catalytically generate ROS at its surface and thereby exerts oxidative stress on KB cells in the vicinity. In addition to this, the rate of ROS generation by Fe₃O₄ NPs is increased drastically in the presence of AMF.⁴⁰ Thus, in our present work, apart from inherent potential Fe₃O₄ NPs to generate ROS, exertion of AMF at 18 and 48 hours further augments its ROS generating potentials. As an outcome of this, KB cells seeded over composite nanofibers are subjected to extensive ROS in the proximity which induces oxidative stress in them. The induction of oxidative stress in KB cells instigates mitochondrial dysfunction and DNA fragmentation which ultimately culminates in apoptosis. In order to assess oxidative stress mediated apoptosis in KB cells, the extent of ROS generated was quantified at different time points by DCFH-DA assay. The KB cell population fluorescence was gated on the basis of untreated negative control and H₂O₂ treated positive control (Fig. 8(a) and (b)). With increase in treatment time, percentage of cells producing ROS subsequently increased as more and more Fe₃O₄ NPs are released from the composite nanofibers at latter time point. In the case of composite nanofibers subjected to AMF at 16 hours ROS producing cells increased quickly to 23.8% at the end of 20 hours whereas with untreated composite nanofibers it is merely 6.43% (Fig. 8(b) and (c)). Under the influence of external AMF, the composite nanofibers undergo progressive disintegration and consequently foster a clear surge in Fe₃O₄ NPs and niclosamide release profile. Thus, higher concentration of free Fe₃O₄ NPs and niclosamide after AMF treatment lead to 17.37% (23.8–6.43%) increase in ROS as compared to that of untreated nanofibers. Similarly, in the case of fibers treated with AMF at 48 hours, cellular ROS increased abruptly to 42.2% at 72 hours from a nominal value of 15.9% at 20 hours (Fig. 7(d) and (g)). Composite nanofibers treated with AMF at 16 hours generated maximum ROS (i.e. 23.8%) as compared to other two cases (i.e. 6.43% and 15.9%) by the end of 20 hours whereas by the end of 72 hours composite nanofibers subjected to AMF at 48 hours (42.2%) surpassed the other two cases (i.e. 28.3% and 39.8%) in extent of ROS generation (Fig. 8(e) and (f)). The observed trend in ROS generation with respect to time corroborates well with Fe₃O₄ NPs and niclosamide release profiles observed in all three cases. Thus, the extent of Fe₃O₄ NPs and niclosamide release and subsequent ROS generation can be easily instigated by external AMF so as to attain desirable therapeutic outcomes. On extended exertion of AMF, Fe₃O₄ NPs tends to get aggregated which leads to decline in surface area per mass which thereby limits the surface reactive sites required for ROS generation.³⁹ The strong dipole–dipole interaction induced between adjacent Fe₃O₄ NPs by AMF also augments aggregation of free Fe₃O₄ NPs.

Fig. 8 Flow cytometry analysis of ROS induction in (a) control KB cells, (b) & (e) KB cells treated with C NF without AMF, (c) & (f) KB cells treated with CNF with AMF at 48 hours, and (d) & (g) KB cells treated with AMF at 16 hours by DCFH-DA based fluorescence assay.

3.11 Nuclear fragmentation analysis by Hoechst 33342

KB cells predisposed to increasing concentration of Fe₃O₄ NPs and niclosamide subsequently undergoes apoptosis in a time dependent manner. The events of nuclear condensation and DNA fragmentation manifested as an outcome of apoptosis were distinctly perceived in cells as small, textured, and irregular Hoechst 33342 stained nucleus.⁴¹ At 72 hours of treatment, control KB cells retained their nucleus intact, whereas in the case of KB cells seeded over C NF, C NF (AMF@16) and C NF (AMF@48) condensed and fragmented nucleus were prominently observed (Fig. S6(a)–(d)†). Thus, apoptosis predominantly contributes to KB cell viability decline in case of C NF, C NF (AMF@16), and C NF (AMF@48) although to a different extent in each case.

