Antibacterial and Antitumor Application of Carbon Dots Based on Natural Products for Photodynamic/Photothermal Effects

Introduction

CDs comprise zero-dimensional (0D) quasi-spherical nanomaterials with sizes below 10 nm.^1,2 These materials exhibit remarkable properties such as low-cost, biodegradability, photo and chemical stability, fluorescence, and a strong capacity for light absorption.^3–5 These and more interesting characteristics have promoted their thorough study since 2004, when they were first reported.⁶ Less than 250 CD-related works were published that year. Nevertheless, the number of studies significantly increased in 2018, almost twenty times higher.⁷ Therefore, CDs have been employed in several fields like solar cells,⁸ catalysis,⁹ metal sensing,¹⁰ biological sensing,¹¹ drug delivery,¹² bioimaging,^13,14 and so on, owing their versatile and tunable optical and physicochemical properties.^15,16

Several routes, either top-down or bottom-up, have been employed to prepare CDs.^17–19 Within this subject, the microwave-assisted method under the bottom-up approach is gaining rapt attention since it consists of a direct synthesis, also called one-pot or single-step synthesis.²⁰ This route only requires a few minutes because microwave irradiation induces faster rates of chemical reactions.⁷ In addition, it allows more uniform heating, which often diminishes the formation of by-products and simplifies the purification steps.²¹ Likewise, bio-inspired trends that employ natural green precursors have complemented the microwave-assisted route.²² Carbon-containing sources from parts of plants such as peels, herbs, roots, leaves, flowers, and fruit can be harnessed as bio-starting materials by this cost-effective methodology.^23–25 These natural products are rich in organic compounds that can provide surface functional groups with antimicrobial and anticancer properties.^26–30 The most common functional groups identified in the CD surface comprise N-H, C=O, -OH, -COOH, C-N, and C-O.^31,32 Besides, the organic molecules can contribute as stabilizing, reducing or capping agents.³³ Considering that these species have heteroatoms, the obtained CDs are mainly constituted not only by carbon (C) and oxygen (O) but also by nitrogen (N) or sulfur (S).^34–36 Nonetheless, it is a fact that the final CD characteristics depend on the synthesis conditions and the precursor nature. Regarding the latter, its specific composition directly impacts the CDs.³⁷

This study used annatto, cinnamon, and curcumin as natural carbon sources regarding their antimicrobial activities, for example that of the annatto has been attributed to phytochemicals alkaloid, phenol, tannins, saponins, mono, and sesquiterpenes.^38,39 The antimicrobial property of cinnamon is based on bioactive phytochemicals such as cinnamaldehyde and eugenol.⁴⁰ Turmeric has been demonstrated that its main components, curcumin and curcumin derivatives, also present antioxidant, anti-inflammatory, antiviral, antifungal, and antimalarial activities.^41,42

It is important to note that this trend of using natural extracts to synthesize CDs is not new. Several works have proposed the preparation of CDs from these kinds of extracts, for example, from leaves of bamboo or paradise tree, flowers of Magnolia liliiflora, and food like tomato and Solanum lycopersicum,^43–45 among others.

Regarding CD properties of author interest, they exhibit strong optical absorption, ie, when irradiated by a light source in the ranges of UV, visible, or even NIR, they emit light and possess tunable excitation/emission properties.^46,47 The emission response will depend on the CDs’ specific features, such as size, composition, presence of surface functional groups, and doping degree. Nevertheless, it is worth noting that the CD emission is also influenced by environmental conditions (temperature, pH, etc.).⁴⁷ The irradiated CDs reach an excited state that can generate reactive oxygen species (ROS), heat, or both.^48,49 Consequently, CDs have been considered excellent candidates as photosensitizers (PS) and photothermal agents (PA) within photodynamic (PDT) and photothermal (PTT) therapies, correspondingly.⁵⁰ Several authors have focused on this field; thus, CDs have been employed by PDT and PTT to combat bacteria and cancerous cells since these photo-activated techniques are considered low side effects and minimally invasive.^51,52

With respect to bacteria, they can cause severe infectious diseases and even death if not treated correctly.^53,54 The main challenge in fighting these microorganisms is their drug resistance, which is triggered by the overuse of antibiotics.⁵⁵ Hence, conventional treatments become ineffective and require higher doses or repeated administration of drugs, which could increase the risk of unwanted side effects.⁵⁶ Resistant Staphylococcus aureus (S. aureus) and Escherichia coli (E. coli) are the two main causes of healthcare-associated infections.⁵⁵ For example, De Kraker et al determined high excess mortality associated with infections produced by methicillin-resistant S. aureus and third-generation cephalosporin-resistant E. coli.⁵⁷ Thus, the impacts that antibiotic-resistant infections cause on health systems have concerned several researchers and have motivated the exploration of alternative treatments against bacteria.^58–65 In this context, the application of CDs within PDT/PPT has emerged as a novel solution. Yan et al and Wen et al demonstrated that a CD nanocomposite (N, S-CDs with curcumin) presented PDT and PTT properties triggered by a dual-wavelength illumination (808 and 405 nm).^66,67 In addition, these nanocomposites showed excellent biocompatibility since they exhibited low cytotoxicity and negligible hemolytic activity. The researchers conducted successful antibacterial tests against E. coli and S. aureus (61,62), concluding that N-CDs and curcumin showed great potential in a multimodal PDT synergistic PTT treatment platform. Likewise, Mei et al prepared a nanocomposite of chitosan oligosaccharide functionalized with graphene quantum dots. They proved that three types of antibacterial activity: photodynamic, photothermal, and chemical can kill Gram-positive and Gram-negative bacteria at the same time under irradiation. This combination was also confirmed by treatment of a bacteria-infected wound of a rat.⁶⁸ With the same purpose, Lv. et al synthesized a nanocomposite with methylene blue and carbon quantum dots, concluding that these results encourage the usage of CDs in clinical applications.^60,69

