Plasmonic/Magnetic Multifunctional nanoplatform for Cancer Theranostics

A multifunctional magneto-plasmonic CoFe2O4@Au core-shell nanoparticle was developed by iterative-seeding based method. This nanocargo consists of a cobalt ferrite kernel as a core (Nk) and multiple layers of gold as a functionalizable active stratum, (named as Nk@A after fifth iteration). Nk@A helps in augmenting the physiological stability and enhancing surface plasmon resonance (SPR) property. The targeted delivery of Doxorubicin using Nk@A as a nanopayload is demonstrated in this report. The drug release profile followed first order rate kinetics optimally at pH 5.4, which is considered as an endosomal pH of cells. The cellular MR imaging showed that Nk@A is an efficient T2 contrast agent for both L6 (r2-118.08 mM−1s−1) and Hep2 (r2-217.24 mM−1s−1) cells. Microwave based magnetic hyperthermia studies exhibited an augmentation in the temperature due to the transformation of radiation energy into heat at 2.45 GHz. There was an enhancement in cancer cell cytotoxicity when hyperthermia combined with chemotherapy. Hence, this single nanoplatform can deliver 3-pronged theranostic applications viz., targeted drug-delivery, T2 MR imaging and hyperthermia.

. TEM image of Nks were spherical in shape with the size range of 11-14 nm exhibiting high crystallinity with aggregation which is illustrated in Fig. 3a(a*). The crystal lattice structure of Nks were distorted on the surface due to the curvature effect. Hence, the gold shell could grow epitaxially on the surface of Nks due to a large lattice mismatch. This leads to the formation of Nk as a core and gold as a shell. As iteration increased, the size of the core-shell is also increased five times from 1 st to 5 th iteration with Au nanoshell. This increment was due to the continued conjugation of nanogold onto the surface of the core Nk, which lead to the formation of Nk@A. HAADF-STEM analysis clearly demonstrated the discrimination between the core and the shell. This is because this contrast is directly proportional to atomic number (Z). In the 1 st iterative step, the Nk was coated with Au nanoshell of around 1-1.5 nm (Fig. 3b(b*)) without any aggregation. This proves that the Au iterations not only forms a shell but also stabilizes the nanoparticles. However, in order to enhance the SPR property, the iteration was continued; thus leading to the formation of a nanoflower, that constituted of the collective core madeof Nk encapsulated by a thick Au shell. The nanocluster showed jagged-like morphology due to non-homogeneously aggregated Nk, which leads to highly asymmetric coating of Au layer (Fig. 3c(c*)) 35 . The thick gold nanoshell formation after multiple iterations in the Nk@A solution could be inferred from the colour change ( Figure S1a in the †ESI). Figure 3c † illustrates the line scan analysis of 3 rd iterated nanoflowers showing the elemental distribution of Co, Fe and Au in a single nanoparticle. The line scan confirms that Au signal is seen on the surface of Nks and Co, Fe signals are enriched in the inner core of the Nks. But as the iterations (5 th iterations) continued, the Au seeds started to fill in the empty space of the knobby structures 22 . This resulted in the formation of separated spherical Nk@A and consecutively the shell size increased to 5-6 nm ( Fig. 3d(d*)). Similar kind of NPs were obtained for Au@Fe 3 O 4 , which had a thin shell of Au 36 . Figure 3d † represents the corresponding line scan, which clearly shows the Co K, Fe K edges in the core and Au L edges in the shell similar to that of 3 rd iterated particles. Additionally to the line scan intensities illustrated in Fig. 3c † ,d † , it is interesting to define the stoichiometry represented by the Au Lα , whose ratio is higher than the core elements such as Co and Fe Kα . This proves that the Au signal is exhibited strongly than the signal from the core. Therefore, these results suggest that the formation of CSNPs expressing typical elemental composition of high Au content than the core elements, is well corroborated from XRD spectra. The colour mapping shown in Fig. 3(a*,b*,c*,d*) proves that the nanostructures are made up of two different metals depending on the electronic density of the atoms. This colour mapping was carried out using the Digital Micrograph 3.7.0 by Gatan software. EDS spectra was carried out to determine the composition of CSNPs for the 1 st , 3 rd and 5 th iterations ( Figure S1b,c,d in the †ESI) showing the signal of Au, Co & Fe 37 . The spectra clearly distinguishes the different Au iterations from 1 to 5 just by increasing order of Au signal intensity.
