Introduction

Bismuth has long been utilized in pharmaceuticals to treat various conditions, including gastrointestinal disorders, hypertension, and syphilis. Its nanostructured forms have recently gained attention due to their unique properties, such as high stability, large surface area, strong diamagnetism, and significant electrical resistivity and magnetoresistance when exposed to magnetic field. Additionally, bismuth exhibits desirable catalytic activity, is easy to functionalize, cost-effective, and chemically inert. It also has a high X-ray attenuation coefficient, strong near-infrared (NIR) absorbance, and excellent photothermal conversion efficiency. These properties have led to its use in a variety of modern applications, including X-ray radiotherapy, biosensors, heavy metal ion detection, antimicrobial treatments, parasitic disease management, combined cancer therapies, bioimaging, and tissue engineering1,2,3,4,5,6. For these purposes, single-component bismuth nanoparticles (BiNPs) can be utilized as precursors to develop other types of BiNPs, including bismuth oxyhalides (BiOX, where X is Cl, Br, or I) and bismuth chalcogenides like bismuth oxide (Bi2O3), bismuth sulfide (Bi2S3), bismuth selenide (Bi2Se3), and bismuth telluride (Bi2Te3)1. The small size of nanoparticles allows more efficient targeting of diseased tissues by accumulating in inflamed and damaged epithelial cells through the enhanced permeation and retention effect. This seems to be a factor of importance in intestine disease treatment2 and cancer therapy7 connected with using small nanoparticle concentration with photothermal therapy, photoacoustic, and photodynamic therapy. In such cases, the low toxicity of nanoparticles was described. However, its toxicity is still under investigation. The good results of combined methods in cancer treatment could suggest the biocompatibility of BiNPs and the small effectiveness of their low concentration. When the concentration was higher the more frequent presence of trafficking nanoparticles was observed in some organs like the liver, spleen, and lungs in animals8. Slower elimination from the body after exposure to BiNPs, which means longer contact with metal ions of tissues, was also found to be associated with persistent neutrophilic and reactive oxygen species (ROS) generation. The inflammatory markers along with ROS generation further promoted oxidative stress and exaggerated additional inflammatory pathways leading to cell death1. These side effects were not observed following single-dose exposure in live animals, as tissue regeneration remained intact. It suggests, similarly to other types of nanomaterials, that BiNPs possess dose-dependent adverse biological effects, and their cytotoxicity is largely dependent on particle diameter, charge, surface modifications, and time exposure9. In this context, comparing the responses of healthy fibroblasts and cancer cells to varying concentrations of BiNPs appears to be a valid approach. For our research we choose bismuth selenide (Bi₂Se₃) nanoparticles due to their unique combination of optical, electrical, and thermal properties, making them promising candidates across various technological domains, including biomedical applications. As a topological insulator, Bi₂Se₃ exhibits exceptional electronic properties, such as high electron mobility and a significant bulk bandgap (~ 0.3 eV), which are leveraged in fields like quantum computing and spintronics10,11. In biomedical applications, their strong optical absorption and tunable bandgap have been explored for bioimaging, photothermal therapy, and drug delivery, with studies demonstrating their efficacy in enhancing therapeutic precision and minimizing side effects12,13,14,15. Moreover, their low thermal conductivity and nanoscale tunability make them suitable for targeted thermoelectric therapy and hyperthermia treatments6,16,17,18. Additionally, Bi₂Se₃ NPs have shown promise as contrast agents for imaging technologies, including photoacoustic imaging, due to their strong optical contrast and biocompatibility19. These applications highlight the significant potential of Bi₂Se₃ nanoparticles in driving advancements in both medical and technological fields.

In the absence of a standardized treatment protocol involving bismuth selenide nanoparticles, evaluating the cytotoxicity of this compound against cancer cells is fully justified. Given that tongue cancer is among the most disfiguring malignancies, the study included a comparative analysis of SCC-25 tongue squamous carcinoma cells and human gingival fibroblasts based on mitochondrial status. Mitochondria are organelles where apoptotic signals from various pathways–activated by factors such as hypoxia, heat shock, or deprivation of growth factors are integrated. They play a central role in coordinating and amplifying the signal for programmed cell self-degradation20.

