Prediction of Drug Synergism between Peptides and Antineoplastic Drugs Paclitaxel, 5-Fluorouracil, and Doxorubicin Using In Silico Approaches

Chemotherapy is the main treatment for most early-stage cancers; nevertheless, its efficacy is usually limited by drug resistance, toxicity, and tumor heterogeneity. Cell-penetrating peptides (CPPs) are small peptide sequences that can be used to increase the delivery rate of chemotherapeutic drugs to the tumor site, therefore contributing to overcoming these problems and enhancing the efficacy of chemotherapy. The drug combination is another promising strategy to overcome the aforementioned problems since the combined drugs can synergize through interconnected biological processes and target different pathways simultaneously. Here, we hypothesized that different peptides (P1–P4) could be used to enhance the delivery of chemotherapeutic agents into three different cancer cells (HT-29, MCF-7, and PC-3). In silico studies were performed to simulate the pharmacokinetic (PK) parameters of each peptide and antineoplastic agent to help predict synergistic interactions in vitro. These simulations predicted peptides P2–P4 to have higher bioavailability and lower Tmax, as well as the chemotherapeutic agent 5-fluorouracil (5-FU) to have enhanced permeability properties over other antineoplastic agents, with P3 having prominent accumulation in the colon. In vitro studies were then performed to evaluate the combination of each peptide with the chemotherapeutic agents as well as to assess the nature of drug interactions through the quantification of the Combination Index (CI). Our findings in MCF-7 and PC-3 cancer cells demonstrated that the combination of these peptides with paclitaxel (PTX) and doxorubicin (DOXO), respectively, is not advantageous over a single treatment with the chemotherapeutic agent. In the case of HT-29 colorectal cancer cells, the combination of P2–P4 with 5-FU resulted in synergistic cytotoxic effects, as predicted by the in silico simulations. Taken together, these findings demonstrate that these CPP6-conjugates can be used as adjuvant agents to increase the delivery of 5-FU into HT-29 colorectal cancer cells. Moreover, these results support the use of in silico approaches for the prediction of the interaction between drugs in combination therapy for cancer.


Introduction
Conventional chemotherapeutic drugs are widely used in cancer therapy; however, their success rate is commonly impaired by the development of adverse side effects and by the development of drug resistance, which together contribute to the inefficiency of chemotherapy, ultimately resulting in patient death [1]. The combination of conventional pools of chemotherapeutic drugs in cancer cells [16]. PEG is the most applied polymer in drug delivery and remains the gold standard for stealth polymers. Taken together, this conjugation aimed to improve the pharmaceutical properties of CPP6, including better water solubility, lower immunogenicity, and prolonged blood circulation [16]. A dipeptide composed of the amino acids Trp and Ser (P4, Scheme 1D) was also included in this work as a comparator due to the ability of these amino acids to interact with the cellular membrane. Therefore, we coupled in silico studies with in vitro approaches to try to establish a relationship between the results obtained using computer-based software and in vitro results, trying to predict the interaction between drugs in combination. their anticancer efficacy. Structurally, P3 is a conjugate consisting of CPP6 (P1) and CFLG (P2) linked by PEG. The conjugation of CPP6 with PEG and GFLG theoretically stabilizes the intracellular pools of chemotherapeutic drugs in cancer cells [16]. PEG is the most applied polymer in drug delivery and remains the gold standard for stealth polymers. Taken together, this conjugation aimed to improve the pharmaceutical properties of CPP6, including better water solubility, lower immunogenicity, and prolonged blood circulation [16]. A dipeptide composed of the amino acids Trp and Ser (P4, Scheme 1D) was also included in this work as a comparator due to the ability of these amino acids to interact with the cellular membrane. Therefore, we coupled in silico studies with in vitro approaches to try to establish a relationship between the results obtained using computerbased software and in vitro results, trying to predict the interaction between drugs in combination. We first performed in silico studies to predict the