Innovative functional polymerization of pyrrole-N-propionic acid onto WS2 nanotubes using cerium-doped maghemite nanoparticles for photothermal therapy

Tungsten disulfide nanotubes (WS2-NTs) were found to be very active for photothermal therapy. However, their lack of stability in aqueous solutions inhibits their use in many applications, especially in biomedicine. Few attempts were made to chemically functionalize the surface of the NTs to improve their dispersability. Here, we present a new polymerization method using cerium-doped maghemite nanoparticles (CM-NPs) as magnetic nanosized linkers between the WS2-NT surface and pyrrole-N-propionic acid monomers, which allow in situ polymerization onto the composite surface. This unique composite is magnetic, and contains two active entities for photothermal therapy—WS2 and the polypyrrole. The photothermal activity of the composite was tested at a wavelength of 808 nm, and significant thermal activity was observed. Moreover, the polycarboxylated polymeric coating of the NTs enables effective linkage of additional molecules or drugs via covalent bonding. In addition, a new method was established for large-scale synthesis of CM-NPs and WS2-NT-CM composites.


Experimental section
Preparation of CAN-mag (CM) nanoparticles in large quantities. The synthesis is similar to the one presented recently 18 , differences are highlighted in boldface. A solution of FeCl 3 ·H 2 O (960 mg, 3.6 mmol) in degassed ddH 2 O water (20 ml) was mixed with an aqueous solution of FeCl 2 ·4H 2 O (390 mg, 1.8 mmol, 20 mL H 2 O). The mixture was kept under nitrogen and ultrasonicated for 1 min at room temperature. Then, a concentrated (28-30 wt.%) NH 4 OH solution (2.4 ml) was added, resulting in the immediate formation of a black precipitate of magnetite (Fe 3 O 4 ) particles. Sonication was continued for an additional 10 min. The liquid was decanted with the help of magnetic separation, using a 0.5 Tesla magnet. The particles were washed with three portions of ddH 2 O (50 mL each) to neutrality. Then, ddH 2 O (50 ml) was added, and the maghemite NPs suspension was set aside for a minimum of 1.5 h at ambient temperature for aging prior to use.
A solution of CAN (2.00 g, 3.648 mmol) in acetone (24 ml) was added to the decanted magnetite NPs, followed by the addition of degassed purified water (96 ml). The resulting mixture was ultrasonicated while stirring for 35 min at 24% amplitude under nitrogen using a high-power sonicator. The acetone and most of the water were removed by rotary evaporation to a final volume of less than 50 ml. The solution was then centrifuged at 4000 rpm for 5 min to remove the liquid, and the residue was redispersed in ddH 2 O, after which the dispersion was transferred into 50 ml Amicon® Ultra-15 centrifugal filter tubes (100 kD, Millipore, Cork, Ireland). The contents were washed with three portions of ddH 2 O (10 ml each) and centrifuged at 4000 rpm for 10 min at 25 °C each time. The washed nanocomposite was dispersed in ddH 2 O (50 ml). The Fe concentration in the dispersion, determined by atomic absorption (AA), was 5.9 mg/ml.

Large-scale synthesis of WS 2 -NT-CM nanocomposit.
The synthesis is similar to the one presented in the previous article 18  Transmission electron microscopy (TEM) images were acquired by a JEM-1400 microscope (JEOL Inc., Peabody, MA, US) equipped with a 2 × 2 k CCD camera (Gatan, Pleasanton, CA, US). Samples for TEM analysis were dispersed in water. A drop of the dispersion was placed on a formvar/carbon film on a 400-mesh copper TEM grid (FCF400-Cu, Electron Microscopy Sciences, Hatfield, PA, US) and dried at ambient temperature for 24 h 18 .
Thermogravimetric analysis (TGA) was performed with a TGA/DSC1 analyzer (Mettler-Toledo, Greifensee, Switzerland). All thermograms were recorded in a nitrogen (50 ml/min) environment at a heating rate of 10 °C·min −1 over the temperature range of 30-800 °C. Weight change and heat flow were measured simultaneously during the analysis. The results were processed using STARe evaluation software (Mettler-Toledo, Greifensee, Switzerland) 18 .