The magnetic property of Nk@A is imperative to have an effective penetration in the cancer cell 38 . SQUID analysis (Fig. 4) shows a decrement in magnetic saturation (MS) from 74 to 45 emu/g, along with the diminishing hysteretic features. As temperature increased from 5 to 312 K, coercivity (HC) and remanence decreased from 146 to 32Oe and from 8 to 2 emu/g, respectively (Table S1 in the †ESI).
This magnetic behaviour was attributed to the thermal relaxation of the magnetic moments of the Nk@A. This behaviour was also ascribed to the re-orientation of the magnetic moments of NPs caused by the thermal energy once it surpassed the magnetic energy imposed by the applied field. Thus, the magnetic characteristics of the Nk@A can be ascribed to those expected from a soft ferromagnetic material, even at 312K. This kind of ferromagnetic character can be understood from MR imaging which showed high relaxivity values without the interference of Au nanoshell. Moreover, in the case of hyperthermia, the Nk@A showed increased heat dissipation under microwave irradiation in a short span of time.
XPS measurements determined the binding energies and composition of Nk@A. The elements viz., Au, Co, Fe, C and O existed within the range from 0 to ≈ 1300 eV. Core level spectra were recorded and represented in Figure S3a in the †ESI. Fe2p3/2 and 2p1/2 peaks from Fig. 5a situated at around 711.7 and 725 eV respectively, were broadened due to the presence of Fe 3+ ions in tetrahedral sites 39,40 . There is also a satellite peak of Fe 3+ at 719.6 eV, which confirms the presence of Fe 3+ 41 . The orbitals of Co (Fig. 5b) showed that Co2p3/2 electrons exhibited binding energies at 781.2 eV, which corresponds to ions. This may be due to the substitution of Fe 3+ ions with Co 2+ in the tetrahedral site. There is also the existence of a peak at 786.6 eV, which again confirms the presence of Co 2+ . This is in accordance with the XRD data, which proves that there are no mixed phases of CoO or Fe 2 O 3 in the Nk samples 41 . Figure 5c shows Au binding energies with doublet peaks at 83.8 and 87.4eV thus denoting the Au state of Au4f7/2 and Au4f5/2, respectively. This shows that gold ions are completely converted into metallic Au 0 , leading to the formation of CSNPs. Interestingly, the spectrum did not show any detectable Cl2p signal 42 which further proved the complete reduction of Au onto the surface of Nk (Fig. 5d). The O1s peak showed in Fig. 5e confirmed the presence of oxygen atoms 43 . The C1s signature markers of carbon atom binding energy at 284.7 eV were taken as a reference (Fig. 5f) 43 . Therefore, XPS pattern was in good agreement with XRD data, TEM-EDS and line scanning results.
