Renally Excretable Molybdenum Disulfide Nanoparticles as Contrast Agents for Dual-Energy Mammography and Computed Tomography

Abstract

Compared with conventional mammography, contrast-enhanced dual-energy mammography (DEM) can improve tumor detection for people with dense breasts. However, currently available iodine-based contrast agents have several drawbacks such as their contraindication for use with renal insufficiency, high-dose requirement, and suboptimal contrast production. Molybdenum disulfide nanoparticles (MoS₂ NPs) have been shown to attenuate X-rays due to molybdenum’s relatively high atomic number while having good biocompatibility. However, work exploring their use as X-ray contrast agents has been limited. In this study, we have developed a novel aqueous synthesis yielding ultrasmall, 2 nm MoS₂ NPs with various small molecule coatings, including glutathione (GSH), penicillamine, and 2-mercaptopropionic acid (2MPA). These nanoparticles were shown to have low in vitro cytotoxicity when tested with various cell lines at concentrations up to 1 mg/mL. For the first time, these particles were shown to generate clinically relevant contrast in DEM. In DEM, MoS₂ NPs generated higher contrast than iopamidol, a commercially available X-ray contrast agent, while also generating substantial contrast in CT. Moreover, MoS₂ NPs demonstrated rapid elimination in vivo, mitigating long-term toxicity concerns. Together, the results presented here suggest the potential utility of MoS₂ NPs as a dual-modality X-ray contrast agent for DEM and CT.

Graphical Abstract

INTRODUCTION

Routine mammography screenings have been shown to lower breast cancer mortality rates.¹ Approximately 50% of people undergoing mammography for breast cancer screening would be characterized as having dense breasts.² Breast density is defined as the ratio of fibroglandular to adipose tissue and can markedly affect the ability of conventional mammography to detect breast cancer.³ Heterogeneously dense breasts may obscure small masses in mammography and extremely dense breasts may lower the sensitivity of mammography.⁴ In addition to making breast cancer more difficult to detect, dense breasts have also been identified as a risk factor for developing breast cancer.⁵ Low efficacy of conventional mammography for people with dense breasts may result in higher rates of false negative cases, presenting an obstacle for timely clinical intervention.

Contrast-enhanced dual-energy mammography (DEM) is a breast cancer detection method recently approved by the Food and Drug Administration that may address the under-performance of conventional mammography for people with dense breasts.⁶ In DEM, image subtraction of mammograms acquired at different X-ray energies creates a dual-energy image that has similar sensitivity to MRI with fewer false positive results.^7–9 DEM requires only minor upgrades to standard mammography equipment, making it accessible and inexpensive compared to other breast cancer detection methods such as contrast-enhanced MRI or ultrasound.⁶

DEM requires the administration of a contrast agent to provide greater contrast between vascularized tumors and normal breast tissue. While efforts have been made toward the development of DEM-specific contrast agents, to date there are none commercially available.^10–13 Iodinated small molecules are currently used as contrast agents in DEM, but these molecules have several drawbacks. They are contraindicated for use with renal insufficiency, have a high-dose requirement, and produce suboptimal contrast in dual-energy mammography.¹⁴ Thus, we sought to develop an alternative contrast agent optimized for use in DEM.

Molybdenum disulfide nanoparticles (MoS₂ NPs) belong to the transition metal dichalcogenide (TMD) class of materials. TMDs have the chemical composition MX₂, where M represents a transition metal ion and X represents a chalcogen ion (S, Te, Se).¹⁵ TMDs have recently gained attention in a wide range of biomedical applications because of their unique structure, photoelectric properties, and high cytocompatibility.^16,17 Among their different biomedical applications, TMDs have been shown to produce X-ray contrast due to the transition metal’s relatively high atomic number.^15,18–20 Molybdenum is an essential element that has highly tolerated doses in the human diet and has been linked to reduced rates of cancer.^21–23 In addition to the demonstrated safety of molybdenum, the resulting MoS₂ NPs are lighter in color than other TMD NPs, which may make them more acceptable to patients. These properties led us to explore the use of MoS₂ NPs as a contrast agent.