permeability and other structurerelated pharmacokinetic (PK) properties of each peptide and antineoplastic agent. Physiologically based pharmacokinetic (PBPK) models have proven useful for integrating all the parameters that influence the parameters related to the properties of the drugs and the physiological parameters specific to the species [19]. The modeling platform GastroPlus ® , a mechanistically based simulation software, can be useful to predict these pharmacokinetic proprieties using oral advanced compartmental and transit (ACAT). The generic GastroPlus ® ACAT model provided reasonable predictions especially for biopharmaceutical classification system (BCS) class 1 compounds [20]. In addition, the ability of PBPK models to predict oral PK will also improve, providing a better tool for the discovery and development of new medicines [21][22][23][24], as drug combinations or drug synergism. We first performed in silico studies to predict the permeability and other structurerelated pharmacokinetic (PK) properties of each peptide and antineoplastic agent. Physiologically based pharmacokinetic (PBPK) models have proven useful for integrating all the parameters that influence the parameters related to the properties of the drugs and the physiological parameters specific to the species [19]. The modeling platform GastroPlus ® , a mechanistically based simulation software, can be useful to predict these pharmacokinetic proprieties using oral advanced compartmental and transit (ACAT). The generic GastroPlus ® ACAT model provided reasonable predictions especially for biopharmaceutical classification system (BCS) class 1 compounds [20]. In addition, the ability of PBPK models to predict oral PK will also improve, providing a better tool for the discovery and development of new medicines [21][22][23][24], as drug combinations or drug synergism.
Next, peptides were evaluated in vitro, both alone and combined with different chemotherapeutic drugs, in three cancer cell lines: MCF-7 (breast), HT-29 (colon), and PC-3 (prostate), to determine if these combinations could be used to enhance the anticancer potential of the antineoplastic drugs. The antineoplastic drugs such as PTX, 5-FU, and DOXO were selected to be used as reference chemotherapeutic drugs for breast, colon, and prostate cancer cells, respectively. Finally, after determining the most promising drug combinations, the interaction between each peptide and chemotherapeutic drug in the combination was evaluated to determine eventual synergistic interactions through the calculation of the combination index using the Chou-Talalay method [25].
Our in silico findings have predicted the peptides P2-P4 to have higher bioavailability and lower T max , as well as the chemotherapeutic agent 5-FU to have enhanced permeability properties over other antineoplastic agents, with P3 having a prominent accumulation in the colon. Therefore, we hypothesized that the combination of 5-FU with these peptides in vitro would lead to increased delivery of 5-FU into colorectal cancer cells, increasing the anticancer activity of this drug and possibly resulting in drug synergism between these two molecules. Indeed, our findings in MCF-7 and PC-3 cancer cells demonstrated that the combination of PTX and DOXO, respectively, with these peptides is not advantageous over a single treatment with the chemotherapeutic agent. Nevertheless, in the case of HT-29 colorectal cancer cells, the combination of P2, P3, and P4 with 5-FU resulted in synergistic cytotoxic effects, as predicted by the in silico simulations.
Taken together, these results demonstrate that the combination of these peptides with 5-FU may enhance the delivery of the antineoplastic agent into colorectal cancer cells. Moreover, these findings support the idea that in silico simulations may be an important tool for the prediction of the interaction between drugs in combination interaction for cancer therapy.

In Silico Evaluation of the PK Properties of 5-FU, DOXO, and PTX
We first simulated the plasma concentration for 24 h of the chemotherapeutic agents 5-FU, DOXO, and PTX as well as the compartmental absorption of these drugs and other PK parameters using the GastroPlus ® software. These simulations were performed for an American individual, 30 years old, with an initial dose of 50 mg for 24 h. The simulations regarding the chemotherapeutic agents and peptides were run separately because these compounds were also administered separately to the cancer cells.