Tethering folic acid linker and Doxorubicin molecules on Nk@A. UV-Visible absorption spectra of different iterations of Nk@A were performed as seen in Fig. 6a. As the number of gold iterations on Nk increased from one to three, there was a bathochromic shift of surface plasmon resonance peak (SPR) from 541 to 546 nm. In addition, as the iteration increased from three to five, this SPR peak further shifted from 546 to 551 nm. This bathochromic shift in the SPR peak was related to the increment in the thickness of gold shells on the surface of the magnetic core 18 . Moreover, as iterations increased, the charge density and the amplitude of the free electron oscillation inside the particles also increased. This may be due to the increased surface coating of CTAB, thus causing enhanced plasmonic absorption 44 . The spectra of Nk and Au seeds are also shown in Figure S2 in the †ESI. UV-Visible spectra of FA attached Nk@A (Fig. 6b) shows distinct peaks at 280.8 and 375.2 nm, which are signature markers of FA. Accordingly, there was a bathochromic shift in the SPR peak after FA attachment which showed a peak at 562.4 nm; this shift confirmed that FA formed a complex with Nk@A 45 . Dox attachment was  confirmed from the peak (inset) at 490.7 nm with minor red-shift of Nk@A peak (538.9 nm) and slight blue-shift of the FA attachment (268.7 & 375.2 nm). These shifts ensured the formation of the Dox-FA-Nk@A complex, which is shown in Fig. 6c. FTIR spectra of functional organic markers on the surface of Nk@A is shown in Fig. 7a. Spectra A represented only CTAB peaks (Table S2 in the †ESI) 46 , which proved that there was an excess amount of CTAB in the solution. However, spectra B which was analyzed after repeated centrifugation and washing the Nk@A solution, showed a minor peak of CTAB which stabilizes the Au shell (Table S2 in the †ESI). This was also demonstrated from its cytocompatibility towards L6 cells, which was evident from MTT assay and confocal microscopy studies. Figure 7b shows FTIR spectra of FA-Nk@A and Dox-FA-Nk@A. Spectra A represents both activated FA, which show bands at 1640.7 cm −1 and 1718 cm −1 expressing -CH stretch and -NH stretch, respectively. FA conjugation to Nk@A was confirmed from -NH and -CH stretch; while the asymmetric stretching of primary amines -NH and bending vibrations of -CO confirmed the formation of amide linkage between FA and Nk@A at 1585.2 cm −1 47 . Spectra B represents the attachment of Dox moieties onto FA-Nk@A. The interaction between these molecules was via amide linkage, which involved -NH amino group of FA and -COOH carboxylic group of Dox. Bands representing these attachments were 1436.8 cm −1 that represents anhydride = CO stretch and 1651.4 cm −1 denotes amide stretch of = CO. The peak at 2915 cm −1 was a classic peak of secondary -NH2 bending, and peak at 3000.68 cm −1 corresponded to primary -NH 2 bending 47 .
TGA analysis was carried out for all samples in the temperature range of 30-900 °C under N 2 flow atmosphere and any change in % weight loss was recorded. TGA graph ( Figure S3b in the †ESI) shows Nk, Nk@A, FA-Nk@A, Dox-FA-Nk@A, activated FA and Dox. Initially, activated FA and Dox showed a gradual weight loss from 30 to 100 °C. The degradation of these moieties was rapid because they are completely organic in nature. Nk showed a rapid degradation at 122 °C, which was due to the complete evaporation of water molecules. Then,  there was a solid plot up to 555 °C and a drop at 728 °C followed by a slow degradation of bound chlorides and hydroxides. In the case of Nk@A, the initial weight loss from 30 to 288 °C was due to the complete desorption of water and CTAB molecules from the surface 48 . The second degradation, from 289 to 595 °C, was a result of the covalent interaction of CTA + ions; the final degradation at a higher temperature, from 596 to 900 °C was most likely due to the electrostatic interaction of the ammonium group from CTAB attached to the Au NPs surface 49 . In FA-Nk@A complex, the weight loss in the range of 30-100 °C was due to the desorption of intercrystalline water molecules. The second degradation, in the range of 101-750 °C, was due to FA covalent attachment, which was seemingly induced by the disintegration of groups like hydroxyl, carboxyl and amino groups present in FA 50 . Dox-FA-Nk@A complex initially showed degradation of water moieties and weak surface interactions of hydrogen with the Nk@A. The second decomposition from 96 to 740 °C, was due to the decomposition of FA and Dox from the NPs complex and the decomposition from 741 to 900 °C is due to slow degradation of Nk@A 51 .

Intracellular localization of Dox & Dox-FA-Nk@A.