For the successful translation of MoS₂ NPs as a contrast agent, it is important to determine not only their high contrast and biocompatibility but also their ability to be rapidly eliminated from the body.^24–26 Clinical translation of inorganic NPs for biomedical applications has been limited in part due to concerns regarding uptake by the mononuclear phagocyte system (MPS) and subsequent long-term retention in the body.^27,28 Long-term retention in organs and tissues may be associated with a higher risk of toxicity. Renal clearance is a rapid excretion process in which the NPs avoid the MPS and are filtered through the kidney’s glomerulus.²⁹ NPs designed to be renally clearable generally have an ultrasmall size (sub 5 nm) that is below the glomerular filtration threshold.^26,29,30 These NPs also incorporate surface coatings designed to avoid serum protein adsorption, which may trigger uptake by macrophages.

To support the search for a DEM-specific contrast agent, we present a unique aqueous synthesis for MoS₂ NPs with sub-5 nm core diameters. We synthesized and characterized a group of three MoS₂ NPs with different coatings: glutathione (GSH), penicillamine, and 2-mercaptopropionic acid (2MPA). GSH and penicillamine are zwitterionic small compounds that have previously been used to develop renally clearable NPs.^31–34 2MPA is a small carboxylic acid that has previously been used to coat NPs for biomedical applications with good long-term colloidal stability.^35–37 Here we present the synthesis and characterization of MoS₂ NPs, their contrast generation, in vitro and in vivo biocompatibility, and renal excretion. MoS₂ NP contrast agents were shown to generate higher contrast in DEM than iopamidol, an iodine-based X-ray contrast agent, and substantial contrast in CT. They were also shown to have low in vitro cell toxicity. Lastly, we report for the first time in vivo mouse imaging with 2MPA-MoS₂ NPs at a dose of 250 mg Mo/kg, which indicated rapid urinary excretion of the particles with biodistribution results indicating low retention in organs.

RESULTS AND DISCUSSION

Synthesis and Characterization of GSH-, Penicillamine-, and 2MPA-Coated MoS₂ NPs.

The challenge of synthesizing renally excretable MoS₂ NPs was developing NPs below the renal clearance size threshold of 5 nm, with coatings selected to avoid serum protein adsorption and provide colloidal stability. To synthesize MoS₂ NPs with these characteristics, we tested different synthetic methods, reaction conditions, and coatings (Table S1). We characterized the library of MoS₂ NPs and selected those that met the size and colloidal stability requirements to further advance their exploration as renally excretable X-ray contrast agents. A schematic of the syntheses leading to the formulations that met the previously described requirements is presented in Figure 1A. TEM micrographs of these GSH-, penicillamine-, and 2MPA-coated MoS₂ NPs are presented in Figure 1B–D. The average core sizes were 2.2 ± 0.7, 1.9 ± 0.6, and 2.6 ± 0.4 nm for GSH-, penicillamine-, and 2MPA-coated MoS₂ NPs, respectively. The z-potential shown in Figure 1E confirms the successful incorporation of the different coatings into the NPs. These values agree with other reports where GSH, penicillamine, and 2MPA coatings provide neutral to slightly negative surface charges.^38–40 A UV–vis spectrum for each sample, showing the characteristic peak of MoS₂ NPs at 215 nm, is presented in Figure 1F.⁴¹ To further confirm the incorporation of the different coatings, FT-IR was performed, and the spectra are presented in Figure 1G. From this data, we observe the characteristic peaks of each coating: 3300 cm⁻¹ band (−NH stretching) and 1650 cm⁻¹ peak (−COOH) characteristic of GSH; 3400 cm⁻¹ (O–H stretch), 1650 cm⁻¹ (−COOH), and 2950 cm⁻¹ (C–H) peak characteristic of penicillamine; and 3400 cm⁻¹ (O–H stretch) and 1650 cm⁻¹ (−COOH) characteristic of 2MPA.^42,43 In addition, we observe weak peaks at 465 cm⁻¹ in all samples, characteristic of Mo–S vibrations.^44,45 Moreover, the lack of a characteristic thiol band at 2500 cm⁻¹ suggests the successful incorporation of the coatings to the NP surface.

Figure 1.