On the other hand, regarding antitumor applications of CDs, it must be highlighted that cancer comprises a threat to human life. In 2020, ~10 million deaths were reported worldwide by the International Agency for Research on Cancer. According to Miller et al, the most prevalent cancers are breast, uterine corpus, and colon and rectum among females, while prostate, skin melanoma, and colon and rectum among males.⁷⁰ Generally, cancer diseases are treated by aggressive procedures such as surgery, radiation, and systemic treatment, which comprise immunotherapy, chemotherapy, targeted and hormonal therapies. Nevertheless, more targeted and less toxic treatments like PDT and PTT, which exhibit reduced long-term morbidity, have also been proposed to fight this illness.⁷¹ In PDT, the abnormal cells could be inhibited by apoptosis, necrosis, or autophagy when ROS are generated, whereas in PTT, the cancerous cells are destroyed by hyperthermia owing to their low tolerance to heat.⁷² To cite a few examples, the study by Yue et al prepared an injectable hydrogel by adding nitrogen-rich CDs (N-CDs). They proved that the N-CDs acted as a PS and an efficient cross-linking agent to form a hydrogel network. Besides, after a laser irradiation of 660 nm, the hydrogel showed a high antitumor effect, referring to the in vitro and in vivo experiments.⁷³ Likewise, Zhang et al prepared a biodegradable poly(γ-glutamic acid) along with glucose oxidase and CDs for simultaneous PDT and PTT under 730 nm laser irradiation. This composite showed a long retention time at the tumor acidic microenvironment and could further target cancer cells. The authors concluded that it is a promising approach for metastasis inhibition.⁷⁴ Thus, several related studies have been carried out in recent years. Lagos et al prepared a review summarizing the main results of some carbon-based materials, including CDs, within PDT and PTT applied to destroy tumors.⁵²

In this manner, PDT and PTT have become desirable therapeutic approaches to battle bacteria and cancerous cells using a variety of PS and PA, respectively. Among these materials are the CDs.

In this work, different CDs were synthesized from water dispersions of natural carbon sources: (i) Bixa Orellana L. (annatto-herbarium collection code QCA237394, identified by botanist Dr. Álvaro J. Pérez), (ii) Cinnamomum verum J. Presl (cinnamon - this specie is native to tropical Asia, especially Sri Lanka. It is not native to Ecuador. Although it is possible that specimens of this species exist in Ecuadorian botanical collections, no specific information on its presence was found in the National Herbarium of Ecuador), (iii) Curcuma longa L. (turmeric or curcumin - herbarium collection code QCA238186, identified by the botanist Dr. Álvaro J. Pérez) and (iv) sucrose. The nanomaterials obtained in this study were called as CDs1, CDs2, CDs3, and CDs4, from annatto, cinnamon, curcumin, and sucrose, respectively. The shape and size of all the CDs were studied by transmission electron microscopy (TEM). The hydrodynamic diameter was determined by dynamic light scattering (DLS), and the chemical composition, referring to CD functional groups, was analyzed by Fourier-transform infrared (FTIR) and Raman spectroscopies. Luminescence emission measurements were also carried out. Besides, the cytotoxicity, photodynamic and photothermal capacities, along with the antibacterial and antitumor properties were also assessed.

Materials and Methods

Synthesis of CDs

CDs were prepared in an open-vessel microwave (General Electric microwave model JES70W) using simple dispersions based on natural products such as annatto, cinnamon, curcumin, and sucrose. First, 100 g of each natural product was poured and mixed into 150 mL of deionized water. Then, the dispersions were irradiated in a microwave with an energy power of 700 watts for 7 min. The color of the liquid changed from colorless to pale yellow, strong yellow, and pale brown, indicating the formation of CDs.⁷⁵ Then, CDs were then purified by repeated dialysis in ultrapure water for two days (double dialysis bag molecular weight 1300 Da). Lastly, the solutions were dried in a hot-air oven to obtain the CDs and then they were stored at 4 °C in the refrigerator before being used in further experiments.

Characterization of CDs (Shape, Size, Zeta Potential and Chemical Composition)

This work’s synthesized CDs were characterized using TEM at 200 kV (Jeol ARM field-emission probe-corrected operating in Transmission and Scanning-Transmission modes) to determine their shape, size and crystalline state. The particle sizes shown in Figure 1 were measured using the ImageJ software, individual particles were identified and analyzed using the software’s particle analysis tool, and the size distribution was obtained by compiling the data. DLS determined the hydrodynamic diameters of the CDs. For this, 1 mg·mL⁻¹ dispersions were prepared with deionized water, and a Particle Size Analyzer (Brookhaven Instruments Corporation, 90Plus) was used. The same dispersions and equipment described for DLS were employed to determine the ZP. Fourier Transform Infrared-FTIR (Perkin Elmer, Spectrum 100) and Raman (Horiba, LabRAM HR Evolution) spectroscopies were used to analyze the chemical composition. FTIR spectra were acquired in the 4000–500 cm⁻¹ range, while Raman spectra were recorded between 1000 and 2000 cm⁻¹.

Figure 1 High-resolution TEM images of the synthesized CDs: (A) CDs1, (B) CDs2, (C) CDs3, (D) CDs4. Additionally, CDs2 were characterized: (E) SAED pattern, (F) bright field TEM images and (G) size distribution. (d=distance).

Luminescence of CDs

The luminescence emission measurements were performed on a Cary Eclipse, Agilent technology spectrofluorometer. The employed concentration of CDs was 50µg·mL⁻¹.

Photodynamic Activity and Evaluation of Singlet Oxygen

The photodynamic activity of CDs was assessed by monitoring the oxidation of TMB in the presence of hydrogen peroxide (H₂O₂) using UV-Vis spectroscopy at 652 nm under blue light irradiation (350 nm). The reaction mixture consisted of 100 μL of TMB (1mM), 200 μL of H₂O₂ (30% v/v), and 1.8 mL of sodium acetate buffer (NaAc, pH 4.5).

To evaluate singlet oxygen species (¹O₂), the system (CDs+TMB) was irradiated t at 350 nm for 30 minutes, after which 10 μL of sodium azide (NaN₃) was added to evaluate its effect as a ¹O₂ scavenger. The absorbance at 652 nm was recorded using a UV-Vis spectrophotometer. Additionally, the absorbance measurements at 630 nm were taken every 5 seconds under 350 nm irradiation to monitor reaction kinetics. The decrease in absorbance upon NaN₃ addition was used to confirm the presence of ¹O₂ as the primary reactive oxygen species responsible for the oxidation of TMB.