CryoTEM analysis confirmed Nk@A internalization and endocytic cavities that contained NPs (Fig. 8a,b) with aggregation, as the pH was acidic. This observation gave an insight towards the mechansim of receptor-mediated endocytosis of Nk@A, which can be attributed to FA and folate receptor interactions 52 . Moreover, the cellular uptake of NPs is also dependent on the surface charge. Zeta potential (ζ ) values for Nk, Nk@A, FA-Nk@A and Dox-FA-Nk@A complex were in the range of + 11.9 to + 12.1 mV, + 33.4 to + 37.7 mV, − 35.9 to − 38.4 mV, + 16.8 to + 19.4 mV, respectively ( Figure S3c in the †ESI) [53][54][55] . This charge dependency is due to the stability of the complex, which conciliates the harsh physiological milieu, such as in the bloodstream or inside the cell. The mechanism of cellular uptake for positively charged nanoparticles is interaction of the positive moieties with the negatively charged cell surface 56 . Rotello and coworkers studied the effect of surface charge on the stability of amine functionalized gold nanoparticles. It was found that net positive charge caused more displacement of ligands in extracellular 57 and intracellular environments 58 .  The tracking of Dox-FA-Nk@A inside the cells was performed using confocal microscopy. Hep2 cell lines were incubated only with Dox (without any fluorescent dye, since Dox itself is a good bioimaging agent) and it showed the internalization. Initially, Dox was distributed all over the cytoplasm ( Figure S4a in the †ESI) and cellular membrane, but after 24 h of incubation, the Dox signal in nucleus was very high due to the migration of Dox to this compartment and its release promoted by internal pH changes ( Figure S4b in the †ESI). This confirmed that the binding of Dox molecules to the nuclei by intercalating into DNA leads to inhibition of macromolecular biosynthesis 59 .
In order to assess the cell viability, MTT assay was carried out by incubating L6 and Hep2 cells with Nk, Nk@A, FA-Nk@A and Dox-FA-Nk@A for 24 h in the concentration range of 10-50 μ g/mL (Fig. 9a,b). Even though the concentrations of all the above complexes increased to 50 μ g/ml, there was not much cytotoxicity observed for L6 cells. Similarly, Hep2 cells also showed less cytotoxicity when exposed to Nk, Nk@A and FA-Nk@A at concentrations as high as 50 μ g/ml. However, the Hep2 cell viability started decreasing rapidly when the concentration of Dox-FA-Nk@A was increased from 10 to 50 μ g/ml. This may be due to the overexpression of folate receptors on the surface of Hep2 cells, which increases receptor mediated endocytosis and overall cellular uptake 60 .
In order to evaluate the cellular and nuclear morphology of L6 and Hep2 cells on incubation with Nk, Nk@A, FA-Nk@A and Dox-FA-Nk@A, the cells were examined using confocal microscopy (Figs 10 and 11). The morphology of cells treated with Nk, Nk@A and FA-Nk@A did not show any noticeable change in both L6 and Hep2 cells. The activity of Dox-FA-Nk@A, illustrated in Fig. 10(d), showed no effect on L6 cells, while Hep2 cells [ Fig. 11(d)] exhibited major apoptosis caused by disruption of nuclear membrane, which leads to total cell damage. This proved that the Dox-FA-Nk@A acted as an efficient nanocargo in delivering the Dox, which finally causes cell death. The increased cytotoxicity of Dox-FA-Nk@A can be due to the active transport of Dox by receptor-mediated endocytosis mechanism as compared to the passive diffusion of free Dox into the cells 61 .
In-vitro Dox release kinetics. The in-vitro drug release behavior of Dox-FA-Nk@A was assessed by using 3 different pH of PBS buffers (5.4, 6.8 & 7.4) for 24 h. We determined that the NPs system was both pH dependent and site-specific; which made it as a unique drug delivery system for cancer therapy. As pH decreased in the cellular organelles, thereby increasing the release of drugs as seen in Fig. 12a. The cumulative release of Dox at 3 different pH after 24 h was 77.9% at pH 5.4, 74.6% at pH 6.8, and 33.6% at pH 7.4, thus confirming pH-dependent release mechanism. The pH 7.4 mimics normal physiological pH; hence, drug release is minimum. Moreover, the tumor microenvironment exhibits a pH of 6.8 62,63 where drug release is comparatively more, which occurs due to the partial dissociation of the amide bonds between FA and Dox molecules. The complex showed maximum amount of drug release at pH 5.4. This pH is a signature marker of endosomal acidic pH consequently leading to the dissociation of the drug from the complex by breaking the amide bond between FA and Dox molecules. Moreover, Dox becomes highly water-soluble as well as hydrophilic at lower pH as compared to neutral pH. Hence, Dox is in its inactive form in normal tissues at neutral pH, while it gains its activity in cancerous tissue at lower pH 64 .
