DOX is a widely used chemotherapy drug for treating breast cancer, but its clinical application is restricted by severe side effects and drug resistance. In our study, we encapsulated the chemotherapy drug DOX in the safe carrier exosomes (Exo), derived from mesenchymal stem cells obtained from human adipose tissue. Additionally, we included melatonin as an antioxidant in the above composition, with the aim of potentially achieving improved results.
Examination of adipose tissue-derived MSCs at passage 3 revealed a fibroblast morphology throughout the culture when observed under an inverted microscope (Fig. 1). Characteristic surface markers on MSCs were identified using flow cytometry. The expression of negative markers, CD45 and CD34, was found to be 1.37% and 2.19%, respectively, while the expression of positive markers, CD44 and CD90, was 99.8% and 99.5%, respectively (Fig. 2). These results, along with the morphology of the cells, confirm that these cells exhibit typical characteristics of stem cells, aligning with findings from similar studies [32, 33].
EXOs were successfully isolated from AMSC-conditioned media by ultracentrifuge method. TEM observations of the EXOs revealed a typical spherical and oval shape for AMSC- EXOs (Fig. 3-A). Additionally, dynamic light scattering (DLS) characterization indicated an average size of 101.7 nm with a peak at 80.72 nm for these EXOs. (Fig. 3-B). Moreover, western blot analysis of surface markers suggested that EXOs were rich in tetraspanins as positive markers such as CD9, CD63 and CD81, while Calnexin was not expressed as a negative marker in EXOs. (Fig. 3-C). These results are consistent with the other results in confirming the characteristics of EXOs in terms of shape, size and expression of surface proteins [34, 35].
Protein content of Exos was measured using BCA method. BCA is a standard method to determine the concentration of Exos. After lysing the Exos, the protein concentration was measured. The results indicated the protein concentration of the Exos, which served as the basis for determining the amount of Exos used in the subsequent stages of the study.
To investigate whether Exos extracted from MSCs are able to bind and enter cancer cells, cancer cells were treated with EXOs whom previously labeled with PKH26 dye. PKH26 as a dye bound to the Exos was observed inside the cell by using a fluorescent microscope; the cellular absorption of the Exos was confirmed. Red and blue colors represent PKH26 and DAPI, respectively (Fig. 4).
Drug entrapment and release rate from Exos and revisited Exos structure
DOX as an anti-cancer drug and Mel as an antioxidant were loaded in Exos and the loading and release capacity of these drugs from Exos was evaluated. Among the drug loading methods evaluated, sonication demonstrated the highest potential for facilitating drug entry into Exos [36]. The quantity of encapsulated DOX was determined using a spectrophotometric method, measuring the autofluorescent characteristics of DOX at 480 nm, while the amount of encapsulated Mel was measured at 285 nm against serial dilutions of known standards. Our results revealed that approximately 37% of DOX and 32% of Mel were successfully loaded into Exos [37].
The drug loading in Exos is influenced by several factors including the loading method, the source of Exos, and the size and solubility of the drug. Various studies have reported different loading and encapsulation percentages of drugs in EXOs [38, 39], while the results of Chunyan Yang et al. were almost similar to our work [40]. TEM and DLS were utilized to assess whether sonication or drug transfer had any impact on the structure of EXOs. TEM images showed that the morphology of Exos remained almost unchanged after drug loading. Also, DLS showed no significant change in the size of EXOs (Fig. 5). The results indicate no change in the nature of the EXOs for drug storage and release, and these results were confirmed by others [38].
DOX is affected by pH due to having amine groups as functional groups. Therefore, the release capacity was investigated at different pH of 5 (environment of cancer cells) and 7.4 (conditions of normal cells). The release rate of DOX and EXO-DOX at pH: 5 were 66.7% and 90.5% after 36 h, respectively, but the release profiles of DOX and EXO-DOX after 36 h at pH: 7.4 were 35.4% and 28.6%, respectively. Also, the release rate of Mel and EXO-MEL at pH: 5 were 25.2% and 33.6%, after 36 h, but the release profiles of Mel and EXO-MEL were 32.3% and 42.4% after 36 h at pH 7.4, respectively.