Photothermal Studies

Thermal studies were carried out by irradiating, in deionized water (control), CDs1, CDs2, CDs3 and CDs4 (0.5 mg·mL⁻¹), with red (630 nm) light and blue (450 nm) light and 30 W.cm² and taking the temperature every 2 minutes until it did not change.

Cell Culture

The cell cultures were maintained weekly using LLC-MK2 cells (strains of rhesus monkey kidney cells, source ATCC). They were maintained in Dubelco Modified Eagle Medium (DMEM) (Gibco), supplemented with 10% fetal bovine serum (Gibco), 1% penicillin/streptomycin (Gibco), and 1% pyruvate (Gibco) in T25 flasks (Thermo Scientific). Cells were passaged weekly, and a T25 flask was inoculated with 2×105 cells. The flasks were then incubated at 37 °C for 24 hours with relative humidity and supplemented with 5% CO₂. This process was performed to harvest the cells and conduct the necessary assays.

Cytotoxicity Assay

LLC-MK2 cells were harvested using trypsin and counted. In a 96-well plate, 2×104 cells were seeded per well from column 1 to column 11. The plate was then incubated for 24 hours at 37 °C with 5% CO₂ and relative humidity. On the following day, the plate was washed with 100 μL of PBS per well, and the 10 concentrations of CDs (4000, 2000, 1000, 500, 250, 125, 62.5, 31.2, 15.6, 7.8 µg·mL⁻¹) were added from column 1 to column 10. Columns 11 and 12 were filled with 100 μL of culture medium and served as the cell viability and cytotoxicity control, respectively. Finally, 10 μL of Resazurin (3 mm) (Thermo Scientific) was added to each well, and the plate was incubated at 37 °C with 5% CO₂ for fluorescence reading at 24 hours and 8 days using the Glomax plate reader (Promega) with excitation wavelengths of 530–560 nm and an emission wavelength of 590 nm.

Antibacterial Activity

MRSA (ATCC: 700682) and ESBL-producing E. coli (ATCC: 43300), multidrug-resistant cryovials, were utilized as test organisms in antimicrobial investigation. Inoculating cryovials with Mueller–Hinton agar (Difco Laboratories) required thawing cryovials at room temperature. A 37 °C overnight incubation of the culture media was performed. After that, the concentration required for the bioassays was reached by diluting each medium according to its absorbance using Mueller–Hinton broth (Difco Laboratories), as assessed by spectrophotometry: For ESBL-producing E. coli, this is 107 CFUs·mL⁻¹, while for MRSA, this is 106 CFUs·mL⁻¹.

Upon reaching the indicated concentration, each medium was dispensed into microtubes: six aliquots of 1 mL per CDs (annatto, cinnamon, curcumin and hyper-concentrated sucrose at 2000 µg·mL⁻¹) and six aliquots of 1 mL as control. Half of the aliquots were used to evaluate the activity of the nanocomposites under irradiation with blue light (450 nm; 40 mW·cm⁻²), while the other half was not exposed to these conditions.

For 10 minutes, the microtubes were centrifuged at 3000 rpm. To wash the cells and remove the remaining culture media, 1 mL of PBS was added after removing the supernatant. Under the same circumstances, the pellet was resuspended and recentrifuged. Each microtube was filled with 1 mL of the substance after discarding the PBS. The cells were once more resuspended in PBS in the control microtubes. Each tube was vortexed to dissolve the pellet once more before being incubated in the dark at 37 °C for 45 minutes.

Three aliquots of each substance and three controls were exposed to blue light at 40mW·cm⁻² after incubation. No radiation was applied to the remaining tubes.

Each aliquot was serially diluted in PBS. Each dilution was inoculated onto an 8-part Petri plate using a 4 µL aliquot and Mueller–Hinton agar. For colony isolation, each inoculum was streaked. Petri plates were incubated for 24 hours at 37 °C, and then the number of colonies in each dilution was counted.

Anticancer Activity

T47D cells (source, ATCC) were dispersed at the concentration indicated for bioassays of 2×105 cells per well with 10% DMEN. CDs3 was added to half of the samples at a concentration of 2000 µg·mL⁻¹, and the other half of T47D cells with DMEN was used as control samples. All samples were incubated for 4 hours with 5% CO₂ and RH (relative humidity).

To evaluate the anticancer capacity of the CDs, one well from each group was taken from the control group and the group with CDs3 to be irradiated with red LED light (630 nm, 65.5 mW·cm⁻²) for 40 min (sample + red light), another to be irradiated with blue LED light (450 nm; 40 mW·cm⁻²) for 30 min (sample + blue light), and another without irradiation (sample + dark).

Subsequently, the contents of each well were distributed in rows in a 96-well plate, and 10 µL of resazurin was to read the oxide-reduction of this compound and incubated in the same conditions. After 24 hours, the plate was read using the Glomax plate reader (Promega) with an excitation wavelength of 520 nm and an emission range of 580–640 nm. Therefore, the assays were duplicated in the 96-well plate and made on three days.

Results and Discussion

Shape, Size and Zeta Potential of the CDs

Based on the results acquired by TEM, it was possible to determine the shape and size below 10 nm of all the synthesized CDs.^1,2

Figure 1A–D shows the corresponding TEM images of CDs1, CDs2, CDs3, and CDs4, respectively, identifying lattice spacings. CDs1 (see Figure 1A) presented an interplanar distance of 0.200 nm and 0.305 nm, corresponding to the (100) and (002) planes of graphite, respectively. These values are characteristic of CDs with a graphitic structure, suggesting the ultrafine crystallite size typical of carbon quantum dots (CQDs). However, the distance d = 0.231 nm has not been commonly reported in previous studies on CDs. It could be related to a less ordered structure or the presence of defects in the crystal lattice of the CDs. CDs2 (see Figure 1B) displayed an interplanar distance of 0.335 nm, which aligns with the (002) plane of the graphite plane. According to Nie et al,⁷⁶ this lattice is related to fluorescent CDs. Concerning CDs3 (see Figure 1C), the interplanar distances were 0.184 nm, 0.185 nm and 0.175 nm, corresponding to the (102) crystalline planes of graphite.⁷⁷ While CDs4 showed lattice distances of 0.367 nm, 0.355 nm and 0.348 nm, corresponding to the (002) crystalline planes of graphite (see Figure 1D).