These results were fitted to different drug kinetic models such as zero order kinetics, first order kinetics, Higuchi model and Hixon-Crowell model (Fig. 12b). The drug release followed first order rate kinetics model which was attributed to the high regression coefficient value (R 2 = 0.9865). This confirms that the drug release is pH as well as concentration dependent mechanism.
Magnetic Resonance Imaging in normal and cancerous cells. Nk@A was employed as an efficient MR imaging contrast agent. It was cultured with both L6 and Hep2 cells. The cells were harvested and then resuspended in PBS with agar gel. These agar phantoms were used to evaluate the T 1 and T 2 -weighted images shown in Fig. 13a (only T 2 -weighted images). As the concentration increased from 0.08-0.64 mM of Nk@A, there was an increment in T 2 -weighed image, which became darker with both cells. The linear relationship was calculated in order to obtain the longitudinal relaxivity r 1 and transverse relaxivity r 2 of both L6 and Hep2 cells incubated with Nk@A (Fig. 13b,c). The r 1 /r 2 and r 2 /r 1 values were calculated and are shown in Table S3 in the †ESI. We found that the r 2 /r 1 value of L6 was 83.15 and for Hep2 it was 120.68. This was much higher than r 1 /r 2 of both, which confirmed that Nk@A were efficient T 2 contrast agent as compared to T 1 . But in the previous report for CoFe 2 O 4 @Au, r 2 /r 1 value was around 33 21 . So, we concluded that the increment in the r 2 /r 1 value was mainly due to Au iterations.
The r 2 value of Nk@A with L6 was 118.08 mM −1 s −1 and Hep2 was 217.24 mM −1 s −1 . This is highly comparable to the clinically used MRI contrast agent such as Feridex 105 mM −1 s −1 65 , which shows noticeable changes after injecting iron oxide NPs 66 . Therefore this confirmed that the Au nanoshell around Nk did not play a role in Nk core spin 67 . But the r 2 value with Hep2 cells showed significant increment when compared to L6 cells because of gold iterations. The other reason might be the uptake of gold coated magnetic NPs by cancerous cells was higher than that of normal cells 68 because the electrostatic interactions of surface charge from gold coating cause more interaction with the cell 69 . Finally, we demonstrated that the Nk@A can be effectively used as an efficient T 2 contrast agent.
Microwave based Hyperthermia therapy. Hyperthermia therapy involves increase in the temperature of tissues or cells, so that they become more susceptible to anti-cancer drugs. Nk@A nanoparticles were tested as a hyperthermal agent under microwave (MW) irradiation by using a microcoaxial double slot antenna as an applicator. The increment of temperature as a function of time was measured under ISM (Industrial, Scientific and Medical) approved frequency of 2.45 GHz in order to induce localized hyperthermia. The applicator was inserted in the phosphate buffered saline (PBS) containing Nk@A of various concentrations (10-125 μ g/ml) and the increment of temperature was measured by using noninterfering fiber optic probes. The MW was irradiated using a home-made setup 70 for 150 sec at 6 W. Interestingly, the temperature increment was very rapid and reached 45 °C in around 75 sec, which was enough to kill cancer cells; the temperature raised upto 50-60 °C within 150 sec for Nk@A (Fig. 14a).