These results clearly demonstrate that the pH of the medium has a significant impact on the release rate of DOX and MEL [41]. Under conditions mimicking those of cancer cells (pH 5), DOX and MEL loaded in EXOs are released more rapidly than free DOX and MEL, indicating an increase in cytotoxicity under acidic conditions. Consequently, there is improved effectiveness and targeting of drug accumulation in cancerous conditions.
Conversely, under conditions mimicking normal cells (pH 7.4), EXO-DOX is released at a slower rate than free DOX, while EXO-MEL is released at a faster rate than free MEL. This discrepancy leads to decreased toxicity in normal cells (reducing side effects of DOX) and an increased presence of antioxidants (MEL) (Fig. 6). Consistent with our findings, other studies have reported similar results, with authors concluding that acidic environments in endosomes and lysosomes can induce drug release [42–44].
Qualitative and quantitative intracellular uptake of Exo-DOX
As illustrated in Fig. 7, both DOX-loaded Exos and free DOX were efficiently absorbed and internalized by both cancer and normal cells within 4 h. The uptake process appeared to be completed within this time frame, as indicated by the high fluorescence intensity observed in the cells. Flow cytometry analysis corroborated these observations, demonstrating that in MCF-7 and MDA-MB231 cells, 76.4% and 45.3% of free DOX, and 96% and 82.1% of EXO-DOX, were absorbed after 4 h, respectively. These findings suggest that Exo-DOX exhibits greater absorption compared to free DOX within the same time period, thus heightening the impact of DOX on these cells. In MCF-10A and A-MSC cells, the absorption rates after 4 h were 80.9% and 74.1% for free DOX, and 65.7% and 69% for EXO-DOX, respectively. These outcomes validate the diminished absorption of EXO-DOX in comparison to free DOX, leading to reduced accumulation in normal cells. Consequently, the initially low absorption of free DOX in cancer cells is largely compensated by the presence of EXO, magnifying the impact of DOX on these cells. However, in normal cells, EXO diminishes DOX absorption, consequently lowering cell death and its associated effects.
In vitro anticancer effect of functional EXOs-mediated co-delivery of DOX and MEL
In order to assess the potential enhancement of the anticancer effect through the combined delivery of DOX and MEL with EXOs, we evaluated the viability of breast cancer cells, normal cells, and stem cells treated with free-DOX, free-MEL, DOX/MEL, free-Exo, Exo-DOX, Exo-MEL, and Exo-DOX-MEL over the course of 24, 48, and 72 h.
As shown in the Fig. 8, the cytotoxicity assay of different treatment showed dose- and time-dependent manner inhibition of cell proliferation. In MCF-7 cells, DOX/MEL, Exo-DOX and Exo-MEL caused greater toxicity than free-DOX and MEL in their free form. However, the most striking difference was observed with Exo-MEL compared to the free form of MEL and Exo. On the other hand, free-Exo exhibited slightly greater effectiveness than free-MEL and achieved a 50% inhibitory concentration within 72 h. Moreover, the combination of Exo-DOX-MEL showed a significantly superior effect on cell death compared to other compounds. In MCF-10A and AMSC cells, MEL and Exo were able to reverse the effect of DOX to a large extent, and free-DOX showed the lowest IC50.
In MDA-MB231 cells, free-DOX could kill 50% of the cells at a concentration of 5.4 µg/ml. However, the combination of DOX with MEL or the loading of DOX in Exos did not exhibit a significant effect on enhancing the efficacy of DOX in causing cell death in this particular cell line. But, when the co-delivery of DOX and MEL took place within Exo, it demonstrated the most pronounced effect in inducing cell death in MDA-MB231 cells.
In general, Exo increases DOX toxicity in MCF-7 and MDA-MB231 cells and decreases it in MCF-10A and AMSC cells. As a result, by targeting cancer cells, it can prevent unwanted side effects on normal cells to some extent. The selectivity index (SI) was calculated as the ratio of cytotoxicity (IC50) observed in normal cells (MCF-10a and AMSC) to that in cancer cells (MCF-7 and MDA-MB-231). An SI value greater than 3 suggests the selectivity of the cytotoxic effect of DOX specifically toward MCF-7 and MDA-MB-231 cells [30].