Experimental method

Reagents and materials

All chemicals utilized in the synthesis of Bi2Se3 NPs, including bismuth nitrate pentahydrate (99.99% purity), elemental selenium powder (99.99% purity), N, N-dimethylformamide (DMF) (99.8% purity), and hexadecyltrimethylammonium bromide (CTAB) (analytical grade) were purchased from Sigma-Aldrich. Ethanol (99.8% purity, analytical grade) was purchased from Pol-Aura (Poland), and sodium hydroxide (NaOH) (analytical grade) was purchased from Warchem (Poland). All these chemicals were used without any further purification.

Synthesis of Bi2Se3 nanoparticles

Bismuth selenide NPs were synthesized using a solvothermal method with a 2:3 molar ratio of bismuth nitrate pentahydrate to elemental selenium powder. DMF served as the solvent (80 ml), and 2 mmol of CTAB was used as a capping agent. NaOH, in a quantity of 4.5 mmol, was used as a reducing agent, with the pH carefully maintained at 10. All reagents were dissolved in DMF and sonicated at room temperature for 1 h. The resulting mixture was transferred to a 100 ml Teflon®-lined container, filled to 80% of its capacity, and then sealed within an autoclave. The autoclave was heated in an oven at 140 °C for 24 h. After the reaction, the solution was allowed to cool inside the autoclave for 24 h. The nanoparticles were then collected by centrifugation, dried at 80 °C for 90 min, and subsequently annealed at 250 °C for 6 h. The prepared material was in the form of a powder, which was then subjected to detailed analysis. The synthesis procedure is shown in Fig. 1.

Fig. 1
figure 1

Preparation of Bi2Se3 nanoparticles.

Full size image

Characterization of Bi2Se3 nanoparticles

X-ray diffraction (XRD) analysis

The XRD patterns of Bi2Se3 NPs powder were recorded with an Empyrean (Malvern Panalytical) powder diffractometer using Ni-filtered CuKα radiation (wavelength 1.54060 Å) and an X’Celerator detector. Diffractograms were recorded at room temperature, in the 20–80° angular range, using a constant step of 0.004° and a 60 s exposure time. Phase analysis was carried out using High-Score Plus XRD Analysis Software (Malvern Panalytical) with the ICDD PDF-4 + database.

Scanning electron microscope (SEM) analysis

The Bi2Se3 NPs powder morphology was studied with an MIRA3 scanning electron microscope (Tescan, Czech Republic). The nanoparticle powder was observed with an accelerating voltage (HV) of 12 kV and a working distance (WD) of 5 mm. The observations were performed using an In-Beam detector compared to secondary electrons (SE). A thin carbon coating layer with a thickness of approximately 10 nm was deposited on samples using the Jeol JEE 4B vacuum evaporator.

Transmission electron microscope (TEM) analysis

The morphology of Bi2Se3 NPs was characterized by TEM using a JEM-1400 Flash Electron Microscope (JEOL) (120 kV). The powder was dissolved in ethanol and sonicated for 30 min. The TEM measurements were performed by depositing a colloidal dispersion of Bi2Se3 NPs by drop casting onto a copper grid covered with a thin carbon layer and placed on a vacuum desiccator overnight.

Particle size and zeta potential

The electrokinetic potential and particle size distribution of Bi₂Se₃ NPs were analyzed using a Malvern Zetasizer Nano-ZS (Malvern Instruments Ltd.) based on dynamic light scattering (DLS) and laser Doppler electrophoresis. Measurements were performed at room temperature with the nanoparticles suspended in a suitable solvent to ensure colloidal stability. Disposable folded capillary cells were used for both zeta potential and size measurements, providing accurate data on the stability and polydispersity of the Bi₂Se₃ nanoparticle suspension.