Synthesis and characterization of GSH-, penicillamine-, and 2MPA-coated MoS2 NPs. (A) Schematic representing the aqueous synthesis of MoS₂ NPs with the different capping agents. Transmission electron micrographs of (B) GSH- (C) penicillamine- (D) 2MPA-coated MoS₂ NPs. Scale bar: 20 nm. (E) Z-potential, (F) UV–vis spectra, and (G) FT-IR spectra of MoS₂ NPs with different coatings. Z-potential data are presented as mean ± SD.

MoS₂ Nanoparticles Do Not Affect the Viability of Cells In Vitro.

We tested the in vitro cytocompatibility of these NPs with two of the cell types that we would expect these NPs to interact with in vivo, i.e., breast cancer and liver cells (other cell types that could be considered include endothelial and kidney cells). Figure 2 shows the viability of HepG2 (liver) and MDA-MB-231 (breast cancer) cells after 4 h of incubation with GSH-, penicillamine-, and 2MPA-coated MoS₂ NPs at concentrations ranging from 0–1 mg of Mo per mL. These results show the biocompatibility of the different types of MoS₂ NPs with the studied cell lines, even at the highest concentration of 1 mg/mL. To the best of our knowledge, this is the first study testing the cytocompatibility of these NPs up to a concentration of 1 mg/mL.^46–48 This concentration is higher than that required by most biomedical imaging applications (since the doses required by X-ray imaging are the highest of any biomedical technique), suggesting that the cytocompatibility of these NPs can lead to their use in other biomedical applications.^49–52

Figure 2.

Cell viability after incubation with GSH-, penicillamine-, and 2MPA-coated MoS2 NPs. Cell viability of HepG2 (blue) and MDA-MB-231 (orange) after incubation with (A) GSH-MoS₂ NPs (B) penicillamine-MoS₂ NPs and (C) 2MPA-MoS₂ NPs at varying Mo concentrations. Data are presented as the mean percentage of viability normalized to control ± SD n.s. = no statistical significance.

We elected to move forward with the 2MPA-MoS₂ NP formulation for further analysis due to the low molecular weight, demonstrated safety and stability, and low cost of 2MPA as a capping agent.³⁷ In addition, the small average core size (2.6 ± 0.4 nm) was expected to allow for both rapid renal clearance and substantial contrast generation.

MoS₂ Nanoparticles Generate Higher Contrast Than Iopamidol in DEM and Substantial Contrast in CT.

As previously discussed, molybdenum has been identified as a material with high contrast generation in DEM via simulations.¹⁴ To verify the results of these simulations and as an example, we used a DEM phantom to test the contrast generation of 2 nm 2MPA-MoS₂ NPs. This phantom is composed of tissue-mimicking materials that closely resemble a range from 100% glandular tissue to 100% adipose tissue in DEM. By inserting a sample in the phantom, we can evaluate its contrast at different breast density levels. Figure 3A shows a representative image of PBS, iodine control in the form of iopamidol at a concentration of 10 mg of iodine/mL, and 2MPA-MoS₂ NPs at different concentrations. For this image, the energies were set at 20 and 32 kV for the LE and HE, respectively. The DE subtraction image highlights the uniform contrast observed regardless of the density level in the phantom. Figure 3B shows the quantification of the absolute signal at the mentioned energy pair. This data shows higher signal generation, at the studied energy pair, for 2MPA-MoS₂ NPs than iopamidol, a clinically used X-ray contrast agent, at 10 mg of the element of interest/mL. The origin of the improved contrast of MoS₂ NPs in DEM is illustrated when comparing the attenuation coefficients of molybdenum and iodine at 30 keV (in the high energy window). The attenuation coefficient of molybdenum at that energy is 28.1 cm²/g, whereas that of iodine is 8.6 cm²/g.⁵³

Figure 3.

DEM and CT phantom imaging. (A) Representative images of PBS, iodine (10 mg/mL) in the form of iopamidol, and solutions containing NPs at varying Mo concentrations (1–10 mg/mL). Images were acquired at LE = 20 kV and HE = 32 kV using a custom-designed DEM phantom. (B) Quantification of the signal generation of the solutions described in panel A. Data is presented as mean absolute signal ± SD. (C) Representative CT phantom images of solutions containing NPs at varying concentrations and iodine in the form of iopamidol. These representative images were acquired at an X-ray energy of 80 kV. (D) Quantification of the attenuation rate from the solutions described in panel C, at varying X-ray energies. Data is presented as mean ± SD **** p < 0.0001. ** p < 0.01. n.s. = no statistical significance.