ATR-FTIR spectra were obtained on a Nicolet iS10 FT-IR spectrometer (Thermo Scientific, Waltham, MA, US) equipped with an iD5 ATR accessory featuring a laminated diamond crystal. Samples were analyzed without further preparation. The data processing was performed using OMNIC 9 spectra software (Thermo Scientific, Waltham, MA, USA).
The temperature profiles for irradiated WS 2 -NT-CM-P[PPA] solutions were measured using a radiometric thermal imaging camera with 320 × 240 pixels, temperature sensitivity of 0.07 °C and spatial resolution of 0.5 mm (FLIR Systems Inc, Boston, MA, model A325). To characterizes the photothermal properties of this www.nature.com/scientificreports/ nanocomposite, concentrations of 20, 50, 100, 200 and 500 ppm were used. Double distilled water (ddH 2 O) was used as a negative control. For each sample, 1 ml was placed into a well with an rea of 2.0 cm 2 . The laser beam was directed at the sample from above, with a diode laser at wavelength of 808 nm (custom built), with maximum output of 6 W. In each experiment, the laser intensity on the place of the sample was divided by the laser spot area on the sample (0.64 cm 2 ). For each experiment, a few seconds of ambient temperature were recorded before irradiating the sample for 120 s. Zeta potential and dynamic light scattering (DLS) measurements were performed using a Zetasizer Nano-ZS device (Malvern Instruments Ltd., Worcestershire, UK). Samples for zeta potential and DLS measurements were dispersed in water (ca. 0.5 mg/ml).
For photothermal therapy (PTT) experiments, we tested a human cancer cell line (HeLa, ATCC, Manassas, VA, USA). Cells were cultured on 24-well glass plates. When the cells reached 80% confluence, freshly prepared aqueous dispersions of WS 2 -NT-CM-P[PPA] or WS 2 -NT-CM (45 µL, 0.8 mg/ml of WS 2 component calculated according to an elemental analysis of sulfur) were added to two of the plates, and a third plate, with no additives, was used for control. After 14 h of incubation, the cells were washed three times with PBS buffer, and fresh DMEM medium was added. For each condition, ten representative frames were imaged under a Zeiss LSM7 inverted two-photon microscope at 10 × magnification in phase-contrast. Next, a square region of 230 µm × 230 µm in the middle of each frame was irradiated with an 808 nm laser (Chameleon Vision II) at 90 mW for 31 s. A dye exclusion test of cell viability was performed using trypan blue for staining. A mixture of trypan blue solution and PBS (1:1 v/v) was added to all the wells after laser irradiation. The same frames were imaged after 5 min 18 .

Results and discussion
The fabrication of WS 2 -NTs coated with polypyrrole (PPy) involves two steps. First, the CM NPs were attached to the NT surface. In the second step, a direct polymerization on the composite surface was implemented due to the step of PPA monomer adsorption and coordinative attachment to the CM NPs (2 h), followed by in situ polymerization by adding the oxidation reagent.
Transmission electron microscopy (TEM) is a very effective tool for tracking the various stages of surface engineering. At the end of each of the two stages of the composite fabrication, we used TEM to test the morphology. The TEM image of WS 2 -NT-CM (Fig. 2c) show that CM NPs (TEM image of only CM NPs, Fig. 2a) attached onto the WS 2 -NTs in small clusters (S-based coordinative chemical linkage). Also, a smaller aggregation level of the obtained composites can be observed in comparison to the "naked" untreated starting WS 2 -NTs (Fig. 2b), even though the synthesis was doneon a large scale. The images of the functional composite WS 2 -NT-CM-P[PPA] (Fig. 2d-f) show the polymeric coating around the nanotubes, more precisely around the CM NPs, demonstrating their role as anchors or linkers between the polymer and the WS 2 -NTs.
