Iron content titanium dioxide nanoparticles as exogenous contrast agent for tissue imaging using swept-source optical coherence tomography

Titanium dioxide (TiO₂) is enormously popular with both industrial and consumer sectors, appearing in dozens of products that people use and see daily.^1–3 TiO₂ is extensively used in paint, catalytic coatings, plastic, paper, pharmaceuticals, and sunscreens, and it is also used in packaging, cosmetics, toothpaste, and food.^4–6 In skincare and makeup products, titanium dioxide is used both as a pigment and as a thickener for creams. As a sunscreen, ultra-fine TiO₂ is used because of its transparency and UV absorbing capabilities.^7,8 Applications using nano-grade TiO₂ have prompted some authorities questioning, as with all nanoparticles (NPs), whether TiO₂ could be a carcinogen. Studies that have linked nano-TiO₂ to cancer risk are based on a lung overload effect observed in rats, involving exposure to very high quantities of TiO₂ by inhalation.^9–12 A monograph detailing the findings as published in 2010 by the International Agency for Research on Cancer (IARC) reclassified that titanium dioxide is “possibly carcinogenic to humans” through inhalation, which included it under the category of the 2B group.¹³ Engineering of TiO₂ nanoparticles with FDA approved metal iron can inhibit toxicity in TiO₂. Fe incorporation slows down the rate of dissolution of TiO₂ by maintaining homeostasis at the cell and mitochondrial organelle membrane and also reduces the reactive oxygen species (ROS), as presented in our previous studies.^14,15

There are many spectroscopic and microscopic methods that have been used previously to examine the behavior and response of nanomaterials with biological matter. However, comparative ex vivo study between TiO₂ and Fe content TiO₂ nanoparticles (NPs) as exogenous contrast agents using swept-source optical coherence tomography (SS-OCT) has not been reported yet. Optical coherence tomography (OCT) is a unique and rapidly evolving non-invasive imaging technique capable of generating high-resolution tomograms through stratified samples like biological tissues.¹⁶ In recent years, OCT has proved to be an efficient tool for imaging superficial tissues of the skin and mucous membranes.¹⁷ However, there are only few studies that investigate the effects of TiO₂ NPs penetrating and accumulating in the human normal lung (NL), lung squamous cell carcinoma (LSCC) tissue, and chicken breast tissue on tissue optical properties, and these reports suggested that TiO₂ can be used as a contrast agent.^18,19 TiO₂ NPs exhibit low absorption in the near-infrared range and strong backscattering properties in the UV-visible spectral range.²⁰ This property makes them suitable for use as an exogenous contrast agent for optical imaging methods. TiO₂ has been mainly used in the cosmetics industry; therefore, it is essential to study its penetration ability into the skin and make it more bio-suitable via Fe incorporation without any toxic effects to cells.

The paper reports on ex vivo tissue imaging efficiency of engineered Fe incorporated TiO₂ nanoparticles as an exogenous contrast agent using the SS-OCT technique. Enhancement of the tissue contrast was quantitatively analyzed in terms of the contrast to noise ratio (CNR).

Ethanol (C₂H₆O, ∼96%), tetraisopropyl (IV) isopropoxide (TTIP, Ti[OCH(CH₃)₂]₄, 99.99%), hydrochloric acid (HCl, 36.5%–38.0%), and anhydrous iron (III) chloride (FeCl₃, ≥99.99%) were purchased from Sigma-Aldrich. Chicken breast tissue for SS-OCT imaging was procured from a nearby slaughterhouse. Other reagents were of analytical reagent grade, and all solutions were prepared with deionized (DI) water. Glassware was washed with dilute nitric acid (HNO₃) and distilled water and was dried in a hot air oven before performing the experiments.

Iron was incorporated into TiO₂ using the sol–gel synthesis method, as shown in Fig. 1. Pure TiO₂ and Fe content TiO₂ powder samples with different concentrations of Fe³⁺ were prepared as reported earlier.^14,15,21 Initially, 4 ml of TTIP was dissolved in 20 ml of ethanol, by continuous stirring for 1 h, in which 1 ml of HCl was added to get a clear transparent solution. FeCl₃ solution of different molar (M) concentrations, i.e., 0, 0.1, 0.5, and 1, was added to the clear transparent solution of TTIP and was kept on a stirrer for 2 h at room temperature, which resulted in four samples named TiO₂, 0.1MFe/Ti, 0.5MFe/Ti, and 1MFe/Ti, respectively. The solution was dried at 80 °C and calcined at 400 °C for 3 h. The synthesized powder was ground and submitted for characterization. The evolution of the morphology and size of the TiO₂ and Fe content TiO₂ (0.1MFe/Ti, 0.5MFe/Ti, and 1MFe/Ti) nanoparticle samples was examined by using a transmission electron microscope (TEM, JEOL/JEM 2100) at 200 kV accelerating voltage. The structural and the chemical state of synthesized nanostructured materials were analyzed through high-resolution x-ray photoelectron spectroscopy (XPS) spectra (PHI 5000 Versa Probe III). The reflectance property of the respective nanoparticles was investigated using a UV–Vis spectrophotometer (V-760 JASCO) in the mode of diffuse reflectance spectroscopy (DRS).