We further evaluated the X-ray contrast generation of the MoS₂ NPs by using a CT phantom designed to mimic the abdominal cavity of human patients. In this phantom, iodine (in the form of iopamidol) and 2MPA-MoS₂ NPs samples at concentrations ranging from 0 to 10 mg of the element of interest were submerged in 21 cm of water. We obtained images at clinically relevant X-ray energies of 80, 100, 120, and 140 kV. Representative images of the different samples, obtained at 80 kV, are presented in Figure 3C. Figure 3D shows the attenuation rate quantification of the different samples and shows that, while less contrast is obtained compared to iopamidol, MoS₂ NPs can generate substantial contrast in CT. This data highlights the contrast generation of MoS₂ NPs in different X-ray imaging modalities. Moreover, these findings highlight the potential use of MoS₂ NPs as DEM and CT contrast agents.

In Vivo Imaging and Biodistribution Show Rapid Clearance of 2MPA-MoS₂ NPs.

For clinical translation of NPs, they must be eliminated from the body in a reasonable amount of time. To test the clearance rate of 2MPA-MoS₂ NPs, tail vein injections were performed in female nude mice at a dose of 250 mg of Mo per kg of body weight. This dose was chosen to be in agreement with that used in other studies focused on contrast agents for dual-energy mammography.^13,31,54,55 CT scans of the mice were obtained preinjection and at 5-, 30-, 60-, and 120 min postinjection. Representative 3D reconstruction images of a mouse are presented in Figure 4A. Quantified attenuation of contrast in mouse organs is presented in Figure 4B. As time progressed, we observed increased NP accumulation in the bladder, where they were transported to be eliminated. This data supports the rapid renal clearance of the 2MPA-MoS₂ NPs.

Figure 4.

In vivo CT imaging and biodistribution. (A) Representative 3D CT image reconstructions of mouse injected with 2MPA-MoS₂ NPs at a dosage of 250 mg Mo/kg at pre- and 5-, 30-, 60-, and 120 min postinjection. Window width (WW): 630 HU and window level (WL): 900 HU. (B) Quantification of attenuation in mouse organs. (C) Biodistribution of 2MPA-MoS₂ NPs in organs of interest represented as percent of injected dose (ID). n = 3 per group. Data are presented as the mean ± SEM.

Given that in vivo imaging suggested rapid renal clearance of the nanoparticles, we further investigated the biodistribution of the nanoparticles after in vivo administration. Mice injected with the 2MPA-MoS₂ NPs were euthanized at 24 h postinjection, and their organs were collected for analysis. Figure 4C and Table S2 show the retained dose in different organs of interest. We did not see significant accumulation of NP in these organs. This data shows an extremely low average of 1.74% ID NP retention in these major organs thus suggesting extensive clearance at 24 h for 2MPA-MoS₂ NPs.

DISCUSSION

In this work, we developed and characterized 2 nm MoS₂ NPs with different coatings: GSH, penicillamine, and 2MPA. Formulations with small core sizes and capping ligands were developed to facilitate rapid renal excretion. All three MoS₂ NP formulations have high biocompatibility with HepG2 (liver) and MDA-MB-231 (breast cancer) cells, even at high concentrations of up to 1 mg/mL. Their excellent biocompatibility indicates their potential success in biomedical applications. In regard to X-ray contrast generation, the commercially available iopamidol contrast agents are not optimized for contrast generation in DEM and have adverse effects in some patients. MoS₂ NPs produce high contrast in DEM, outperforming iopamidol while also showing substantial contrast in CT. Due to the small size and low cost of 2MPA as a capping agent, we proceeded with 2MPA-MoS₂ NPs for analysis of their in vivo clearance and biodistribution. 2MPA-MoS₂ NPs were rapidly cleared from mice injected with a dose of 250 mg/kg. The rapid renal excretion decreases the level of NP accumulation in organs and tissues, further decreasing the concern for long-term toxicity. The high contrast generation, biocompatibility, and excretion of MoS₂ NPs can lead to their application as a DEM-optimized contrast agent.