The detailed PK parameters obtained for 5-FU, DOXO, and PTX are described in Table 1 and the plasma concentration and compartmental absorption of each drug are represented in Figures 1-3. Table 1. Summary of simulated mean plasma PK parameters for GastroPlus ® simulations of 5-FU, DOXO, and PTX. The pharmacokinetic parameters are %F (% bioavailability), which is the percent of drug that reached the systemic circulation, T max , C max , and AUC (area under the curve).      The results from Table 1 regarding the PK parameters for the GastroPlus ® simulation of 5-FU, DOXO, and PTX demonstrate that 5-FU has a higher Peff (2.81 cm/s × 10 4 ) than DOXO and PTX (0.26 and 0.21 cm/s × 10 4 , respectively), which may indicate that this drug will have enhanced ability to enter cancer cells and to exert cytotoxic effect than DOXO and PTX. Moreover, the analysis of the bioavailability (F) of 5-FU demonstrates it has a higher bioavailability than DOXO and PTX (55.66 vs. 49.40 and 5.47%, respectively), which may indicate this drug will be more available to interact with other molecules such as peptides than DOXO and PTX.  The results from Table 1 regarding the PK parameters for the GastroPlus ® simulation of 5-FU, DOXO, and PTX demonstrate that 5-FU has a higher P eff (2.81 cm/s × 10 4 ) than DOXO and PTX (0.26 and 0.21 cm/s × 10 4 , respectively), which may indicate that this drug will have enhanced ability to enter cancer cells and to exert cytotoxic effect than DOXO and PTX. Moreover, the analysis of the bioavailability (F) of 5-FU demonstrates it has a higher bioavailability than DOXO and PTX (55.66 vs. 49.40 and 5.47%, respectively), which may indicate this drug will be more available to interact with other molecules such as peptides than DOXO and PTX.

5-FU DOXO PTX
Analyzing the plasma concentration representations (Figures 1-3), it is possible to verify that 5-FU ( Figure 1) has an improved PK profile compared to DOXO and PTX (Figures 2 and 3, respectively), characterized by a rapid absorption as well as a rapid elimination rate. Rapid elimination is especially important for chemotherapeutic drugs to avoid a prolonged accumulation of the drugs inside the organism that could further lead to the development of undesired side effects.
The compartmental absorption of 5-FU, DOXO, and PTX is represented in Figures S1-S3. These results demonstrate that 5-FU and DOXO have higher absorption in jejunum (Figures S1 and S2, respectively), while PTX accumulates most in the ascending colon ( Figure S3).
Taken together, these results demonstrate that 5-FU may have an improved PK profile over DOXO and PTX, which may indicate better anticancer effects in cancer cells.

In Silico Evaluation of the PK Properties of Peptides P1-P4
Then, we simulated the plasma concentration for 24 h of peptides P1-P4 as well as their compartmental absorption and other PK parameters using the GastroPlus ® software. These simulations were performed using the same parameters as previously described for 5-FU, DOXO, and PTX. These results are shown in Figures 4-7 and Table 2.         The analysis of PK parameters from Table 2 demonstrates that P2-P4 are characterized by having enhanced bioavailability (32.77, 17.73, and 37.45%, respectively) than P1 (13.15%), which indicates that these peptides will have more availability to interact with the chemotherapeutic agents when administered in combination than P1.  The analysis of PK parameters from Table 2 demonstrates that P2-P4 are characterized by having enhanced bioavailability (32.77, 17.73, and 37.45%, respectively) than P1 (13.15%), which indicates that these peptides will have more availability to interact with the chemotherapeutic agents when administered in combination than P1.  The analysis of plasma concentration over time demonstrates that peptides P1, P2, and P4 (Figures 4-6, respectively) are characterized by having a rapid absorption rate as well as an elimination profile. P3 ( Figure 7) has a slightly different PK profile, being rapidly absorbed but having a reduced elimination rate compared to peptides P1, P2, and P4. A slower elimination rate can be specifically helpful when using peptides for enhancing drug transport to increase their time inside the organism. Some studies have suggested that some CPPs have enhanced uptake efficacy and delivery efficiency than other similar treatments, such as nanoparticles, while presenting less cytotoxicity [26]. Indeed, several peptides have already entered Phase I, Phase II and even, Phase III clinical trials [2]. Therefore, assuming these peptides also have an acceptable biosafety profile, their increased time inside the organism may not result in side effects.