A diagram representing the SS-OCT scheme applied in this experiment is shown in Fig. 2. The experimental setup comprises a laser source equipped with a fiber coupler (FC 50:50) and a circulator. The reference arm and sample arm consists of a fiber collimator lens, double-axis galvo scanners, and an achromatic doublet lens. The back-reflected light through both arms is impeded with the 50:50 FC and captured with a balanced photodetector (BPD). The analog OCT signal is further converted into a digital signal. The developed signal is stored in a computer for post-signal-processing. Subsequently, optimal alignment is achieved, and then, the desired experiment is performed.

To investigate the effect of TiO₂ and Fe content TiO₂ nanoparticles as an exogenous contrast agent, water suspended TiO₂ and Fe content TiO₂ nanoparticles (50 µg/ml) were applied on the chicken breast tissue. The chicken breast tissue was washed and sliced to obtain a consistently flat surface. The tissue sample was then imaged using the SS-OCT system by fixing it at the sample arm to generate a control data set. Following this, TiO₂ and Fe content TiO₂ nanoparticles were applied to the tissue sample, and imaging was performed at every 5 min time intervals. Exposure times of 5 min, 10 min, 15 min, 20 min, 25 min, and 30 min were set to analyze the level of penetration. Scanning and data acquisition were performed with custom made software. After receiving the primary SS-OCT signals, signal polarization controllers were used for fine adjustment of the signal. The scanning area chosen for the experiment was a 6 mm line scan, where each B-scan comprises 2000 A-scans and the SS-OCT images acquired after averaging 100 B-scan frames.^18,19 In the SS-OCT device, near-infrared (NIR) light propagates through the biological tissue sample, generating backscattered light, which is observed via a photodetector. The NIR light (1060 nm) is used as the swept-source light. Quantification of the scattering coefficient (μs) and contrast-to-noise ratio (CNR) was performed. Both parameters were estimated for the OCT images recorded at different exposure times. The image quality was determined by measuring the CNR given by

where S_a and S_b are the mean of the signal strengths at the region of interest. S_a was calculated by taking the region of interest selected from the upper layer of the tissue skin, whereas for S_b, the inner-layer of the tissue was selected as the region of interest. σ_N is the standard deviation of the background noise signal, and N was selected from the background noise present in the B-scan.^18,22 The localized scattering coefficients were estimated by averaging 50 adjacent A-scans. The images were recorded over five different locations, selected randomly within the image over an area of 100 × 100 (pixel)² under the region of interest.

Figure 3 depicts the TEM images of pure TiO₂ and Fe content TiO₂ nanoparticle samples. The micrographs of pure TiO₂ and 0.1MFe/Ti nanoparticles indicate the occurrence of spherical particle aggregations that are homogeneously dispersed through the matrix of the oxide, which are shown in Figs. 3(a) and 3(b). After careful examination, one can notice that the morphology of the samples is very sensitive to the Fe content in each sample. The micrograph of the 0.5MFe/Ti and 1MFe/Ti samples indicates the existence of a nanorod structure in both samples due to an increase in the Fe concentration in TiO₂, as shown in Figs. 3(c) and 3(d). TEM images of all the samples clearly indicate the transformation of morphology from spherical nanoparticles to nanorods. The previous reports of the authors suggested that the shape-dependent penetration into cells and tissue is a very important aspect for toxicity assessment.^14,15 Spherical TiO₂ nanoparticles can easily penetrate the membrane of cells by the endocytosis process after forming a phagocytic cup on the cellular membrane whereas cellular internalization of Fe content TiO₂ nanorods is challenging because rod-shaped particles need a higher binding surface on the membrane, which leads to creating a hurdle and slows down the penetration and dissolution rate of TiO₂ in the medium. Figure 3(e) presents the size distribution graph of TiO₂ and Fe content TiO₂ nanoparticle samples calculated from the Image J program. The average diameter (nm) seems to decrease from 2.8 nm for TiO₂ to 1.5 nm for 1MFe/Ti with an average length (nm) of 26.5 nm and 25.06 nm for 0.5MFe/Ti and 1MFe/Ti nanorods, respectively.