3.12 Interpretation of autophagy by lysotracker red and MDC staining

Substantial preexisting literature provides comprehensive insight to diverse autophagy signaling pathways instigated by niclosamide and Fe₃O₄ NPs independently. Thus, in the present work, autophagy was merely ascertained by lysotracker red and MDC staining. During the initial stages of autophagy, a portion of cytoplasm, including organelles, is enclosed by a phagophore or isolation membrane to form an autophagosome. The outer membrane of the autophagosome subsequently fuses with the lysosome to form autophagolysosome where all the internalized material undergoes enzymatic degradation. The fluorescent dyes MDC and lysotracker red selectively labels autophagosomes and lysosomes, respectively. Upon fusion of autophagosome with lysosomes both lysotracker red and MDC colocalize indicating formation of autophagolysosome.⁴² In the case of KB cells seeded over control NF, lysotracker red labelled lysosomes were present predominantly along with a few MDC positive autophagic vacuoles which corresponds to basal autophagic flux in KB cells (Fig. S6(e)†). In the case of KB cells seeded over C NF, C NF (AMF@16), and C NF (AMF@48), MDC positive structures increased drastically and colocalized with lysotracker positive structures which indicates formation of autophagolysosome a hallmark event of autophagy (Fig. S6(f)–(h)†). Owing to progressive accumulation of autophagolysosomes, Fe₃O₄ NPs mediated pro-death role of autophagy takes over its pro-survival role and gives rise to further decline in KB cell viability with C NF, C NF (AMF@16), and C NF (AMF@48).

3.13 Gene expression studies

As evident from this preliminary study, Fe₃O₄ NPs mediates cell death by apoptosis, oxidative stress, and pro-death autophagy whereas niclosamide prominently effectuates cell death by apoptosis alone. Apoptosis and oxidative stress are initially manifested by altered expression of various genes involved in the respective pathways. Such up-regulation or down-regulation of genes can be effectively estimated by semi-quantitative reverse transcription polymerase chain reaction (RT-PCR). Gene specific amplification of apoptotic genes (c-myc, p53, bax, bad and caspase 3), anti-apoptotic genes (bcl-2 and bcl-xL) and oxidative stress response genes (GADPH, catalase, Mn SOD, Cu/Zn SOD1, HO1, NADPH and GPx) was carried out in KB cells subjected to composite nanofibers for 72 hours. The expression profiles of all genes were normalized with respect to β-actin as endogenous control.

Concerted increase in expression of apoptotic genes was observed in case of KB cells treated with composite nanofibers as compared to that of control nanofibers (Fig. 9(a)). Although nanofibers loaded with niclosamide alone and Fe₃O₄ NPs alone also could manifest significant up-regulation of apoptotic genes, they were not as prominent as those attained with composite nanofibers. Such differences in gene expression profiles clearly points out to the fact that in the case of KB cells subjected to composite nanofibers, apoptosis was mediated by both Fe₃O₄ NPs and niclosamide. As depicted in Fig. 9(i), with KB cells subjected to composite nanofibers, pro-apoptotic genes specifically caspase 3, p53, and c-myc attained a drastic surge in their expression profile as compared to the others. Activation of p53 subsequently inducts a cascade of signaling moieties involved in apoptosis and one such immediate effector gene, which is up-regulated upon activation of p53, is bax(bcl-2-associated X protein). The overexpressed bax protein localizes itself onto mitochondrial membrane and triggers mitochondrial outer membrane permeabilization (MOMP) by orchestrating oligomeric pores in the mitochondrial membrane.⁴³ The loss in mitochondrial membrane integrity leads to an efflux of cytochrome c into the cytosol. The free cytosolic cytochrome c consequently activates the caspase family (caspase 3 (cysteine–aspartic acid proteases)) of genes which causes nuclear DNA fragmentation and cell death.

Fig. 9 Semi-quantitative RT-PCR analysis of (a) apoptosis and (b) ROS dependent genes expression. Statistical significant values are denoted by *p < 0.05, **p < 0.005, and ***p < 0.001.

The folate functionalized Fe₃O₄ NPs (FA-bPEI@Fe₃O₄ NPs) released from composite nanofibers sequesters itself on KB(FR+) cells and catalytically generates free radicals (ROS) which induces oxidative stress in cells. Cellular response to ROS is instigated at the transcriptional level by activating MAP protein kinases and other transcriptional activators like AP-1, ATF, and NF-κB. These signaling moieties mount a primary defense against oxidative stress at later stages by regulating intracellular antioxidant enzymes like glutathione peroxidase, SOD, and catalase. Although the cellular antioxidant system circumvents ROS mediated cytotoxicity at lower concentrations, their catalytic activity ceases at higher ROS concentrations. KB cells treated composite nanofiber loaded with Fe₃O₄ NPs and niclosamide manifested drastic down-regulation of catalase, MnSOD, Cu/Zn SOD1, NADPH, and GPx genes apparently due to excessive ROS production (Fig. 9(b)). Such perturbation in antioxidant enzymes renders cells even more susceptible to oxidative damage by free radicals.⁴⁴ Niclosamide loaded nanofiber could attain only marginal decline in expression of Mn SOD and NADPH and thus barely plays any role in ROS generation whereas in the case of Fe₃O₄ NPs loaded nanofibers substantial declines in antioxidant genes expression were observed. Such a difference clearly points to the fact that perturbation of antioxidant gene expression with composite nanofibers is predominantly contributed by Fe₃O₄ NPs.