The analysis of PK parameters from Table 2 demonstrates that P2-P4 are characterized by having enhanced bioavailability (32.77, 17.73, and 37.45%, respectively) than P1 (13.15%), which indicates that these peptides will have more availability to interact with the chemotherapeutic agents when administered in combination than P1. Moreover, peptides P2-P4 are characterized by having lower T max (2.8, 3.84, and 2.72 h, respectively) than peptide P1 (4.24 h), which indicates that these peptides will also interact faster and produce a rapid therapeutic effect when combined with the chemotherapeutic agents than P1.
The representation of the Blood/Plasma concentration ratio versus P eff for peptides P1-P4 ( Figure 8) demonstrates that P2 and P4 demonstrate higher values of these parameters, which are in line with the enhanced profiles of bioavailability of these peptides.
Taken together, these results suggest that drug combinations using 5-FU in HT-29 colorectal cancer cells will result in enhanced anticancer effects than using DOXO and PTX in PC-3 prostate and MCF-7 breast cancer cells, respectively. Furthermore, these results also predict that due to the higher bioavailability and lower T max of P2-P4, these peptides will be able to interact most strongly and faster with 5-FU, which can lead to synergistic interactions. As P3 is a conjugate of CPP6 (P1) and CFLG (P2) connected by a linker (PEG), we hypothesized this peptide would also result in enhanced drug transport and internalization of 5-FU into cancer cells than P1 and P2.
The representation of the Blood/Plasma concentration ratio v P1-P4 ( Figure 8) demonstrates that P2 and P4 demonstrate hi parameters, which are in line with the enhanced profiles of bi peptides. The results from compartmental absorption also demonstrate ( Figure S5), and P4 ( Figure S7) have higher absorption in the jejunum respectively) than P3 (Figure S6, 1.5%), while this peptide has an in in the ascendent colon (17.2%) compared to other peptides (3.4, 1.1, P1, P2, and P4, respectively). Therefore, we predicted that P3 would in HT-29 colon cancer cells than in MCF-7 and PC-3 prostate cance Taken together, these results suggest that drug combinations colorectal cancer cells will result in enhanced anticancer effects than

In Vitro Studies on the Anticancer Activity of Peptides P1-P4 Alone and Combined with Chemotherapeutic Agents
To further confirm our previous in silico predictions, we next evaluated all peptides, both alone and in combination with 5-FU, DOXO, and PTX, in HT-29, PC-3, and MCF-7 cancer cells, respectively. After finding the most promising drug combinations, the calculation of the combination index was performed using the Chou-Talalay method to assess for synergic interactions.

Anticancer Activity of Peptides P1-P4
First, the four peptides (P1-P4) were tested alone on three different cell lines (MCF-7, HT-29, and PC-3), using concentrations of 0.01, 0.1, 1, 10, 25, and 50 µM for 48 h. The results for the MCF-7 cell line regarding cell viability and morphological images are represented in Figures 9 and 10, respectively. P1 and P2 had no effects on cell viability ( Figure 9A,B, respectively), which made it impossible to obtain an IC 50 value with the concentrations used. Meanwhile, P3 and P4 showed a statistically significant decrease in cell viability when cells were treated with the highest concentration (50 µM) ( Figure 9C,D, respectively). However, this decrease was only about 20% of cell viability reduction, both for P3 and P4, so no IC 50 could be obtained for these concentrations as well. These results were in accordance with the morphological analysis of MCF-7 cells treated with each peptide (Figure 10), where it was found that the number of MCF-7 cells decreased in the treatments with P2-P4 at a concentration of 50 µM ( Figure 10).