Among the various potential biomedical applications of MoS₂ NPs, they have not yet been widely explored in X-ray imaging. To date, only two other studies have been published exploring the use of MoS₂ NPs as a CT contrast agent, although at very low doses.^46,48 In one study, MoS₂ nanohybrids composed of MoS₂ quantum dots and polyaniline were injected at a low dose (200 μL, 8 mg/mL) into breast cancer tumor-bearing mice via tail vein to evaluate their in vivo CT contrast.⁴⁶ This study showed an average change in attenuation of 108.7 HU 8 h postinjection. However, although the quantification was not presented, CT images reveal a high accumulation of the nanohybrids in the liver at 8 h postinjection. In another study, MoS₂ quantum dots were loaded into silica oxide nanoparticles as multimodal agents where a low dose of the nanoparticles (2 mg/mL, 200 μL) was administered to mice to test their in vivo CT contrast.⁴⁸ Maximum contrast was reported at 12 h postinjection with an average change in attenuation of 105.9 HU. Biodistribution experiments showed high nanoparticle accumulation in the liver and spleen of the mice. The low doses used in these studies, while relevant to other imaging and therapeutic applications, including fluorescence and photoacoustic imaging and photothermal therapy, are lower than those used in clinical CT imaging. Thus, these studies do not necessarily reflect the CT imaging capabilities of MoS₂ NPs or their safety in a clinical setting. Moreover, the overall large size of the nanoparticles reported in these studies leads to their accumulation in the liver and spleen, resulting in potential long-term toxicity concerns.

While the data presented in this study show the potential capabilities of ultrasmall MoS₂ NPs in biomedical applications for the first time, we recognize several limitations of this work. Biocompatibility analysis is limited to the two cell lines and the mouse model used. A broader assessment of biocompatibility and toxicology would be necessary for clinical translation. In addition, mice were sacrificed at 24 h postinjection with 2MPA-MoS₂ NPs. Though we do not expect long-term toxicity due to the demonstrated rapid clearance of MoS₂ NPs, a long-term analysis of safety would also be necessary for clinical translation.

In addition to providing higher contrast than iodinated agents, MoS₂ NPs were designed to be safer for those with renal insufficiency for whom iodinated agents are contraindicated. A 2.6 nm core MoS₂ NP contained 175 Mo atoms. Therefore, for an equivalent dose of contrast agent, 58 times fewer MoS₂ NP would be injected than an iodinated agent such as iopamidol (since iopamidol has three iodine atoms). This should place a lower burden on the kidneys in terms of the number of excretion events needed, so nanoparticles may be more compatible with patients with poor kidney function than iodinated agents. However, detailed studies will need to be performed to elucidate this point.

Other future experiments include analysis of the contrast generation of MoS₂ NPs in tumor-bearing mice using DEM or dual-energy CT to mimic optimal clinical imaging parameters. These experiments have the potential to further demonstrate the use of MoS₂ NPs as contrast agents for breast cancer screening.

CONCLUSIONS

In this work, we developed a novel aqueous synthesis, leading to 2 nm MoS₂ NPs with different coatings. Among these, we further characterized formulations with colloidal stability expected to have high biocompatibility: GSH-, penicillamine-, and 2MPA-coated MoS₂ NPs. All three types of particles have excellent biocompatibility with the studied cell lines even at high concentrations. As described above, to the best of our knowledge, this is the first study measuring the biocompatibility of these particles at concentrations in the mg/mL range. Moreover, we were able to confirm, for the first time, that MoS₂ NPs produce high contrast in DEM. 2MPA-MoS₂ NPs, at optimal energy pairs of 20 kV (LE) and 32 kV (HE), show higher contrast in DEM than the commercially available iopamidol X-ray contrast agent while also showing substantial contrast in CT. 2MPA-MoS₂ NPs were rapidly cleared from mice injected with a relatively high dose of 250 mg/kg. Together, the results highlight the biocompatibility, X-ray contrast generation, and excretion of MoS₂ NPs, and thus present a new potential application for MoS₂ NPs.