The electron pattern of CDs2 is shown in Figure 1E. According to Diwan et al,⁷⁸ the broad diffused ring in the Selected Area Electron Diffraction (SAED) pattern reveals that a significant body of the synthesized CDs is amorphous. Nevertheless, well-defined crystalline CDs are detected in the indexed ring pattern shown in Figure 1E. The inter-layer d-spacing is 0.36 nm, corresponding to (002), and the 0.23 nm inter-layer d-spacing to plane (100) of graphite.⁷⁸

The particle sizes of CDs exhibited an overall mean diameter of 2.82 nm, with a global standard deviation (SD) of 0.61 nm, indicating moderate variability in size. Each type of CDs displayed distinct mean sizes: CDs1 had an average size of 2.867 nm (SD = 0.673), CDs2 had a mean size of 2.428 nm (SD = 1.057) (see Figure 1F), CDs3 measured 2.336 nm (SD = 0.701), and CDs4 presented the largest average size at 3.665 nm (SD = 1.114). The distribution of particle sizes was somewhat heterogeneous, as evidenced by kurtosis values, suggesting a spread that includes smaller and larger particles relative to the mean. Specifically, CDs1 exhibited a positive kurtosis (1.044), indicating a slightly peaked distribution, while CDs2, CDs3, and CDs4 displayed negative kurtosis values (−0.805, −1.259, and −0.577, respectively), suggesting a broader distribution. The minimum and maximum sizes observed were 0.278 nm (CDs2) and 6.017 nm (CDs4), respectively. The median particle sizes were close to the mean for each sample, further supporting a relatively symmetric distribution. These results are in agreement with the data previously reported in literature.^1,2

The histogram (see Figure 1G) revealed that the particle size distribution of CDs generally follows a normal distribution; however, some atypical measurements, particularly larger particles up to 5 nm or more, were noted beyond the typical range. Table 1 summarizes these results, including the global mean particle size of 2.824 nm and the global SD of 0.61 nm, along with specific metrics (standard deviation, kurtosis, minimum, median, and maximum) for each type of CDs. This study found that the synthesized CDs had irregular shapes, probably due to an uncontrollable synthesis process.⁷⁹ Besides, using the microwave-assisted technique makes it difficult to have monodispersed nanoparticles. The aggregation of CDs in this process is inevitable since the raw liquid was dried entirely in the microwave oven.⁷⁹ For instance, Vasimalai et al synthesized CDs from cinnamon with sizes of 3.5 ± 0.1 nm,⁸⁰ which, compared with our results, seems smaller. Nevertheless, the effect of agglomeration also must be considered.

Table 1 Descriptive Statistics of the Particle Size Distribution of CDs

By DLS were measured the hydrodynamic diameters of the CDs on different days. The results are presented in Table 2. Consider that the sizes measured by DLS correspond to a mean hydrated diameter assuming spherical particles. This technique makes it impossible to distinguish if the measured value corresponds to an agglomerate or single particle since an average is calculated.⁸¹ These results could be a consequence of three factors: (i) agglomeration degree, (ii) attached molecules in the CDs, and/or (iii) the presence of bigger organic by-products.⁸¹ In any case, considering the data of day 7, the size of the CDs agglomerates was established as CDs4>CDs1>CDs3>CDs2. The CD hydrodynamic diameters in Table 2 also indicate a tendency to disaggregate from day 0 to day 7. This shows that on day 7, hydrodynamic diameters had a lower value in relation to day 0, except for CDs1.

Table 2 CDs Hydrodynamic Diameters After a week

Regarding Zeta potential (ZP) results, they are also shown in Table 2. The values of CDs1, CDs2, and CDs3 were close to −30 mV, and according to Sahiner et al^82,83 and Hua et al^82,83 this indicates that CDs possess good stability owing to the interparticle electrostatic repulsion.⁸⁴ Thus, the negative charge is sufficient for CDs1, CDs2, and CDs3 to form stable colloidal dispersions at neutral pH. Considering the absolute values of Table 2, the CDs4 presented the lowest ZP, which is coherent with Figure 1 since the CDs4 had the most considerable particle size.

Chemical Composition of the Synthesized CDs

Figure 2a presents the FTIR spectra of CDs1, CDs2, CDs3, and CDs4. Remember that the intermediate species (species that were formed during the synthesis but disappeared during the process), which form the CDs in each mechanism through diverse pathways, are different and directly dependent on the chemical nature of the precursor.^85,86 However, some common functional groups were identified, especially in the 2000–800 cm⁻¹ wavenumber range fingerprint region.

Figure 2 (a) FTIR and (b) Raman spectra of the synthesized CDs.

In the zones marked with red, between ~3550-2900 cm⁻¹, are situated broad and intense bands of the functional groups O-H and N-H (stretching vibration). The first one might indicate the presence of hydroxyl moieties.^87–89 Only in CDs3 was distinguished the C=O group, in ~1725 cm⁻¹ highlighted in blue, denoting the existence of carboxyl groups.⁹⁰ In all CDs were identified C=C bonds, marked in green, they are situated at ~1650–1600 cm⁻¹. This band possibly corresponds to the benzene ring of the CDs.^91,92 Besides, C-N bonds were identified, pointed in sky-blue, which should demonstrate amine groups in CDs1, CDs2, and CDs4.⁹³ In all samples detected, C-H and C-O functionalization, marked with orange (~1500–1450 cm⁻¹) and purple (~1250–1125 cm⁻¹), correspondingly.⁹⁴ Moieties such as hydroxyls, amines, and carboxyl groups contribute to the water dispersibility of the CDs due to these functionalization’s attributes of hydrophilic nature.^95–97 This functionalization, along with the strong negative charge, could be related to the antimicrobial and antitumoral capacities of the CDs.⁸²