The PBS was used as a control which showed maximum rise in temperature as water has the highest absorption of MW 71 . This temperature increment was purely based on the Au iterations on Nks. Au nanoshell as well as   superparamagnetic core led to energy increment which may be due to magnetic anisotropy 72 as compared to Nks in xylene at 2.45 GHz 73 . The temperature increment was much higher when compared with Au@γ -Fe 2 O 3 , which was around 38 °C at 2.45 GHz in water, for 10 mins at a power of 120 W 74 .  In-vitro Hyperthermia and Chemo-Hyperthermia therapy. Chemo-hyperthermal effect or thermo-chemosensitization is quantified by the interaction of anti-cancer drugs with cells at elevated temperature. The thermal enhancement ratio (TER) for Dox is 1 at two different temperatures (41.5 °C versus 43.5 °C, respectively) 75 . There is a relationship between drug-heat interactions on cell cytotoxicity. This pharmacodynamics is responsible for enhanced killing of cells. The mechanism of thermal enhancement for drug cytotoxicity includes enhanced drug uptake as well as DNA damage and retardation of DNA repair 33 . There are previous studies, which reported the thermal enhancement of cellular cytotoxicity when drug interaction with cells takes place at higher temperatures 76 . In this report, an increase in Nk@A concentration shows an increased cell mortality in addition to MW radiation exposure for 50 sec. The cells interacting with increased Nk@A concentration, but without any exposure to MW irradiation (Fig. 14b), showed comparatively lesser mortality than the exposed ones. Both Au shell and superparamagnetic core absorb MW, due to which hyperthermia is induced in the cells, thus consequently leading to cell death. The viability of Hep2 cells was suppressed more by Dox-FA-Nk@A when exposed to microwave as compared to unexposed ones (Fig. 14c). The Dox concentration was considered based on its half-maximal inhibitory concentration (IC50) value of 12.5 μ g. Hence, here the Dox tethered Nk@A concentration was in the range of 2-14 μ g/ml. The IC50 value of Dox-FA-Nk@A-MW was 8 μ g as compared to the IC50 value of Dox-FA-Nk@A, which is 12 μ g. This clearly shows thermal enhancement of Dox cytotoxicity at lower IC50 values in comparison with Dox alone. Hence, Dox orchestrated Nk@A efficiently inhibited the cell viability at elevated temperatures and very low concentrations of Dox, thus improving the therapeutic proficiency along with minimum side-effects.
Concurrently, this combined therapy shows efficient synergism, thus exhibiting inimical effect on Hep2 cells.
In summary, we have developed SPR enhanced Nk@A by multiple iterative method, which proved to be a promising nanomaterial with multifunctional properties specifically in the field of cancer nanotheranostics. These multiple iterations provided a new platform for high surface functionalization. This helped in the efficient delivery of drugs, following first order rate kinetics. This Nk@A was also used as a competent MRI contrast agent and proved to be an effective T 2 agent with high relaxivity values in the presence of both L6 and Hep2 cells. Finally, Nk@A was used as a hyperthermal agent. In an in-vitro study using Hep2 cells, both Nk@A and Dox-FA-Nk@A on exposure to microwave irradiation using 2.45 GHz for 50 sec showed cell mortality. It was found that there is an enhancement of cell mortality when MW based hyperthermia collates with Chemotherapy. With this efficiency, Nk@A can be used for potential applications as a single nanomaterial for 3 different uses, from tracking, diagnosing to therapeutics.

Synthesis of Cobalt ferrite (Nk) and gold-coated Nk (Nk@A) NPs. Nk preparation. Magnetic Nk
were synthesized using co-precipitation method. The precursors 0.5 M of ferric (III) chloride and 0.25 M of cobalt (II) nitrate hexahydrate were taken in the ratio of 1:0.5 and in order to avoid the precipitation of the salts, they were initially dissolved separately in 10 ml of nitrogen (N 2 ) degassed DI water and mixed with 1.5 M solution of 40 ml NaOH, which was used as a reducing agent. The entire synthesis process was carried out under N 2 atmosphere for 1.5 h at 80 °C (pH 12). The black resultant precipitate was separated using a strong magnet and it was washed 3 times with DI water. This pure Nk were further used for Nk@A formation.
Nk seed preparation. 200 μ l of synthesized Nk was centrifuged and dispersed in 1 ml of DI water (pH adjusted to 7). Then, 400 μ l of DMSO was added and the mixture was stirred continuously under N 2 atmosphere at 75 °C for 3 h to reduce the aggregation of NPs.
Preparation of gold seed solution. Au coating was carried out for 5 iterations, 5 aliquots of gold seed solution were prepared freshly by mixing 0.5 ml (1 M) of CTAB, 1 ml (50 mM) of ascorbic acid, and 100 μ l (1M) of HAuCl 4 solution. This solution complex mixture was sonicated for 15 mins. The golden yellow color immediately changed to a milky orange color and then to a milky white color ( Figure S1b, (b*) in the †ESI).