MATERIALS

Sodium molybdate dihydrate (Na₂MoO₄), 2-mercaptopropionic acid (2MPA), l-glutathione reduced (GSH), D-penicillamine, and sodium sulfide were purchased from Sigma-Aldrich (St. Louis, MO). Nitric acid (HNO₃) was purchased from Fisher Chemical (Thermo Fisher, Waltham, MA). Milli-Q deionized water (18.2 MΩ·cm) was used throughout the experiments.

METHODS

MoS₂ NP Synthesis and Purification.

To synthesize 2 nm MoS₂ NPs, the different ligands used were dissolved in 7 mL of DI water (30 mM GSH, 30 mM penicillamine, and 90 mM 2MPA). Next, 1 mL of a sodium molybdate aqueous solution (10.88 mg/mL) and 1 mL of a sodium sulfide aqueous solution (7.1 mg/mL) were added. The reaction was allowed to proceed at room temperature for 10 min. 0.5–1 kDa dialysis tubing was used to purify the MoS₂ NPs overnight. Three kDa molecular weight cutoff tubes were used to concentrate the GSH-MoS₂ NPs by centrifugation at 4000 rpm. The particles were concentrated to 1 mL, filtered using 20 nm filters, and stored at 4 °C until further use. The penicillamine- and 2MPA-coated MoS₂ NPs were filtered with a 20 nm filter following dialysis. To concentrate these, the nanoparticles were lyophilized for 5 days, resuspended in 1 mL of DI water, and stored at 4 °C until further use.

MoS₂ NP Characterization.

Transmission Electron Microscopy (TEM).

To measure the nanoparticle core size, we obtained TEM micrographs of the MoS₂ NPs were obtained. In brief, 10 μL of a sample concentrated to 0.5 mg/mL were dried into a carbon-coated copper grid with 200 mesh (Electron Microscopy Sciences, Hatfield, PA). The samples were imaged using a Tecnai T12 electron microscope operated at 100 kV. The core diameter was measured using ImageJ (National Institutes of Health, Bethesda, MA).

UV–Vis Spectrometry.

UV–vis spectra of the MoS₂ NPs were recorded using a Genesys 150 UV/vis spectrophotometer (Thermo Scientific, USA) from solutions of 5 μL of the concentrated MoS₂ NPs diluted in 995 μL of DI water.

Zeta-Potential.

Z-potential measurements of the MoS₂ NPs were acquired using a Zetasizer Nano (Malvern Instruments), using the solutions described above.

Fourier-Transformation Infrared (FTIR) Spectrometry.

For FT-IR analysis, samples were prepared by combining 100 mg of KBr with 5 mg of MoS₂ NPs. The samples were then dried in an oven and a pellet press was used to form the KBr pellets. The FT-IR spectra were obtained using a Jasco FT/IR 480 plus spectrometer.

Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES).

A Spectro Genesis ICP-OES instrument was used to determine the molybdenum concentration in each sample. ICP samples were prepared by digesting 10 μL of the concentrated MoS₂ NP solution in 500 μL of nitric acid for at least 1 h followed by addition of 9.49 mL of DI water to each sample.

In Vitro Biocompatibility Assays.

Cell Culture.

HepG2 (liver) and MDA-MB-231 (breast cancer) cell lines were cultured following the ATCC recommendations. In brief, 5% CO₂ and a temperature of 37 °C were used to incubate the cells. The cell media recommended for each cell line were supplemented with 10% fetal bovine serum (FBS) (Corning, Tewksbury, MA) and 1% penicillin-streptomycin (Thermo Fisher Scientific, Waltham, MA).

LIVE/DEAD Staining.

Cell viability studies using LIVE/DEAD staining were performed as described elsewhere.¹³ In brief, 24 h before the experiment, cells were plated at a density of 80,000 cells in a 35 mm glass bottom dish and kept in cell media. The next day, the cells were treated with MoS₂ NPs at concentrations ranging from 0 to 1 mg/mL for 4 h. Then, the cells were washed with PBS and incubated with 400 μL of a staining solution for 20 min. The staining solution contains Calcein AM, Ethidium-1 homodimer, and Hoechst 33342 for live cells, dead cells, and nucleus staining, respectively. Next, the cells were imaged by using a Zeiss Axiovert fluorescence microscope equipped with FITC, Texas Red, and DAPI filters. In each plate, we acquired 4 images per filter. The percentage of live cells compared to dead cells was determined using FIJI (ImageJ). The data was normalized to the control and is presented as % of control.