Raman spectroscopy results of the CDs are depicted in Figure 2b, where it is easy to distinguish the characteristic bands of the carbon-based materials, reported elsewhere.^98,99 D-band (defect or disorder at ~1350 cm⁻¹) and G-band (graphite at ~1580 cm⁻¹), marked in black and grey, correspondingly.^98,99 The D-band corresponds to sp³ hybridized carbon atoms and the presence of structural defects, while the G-band is related to sp² hybridized carbon atoms associated with well-organized carbon.¹⁰⁰ Only in CDs3 the intensity of the D-band is slightly higher than that of the G-band, indicating that these CDs have some defects sp² carbons. The rest of the CDs presented a superior G-band than D-band. The relative intensity ratio of D-band to G-band, as ID/IG, measures the extent of disorder or a degree of graphitization.¹⁰¹ The Raman result of CDs1 (lowest ID/IG value, low degree of graphitization of all CDs) agrees with the interplanar distance of CDs1, which does not indicate the presence of graphitic material that could confirm the presence of defects in the crystalline structure (TEM analysis). These defects could produce greater dispersion of the CDs, showing this in the smallest hydrodynamic size of all the CDs (DLS analysis after 7 days later). Conversely, the higher degree of graphitization corresponds to CDs3, which means that this sample has the most “ordered structure” since graphitization is the degree of rearranging carbon atoms to fill atom vacancies and improve atom layout. Consequently, if the crystalline structure changes, it has antibacterial and antitumoral features.¹⁰²

Luminescence of CDs

The photoluminescence emission spectrum obtained from 400 to 700 nm of all the CDs are depicted in Figure 3 (A: CDs1, B: CDs2, C: CDs3 and D: CDs4). The CDs showed strong and stable photoluminescence dependent on the excitation wavelength. The longer the excitation wavelength, the more emission bands of longer wavelength are observed, but with lower intensity. The highest intensity was detected at an excitation wavelength of around 420 nm, with emission at around 530 nm for all CDs. It was also confirmed that at an excitation wavelength of 600 nm, there is still a band with a low emission intensity of around 650 nm. This data shows that all CDs presented the highest luminescence in the visible region of the spectrum. The divergent peak position, spectral width, and shape might be related to factor such as size, surface state, and/or sp²-sp²-hybridized carbon domains (located in the carbon core).^103–105 The highest emission was detected at ~530 nm in all the cases.

Figure 3 Emission spectrum of the synthesized CDs. (a) CDs1, (b) CDs2, (c) CDs3 and (d) CDs4.

These results can motivate future studies since they encourage the application of CDs as antibacterial agents by exciting with blue light of around 400 nm (see Figure 4), which is related to soft light penetration into the skin, as well as tumor elimination at 600 nm that results in greater light penetration into the skin.^106,107

Figure 4 Photographs of CDs under day light and blue light.

Photodynamic Analysis of CDs

The photodynamic activity of CDs was evaluated based on the oxidation of 3,3′,5,5′-tetramethylbenzidine (TMB). The oxidation process leads to the formation of oxidized TMB (oxTMB), which is characterized by an increase in absorbance over time. Figure 5 illustrates the absorbance difference at 652 nm between the TMB/H₂O₂ + CDs system and the control (TMB/H₂O₂) under light irradiation (A_Light) and dark conditions (A_DC) as a function of time. Each bar represents the standard deviation (0.0029) based on triplicate measurements.

Figure 5 POD-like activity of CDs under light irradiation: Difference of absorbance vs Time.

Each CDs sample exhibited a distinct oxidation rate, with saturation times as follows: CDs1 at 120s, CDs2 at 55s, CDs3 at 160s, and CDs4 at 50s. Among all samples, CDs3 demonstrated the highest absorbance, followed by CDs1, CDs2, and then CDs4. The negligible absorbance observed in the TMB/H₂O₂ control confirms that the CDs play a crucial catalytic role in the process of oxidation under light irradiation. This steady increase in absorbance over time, particularly under blue light irradiation at 450 nm, highlights the ability of each CD samples to generate ROS which catalyze the oxidation of TMB to oxTMB.¹⁰⁸

To confirm the involvement of singlet oxygen (¹O₂) in the oxidation of TMB, sodium azide (NaN₃) was introduced as a selective scavenger. This experiment was performed by irradiating each sample under blue light irradiation at 450 nm for 30 minutes, followed by the addition of 10 μL of NaN₃. The results, described in Figure 6, show a significant reduction in absorbance at 652 nm upon the introduction of NaN₃, confirming the presence of ¹O₂ as the primary ROS responsible for the oxidation process. NaN₃ is a well-established quencher of ¹O₂ due to its rapid reaction through both physical and chemical deactivation mechanisms. The primary pathway involves converting ¹O₂ into a less reactive species via the formation of azide radicals (N₃⁻). Wu et al reported that this reaction effectively suppresses TMB oxidation by depleting ¹O₂, resulting in a decreased absorbance at 652 nm. Since ¹O₂ is crucial for initiating TMB oxidation, the significant reduction in oxTMB formation upon NaN₃ addition further confirms the predominant role of ¹O₂ in the photodynamic activity of the CDs.¹⁰⁸ The most pronounced decrease in absorbance was observed for CDs3, indicating its superior ¹O₂ generation capability. This suggests that the structural and chemical properties of CDs3, synthesized from curcumin, enhance its photodynamic performance. According to Najaflu et al, the presence of π-π interactions within the benzene ring contributes to the enhanced photodynamic activity of CDs.¹⁰⁹ Similarly, CDs2 and CDs1 also exhibited notable reductions in absorbance, indicating substantial ¹O₂ production under blue light irradiation. In contrast, CDs4 showed the least reduction, suggesting lower photodynamic activity compared to the other CDs. This reduced efficiency may be attributed to the chemical composition of sucrose-derived CDs, which lack aromatic structures that facilitate efficient ROS generation.¹¹⁰

Figure 6 Irradiation of CDs1, CDs2, CDs3 and CDs4 under blue light (450 nm) for 30 minutes and addition of NaN₃.