To analyze the chemical composition of TiO₂ and Fe content TiO₂ nanoparticles and to identify the chemical status of the incorporated element, the samples were characterized by XPS.²³ The XPS survey spectra of TiO₂, 0.1MFe/Ti, 0.5MFe/Ti, and 1MFe/Ti samples calcined at 400 °C are shown in Fig. 4(a). The peaks corresponding to Ti2p and O1s were observed in the TiO₂ sample, whereas the peak for Fe starts originating from the rest of the Fe content samples. The Ti2p_3/2 of pure TiO₂ is at about 457.36 eV, and the peak assigned to Ti2p_1/2 appeared at 463.15 eV, which suggests that titanium is in the fourth coordination state in the TiO₂ sample. The peak position of Ti⁴⁺2p_3/2 in the Fe content TiO₂ samples are observed slightly at higher binding energies of 457.38 eV, 457.58 eV, and 457.59 eV for 0.1MFe/Ti, 0.5MFe/Ti, and 1MFe/Ti, respectively, which is shown through the high-resolution spectra of Ti2p in Fig. 4(b). The peak position of Ti⁴⁺2p_1/2 in the Fe content TiO₂ samples is situated at higher binding energies of 463.43 eV, 463.50 eV, and 463.51 eV for 0.1MFe/Ti, 0.5MFe/Ti, and 1MFe/Ti, respectively, after Fe incorporation. After an increase in Fe concentration in TiO₂, doublet Ti2p peak intensity gets reduced. Meanwhile, the decreasing area of Ti⁴⁺ indicates a reduction in TiO₂ in the sample and probably the formation of a Ti–O–Fe structure in the TiO₂ lattice through substitution of transition metal ions.

Figure 4(c) shows the high resolution spectrum of Fe2p for the 0.1MFe/Ti, 0.5MFe/Ti, and 1MFe/Ti samples. Two main peaks can be seen, which are affiliated with Fe2p_3/2 and Fe2p_1/2. Binding energies for Fe2p_3/2 was observed at 710.07 eV, 710.22 eV, and 710.39 eV for 0.1MFe/Ti, 0.5MFe/Ti, and 1MFe/Ti, respectively. This is an indication that Fe ions may exist mainly in the form of Fe³⁺ in Fe content TiO₂ samples. The binding energies of the Fe2p_1/2 spectrum are observed at 723.43 eV, 723.84 eV, and 723.91 eV for 0.1Fe, 0.5Fe, and 1Fe, respectively, with an increase in the Fe content in the samples. It has been also reported that the binding energy of Fe³⁺ and Fe²⁺ are 710.9 eV and 709.4 eV, respectively.²⁴ The peak intensities of Fe2p_3/2 and Fe2p_1/2 get increased after Fe incorporation into TiO₂.

Figure 4(d) shows the high-resolution spectrum of O1s for TiO₂ and Fe content TiO₂ samples. Peaks affiliated to the O1s are placed at binding energies of 528.58 eV, 528.54 eV, 528.77 eV, and 528.78 eV for TiO₂, 0.1MFe/Ti, 0.5MFe/Ti, and 1MFe/Ti, respectively. The peak at 529.87 eV of pure TiO₂ is attributed to the lattice oxide. The peaks at 530.26 eV, 530.53 eV, and 530.16 eV are attributed to 0.1MFe/Ti, 0.5MFe/Ti, and 1MFe/Ti, respectively, which are due to the existence of defective oxygen or surface hydroxyl groups. In the pure TiO₂ sample, the amount of lattice oxide is dominant. When the Fe element is incorporated into the TiO₂ matrix, the amount of oxygen vacancy is significantly increased, thus leading to the appearance of Ti³⁺.²⁵ 

The paper deals with the tissue imaging efficacy of TiO₂ and Fe incorporated TiO₂ nanoparticles for chicken breast tissues using OCT. OCT is a very unique facility used for tissue imaging for which a contrast agent is required to enhance the image intensity. Therefore, the interaction of electromagnetic radiation with the nanoparticles induced in the tissue plays an important role. Near infra-red (NIR) light of a wavelength of 1060 nm is used as the swept-source light for tissue imaging in OCT. The optical response (reflectance) of the NPs in the NIR spectrum needs to be evaluated for imaging efficiency. Figure 5 shows the diffuse reflectance spectrum for the synthesized nanoparticles.