According to the results of Tables 2 and 3, free DOX cannot act selectively on any of MCF-7 and MDA-MB231 cell lines. Although the combination of DOX with MEL was able to reduce the IC50 compared to free DOX, it was selective only in MCF-7 cells over A-MSC cells. Furthermore, the outcomes from the encapsulation of DOX in EXOs were similar to the results of combining DOX with MEL. Co-delivery of DOX and MEL in EXOs (Exo-Dox-Mel) amplified the selectivity index in MCF-7 and MDA-MB-231 cells. Subsequently, it can be concluded that, in contrast to free DOX, unlike free DOX, which affects not only cancer cells but also rapidly dividing normal cells, the Exo-DOX-Mel demonstrates selectivity toward breast cancer cells while exhibiting less impact on normal cells.
Table 1
IC50 concentration of free-DOX, free-MEL, DOX/MEL, free-Exo, Exo-DOX, Exo-MEL, and Exo-DOX-MEL against Saos-2, MG63, and hBM-MSC for 24, 48, and 72 h.
IC50 | | free-DOX (µg/mL) | free-MEL (µg/mL) | DOX/MEL (µg/mL) | free-Exo (µg/mL) | Exo-DOX (µg/mL) | Exo-MEL (µg/mL) | Exo-DOX-MEL (µg/mL) |
MCF-7 | 24h | 2.21 | - | 2.08 | 50 | 1.83 | 5.65 | 0.89 |
48h | 1.93 | - | 1.49 | 37.5 | 0.98 | 5.07 | 0.71 |
72h | 1.2 | 33.54 | 0.86 | 18.75 | 0.91 | 3.64 | 0.38 |
MCF-10A | 24h | 2.11 | - | 4.04 | - | 4.6 | - | 5.35 |
48h | 1.06 | - | 1.1 | - | 1.25 | - | 2.49 |
72h | 0.64 | - | 0.71 | - | 0.76 | 34.86 | 1.79 |
MDA-MB231 | 24h | 5.4 | - | 4.68 | - | 3.48 | 12.22 | 1.76 |
48h | 2.88 | - | 3.13 | - | 2.86 | 8.09 | 0.81 |
72h | 1.67 | - | 3.08 | 39.49 | 1.47 | 6.17 | 0.57 |
A-MSC | 24h | 4.8 | - | 5.65 | - | 6.49 | - | 8.52 |
48h | 3.95 | - | 6.34 | - | 6.41 | - | 9.15 |
72h | 3.06 | - | 4.14 | - | 5.73 | - | 10.07 |
Table 2
Selectivity index (SI) of treatments for MCF-7 and MDA-MB231 cells.
SI | free-DOX (µg/mL) | free-MEL (µg/mL) | DOX/MEL (µg/mL) | free-Exo (µg/mL) | Exo-DOX (µg/mL) | Exo-MEL (µg/mL) | Exo-DOX-MEL (µg/mL) |
MCF-10A/MCF-7 | 24h | 0.95 | - | 1.9 | - | 2.51 | - | 6.01 |
48h | 0.54 | - | 0.73 | - | 1.27 | - | 3.50 |
72h | 0.53 | - | 0.82 | - | 0.83 | 9.57 | 4.71 |
A-MSC /MCF-7 | 24h | 2.17 | - | 2.71 | - | 3.54 | - | 9.57 |
48h | 2.04 | - | 4.25 | - | 6.54 | - | 12.88 |
72h | 2.55 | - | 4.81 | - | 6.29 | - | 26.5 |
MCF-10A/MDA-MB231 | 24h | 0.39 | - | 0.86 | - | 1.32 | - | 3.03 |
48h | 0.36 | - | 0.35 | - | 0.43 | - | 3.07 |
72h | 0.38 | - | 0.23 | - | 0.51 | 5.6 | 3.14 |
A-MSC/MDA-MB231 | 24h | 0.88 | - | 1.20 | - | 1.86 | - | 4.84 |
48h | 1.37 | - | 2.02 | - | 2.24 | - | 11.29 |
72h | 1.83 | - | 1.34 | - | 3.89 | - | 3.14 |
Gene expression changes as a result of the combination of MEL with DOX and loading in EXO
Since DOX interacts with double-stranded DNA and increases free radicals, it causes double-stranded breaks in DNA structure, DNA damage and induction of apoptosis [45]. Therefore, we analyzed the mechanistic effects of DOX in combination with MEL loading in EXOs at the gene level by assessing the expression of p53, NANOG, and MIR-34a genes in MCF-7, MCF-10A, MDA-MB231, and A-MSC cell lines. P53 is recognized as the principal regulator of the cell cycle and apoptosis [46].