Synthesis of Nk@A CSNPs. The seed solution of Nk and gold were used to synthesize Nk@A. In this case, the ratio of 1:5 was taken. This was because we had already optimized the ratio with 1:5, 1:7, and 1:9 and found that 1:5 was appropriate for the Nk@A. Initially, 1 part of milky white colored gold seed solution was added dropwise to the faint brown-colored Nk seed solution. This solution mixture was stirred for 2 h continuously until the brown colored solution turned into purple. Then, the gold iteration was continued for 5 times for every 2 h. Finally, a dark purple color Nk@A solution was obtained. Then, these Nk@A were magnetically separated (after each iteration) by magnetic separation technique and they were washed twice with a mixture of hexane and ethanol to obtain high purity NPs excluding excess gold NPs. The washed NPs were centrifuged again twice to remove excess CTAB from the solution.
FA Activation and attachment to Nk@A. Activated FA was used for the attachment onto the surface of Nk@A. The reason behind the activation of FA was to activate the carboxylate group. This was carried out by dissolving FA (0.25 g) into 20 ml of DMSO which was subjected for 1 h sonication to ensure complete dispersion. Later, carboxylate group present in FA was activated by mixing 0.125 gm of DCC and 0.225 gm of NHS. The complete reaction was carried out under N 2 atmosphere at 30 °C for 12 h (FA/DCC/NHS molar ratio 2:1:2). The resultant product was filtered using Whatman filter paper; then it was further used for attach onto Nk@A.
Activated FA was used to attach onto the Nk@A. This attachment was carried out by mixing 1ml of activated FA and 10 ml of Nk@A under N 2 atmosphere and stirring the mixture continuously for 5 h. Then, N 2 atmosphere was detached and the reaction was continuously stirred for 24 h. Finally, this reaction mixture was filtered using Whatman filter paper. Then the process of dialysis was carried out to eliminate unreacted FA using a 3000 kDa dialysis membrane in PBS (pH 7.4). After centrifugation, the pellet was again dialyzed in DI water for a period of 24 h. The activated FA binding onto the surface of Nk@A was studied using UV-Vis spectroscopy analysis.
Synthesis of Dox-FA-Nk@A CSNPs complex. Anthracycline antibiotic Dox was used to kill cancerous cells using FA-Nk@A. Activated FA functionalization on Nk@A acted as an anchor for the binding of Dox. To bind Dox, 5 ml FA-Nk@A was mixed with 1ml of TEA and ml of DMSO as a solvent; finally, 400μ l of 2.36 mM Dox solution were added. This mixture was purged using N 2 gas under continuous stirring at 60 °C for 5 h. The final complex solution was dialyzed to remove the unbound or excess of Dox. The Dox binding was studied by characterization techniques, like FTIR and UV-Vis spectroscopy.
Dox loading efficiency and in-vitro Dox release. Before initiating the drug release studies, it was very significant to determine the Dox loading efficiency because the exact amount of Dox bound onto FA-Nk@A complex was so decisive to calculate the proper drug delivery study (in the †ESI). Dox release studies were carried out by dialysis process. Drug release study was determined at 3 different PBS at 37 °C in an incubator of pH 5.3, 6.8 and 7.4 with continuous stirring to simulate the intercellular, intracellular and external environment of cancer cells. To carry out this study, 2 ml of dialyzed Dox-FA-Nk@A complex was used. Then, 1ml of sample was withdrawn from the system every 40mins to determine the drug content. To compensate the PBS solution as soon as 1ml was drawn, it was replaced with the same equivalent volume. The amount of Dox released was determined using UV-Vis spectrophotometer at 485nm which was the signature absorbance of Dox. All the experiments were repeated thrice for all the samples. This drug release study was also explained with different drug kinetics models in order to explain the release mechanism.
Thermogravimetric analysis (TGA). Thermal analysis was carried out for liquid samples using a TGA Q50 (TA Instruments) from 30 to 900 °C under nitrogen flow with a heating rate of 10 °C/min.