The average size of the CDs was equally determined to be 2.82 nm as reflected in Table 1. This gives a huge specific area of surface, which is critical for superb catalytic activity. Moreover, the presence of benzene rings of CDs allows the interaction with biphenyl of TMB molecules by means of some π-π stacking, promotes the favorable vicinity for the surface of the substrate and hence improves the catalytic function.^111,112 Also, the high contents of carbon within the CDs indicate that the carbon nuclei that developed were well formed and stable for optimal reactivity and stability needs.¹¹¹ Previous research also highlights that the efficiency of photodynamic activity in CDs is enhanced by factors such as particle size, surface functionalization, and the presence of conjugated systems (eg benzene rings).^111–113 The small size of the CDs and their high surface area-to-volume ratio provide wide active sites for the catalytic reaction, making them competitive with other nanocatalysts.

Photothermal Analysis

The heating profiles of CDs1, CDs2, CDs3, and CDs4 were compared to deionized water as a control (see Figure 7). The results demonstrated that CDs1, CDs2, and CDs3 exhibited higher temperature increases under both irradiation conditions, indicating strong photothermal conversion properties. In contrast, CDs4 showed a limited rise in temperature, suggesting weaker photothermal activity.

Figure 7 Heating profiles of CDs1, CDs2, CDs3, and CDs4 at (A) Blue light and (B) Red light.

Under blue light irradiation (450nm), CDs1, CDs2, and CDs3 reached temperatures of approximately 50 °C (see Figure 7A), whereas CDs4 only increased to ~30–35°C. The control exhibited minimal temperature changes, confirming that the observed heating was due to the photothermal properties of the CDs. Red light irradiation (630 nm), which is more relevant for biomedical applications due to its deeper tissue penetration,¹¹⁴ resulted in maximum temperatures around 40 °C for CDs1, CDs2, and CDs3, while CDs4 displayed only an increase, not exceeding 30°C (see Figure 7B). Sun et al investigated CDs synthesized from different organic precursors and found that those with π-π conjugated structures exhibited superior photothermal properties. This supports our hypothesis that CDs1, CDs2, and CDs3 possess structural features favorable for photothermal conversion, while CDs4 may lack these key characteristics.¹¹⁴ These results suggest that CDs1, CDs2, and CDs3 possess effective photothermal conversion capabilities, making them promising candidates as photothermal agents for PTT applications.

Cytotoxicity Analysis in LLC-MK2 Cells

The percentage of cell viability of monkey kidney cells (LLC-MK2), at concentrations up to 4000 µg·mL⁻¹ (LOG concentration = 3.60) of all CDs, did not present significant changes at 24 h and 48 h (see Figure 8a and b). The cell viability of LLC-MK2 cells at concentrations of up to 4000 µg·mL⁻¹ of CDs2, CDs3, and CDs4 did not present significant changes (80–100%) at 24 h, as well as concentrations of up to 2490 µg·mL⁻¹ of CDs1. In contrast, after 48 h, the CDs1, CDs2, CDs3, and CDs4 samples did not show significant differences in cell viability (80–100%) for concentrations up to 2000 µg·mL⁻¹. This result is consistent with studies at high concentrations of CDs (1000–4000 µg·mL⁻¹), where negligible cytotoxicity is reported.

Figure 8 Cell viability percentage in monkey kidney cells by Resazurin cell viability assay, using CDs1, CDs2, CDs3, and CDs4, at concentrations between 7.81–4000 µg·mL−1 for (a) 24 h and (b) 48 h. Data represent the mean ± SD (n= 3 per group).

The IC₅₀ corresponding to 50% cellular inhibition caused by the presence of CDs2, CDs3, and CDs4 were calculated in OriginLab software with a 95% confidence interval with a sigmoidal fit Dose-Response. This information is detailed in Table 3. It is worth noting that the IC₅₀ values obtained are not within the concentration range used in this assay. Within the results obtained, it can be observed that the IC₅₀ of CDs3 doubles the other CDs, which agrees with the literature since CDs based on curcumin have been shown to have high biocompatibility with normal cells and almost no cytotoxicity at high concentrations (10 µM).⁶⁶

Table 3 IC50 of CDs at 24 h and 48 h

The robust catalytic activity observed in the POD-like studies, especially for CDs3, suggests enhanced oxidative stability and high surface reactivity, likely responsible for its biocompatibility. The structural stability and lower cytotoxicity of CDs3 can be advantageous, as higher biocompatibility permits application in biological systems at significant concentrations without adverse cellular effects. This property further aligns with findings that curcumin-based CDs show high biocompatibility, supporting their suitability in therapeutic applications without substantial toxicity.^110,115

Antibacterial Studies

The antimicrobial effect of the CDs was evaluated against the microorganism’s methicillin-resistant S. aureus (MRSA) and Extended-Spectrum β-Lactamase (ESBL)-producing E. coli, using a CDs concentration of 2000 μg·mL⁻¹, and blue light irradiation (450 nm) at 40 mW cm⁻² for 30 min.

The results obtained for each colony are shown in Figure 9. The histograms present the LOG (CFUs·mL⁻¹) evaluation standardized to 1. The established concentration for bioassays was used, which is 107 CFUs·mL⁻¹ for MRSA and 106 CFUs·mL⁻¹ for ESBL-producing E. coli. These results show that light alone cannot eliminate microorganisms (control + light samples); to combat the microorganisms, it was necessary to irradiate the samples in the presence of CDs (CDs1+ light, CDs2 + light, CDs3+ light, and CDs4+ light).

Figure 9 PDT/PTT antimicrobial effect. Standardized result of LOG (UFC/mL) by (a) MRSA and (b) ESBL-producing E. coli, based on CDs1, CDs2, CDs3 and CDs4. The concentration of CDs was 2000 μg·mL−1. Time of irradiation of CDs was 30 min. Significant differences in means according to the Tukey’s test (**p ≤ 0.01, and *** p ≤ 0.001).