Two main comparisons were conducted: one toward untreated cells (control) and the other toward cells that received only DOX (free DOX), aimed at more precisely elucidating the mechanism of the drug's impact when using the EXO carrier and MEL antioxidant. In the current research, we demonstrated that the p53 gene, recognized as the primary regulator of the cell cycle and apoptosis, exhibited increased expression in cells treated with DOX, aligning with our expectations. In the presence of free DOX, MCF-10A and MDA-MB231 cells recorded the highest and lowest p53 gene expression, respectively, which indicates the inability of free DOX to effectively destroy resistant breast cancer cells (MDA-MB231) and side effects on the cells. In addition, free DOX caused a significant increase in p53 gene expression in normal breast tissue cells (MCF-10A) and mesenchymal stem cells (MSCs) compared to the control group, which indicates the beginning of apoptosis. EXO-loaded DOX showed a significant increase in p53 gene expression compared to free DOX in MCF-7, but had a protective role in MCF-10A and MSC cells, which caused the reduction of apoptosis due to the reduction of p53 gene expression. But in MDA-MB231 cells, EXO-DOX did not cause much change compared to free DOX. Co-delivery of DOX with MEL in EXOs significantly increases the expression of p53 gene in cancer cells and has a much greater protective effect in normal cells compared to the control group and the group receiving free DOX (Fig. 9).
p53 is directly related to NANOG; a key transcription factor for the maintenance of embryonic stem cell pluripotency, which has recently been shown to be overexpressed in many types of human cancers, including breast cancer and it is related to increasing drug resistance properties of cancer cells. NANOG expression is modulated through p53 deacetylation, and induction of p53 expression by DNA damage represses NANOG and induces embryonic stem cells (ESC) to differentiate into cell types that can undergo cell death to efficiently remove damaged DNA, which ultimately prevents tumorigenesis [47, 48].
Our findings indicate that the expression of NANOG significantly decreased in MCF-7, MCF-10A, and A-MSC cells treated with free DOX, in comparison to the control group. This reduction in NANOG expression has the potential to induce P53-mediated cell death [49]. In the case of MDA-MB231 cells, the effect of free DOX on NANOG expression was observed to exhibit a slight decrease compared to the control group, which was not statistically significant and indicated a possible connection with the inherent resistance of these cells to chemotherapy agents. In both the DOX-MEL and EXO-DOX groups, the expression of NANOG was significantly downregulated in MDA-MB231 and MCF-7 cells, in comparison to the free DOX group. Conversely, an elevated expression of NANOG was observed in MCF-10A and A-MSC cells when compared to the free DOX group. In the DOX-MEL, EXO-DOX, and EXO-DOX-MEL treatment groups, a notable decrease in the expression of NANOG was observed in MDA-MB231 and MCF-7 cells when compared to the free DOX group. Conversely, increased NANOG expression was detected in MCF-10A and A-MSC cells when compared to the free DOX group. Of particular significance, the EXO-DOX-MEL group exhibited the most pronounced changes, highlighting its effectiveness in comparison to the other treatment groups.
These results demonstrate that the loading of DOX in EXOs in combination with MEL (EXO-DOX-MEL) leads to a reduction in NANOG expression in cancer cells, thereby amplifying the toxicity of DOX in these cell lines and inducing heightened apoptosis through P53 activation. Moreover, increasing Nanog expression in normal cells may partially reduce the DOX cytotoxic effects and prevent the destruction of these cells [50].
In this regard, Ebeid et al showed that free DOX combined with berberine can significantly reduce the expression of NANOG compared to the control group [49]. Also, similar studies emphasized that inhibiting Nanog, reverses resistance to chemotherapy and radiation therapy [51]. It was also found that removing NANOG by reducing the expression of MDR1 increases the chemical sensitivity of liver cancer cells to DOX [52].