4. Conclusion

In the present work, FA functionalized monodisperse octagonal Fe₃O₄ NPs were synthesized initially by an oxidation–precipitation method and then co-incorporated along with bPEI–niclosamide complexes within PEO–bPEI nanofibers by blend electrospinning. Time dependent release of FA-bPEI@Fe₃O₄ NPs and bPEI–niclosamide from composite nanofibers was studied for 96 hours in the absence and presence of AMF. When triggered by external AMF (at 16 hours and 48 hours), the composite nanofibers demonstrated a sudden surge in FA-bPEI@Fe₃O₄ NPs and bPEI–niclosamide release profiles which renders them versatility to meet the prerequisite of cancer prognosis. Furthermore, the targeted anticancer potential of composite nanofibers loaded with FA-bPEI@Fe₃O₄ NPs and bPEI–niclosamide complexes was consequently investigated against KB cells and L132 cells on the basis of both qualitative and quantitative assays in a time dependent manner. The Fe₃O₄ NPs cellular uptake studies confirmed folate targeted delivery of FA-bPEI@Fe₃O₄ NPs in KB cells correlates well with lower cell viability attained in case of KB cells (at 48 and 96 hours). The niclosamide and Fe₃O₄ NPs released from the composite nanofibers concertedly effectuated apoptosis (MTT assay, Hoechst 33342 nuclear staining, FE-SEM observation, and gene expression studies), autophagy (MDC and lysotracker red staining) and ROS (DCFH-DA and rhodamine 123 based fluorescence assay and gene expression studies) mediated cell death in KB cells. Thus, the composite nanofiber fabricated in this work could simultaneously manifest AMF triggered targeted delivery of Fe₃O₄ NPs and bPEI–niclosamide complexes which makes them a versatile option for efficient cancer management.

Acknowledgements

This work was supported by the Science and Engineering Research Board (no. SR/FT/LS-57/2012), and Department of Biotechnology (no. BT/PR6804/GBD/27/486/2012), Government of India. S.U.K. is thankful to the Ministry of Human Resource Development, Government of India, for the fellowship. Sincere thanks to Department of Chemistry and Institute Instrumentation Centre, IIT Roorkee, for the various analytical facilities provided.