Figure 9 presents the quantitative result of the colony counts of (A) MRSA and (B) ESBL-producing E. coli., before and after being irradiated with blue LED light (450 nm, 40 mW·cm⁻²). This figure indicates that the colonies of MRSA and ESBL-producing E. coli (samples CDs1+ dark, CDs2+ dark, CDs3+ dark, and CDs4+dark) were not different from the control group. In contrast, the bacteria colony count in samples CDs1 + light, CDs2 + light, CDs3+ light, and CDs4 + light was significantly lower than the control group (*** p ≤ 0.001) for ESBL-producing E. coli and CDs1 + light, CDs2 + light, and CDs3 + light, was significantly lower than the control group (*** p ≤ 0.001) for MRSA. This means that after irradiation, CDs1, CDs2, and CDs3 eliminated all MRSA and ESBL-producing E. coli, and CDs4 can only completely remove ESBL-producing E. coli. Therefore, all CDs can eliminate some of the microorganisms used in this study, with CDs1, CDs2, and CDs3 being the best due to their ability to eliminate the two types of bacteria. According to Figure 10, the elimination occurs by bimodal strategy PDT/PTT, where it is observed that the temperature after irradiation rises to around 50 °C evidencing the PTT effect, and from the literature, it is known that CDs can act as PS-based PDT effect.^116,117 Yan et al, Romero et al, and Lin et al showed that the rise in temperature changes bacterial outer membrane permeability through PTT, significantly weakening bacterial activity, promoting PDT singlet oxygen diffusion into cells, disrupting the homeostatic state of cells, and synergistically enhancing the antibacterial effect.^66,118,119

Figure 10 Sample temperatures of (a) MRSA and (b) ESBL-producing E. coli, based on CDs1, CDs2, CDs3, and CDs4 before (To) and after (Tf) being irradiated with blue light. The concentration of CDs was 2000 μg·mL−1. Time of irradiation of CDs was 30 min. Significant differences in means according to the Tukey’s test (***p ≤ 0.001).

CDs can generate ROS, such as singlet oxygen, which is highly reactive and can induce microbial cell death by necrosis/apoptosis.¹²⁰ The ROS causes oxidative damage to various cellular components, including proteins, nucleic acids, and lipids, leading to the destruction of microorganisms.^120,121

The antimicrobial activity of the CDs used in this research based on the PDT/PTT strategy exhibits a 100% inhibition of ESBL-producing E. coli in comparison with other types of CDs without using phototherapy such as CDs coated CaCO₃ nanocarrier conjugated with levofloxacin (CDs-CCNC- Lvx) obtained in the study by Kanwal et al where inhibition of 68% is achieved.¹²² In addition, CDs1, CDs2, and CDs3 eliminate MRSA by applying the PDT/PTT strategy using a red light for 30 min, which, in comparison to the study without phototherapy by Kung et al, presents an improvement in treatment time and concentration of CDs, since this study shows elimination of MRSA after 24 hours of application of citric acid CDs at a concentration of 2500 µg·mL⁻¹.¹²²

ESBL-producing E. coli are classified as Gram-negative bacteria and have a relatively thinner peptidoglycan layer in their cell walls, making them more vulnerable to heat and ROS,¹²³ allowing the CDs studied to eliminate this type of colony. MRSA strains are classified as Gram-positive bacteria and have a thicker peptidoglycan layer in their cell wall, which provides some protection against heat.¹²⁴ In this case, CDs1, CDs2, and CDs3 present antimicrobial molecules that effectively eliminate MRSA in addition to the PDT/PTT effect. CDs4, being made up mainly of sucrose, does not demonstrate antibacterial properties, which agrees with the literature where it has been shown that high concentrations of sucrose do not cause microbial inhibition.¹²⁵

The photothermal performance of the CDs was further correlated with their antimicrobial activity. The ability of CDs1, CDs2, and CDs3 to generate significant heat under irradiation aligns with their complete elimination of MRSA and ESBL - Escherichia coli. In contrast, CDs4, which displayed a weaker photothermal response, was only effective against ESBL-producing E. coli but failed to eliminate MRSA.

The antimicrobial activity of the synthesized CDs was evaluated in correlation with their PDT efficacy. The mechanism of microbial elimination involves the generation of ROS, particularly ¹O₂, which induces oxidative stress, leading to cell membrane damage and microbial death. CDs3, derived from Curcuma longa L., exhibited the highest ¹O₂ generation, as evidenced by its superior oxidation rate and significant reduction of absorbance at 652 nm with the presence of NaN₃. This enhanced ROS generation directly correlated with its antimicrobial efficacy, leading to the complete elimination of MRSA and ESBL-producing Escherichia coli. The effectiveness of CDs3 against these pathogens suggests that its structural properties, including the presence of conjugated benzene rings, facilitate efficient ROS-mediated bacterial inactivation. Similarly, CDs1 and CDs2 demonstrated substantial ROS production and notable antimicrobial activity, whereas CDs4 showed the weakest photodynamic response and lower microbial inhibition, likely due to its lack of aromatic structures that enhance ROS generation. CDs3, in particular, demonstrated the most favorable combination of high photothermal conversion, ROS generation, and biological efficacy, making it an excellent candidate for further investigation in photothermal and photodynamic therapy applications.¹¹⁸

The photothermal performance of the synthesized CDs was directly correlated with their antimicrobial efficacy. CDs1, CDs2, and CDs3 exhibited significant heat generation under irradiation, leading to the complete elimination of MRSA and ESBL-producing Escherichia coli. In contrast, CDs4, which demonstrated a weaker photothermal response, effectively inhibited ESBL-producing E. coli but failed to eradicate MRSA. This trend aligns with findings from previous studies by Bhavikatti et al and Wang et al, where CDs synthesized from ginger and garlic demonstrated high photothermal conversion efficiency, achieving superior antibacterial effects (90%) through localized hyperthermia-induced membrane disruption. This enhanced antibacterial activity is attributed to the intrinsic properties of the carbon source, which influence the structural and electronic characteristics of the CDs, thereby optimizing their photothermal and antimicrobial performance.^126,127

Furthermore, the antimicrobial activity of the CDs was closely associated with their photodynamic efficiency. The primary mechanism of microbial inactivation involved the generation of ROS, particularly ¹O₂, which induced oxidative stress and membrane damage, ultimately leading to bacterial cell death. CDs3, synthesized from Curcuma longa L., exhibited the highest ¹O₂ generation, as demonstrated by its superior oxidation rate and significant reduction in absorbance at 652 nm upon the addition of NaN₃. This result is consistent with previous research by Yan et al, who synthesized CDs from the turmeric plant as a carbon source and demonstrated that CDs with high aromatic content and conjugated π-systems facilitate efficient ROS production, thereby enhancing PDT efficacy against bacterial pathogens.^112,113

Similarly, CDs1 and CDs2 displayed substantial ROS production and antimicrobial activity, reinforcing the role of PDT in microbial elimination. In contrast, CDs4 exhibited the weakest photodynamic response, which correlated with its lower antibacterial activity. This diminished effect may be attributed to its structural composition, as CDs lacking aromatic systems tend to generate ROS less efficiently, limiting their photodynamic potential.¹¹³ Overall, CDs3 demonstrated the most favorable combination of high photothermal conversion, robust ROS generation, and potent antimicrobial efficacy, positioning it as a promising candidate for further applications in photothermal and photodynamic antimicrobial therapies.