Cell culture

In this study, human gingival fibroblasts and human squamous carcinoma cells (SCC-25, CRL-1628™, ATCC) were used. Gingival fibroblasts were extracted from gingival tissue collected from five generally healthy patients during a routine surgical procedure. This research was carried out with the approval of the bioethics committee (Consent no. 150/17 from 02/03/2017 by decision of the Bioethics Commission at the Medical University of Karol Marcinkowski in Poznan), have been performed in accordance with the Declaration of Helsinki, and all subjects gave their informed consent for inclusion before they participated in the study. Gingival fibroblasts were transferred under sterile conditions from a freezing medium containing Dulbecco’s modified Eagle’s medium (DMEM)/F12 (1:1), 10% fetal bovine serum (FBS), and 10% dimethyl sulfoxide (DMSO). The cell culture was maintained in DMEM/F12 medium (1:1) supplemented with 10% FBS and 1% antibiotic/antimycotic solution (50 µg/ml gentamicin sulfate, 100 IU penicillin, and 50 µg/ml streptomycin). Cultivation occurred at 37 °C in a humid environment with 5% CO2, under sterile conditions. The medium was refreshed every 3 days. When cultures reached 70–80% confluence, they were passaged using a trypsin–EDTA solution (0.25% trypsin, 0.02% EDTA).

Before the addition of bismuth selenide NPs, the cells were detached from TC flasks via trypsinization and resuspended in an un-supplemented medium to achieve a concentration of 5 × 104 cells/ml. A 100-μl aliquot of the cell suspension was then transferred to 96-well plates. Cell suspensions exposed to various concentrations of bismuth selenide, along with control groups without the additive, were analyzed using the Scepter™ 2.0 Cell Counter. Cultivation in the 96-well plates was carried out for 24 h before adding bismuth selenide. The experiment assessed the impact of different concentrations of bismuth selenide (6.25 µg/ml, 12.5 µg/ml, 25 µg/ml, 50 µg/ml, 100 µg/ml, 200 µg/ml) on the viability of gingival fibroblast cells and SCC-25 cells after 24, 48, and 72 h of incubation at each concentration. All experiments were conducted in three replicates.

Cell viability assay (CCK-8 assay)

Following each incubation period, the mitochondrial activity of fibroblast cells and human squamous carcinoma cells (SCC-25) was evaluated using the Cell Counting Kit-8 (CCK-8) assay. This assay facilitates convenient testing with WST-8 (2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2Htetrazolium monosodium salt), which, in the presence of an electron carrier, 1-Methoxy PMS, is reduced by cellular dehydrogenases to form a water-soluble orange formazan dye. The amount of formazan produced is directly proportional to the number of viable cells. The Cell Counting Kit-8 enables sensitive colorimetric assays to determine the number of viable cells in proliferation and cytotoxicity assays. Measurements were taken for each 96-well plate at 24, 48, and 72 h after the addition of bismuth selenide. The CCK-8 assay procedure was as follows: 10 μl of CCK-8 reagent was added to each well, and the plate was incubated for 2 h at 37 °C. The absorbance was then measured at 450 nm using an Infinite M200 PRO microplate reader (Tecan). The cell number results obtained from the CCK-8 assay were statistically analyzed and presented as bar graphs. The study of changes in cell morphology of both cell lines was performed using the Axiovert 200 inverted light microscope (Zeiss) and ZOE Fluorescent Cell Imager device (Bio-Rad).

Results

X-ray diffraction (XRD)

In recent years, there has been significant interest in utilizing coordination compounds for the synthesis of nanomaterials21,22. The use of novel compounds offers great potential, as they can provide innovative approaches for synthesizing nanomaterials while enabling precise control over the shape and size distribution of nanostructures. Powder XRD patterns are employed to analyze the structural parameters and assess the phase purity of Bi₂Se₃. The XRD pattern of the samples was recorded over a scanning angle range of 20°–80°. The X-ray diffraction data were analyzed using Origin 2018 software. Various peaks were identified and analyzed to uncover distinct characteristics of the nanoparticles. Using Origin software, these peaks were labeled accordingly. The positions of the Bragg peaks were compared with the reference peaks listed in the Joint Committee on Powder Diffraction Standards (JCPDS) database to confirm the synthesized materials. All the peaks were indexed according to the crystal structure. The XRD pattern for the Bi₂Se₃ sample is shown below in Fig. 2. Nine XRD peaks were observed at 2θ values of 25.00°, 27.87°, 29.35°, 32.47°, 47.84°, 53.54°, 57.53°, 68.79°, and 77.96°. The corresponding planes for these peaks were identified as (101), (104), (015), (107), (116), (205), (1016), (0021), and (2110), respectively. This confirms the formation of a hexagonal structure with the space group R-3 m (JCPDS reference code 00-033-0214), consistent with the standard values from the International Center for Diffraction Data (ICDD) reference code 04–008-3239. The (107) plane, observed at a Bragg angle of 32.47°, exhibits a strong diffraction intensity, indicating that the preferred orientation of bismuth selenide is along the [107] direction. The spectra show no additional peaks for Bi or Se, confirming that all the elements have reacted completely to form the Bi₂Se₃ phase. From the XRD data, the crystallite size (D) of the Bi₂Se₃ nanoparticles was calculated to be 15 nm using the Scherrer equation (Eq. 1), as shown in Table 123,24,25:

$$D = \frac{k\lambda }{{\beta cos\theta }}$$
(1)

here \(\beta\) represents the width of the observed diffraction line at half its maximum intensity, \(k\) is the shape factor, typically valued at around 0.9, and \(\lambda\) denotes the wavelength of the X-ray source used in the XRD analysis25,26. The lattice constants were calculated using the equation:

$$\frac{1}{{d^{2} }} = \frac{4 }{3}\left[ {\frac{{h^{2} + hk + k^{2} }}{{a^{2} }}} \right] + \frac{{l^{2} }}{{c^{2} }}$$
(2)
Fig. 2
figure 2

XRD pattern of Bi2Se3 NPs powder showing the characteristic peaks with assigned crystallographic planes.

Full size image
Table 1 Values of crystalline size (D), lattice constant, x-ray density (dX) and volume of the cell of Bi2Se3 sample (V).
Full size table

The obtained values for the lattice constants (\(a\) and \(c\)) and the unit cell volume (V) of bismuth selenide with a hexagonal structure (a = b ≠ c, α = β = 90º, γ = 120º) were calculated and are provided in Table 1.

Scanning electron microscopy (SEM)

Figure 3a shows SEM images of the microstructure of the prepared Bi₂Se₃ sample. It is observed that the bismuth selenide sample exhibits an irregular shape, with agglomeration and a non-uniform surface27. Using ImageJ software, the typical powder grain size of the sample was calculated to be 0.4 μm28, as shown in Fig. 3b. This result aligns with previously published findings25.

Fig. 3
figure 3

(a) SEM image of Bi2Se3 nanoparticle powder (SEM HV: 12 kV, WD: 5 mm, detector: In-Beam SE) (b) histogram of SEM image.

Full size image

Transmission electron microscopy (TEM)

Due to the limited spatial resolution of the SEM technique, observing the crystallite grains that form the nanoparticles is significantly constrained. To investigate the morphology, shape, and size of the Bi₂Se₃ NPs, the TEM method was employed. A representative TEM image of a Bi₂Se₃ NPs is shown in Fig. 4a. The histogram in Fig. 4b illustrates the particle size distribution of Bi₂Se₃, with an average particle size of 9.35 nm29. It is evident that the nanoparticles consist of numerous homogeneous crystallite grains with random orientations, and their dimensions are consistent with those calculated using the XRD method.

Fig. 4
figure 4

(a) TEM image of Bi2Se3 nanoparticles composed of crystalline grains (b) histogram of TEM image.

Full size image

Particle size and zeta potential

Bi₂Se₃ samples were prepared in a 10% DMSO solution (v/v, analytical.

grade) and subjected to sonication for over 2 h to achieve a homogeneous suspension. Dynamic light scattering (DLS) measurements were performed on three independent samples. The size distribution obtained were 167.3 nm, 180.4 nm, and 204.2 nm, respectively. The average particle size was 184 ± 15 nm. The polydispersity index (PDI) values were 0.281, 0.381 and 0.474, indicating a moderate level of particle size heterogeneity.

During observation, fast sedimentation was noted, suggesting low nanoparticles stability in the presence of 10% DMSO. Zeta potential measurements showed consistently negative surface charges: –16.0 mV, –16.6 mV, and –16.3 mV. The average zeta potential was 16.3 ± 0.3 mV.