In Vitro Imaging.

Dual-Energy Mammography (DEM) Phantom Imaging.

DEM phantom imaging was performed as described elsewhere.¹³ In brief, PBS, iopamidol (10 mg/mL), and 2MPA-MoS₂ NP solutions of varying concentrations (0–10 mg/mL) were loaded into polyethylene tubes. These tubes were then inserted into a custom-designed phantom composed of tissue-equivalent materials mimicking breast tissue in a range from 100% glandular tissue to 100% adipose tissue. Images of the phantom were acquired, in triplicate, using a DE Hologic Selenia Dimensions mammography system at low energy (LE) of 20 kV and high energy (HE) of 32 kV. DE images of the phantoms were obtained by a weighted logarithmic subtraction of the HE and LE image pairs. A Gaussian denoising method was applied to the DE subtraction image.

Computed Tomography (CT) Phantom Imaging.

CT phantom images were acquired as described elsewhere.¹³ In brief, triplicate solutions of varying concentrations (0–10 mg of the element of interest/mL) of 2MPA-MoS₂ NPs, Na₂MoO₄, and iopamidol as a positive control were prepared in 300 μL tubes. Tubes with PBS were imaged as negative control. The tubes were placed in a plastic rack, covered with parafilm, and submerged in 21 cm of water to simulate the average human body thickness. CT images were acquired using a Siemens SOMATOM Force clinical CT scanner using the following parameters: tube voltages of 80, 100, 120, and 140 kV, 360 mAs tube current, 37 cm × 37 cm field of view, 0.5 mm thickness, and a 512 × 512 matrix. CT images were analyzed by drawing a region of interest (ROI) in the tubes using the OsirixMD software and recording their CT attenuation. The data are presented as the average of three slices per tube.

In Vivo Imaging.

Animal Model.

All animal procedures were performed following the Public Health Service (PHS) Policy on Humane Care and Use of Laboratory Animals (Public Law 99–158) per the University of Pennsylvania Guidelines for Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Pennsylvania under protocol number 805593.

Female nude mice were selected as the models for the experiments presented. Three mice were used. A dose of 250 mg of Mo per kg of body weight of 2MPA-MoS₂ NPs was injected into the mice via the tail vein.

Computed Tomography (CT) Imaging.

In vivo CT imaging was performed by using a MILabs micro-CT scanner (Utrecht, The Netherlands). Scans were acquired preinjection, 5, 30, 60 and 120 min postinjection. The imaging parameters used were as follows: 50 kVp, 210 uA, 75 ms exposure, 0.75-degree step, and 360-degree acquisition. The images were reconstructed as follows: 100 μm reconstruction with 150 um Gaussian postfilter. Analysis of scans was performed using OsirixMD: ROIs were created in organs of interest and the attenuation was recorded. The data is presented as the change in attenuation from preinjection images ± standard error of mean (SEM).

Biodistribution.

Female nude mice were euthanized using CO₂ gas for 5 min at 24 h postinjection. Organs including the lungs, kidneys, heart, spleen, and liver were collected and their weights were recorded. To prepare the organs and tissues for analysis, they were digested in 5 mL of nitric acid and placed overnight in an oven at 90 °C. The samples were then diluted with DI water to a final volume of 10 mL. ICP-OES was used to determine the molybdenum concentrations in each sample. Biodistribution data is presented as mean percentage of injected dose (%ID) ± SEM.

Statistical Analysis.

All experiments were performed at least three times independently. One-way ANOVA was used to compare the data in the DEM phantom study. Multiple t test (Welch test) was used to compare the data in the cell viability and CT phantom study. In all figures, data is presented as the mean value ± standard deviation (SD) or standard error of the mean (SEM), as specified on each figure legend. Statistical analysis was performed using Prism Graphpad 9 (San Diego, California USA) where p values <0.5 were considered statistically significant.