Figure 5(a) shows the reflectivity (R%) vs wavelength graphs for the synthesized nanoparticles. The TiO₂ sample showed wide absorbance in the UV region, with high reflectance beyond 350 nm. The sample showed the highest reflectance for the complete spectrum of 200 nm–1200 nm as compared to Fe content TiO₂ samples. The wide absorbance in TiO₂ nanoparticles between 200 nm and 350 nm is related to the ligand-to-metal charge transfer transitions.

For the rest of the samples, the reflectance showed a gradual decrease with an increase in the Fe content following a shift in the absorption band edge toward higher wavelength. The features perceived for 0.5MFe/Ti and 1MFe/Ti spectra showed the presence of several humps, signifying absorbance related to Fe³⁺ ligand field transitions.²⁶ Table I details the observed ligand field transitions assigned to different wavelength regions. Figure 5(b) shows the reflectivity % for the samples in particular to the 1060 nm wavelength (OCT light source wavelength). The figure shows the decline in the reflectivity of the synthesized samples with Fe incorporation. The reflectance of the nanoparticles at this region of the spectrum will determine the light attenuation when it travels through the nanoparticle treated tissue samples. The backscattered light intensity is captured, and the image is viewed in terms of photocurrent through the photodetector.

The bandgap of the semiconductor materials was calculated from the UV–Vis DRS spectrum directly using the Kubelka–Munk equation given as

The optical bandgap got reduced due to Fe incorporation in the TiO₂ matrix, with values changing from 3.49 nm for TiO₂ to 2.50 nm for 1MFe/Ti samples, as presented in the inset of Fig. 5(b). The present paper reports the use of Fe content TiO₂ nanomaterial for SS-OCT imaging. Nano-TiO₂ has the potential to trigger reactive oxygen species generation, which is one of the chief mechanisms behind nanotoxicity, and the band structure is the crucial material property that controls ROS producing ability. Incorporating Fe³⁺ into the TiO₂ matrix is to inhibit the ROS generation through band tuning. Figure 5(c) explains the detailed mechanism of toxicity prevention of TiO₂ via the introduction of Fe³⁺.

The potential of TiO₂ and Fe content TiO₂ nanoparticles as an exogenous contrast agent for chicken breast tissue was imaged with and without nanoparticles using an SS-OCT system. Initially, the OCT image of the tissue was taken without exposing it to nanoparticles. Each of the scans consists of 1000 A-scans. The final OCT intensity B-scan images were obtained by averaging 100 B-scan frames. SS-OCT noticeably distinguishes three different layers in the sample: (1) the top-most layer, (2) the middle layer (epimysium layer of the tissue), and (3) the bottom layer. The layered structure of the OCT images is due to the combination of tissue birefringence and the polarization sensitivity of the OCT setup.¹⁷ 

The upper surface of the tissue was then treated with TiO₂ and Fe content TiO₂ nanoparticles. SS-OCT B-scan images were captured for different exposure times of 0 min, 5 min, 10 min, 15 min, 20 min, 25 min, and 30 min. For interpretation of data, here, we have shown in Figs. 6(a)–6(d) the SS-OCT images captured at 0 min, 15 min, and 30 min for analysis. It was observed that with the increase in the exposure time of TiO₂ nanoparticles, the OCT signal intensity increased as the brightness of the deeper layer gradually increased. This reveals that the TiO₂ NPs gradually penetrate into the tissue over time, which is shown by the yellow arrow in Fig. 6(a). The excellent backscattering property of the nanoparticles is responsible for the strong signal whereas Fe content TiO₂ nanoparticles showed a slow penetration rate into the tissue. Fe incorporation reduces the dissolution rate of the parent TiO₂ nanomaterial and hence reduces the light penetration ability.

The application of 0.5MFe/Ti and 1MFe/Ti NP samples on the tissue showing a low penetration rate gets stabilized after 10 min–15 min, which can be visibly seen as shown in Figs. 6(c) and 6(d) as compared to TiO₂ NPs. This is attributed to the rod shape morphology of both samples, as discussed in the TEM results. The change in the shape of TiO₂ from spherical to nanorods after Fe³⁺ addition is well-intentioned to slow down the penetration rate of Fe content TiO₂ nanoparticles into the tissue layers.