Cytotoxicity

The influence of bismuth selenide on SCC-25 and human gingival fibroblasts is presented in Figs. 5 and 6. The addiction of all prepared bismuth selenide concentrations (6.25 µg/ml, 12.5 µg/ml, 25 µg/ml, 50 µg/ml, 100 µg/ml, 200 µg/ml) caused a significant viability to decrease of cells in the following days of observation compared to the control group (without bismuth selenide addiction). In the case of the cells examined 24 h after Bi2Se3 NPs addiction, for all concentrations used, the SCC-25 and the gingival fibroblast cells viability decreased on average 6.6 times and 8.0 times, respectively, compared to the control. The corresponding decreases were 6.2 times and 5.2 times for the cells examined 48 h after the nanoparticles addiction, and 5.9 times and 7.6 times for the examination 72 h after the addiction. The changes in cell morphology of both cell lines (SCC-25 and gingival fibroblasts) observed in both inverted light microscope and fluorescent cell imager device revealed the death of cells 24 h after the addition of Bi2Se3 NPs even at the smallest concentration used (Fig. 7).

Fig. 5
figure 5

Analysis of SCC-25 cell viability measured 24 h, 48 h, and 72 h after bismuth selenide addiction. Data are expressed as an absorbance value of the control (no addiction) and the mean of pooled results from experiments performed in triplicate. The lowercase superscript letters illustrate significant differences between the samples (α = 0.05).

Full size image
Fig. 6
figure 6

Analysis of gingival fibroblast cell viability measured 24 h, 48 h, and 72 h after bismuth selenide addiction. Data are expressed as an absorbance value of the control (no addiction) and the mean of pooled results from experiments performed in triplicate. The lowercase superscript letters illustrate significant differences between the samples (α = 0.05).

Full size image
Fig. 7
figure 7

(a) Human squamous carcinoma cells (SCC-25, CRL-1628™, ATCC) before bismuth selenide addiction. (b) Human squamous carcinoma cells (SCC-25, CRL-1628™, ATCC) 24 h after bismuth selenide 6.25 µg/ml addiction.

Full size image

Discussion

Nanomedicine is an emerging field focused on the personalization of therapy and the targeted delivery of therapeutic compounds directly to the site of action, at doses that minimize toxicity to surrounding tissues. Metal-based nanoparticles represent a promising area within this field. The synthesis of particles with precise and expected size appears to be a key factor in achieving effective therapeutic outcomes30,31,32.

The basic method for Bi2Se3 nanoparticles’ size analysis is SEM, however, the resolution of this method is generally too small to observe their grainy structure below 500 nm, which is also visible in our study (Fig. 3a,b). For grainy structure analysis transmission electron microscopy (TEM) is dedicated33. A typical TEM analysis of a nanoparticles powder allows the observation of crystallite grains with sizes smaller than a dozen nanometers. In our investigations, the nanoparticles were formed from such crystallite grains. It was also possible to observe that the nanoparticles were mostly interconnected into clusters of different shapes and sizes, which can be important for biological activity and the ability for transmembrane diffusion. As smaller the cluster is, the better is its biological activity. In our study, nanoparticles of different shapes resembling nanoflakes or nanoflowers were observed. TEM analysis revealed that the Bi2Se3 NPs are composed of crystallite grains with sizes of the order of 9.35 nm (Fig. 4a,b) which is consistent with the XRD results. Regardless of the size of the nanoparticles (0.4 μm), their grainy structure and the resulting roughness of the surface may influence the effectiveness of their interaction with biological structures.

It is worth mentioning that the increase in the size of the crystallite grains can be attributed to an increase in the synthesis temperature. Bi2Se3 powder was reported to have a larger crystallites size after hydrothermal synthesis than the size of crystallite grains in our case34. Moreover, upon increasing the temperature from 130 to 200 °C, the nanocrystals grew from 30 to 100 nm in diameter.

The size of Bi2Se3 NPs determined in the TEM study as well as their purity and the size of the crystallite grains that form the NP might have an impact on the cytotoxicity bismuth selenide.

The zeta potential is an important indicator of the surface charge and stability of solid nanoparticles in a suspension. Higher absolute values of the zeta potential are associated with stronger electrostatic repulsion which typically indicates better stability of the suspension and prevents aggregation.

The measured zeta potential was relatively low, which may slightly influence the cellular response. In our case, the high cytotoxicity of the nanoparticles against Scc-25 cells and gingival fibroblasts suggests that it may be related to the small negative zeta potential value.