Antitumoral Studies

The compounds CDs1, CDs2, CDs3 and CDs4 have similar characteristics. This was observed in the chemical, morphological, photophysical and bacterial elimination analysis using the CDs’ synchronous PDT/PTT strategy. In this way, for antitumor studies of kidney CDs in T-47D cells (human breast cancer cell line) the compound CDs3 was chosen. In addition, the compound CDs3 shows the highest IC₅₀ according to the cytotoxicity studies of monkey cells (LLC-MK2), which indicates that it is the least cytotoxic of all CDs.

The results obtained are shown in Figure 11. The histograms present the percentage of cell viability. A concentration of 2×105 cells per well-established bioassays was used. These results show that light alone does not have anticancer effects, which can be observed in the control bars, which indicates that to eliminate cancer cells, it is necessary to irradiate the samples in the presence of CDs3.

Figure 11 PDT/PTT anticancer effect. Samples Control (T47D cells+DMEN) and Samples with CDs3. The concentration of CDs3 was 2000 μg·mL−1. Time of irradiation of CDs was 40 min with red LED light and 30 min with blue LED light. Significant differences in means according to the Tukey’s test (*p ≤ 0.05 and *** p ≤ 0.001).

Furthermore, a significant difference was observed between cells irradiated with red and blue light, with lower cell viability (greater cell death) in the presence of CDs3 and red light. This result is consistent with previous research by Tiron et al, where carbon dots derived from herbal medicine demonstrated enhanced photodynamic efficacy against breast cancer cells, resulting in cell viability below 10% under red light irradiation. This effect is attributed to the superior absorption of red light by the CDs, which enhances ROS generation—particularly singlet oxygen (¹O₂)-leading to increased oxidative stress and apoptosis in cancer cells. The extended penetration depth of red light also allows for more effective activation of CDs in deeper tissue layers, further enhancing their photodynamic therapeutic potential.^128,129

The photodynamic and photothermal properties of the CDs were investigated for their potential in antitumoral applications. PDT relies on the generation of ROS, primarily ¹O₂, to induce oxidative stress, leading to apoptosis or necrosis in cancer cells. Under red light irradiation (630 nm), CDs3 demonstrated remarkable Photothermal conversion, reaching temperatures of approximately 40 °C—a range known to induce hyperthermia-mediated cancer cell death. This finding is consistent with the studies conducted by Geng et al, Kim et al, and Najaflu et al, which demonstrated the effectiveness of CDs synthesized from biomass in photothermal cancer therapy under low-power red light irradiation. Their conclusions indicated that increasing the temperature to approximately 50 °C led to efficient tumor elimination through a synergistic photothermal effect, triggering irreversible protein denaturation and cell membrane disruption.^109,130,131 Likewise, the dual PDT-PTT effect observed in CDs3 highlights its strong potential for tumor ablation, where ROS-mediated oxidative stress works in tandem with hyperthermia-induced apoptosis, thereby enhancing overall therapeutic efficacy.^109,131,132

CDs1 and CDs2 also displayed notable antitumoral effects, albeit slightly lower than CDs3. Their ability to generate ¹O₂ and increase local temperature under irradiation contributed to their cytotoxicity, demonstrating a synergistic PDT-PTT mechanism. In contrast, CDs4 exhibited minimal singlet oxygen generation and weaker photothermal effects, resulting in reduced antitumoral activity. This suggests that the structural properties of CDs, including aromaticity and surface functionalization, play a crucial role in determining their effectiveness in cancer therapy.

Overall, these results demonstrate that CDs3, followed by CDs1 and CDs2, possess significant photodynamic and photothermal properties that can be leveraged for antitumoral applications. Their ability to generate ROS, induce oxidative stress, and achieve localized hyperthermia under irradiation positions them as promising candidates for PDT and PTT in both bacterial infections and cancer treatment.^132,133

Conclusions

CDs were synthesized from natural green carbon sources of annatto, cinnamon, curcumin, and sucrose. The obtained CDs products were identified as CDs1, CDs2, CDs3, and CDs4. CD particle size showed a diameter mean size of 2.30 nm. Regarding the chemical composition by FTIR analyses, hydroxyls, amines, and carboxyl functionalized groups promote antimicrobial and antitumoral capacities. The CDs´s cytotoxicity was evaluated against monkey kidney cells, and it was confirmed that the employed concentration of 2000 μg·mL⁻¹ was lower than that of the IC50 at 24 h (<5000 μg·mL⁻¹). The antibacterial capacity of the CDs was positively verified against microorganism methicillin-resistant S. aureus (MRSA) and Extended-Spectrum β-Lactamase (ESBL)-producing E. coli since complete inhibition was achieved after blue light irradiation (450 nm; 40 mW·cm⁻²). Concerning PTT characteristics in all cases, there was evidence of an increment in temperature over 40 °C after 30 min of irradiation. Finally, the antitumoral properties of CDs3 were successfully evaluated against T-47D cells (human breast cancer cell line).

This study demonstrated that is viable to employ CDs1, CDs2, CDs3 and CDs4 as photosensitizers (PS) or photothermal agents (PA) within PDT and PTT strategies for antibacterial and antitumoral applications. This approach can be used in the future for bacterial and tumor elimination in the dermatological area. In addition, we proved that the employed carbon sources (Bixa Orellana L., Cinnamomum verum J. Presl, and Curcuma longa L). can be used to develop antibacterial/antitumor nanomedicine, despite some differences could appear in the natural extracts due to the obtaining process.