The scattering A-scan profiles acquired for all the samples are shown in Fig. 7. Detailed quantitative analysis of the images was performed using the A-scan profile, and the CNR ratio was determined. The single A-line profile on the linear scale was extracted from the B-scan. Figure 7(a) shows that the signal intensity increases with exposure time for TiO₂ NPs treated samples, which are indicated by the black arrow whereas the penetration rate of Fe incorporated TiO₂ NPs treated samples, shown in Figs. 7(b)–7(d), decreases as compared to pure TiO₂ treated samples. It could be presumed that TiO₂ NPs provide more active sites of interaction, which is inhibited after Fe incorporation in the crystal lattice of TiO₂. The light penetration ability reduces after this difference is attributed to Fe incorporation, which leads to a change in the surface reactivity of TiO₂, inducing a greater number of active interaction sites with the surrounding tissue in the vicinity and thereby reduces the rate of penetration. It was observed that the penetration of TiO₂ and Fe content TiO₂ NPs inside the tissue is not similar. However, the penetration rate of TiO₂ is higher due to its smaller structure. The signal intensity was not found to get increased or stabilized after application of Fe content TiO₂ nanoparticles.

The CNR was calculated for different exposure times using an A-scan and Eq. (1). Figure 8(a) presents a graph of the CNR and the different exposure times (min). It shows that the CNR gets enhanced with the increase in exposure time. The CNR (db) was increased up to 37.47 over 30 min of exposure of TiO₂ NP application. This indicates that at a maximum exposure time, the CNR is also maximum, which results in better imaging, but as far as toxicity is concerned, TiO₂ NPs cannot be suggested as a good contrast agent. TiO₂ can easily penetrate in the upper layer of the tissue within 30 min as compared to Fe content TiO₂ NP samples. The CNR of Fe content TiO₂ samples is low or decreased as compared to pure TiO₂ samples. The CNR (db) of 0.1MFe/Ti, 0.5MFe/Ti, and 1MFe/Ti are 24.45, 25.28, and 29.47, respectively, for over 30 min of exposure because of their low penetration rate in tissue layers. Therefore, Fe content TiO₂ NPs can be suggested as good contrast agents with a low toxicity level.

The graph shown in Fig. 8(b) presents the improvement in the scattering coefficient with respect to time. The values are the average of five scattering coefficients calculated from a single B-scan, and the error bar presents the standard deviation between those values. The data are captured for every 5 min time interval, which shows gradual improvement, and after 25 min of time interval, it starts saturating in case of Fe content TiO₂ samples. However, after TiO₂ application on the tissue, the data get enhanced with time without any saturation point. The result of the scattering coefficient confirms that TiO₂ has a high scattering ability.

The penetration depth (μm) on the tissue layer after application of nanoparticles at 30 min is presented in Fig. 8(c), and it was calculated from the A-scan profile graph. It can be observed that TiO₂ can penetrate up to a depth of 780 µm. However, 0.1MFe/Ti, 0.5MFe/Ti, and 1MFe/Ti samples penetrated into the tissue layer up to a depth of 650 µm, 320 µm, and 210 µm, respectively. The 1MFe/Ti sample showed the lowest penetration rate. This is attributed to the fact that Fe helps reduce the rate of dissolution of parent TiO₂ nanoparticles on tissue layers.

In the current study, we have investigated the ex vivo imaging of chicken breast tissue with application of TiO₂ and Fe content TiO₂ nanoparticles as an exogenous contrast agent with non-specific binding. TiO₂ and different concentrations of Fe content TiO₂ nanoparticles were successfully synthesized by the sol–gel method and then characterized with XPS, DRS, and TEM. Fe incorporation reduces the bandgap of nano-TiO₂, and the morphology was changed from a spherical to elongated nanorod structure. We have observed that nano-TiO₂ has an effect on the optical properties of tissue, which results in an enhancement in the CNR for the SS-OCT image with an increase in the exposure time. The improvement in the scattering coefficient with respect to time is observed. The absorption of TiO₂ nanoparticles in chicken breast tissue over different layers and different exposure times is not uniform. TiO₂ showed penetration up to a depth of 780 µm inside the tissue layer, whereas Fe content TiO₂ samples showed a lower and slower penetration rate. Therefore, Fe content samples can be used as a safe or stable contrast agent because incorporation of Fe can prevent the toxic effects of TiO₂ on tissue or cells. Hence, Fe content TiO₂ nanoparticles can be suggested for use as a non-toxic exogenous contrast agent.

The authors acknowledge the Central Instrumentation Facility (CIF), Central University of Gujarat, Gandhinagar, for providing the instrumentation facility for the present work. T.B. is thankful to the Birla Institute of Technology, Mesra, and the Indian Institute of Technology Roorkee, for providing characterization facilities.

There are no conflicts of interest to declare.

The data that support the findings of this study are available from the corresponding author upon reasonable request.