The results for the HT-29 cells regarding cell viability and morphological evaluation of cells from peptides tested alone at concentrations of 0.01, 0.1, 1, 10, 25, and 50 µM for 48 h can be found in Figures 11 and 12, respectively. The results are similar to those obtained for MCF-7, with P1 and P2 having no significant effect on the reduction of cell viability ( Figure 11A,B, respectively); on the other hand, P3 and P4 caused a significant decrease in cell viability but only at the highest concentration of 50 µM ( Figure 11C,D, respectively). Likewise, no IC 50 could be calculated for any peptide with the concentrations tested, meaning that it must be higher than 50 µM. The results of the morphological analysis ( Figure 12) are in accordance with the MTT results and demonstrate a reduction in the number and size of cellular aggregates for treatments with P3 and P4 at a concentration of 50 µM.
The following step was to assess the cytotoxicity of each peptide in the PC-3 prostate cells. Concentrations of 0.01, 0.1, 1, 10, 25, and 50 µM were tested for 48 h, and the results of cell viability and morphological evaluation images can be seen in Figures 13 and 14, respectively. The results demonstrate that there was a significant decrease in cell viability for all peptides at the concentration of 25 µM ( Figure 13); moreover, P2 and P4 also induced a significant reduction in the number of viable cells at the concentration of 0.01 µM ( Figure 13B,C, respectively). Based on these results, no IC 50 could be obtained for any peptide in this cell line. The results regarding morphological evaluation ( Figure 14) are in accordance with the MTT results and demonstrate a reduction in the number of cells in the treatments with all peptides at a concentration of 25 µM.
The summary of the results regarding the viability of all cell lines treated with increasing concentrations of each peptide is represented in Figure 15. It can be concluded that P1 has the least influence on MCF-7 cell viability, while P3 together with P4 had the strongest influence on the viability of this cell line ( Figure 15A). Regarding HT-29 cells, like MCF-7 cells, P1 had the smallest effect on cell viability, while P4 had the greatest ( Figure 15B). Generally, for PC-3 cells, all the peptides had a comparable profile, with P4 having slightly lower cell viability values ( Figure 15C).
These results demonstrate that peptides P1-P4 do not induce significant anticancer effects in HT-29 colon and MCF-7 breast cancer cell lines at lower concentrations (<25 µM).
In the case of PC-3 cells, these peptides are also relatively safe for this cell line and do not induce a reduction of cell viability of more than 30%. Taken together, these results also support that these peptides have an acceptable safety profile, making them ideal candidates to increase drug transport and delivery into cells.
in PC-3 prostate and MCF-7 breast cancer cells, respectively. Furthermore, these resu also predict that due to the higher bioavailability and lower Tmax of P2-P4, these pepti will be able to interact most strongly and faster with 5-FU, which can lead to synergi interactions. As P3 is a conjugate of CPP6 (P1) and CFLG (P2) connected by a linker (PE we hypothesized this peptide would also result in enhanced drug transport a internalization of 5-FU into cancer cells than P1 and P2.

In Vitro Studies on the Anticancer Activity of Peptides P1-P4 Alone and Combined with Chemotherapeutic Agents
To further confirm our previous in silico predictions, we next evaluated all peptid both alone and in combination with 5-FU, DOXO, and PTX, in HT-29, PC-3, and MC cancer cells, respectively. After finding the most promising drug combinations, calculation of the combination index was performed using the Chou-Talalay method assess for synergic interactions.  Figure 9A,B, respectively), which made it impossible to obtain an IC50 value with concentrations used. Meanwhile, P3 and P4 showed a statistically significant decreas cell viability when cells were treated with the highest concentration (50 μM) ( Figure 9C respectively). However, this decrease was only about 20% of cell viability reduction, b for P3 and P4, so no IC50 could be obtained for these concentrations as well. These res were in accordance with the morphological analysis of MCF-7 cells treated with e peptide (Figure 10), where it was found that the number of MCF-7 cells decreased in treatments with P2-P4 at a concentration of 50 µM ( Figure 10).   The results for the HT-29 cells regarding cell viability and morphological evaluation of cells from peptides tested alone at concentrations of 0.01, 0.1, 1, 10, 25, and 50 μM for 48 h can be found in Figures 11 and 12, respectively. The results are similar to those viability ( Figure 11A,B, respectively); on the other hand, P3 and P4 caused a significant decrease in cell viability but only at the highest concentration of 50 μM ( Figure 11C,D,  respectively). Likewise, no IC50 could be calculated for any peptide with the concentrations tested, meaning that it must be higher than 50 μM. The results of the morphological analysis ( Figure 12) are in accordance with the MTT results and demonstrate a reduction in the number and size of cellular aggregates for treatments with P3 and P4 at a concentration of 50 µM.