References

1. S. Mura, J. Nicolas and P. Couvreur, Nat. Mater., 2013, 12, 991.
2. S. Ganta, H. Devalapally, A. Shahiwala and M. Amiji, J. Controlled Release, 2008, 126(3), 187.
3. O. Onaca, R. Enea, D. W. Hughes and W. Meier, Macromol. Biosci., 2009, 9(2), 129.
4. Q. Yuan, R. Venkatasubramanian, S. Hein and R. D. Misra, Acta Biomater., 2008, 4(4), 1024.
5. J. Pearce, A. Giustini, R. Stigliano and P. Jack Hoopes, J. Nanotechnol. Eng. Med., 2013, 4(1), 0110071.
6. R. J. Wydra, C. E. Oliver, K. W. Anderson, T. D. Dziubla and J. Z. Hilt, RSC Adv., 2015, 5(24), 18888.
7. S. Acharya and S. K. Sahoo, Adv. Drug Delivery Rev., 2011, 63(3), 170–183.
8. M. Wu, D. Zhang, Y. Zeng, L. Wu, X. Liu and J. Liu, Nanotechnology, 2015, 26(11), 115102.
9. S. F. Medeirosa, A. M. Santosa, H. Fessib and A. Elaissarib, Int. J. Pharm., 2011, 403(1), 139.
10. Y. Zhu, Y. Fang and S. Kaskel, J. Phys. Chem. C, 2010, 114, 16382.
11. P. S. Low, W. A. Henne and D. D. Doorneweerd, Acc. Chem. Res., 2008, 41, 120.
12. X. Pan and R. J. Lee, Expert Opin. Drug Delivery, 2004, 1, 7.
13. J. Sudimack and R. J. Lee, Adv. Drug Delivery Rev., 2000, 41, 147.
14. Y. Li, P. K. Li, M. J. Roberts, R. C. Arend, R. S. Samant and D. J. Buchsbaum, Cancer Lett., 2014, 349(1), 8.
15. S. U. Kumar and P. Gopinath, Colloids Surf., B, 2015, 131, 170.
16. B. Bhushan, P. Dubey, S. U. Kumar, A. Sachdev, I. Matai and P. Gopinath, RSC Adv., 2015, 5, 12078.
17. U. K. Sukumar, B. Bhushan, P. Dubey, I. Matai, A. Sachdev and G. Packirisamy, Int. Nano Lett., 2013, 3, 45.
18. Y. W. Guang, Z. T. Lai, Q. X. Jing, Z. J. Guo and Y. Li, Mater. Sci. Eng., B, 2007, 136, 101.
19. A. Roldan, R. G. Crespo and N. H. de Leeuw, Phys. Chem. Chem. Phys., 2014, 16, 21082.
20. M. Khalkhali, K. Rostamizadeh, S. Sadighian, F. Khoeini, M. Naghibi and M. Hamidi, Daru, J. Pharm. Sci., 2015, 23(1), 45.
21. L. Li, Y. Yang, J. Ding and J. Xue, Chem. Mater., 2010, 22, 3183.
22. F. Huang, Z. Wang, X. Lu, J. Zhang, K. Min, W. Lin, R. Ti, T. Xu, J. He, C. Yue and J. Zhu, Sci. Rep., 2013, 3, 2907.
23. A. Mukhopadhyay, N. Joshi, K. Chattopadhyay and G. De, ACS Appl. Mater. Interfaces, 2012, 4(1), 142.
24. S. U. Kumar, I. Matai, P. Dubey, B. Bhushan, A. Sachdev and P. Gopinath, RSC Adv., 2014, 4, 38263.
25. P. S. Claire, Macromolecules, 2009, 42, 3469.
26. N. Lee, D. Yoo, D. Ling, M. H. Cho, T. Hyeon and J. Cheon, Chem. Rev., 2015, 115(19), 10637.
27. U. K. Sukumar and G. Packirisamy, ACS Appl. Mater. Interfaces, 2015, 7(33), 18717.
28. A. Jayakrishnan and S. R. Jameela, Biomaterials, 1996, 17(5), 471.
29. Y. Park, E. Kang, O. J. Kwon, T. Hwang, H. Park, J. M. Lee, J. H. Kim and C. O. Yun, J. Controlled Release, 2010, 148(1), 75.
30. X. Qiang, T. Wu, J. Fan, J. Wang, F. Song, S. Sun, J. Jianga and X. Peng, J. Mater. Chem., 2012, 22, 16078.
31. H. Pua, F. Jianga and Z. Yanga, Mater. Chem. Phys., 2006, 100(1), 10.
32. R. A. Binia, R. F. C. Marquesb, F. J. Santosa, J. A. Chakerc and M. J. Jafelicci, J. Magn. Magn. Mater., 2012, 324(4), 534.
33. C. Sun, R. Sze and M. Zhang, J. Biomed. Mater. Res., Part A, 2006, 78(3), 550.
34. R. J. Lee and P. S. Low, J. Biol. Chem., 1994, 269(5), 3198.
35. M. T. Zhu, Y. Wang, W. Y. Feng, B. Wang, M. Wang, H. Ouyang and Z. F. Chai, J. Nanosci. Nanotechnol., 2010, 10(12), 8584.
36. I. Marzo, S. A. Susin, P. X. Petit, L. Ravagnan, C. Brenner, N. Larochette, N. Zamzami and G. Kroemer, FEBS Lett., 1998, 427(2), 198.
37. A. Baracca, G. Sgarbi, G. Solaini and G. Lenaz, Biochim. Biophys. Acta, 2003, 1606(1), 137.
38. S. Klein, A. Sommer, L. V. Distel, W. Neuhuber and C. Kryschi, Biochem. Biophys. Res. Commun., 2012, 425(2), 393.
39. R. J. Wydra, C. E. Oliver, K. W. Anderson, T. D. Dziublaa and J. Z. Hilt, RSC Adv., 2015, 5, 18888.
40. I. Hilger, E. Dietmar, W. Linss, S. Streck and W. A. Kaiser, J. Phys.: Condens. Matter, 2006, 18, S2951–S2958.
41. E. C. Filippi-Chiela, M. M. Oliveira, B. Jurkovski, S. M. Callegari-Jacques, V. D. da Silva and G. Lenz, PLoS One, 2012, 7(8), 42522.
42. Y. Tang, Y. Chen, H. Jiang and D. Nie, Cell Death Differ., 2011, 18(4), 602.
43. P. Gopinath, S. K. Gogoi, P. Sanpui, A. Paul, A. Chattopadhyay and S. S. Ghosh, Colloids Surf., B, 2010, 77(2), 240.
44. B. Bhushan and P. Gopinath, J. Mater. Chem. B, 2015, 3, 4843.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra05006a

---