The type of cancer may also influence the cell response to chemical compounds. In our study, we used the tongue cancer cell line while the lung adenocarcinoma cell line or colon carcinoma cell line was mostly described in the literature35,36. Although we used the same concentration of bismuth selenide nanoparticles as You et al.35, we obtained contrasting results. After 24 h incubation SCC-25 cell viability decreased on average 6.6 times and gingival fibroblasts cell viability decreased on average 8 times, while You et al. concluded that bismuth selenide nanoparticles were nontoxic for A549 cells even at concentrations up to 200 µg/ml. It can relate to nanoparticle size as mentioned above (in our study it was about 0.4 μm and in another study 2.750 μm) but also might suggest the application of individual dosages for different types of cancer. The second hypothesis is confirmed by Shakibaie et al.36, who found the decreased viability of HT-29 cells above 10, 20, 30, 70, and 80% after 24 h observation using 5, 10, 20, 40, and 80 µg/ml concentrations of bismuth nanoparticles (without specification), respectively, and about 20% more for each concentration after 48 h observation. In our study similar results, we obtained after 24, 48, and 72 h of observation, and relative to fibroblasts the viability lost was higher than to cancer cells. In our research, we used mitochondrial activity as a marker of both line cell’s vitality. Mitochondria are core regulators of tumor cell homeostasis, and their damage has become an arresting therapeutic modality against cancer. The results we obtained, indicate the mitochondrial toxicity of Bi2Se3 NPs. To evaluate the mitochondrial activity of fibroblasts and cancer cells, we employed an assay that measures the ability of mitochondrial dehydrogenases to convert WST-8 into formazan. Mitochondrial activity directly relates to the applied cell viability assay, which is based on the conversion of tetrazolium salt by cellular dehydrogenases to a water-soluble orange formazan dye. The amount of formazan produced is directly proportional to the number of viable cells. Compared to other assays that use different tetrazolium salts (such as MTT, XTT, MTS, or WST-1), the CCK-8 assay offers higher detection sensitivity and lower toxicity. The lower impact of Bi2Se3 NPs on cancer cells we observed might be compared to the current research about the function of mitochondria in tumor metabolism37,38. Upon induction of mitochondrial apoptosis, mitochondrial outer membrane permeabilization usually commits a cell to die. However, the new findings demonstrate that during apoptotic process besides eliciting caspase activation, mitochondrial outer membrane permeabilization engages various pro-inflammatory signaling functions, which may also have non-lethal functions39. In healthy cell the irritation of mitochondria results in co-aggregation. In human cancer mitophagy as a more specialized autophagy is observed, what prevents the accumulation of damaged mitochondria37. On the other hand, the oral cancer is recognized as having specific lipid metabolism and what is more interesting the more active process of mitochondrial fission than fusion. The mitochondrial fission supports the survival of cancer cells40. Our results also suggest some clinical implications. Bi2Se3 nanoparticles in used size should be rather applied as external treatment, what is in similar with observation done for bismuth lipophilic nanoparticles41 and are in concordance with proposal that proper use of nano-carriers with more understanding of the interaction between them and internal body, in the in vivo environment, might mitigate potential unforeseen complications during clinical trials42.

Conclusions

This study highlights the cytotoxic behavior of Bi₂Se₃ NPs, particularly in tongue cancer cells (SCC-25) and gingival fibroblasts, where significant viability reductions were observed after 24 h of incubation. In contrast to previous studies reporting no toxicity in A549 lung adenocarcinoma cells at similar nanoparticle concentrations, our findings suggest that the smaller nanoparticle size (0.4 μm vs. 2.750 μm) plays a key role in enhanced cytotoxicity. TEM and SEM analyses confirmed 0.4 μm and 9.35 nm powder grain (NP) size and crystallite grain in our samples, indicating. in relation to other reports, that synthesis conditions, especially temperature, influence grain size and biological activity. The results also reveal differential sensitivity among cell types, with gingival fibroblasts showing higher vulnerability, possibly due to increased nanoparticle interaction and mitochondrial toxicity. These findings emphasize the importance of optimizing nanoparticle size, dosage, and cell-specific targeting strategies for safe and effective cancer therapies.