Anticancer Efficacy of the Combination of Peptides P1-P4 with Chemotherapeutic Drugs
After finding that peptides P1-P4 do not demonstrate promising anticancer effects on these cell lines, we hypothesized that their combination with conventional chemotherapy could help increase the transport of such drugs into different cancer cells and therefore increase the anticancer potential of these antineoplastic agents.
To do so, we combined increasing concentrations of peptides P1-P4 with conventional antineoplastic agents commonly used for breast, colon, and prostate cancer, namely PTX, 5-FU, and DOXO, respectively, at their IC 50 ( Table 3). As these drugs are widely used for cancer therapy, their IC 50 value was obtained from the literature [27][28][29].   respectively. The results demonstrate that there was a significant decrease in cell viability for all peptides at the concentration of 25 μM ( Figure 13); moreover, P2 and P4 also induced a significant reduction in the number of viable cells at the concentration of 0.01 μM ( Figure 13B,C, respectively). Based on these results, no IC50 could be obtained for any peptide in this cell line. The results regarding morphological evaluation ( Figure 14) are in accordance with the MTT results and demonstrate a reduction in the number of cells in the treatments with all peptides at a concentration of 25 µM.   Figures 16 and 17, respectively. For all peptides, the combination was more effective than the peptide alone at all ranges of concentrations ( Figure 16). Meanwhile, when compared with PTX alone, the combinations were not as effective in decreasing MCF-7 cell viability, and the combined effect seems to reflect the anticancer activity of the chemotherapeutic agent; nevertheless, it is clear that increasing peptide concentrations also interfered with the combined effect. The results of the morphological analysis ( Figure 17) are in accordance with these results and demonstrate a pronounced reduction in cell number and changes in the size and shape of MCF-7 cells after all treatments, compared to vehicle.
Combination studies were then performed in PC-3 cells treated with each peptide and DOXO. The results of cell viability and the morphological evaluation images from a combination of 8 µM of DOXO with increasing concentrations (0.01, 0.1, 1, 10, 25, and 50 µM) of each peptide and their comparison against DOXO and the peptides alone are shown in Figures 18 and 19, respectively. All combinations induced a significant reduction of cell viability compared to each peptide alone ( Figure 18); nevertheless, similar to the MCF-7 results, no significant results were found between the anticancer effect of DOXO alone and the combined pairs, demonstrating that the combined anticancer effect may be derived from the anticancer drug alone. These results are supported by the analysis of the cellular morphology represented in Figure 19, where cells become smaller, rounder, and fewer in number after treatments with DOXO and all drug combinations, compared to control cells.  The summary of the results regarding the viability of all cell lines treated with increasing concentrations of each peptide is represented in Figure 15. It can be concluded that P1 has the least influence on MCF-7 cell viability, while P3 together with P4 had the strongest influence on the viability of this cell line ( Figure 15A). Regarding HT-29 cells, like MCF-7 cells, P1 had the smallest effect on cell viability, while P4 had the greatest ( Figure 15B). Generally, for PC-3 cells, all the peptides had a comparable profile, with P4 having slightly lower cell viability values ( Figure 15C). These results demonstrate that peptides P1-P4 do not induce significant anticancer effects in HT-29 colon and MCF-7 breast cancer cell lines at lower concentrations (< 25 µM). In the case of PC-3 cells, these peptides are also relatively safe for this cell line and do not induce a reduction of cell viability of more than 30%. Taken together, these results also support that these peptides have an acceptable safety profile, making them ideal candidates to increase drug transport and delivery into cells.