High Resolution Transmission electron microscopy (HRTEM)
. TEM images were taken on a HRTEM (JEOL, JEM-ARM200F) equipped with HAADF-STEM (high-angle annular dark-field scanning transmission electron microscopy) detector and Oxford XMax 80 Energy Dispersive X-Ray Spectrometer (EDS). The sample was prepared in such a way that aggregation in the grid was avoided. Then, 10 μ l of sample were dispersed in 100 μ l of isopropanol, which was sonicated for 30 mins; then a drop of sonicated nanoparticle dispersion was placed onto the amorphous carbon-coated 200 mesh copper grid (Ted Pella, Inc.). Finally, the sample was allowed to dry at ambient temperature before it was loaded into the microscope.
CryoTEM. The internalization of NPs inside the cells was confirmed by cryoTEM analysis. 5 μ l of Nk@A nanoparticle suspensions were made into a thin liquid film which was prepared on lacy carbon grid (Pelco, USA) and then quenched into liquid ethane to freeze the sample using a Leica EM-CPC chamber. Tecnai F20 (FEI) operated at 200 kV to obtain the images which were recorded with a USC1000 slow scan CCD camera (Gatan) at 50000x.
X-Ray photoelectron spectroscopy (XPS) Analysis. XPS analysis were performed using a K-Alpha X ray Photoelectron Spectrometer (XPS) System (Surface Analysis, Thermo Scientific). Monochromated, Micro-focused Al Kα was used as an X-ray source type. The binding energy of C1s carbon at 284.7 eV was used for calibration.
Magnetic measurements. Nk and Nk@A magnetic properties were measured using a superconducting quantum interference device (SQUID), (Quantum Design, MPMS3). The magnetization hysteresis of the samples was obtained by changing H between + 70 to − 70 kOe at 3 different temperatures, like 5 K, 300 K and 312 K. The hysteresis of the samples was also obtained at 100 kOe using a temperature interval from 1.8 to 312 K.
Zeta potential measurements. The zeta potential values were determined by using a Zetasizer Nano ZS90 (Malvern instruments) at 25 °C at a wavelength of He-Ne laser 633 nm, Max 4 mW at a scattering angle of 90° using a Universal 'dip' cell kit. Data were obtained using a monomodal acquisition according to the Smoluchowski theory. The measurements were repeated 3 times. Before the analysis the samples were well sonicated for 1 h to avoid aggregation.
Magnetic Resonance Imaging (MRI) experiments. MR imaging was performed with a 7T clinical Signa HDxt scanner (Varian). T 2 -weighted images were acquired using the following parameters: 7T, Repetition time TR = 2000 ms, fast spin echo, FOV = 3*3 cm, Echo time TE = 15-250 ms, slice thickness = 4 mm and resolution 256 × 256 points. For T 1 measurements, coronal spin-echo sequences with fixed echo time (TE) = 24 ms and varying repetition time (TR) (25 ms to 4 s) were used. Nk@A suspensions was taken at varying concentrations.
Microwave (MW) experiment setup. The setup used to apply the MW electromagnetic field to perform the experiments consisted of a generator (SML 03, Rhode & Schwarz, Germany) set to a frequency of 2.45 GHz. This signal was then amplified using an RF & MW module power amplifier (1164-BBM3Q6AHM, Empower, USA). The output power was then monitored through the use of a dual direct coupler (DC7154M, Amplifier Research, USA) and a power meter (PM2002, Amplifier Research, USA) to ensure an output power of 6W and to monitor the reflected power of the system. To adjust the standing wave ratio (SWR) to an optimal value, a coaxial stub tuner (1878C, Maury Microwave Corp., USA) was used along a network analyzer (E5071B, Agilent Technologies, USA) to measure and reduce the SWR to a minimum prior to each experiment.
Temperature Measurements. Non electromagnetic interfering optical fiber probes temperature sensors (M3300, Luxtron, USA) were used to record temperature increment. The temperature was measured inside the PBS. Each test lasted 200 sec in order to study the temperature response as a function of time. The temperatures were recorded every second during the experiments using True Temp software (Luxtron, USA).
Confocal Imaging. Z-series confocal images were collected using Zeiss LSM 700 confocal microscope fitted with a 40X oil-immersion lens. Images were processed using Zen 2012 software.
All the above analysis and characterization techniques used liquid sample (FTIR, TGA, UV-Vis spec, HRTEM, Zeta potential, Flow cytometry analysis), powder form (SQUID), and thin film (XRD).