Cardiotoxicity is a known side effect of DOX and has been linked to the activity of MiR-34a. MiR-34a plays a crucial role in the p53-mediated antitumor process, and its upregulation in tumor cells can enhance their sensitivity to chemotherapy drugs [47, 53]. Moreover, research findings indicate that miR-34a acts as a barrier to reprogramming by suppressing pluripotency marker genes, including NANOG. The repression of Nanog by p53 occurs through direct transcriptional silencing as well as indirect post-transcriptional silencing mediated by miR-34 [54, 55].
By assessing the levels of MiR-34a, our study approved the findings of p53. Following DNA damage, the activation of miR-34a by p53 arrested the cell cycle and enhanced apoptosis in breast cancer cells upon treatment with EXO-DOX-MEL. Conversely, in normal cells, miR-34a played a protective role, potentially guarding against DOX-induced cardiotoxicity.
In the study of Lee et al., miR-34a by inhibiting Sirt1 expression leads to increased activity of p53, suppression of cell reprogramming by Nanog and finally apoptosis [47]. In this regard, Zheng et al. stated that the overexpression of MiR-34a enhances the inhibitory effect of DOX on HepG2 cancer cells [55]. In addition, Correia Marques et al published that high expression of miR-34a improves the response to DOX in diffuse B-cell lymphoma [56]. In addition, Deng et al found that miRNA-34a can enhance the antitumor activity of DOX in the treatment of triple negative breast cancer [57].
The co-delivery of DOX and MEL within EXOs enhances the rate of apoptosis.
The apoptosis rate was evaluated using Annexin V and PI staining. MCF-7, MCF-10A, MDA-MB231, and A-MSC lines were treated with free-DOX, free-MEL, DOX/MEL, free-Exo, Exo-DOX, Exo-MEL, and Exo-DOX-MEL. The control group consisted of untreated cells (Fig. 10).
Treatment with free DOX in four cell lines include MCF-7, MCF-10A, MDA-MB231, and A-MSC caused 47.9%, 52.5%, 18.19%, and 46.1% apoptosis (sum of late and early apoptosis), respectively. Among the cell lines tested, the highest and lowest cell death was occurred in MCF-10A and MDA-MB231 cell lines, respectively, indicating the toxicity of DOX on normal cells and the relative resistance of triple negative cancer cells. Interestingly, the combination of MEL with DOX not only increased the death rate in MCF-7 cells (66.4%) but also enhanced toxicity in normal cells (MCF-10A: 38.2% and A-MS: 39.5%). In contrast, its impact on MDA-MB231 cells was relatively modest (23.05%).
In our study, Exo, primarily utilized as a drug carrier, exhibited a notable self-healing effect. It demonstrated significant apoptotic effects of Exos in MCF-7, MCF-10A, MDA-MB231, and A-MSC cells, resulting in respective cell death rates of 37.68%, 8.05%, 13.58%, and 6.59%. These findings suggest a higher susceptibility of cancer cells to Exo-induced cell death compared to normal cells. Exo in combination with DOX or MEL could slightly enhance the effects of these drugs alone. But the highest rate of cancer cell death was 80.82% and 56.1% apoptosis in MCF-7 and MDA-MB231, respectively, in treatment with Exo-DOX-MEL. Also, treatment with Exo-DOX-MEL in normal cells resulted in 18.13% and 7.14% apoptosis in MCF-10A and A-MSC, respectively. Subsequently, co-delivery of DOX and MEL in Exo could increase the toxicity in cancer cells, especially MDA-MB231 cells, which were resistant to free DOX. It also greatly protected normal cells from damage caused by DOX. This function can be partially attributed to melatonin, which plays a role as an antioxidant in destroying cancer cells and protecting normal cells [58].
Figure 11 shows the images of apoptosis evaluation and apoptotic cells using AO/EB staining. Live cells are uniformly green in color. Early apoptotic cells have a yellow color, chromatin density and membrane bubbles, while late apoptotic cells have orange to red pieces, fragmentation of chromatin and nucleus, and the presence of apoptotic bodies. Our results showed the toxicity of free DOX on MCF-7, MCF-10A, and A-MSC cells without significant effect on MDA-MB231 cells. Treatment with DOX/MEL, Exo-DOX and Exo-DOX-MEL enhances the effect of DOX in cancer cells and protects normal cells from DOX toxicity, among them, the most effective combination is Exo-DOX-MEL, which has shown a significant effect even on MDA-MB231 cells.