In order to quantify the polymer coating, TGA analyses were performed using a temperature profile of 30-800 °C at 10 °C/min under an airflow of 50 ml/min (Fig. 3). In the temperature range of 120-800 °C, the The untreated WS 2 -NTs exhibited a weight loss of 5.9% in the range of 400-520 °C due to the oxidation of WS 2 to WO 3 and the evolution of SO 2 . The nanotubes functionalized with CAN-mag (WS 2 -NT-CM) exhibited a continuous and moderate weight decline in the range of 30-354 °C. This weight loss is related mainly to the release of H 2 O, but also to the decomposition of adsorbed organic materials, and possibly to the scission of the Ce-ligands with their consequent release as nitrogen oxides. A weight loss of 5.9% was observed in the temperature range of 354-520 °C, consistent with the oxidation of WS 2 to WO 3 and with the subsequent release of SO 2 . WS 2 -NT-CM-P[PPA] exhibited the same moderate weight decline in the low temperature range as WS 2 -NT-CM, as well as the characteristic weight loss of 6.2% in the range of 385-580 °C. More importantly, it exhibited a peak weight loss of 6.2% in the temperature range of 206-385 °C, attributed to polymer decomposition and release of combustion products.
As mentioned in the introduction, light absorbance in the near IR range is a fundamental property that indicates the potential to exert a photothermal effect. In this work, UV-Vis spectrometry was performed on the three types of WS 2 -NTs-untreated, CAN-mag (CM)-decorated, and PPA-polymerized-in order to measure and compare their IR absorbance, thus predicting their photothermal activity (Fig. 4a). As the diagram shows, all three composites-WS 2 -NTs, WS 2 -NT-CM, and WS 2 -NT-CM-P[PPA]-demonstrated absorbance peaks around 700 nm originating from the WS 2 -NT core, consistent with our recent results 18 , which demonstrated photothermal activity at 700 nm in WS 2 -NTs and WS 2 -NT-CM. As can be seen for these polymer-free composites, the absorbance declines significantly above 700 nm, rendering them inactive in the deeper penetrating, higher wavelength region of 800-900 nm. However, in the case of the polyPPA coated NTs (WS 2 -NT-CM-P[PPA]), www.nature.com/scientificreports/ absorbance remained almost the same around 800 nm, and only a minor decline was measured around 900 nm. This observed prolongation of the absorbance range toward 900 nm is attributed to the PPA coating and suggests the possibility of photothermal activity in deeper tissues, while utilizing lower energy (and thus less harmful) light beams.
To characterize the photothermal activity of WS 2 -NT-CM-P[PPA], a radiometric thermal imaging camera was used to trace the temperature elevation at the irradiated spot. Temperature elevation of 500 ppm nanocomposite samples were measured at different laser intensity of 1.6, 2.7 and 4 W/cm 2 (Fig. 4b), indicating a nice correlation between the heating profile and the laser intensity. Figure 4c shows the temperature profiles for irradiated solutions as a function of time for different concentrations of the WS 2 -NT-CM-P[PPA] composite at constant intensity (4 W/cm 2 ), showing again a positive correlation between the NT concentrations and the temperature elevation. For the highest concentration (500 ppm), a temperature difference of about 16 °C was observed. It can be seen that all graphs have two slopes; the first one is sharper and related to the instant temperature rise of the laser spot on the sample. The second slope represents the temperature rise of the entire sample volume due to heat diffusion from the laser spot area, and it reaches steady state (linear temperature rise) 20 s after the laser application. This is confirmed in Fig. 4d where the temperature rise was measured in two points, on the spot area and on the surrounding simultaneously; one can see that both curves rise in parallel after ~ 20 s (see also the difference curve).
Moreover, the photothermal conversion efficiency (PTCE) of the WS 2 -NT-CM-P[PPA] composite was calculated based on the rate of heat absorbed by the water and relative to the laser intensity, leading to efficiency of 33.2%, very similar to the value found in the literature regarding WS 2 nanosheets (32.8%) 33 . Similar PTCE was obtained by black phosphorus quantum dots (BPQDs)-up to 28.4% 34 . In vitro experiments showed that at a low concentration (50 ppm), BPQDs generated sufficient heat to kill tumor cells almost completely under irradiation with an 808 nm laser 35 , and were successfully combined with immunotherapy 36 . Recently, successful nanomaterials such as tin-sulfide nanosheet-based dual-therapy nano-platforms (SDTNPs) 37 , gold nanoparticles 38-40 and 2D titanium nanosheets (TiNSs) 41 showed even higher photothermal performance owing to localized surface plasmon resonances. In the latter case, an exponential temperature increase is evident (see Fig. 3D in Ref 41 ), while the gold nanoparticles exhibit both a two-slope (see Fig. 3 in Ref 39 ) and a largely linear behavior (see Fig. 3 in Ref 38 ). Figure 5 shows the FTIR absorbance spectrum of each stage of the synthesis of the targeted composite. In addition, it shows the spectrum of the PPA polymer, which was prepared separately under the same composite polymerization protocol. A full characterization of WS 2 -NT and WS 2 -NT-CM can be found in our previous article, indicating the presence of the CM NPs in the WS 2 -CM composite 18    Apparently, these carboxylic acid moieties face the nanotubes and are involved in the coordination to the Ce, and thus have no influence on the surface zeta potential. Moreover, it is known that the zeta potential of polypyrroles is pH dependent; indeed, all zeta potential measurements of the four samples were performed in the pH range of 5-6. Within this range, the positive zeta potentials increase only slightly, as demonstrated by Zhang et al. 42 however significant changes appear at pH below 3 and above 8. Therefore, in this work, the positive zeta potential value corresponds to the polymer backbone chain. During oxidative polymerization (chemical or electrochemical) of conjugated polymers such as polypyrroles, electrons are abstracted from the backbone of the polymer chain, creating p-type (positive) charge carriers (see Fig. 7). To maintain charge neutrality, some of the counter anions present in solution (i.e., Cl − from the oxidant, FeCl 3 ·6H 2 O) are incorporated into the growing polymer during polymerization. However, in aqueous media, the doped counter anions (Cl − ) dissociate from the surface of the polymer and transfer into the bulk solution, leaving a positively charged surface and thereby a positive zeta potential.
In order to examine the potential of WS 2 -NT-CM and WS 2 -CM-P[PPA] as photothermal agents, human HeLa cells were incubated with both nanocomposites and treated with irradiation at 808 nm (IR laser) for 31 s.

Conclusion
A new core shell composite of WS 2 -pyrrole-N-propionic acid (PPA) was presented and evaluated for photothermal therapy (PTT). In addition, this work displays an improved protocol for the synthesis of both CM NPs and WS 2 -NT-CM composites in far greater amounts than in the recent publication 18 . The use of Ce-doped maghemite (CM) NPs solved the two major problems regarding functional polymerization on WS 2 -NTs: (1) the difficulty of polymerization onto an insoluble particle in most known solvents; and (2) the need for a linker between the monomers and the nanotubes preceding the polymerization. The CM NPs produce a much more stable nanotube suspension with strongly coordinating Ce sites on the NT surface, available for attachment of the polymerized www.nature.com/scientificreports/ monomers. The successful polymerization process is confirmed by TEM images and by TGA, FTIR, UV-visible spectroscopy, and zeta potential results, indicating a stable nanocomposite with high positive zeta potential value and a core-shell structure with 10-50 nm of polymer coating, which is 6.2% of the total composite (based on its weight loss in TGA). The photothermal characterization of WS 2 nanotubes was investigated for the first time, and their efficiency was calculated to be about 33%, in excellent agreement with WS 2 nanosheets. In the PTT in vitro assay, we expected a larger effect with WS 2 -NT-CM-P[PPA], namely greater cell death after irradiation compared with uncoated WS 2 -NT-CM. Nonetheless, the addition of a polymer did not reduce the PTT activity. Both composites had a rather significant PTT effect, with about 70-75% cell death after only 31 s. Furthermore, the polycarboxylated polymer enables the linkage of numerous materials in a covalent chemical bond (especially drugs and biomolecules), thus making the polymerization process worthwhile. Additional materials may be added through the covalent chemical bond, including substances that target cancerous growth and other types of light-activated therapies such as photodynamic therapy. Owing to the magnetic CM middle phase, two additional benefits-high MRI imaging ability and magnetic delivery-are anticipated. Thus, these hybrid NTs have a high potential to act as a multidrug platform for targeted treatment while enabling imaging of the treated area. WS 2 nanoplates and nanosheets shall be investigated.

Data availability
All data and materials support the published claims and comply with field standards. All research data are attached and are in concordance with disciplinary norms and expectations of the journal.