Anticancer Efficacy of the Combination of Peptides P1-P4 with Chemotherapeutic Drugs
After finding that peptides P1-P4 do not demonstrate promising anticancer effects on these cell lines, we hypothesized that their combination with conventional chemotherapy could help increase the transport of such drugs into different cancer cells and therefore increase the anticancer potential of these antineoplastic agents.
To do so, we combined increasing concentrations of peptides P1-P4 with conventional antineoplastic agents commonly used for breast, colon, and prostate cancer, namely PTX, 5-FU, and DOXO, respectively, at their IC50 (Table 3). As these drugs are widely used for cancer therapy, their IC50 value was obtained from the literature [27][28][29]. To perform the combination studies in MCF-7 cells, 3 nM of PTX was combined with concentrations of 0.01, 0.1, 1, 10, 25, and 50 μM of each peptide. The results of cell viability of combinations against each compound alone and the morphological evaluation are represented in Figures 16 and 17, respectively. For all peptides, the combination was more effective than the peptide alone at all ranges of concentrations ( Figure 16). Meanwhile, when compared with PTX alone, the combinations were not as effective in decreasing MCF-7 cell viability, and the combined effect seems to reflect the anticancer activity of the chemotherapeutic agent; nevertheless, it is clear that increasing peptide concentrations also interfered with the combined effect. The results of the morphological analysis ( Figure 17) are in accordance with these results and demonstrate a pronounced reduction in cell number and changes in the size and shape of MCF-7 cells after all treatments, compared to vehicle. Control cells were treated with 0.01% DMSO (vehicle). Cell viability was obtained using the MTT assay, and the results are given as the mean ± SEM (n = 3). * Statistically significant vs. drug alone at p < 0.05, ** statistically significant vs. drug alone at p < 0.01, *** statistically significant vs. drug alone at p < 0.001, **** statistically significant vs. drug alone at p < 0.0001. Control cells were treated with 0.01% DMSO (vehicle). Cell viability was obtained using the MTT assay, and the results are given as the mean ± SEM (n = 3). * Statistically significant vs. drug alone at p < 0.05, ** statistically significant vs. drug alone at p < 0.01, *** statistically significant vs. drug alone at p < 0.001, **** statistically significant vs. drug alone at p < 0.0001.  viability compared to each peptide alone ( Figure 18); nevertheless, similar to the MCF-7 results, no significant results were found between the anticancer effect of DOXO alone and the combined pairs, demonstrating that the combined anticancer effect may be derived from the anticancer drug alone. These results are supported by the analysis of the cellular morphology represented in Figure 19, where cells become smaller, rounder, and fewer in number after treatments with DOXO and all drug combinations, compared to control cells. Control cells were treated with 0.01% DMSO (vehicle). Cell viability was obtained using the MTT assay, and the results are given as the mean ± SEM (n = 3). **** Statistically significant vs. drug alone at p < 0.0001.  Figure 20C) resulted in a higher reduction of HT-29 cell viability than both the corresponding peptides and 5-FU alone, demonstrating these peptides can be used to enhance the delivery of 5-FU into cancer cells when presented at higher concentrations. The results of the analysis of cell morphology ( Figure 21) support these results and demonstrate a reduction in the number and size of cell aggregates after all treatments, compared to vehicle. Figure 22 denotes the comparison of cell viability among all cell lines against the concentration of combinations for all peptides. There was practically no difference in values among the different peptides in MCF-7 cells ( Figure 22A). Combining the different peptides with 5-FU showed similar effects on cell viability between themselves in HT-29 cells ( Figure 22B). Comparing the combinations with each other in PC-3 cells shows a relatively uniform effect among them, with the DOXO and P2 combination being marginally stronger for higher concentrations of P2 ( Figure 22C).
Taken together, the results previously obtained in MCF-7 breast cancer cells and PC-3 prostate cancer cells demonstrate that the combination of peptides P1-P4 is not advantageous over the use of the chemotherapeutic agents alone. Nevertheless, in the case of HT-29 colorectal cancer cells, it was possible to verify that the combination of P1-P3 with 5-FU at higher concentrations (25 and 50 µM) induces higher cytotoxicity than 5-FU and each peptide alone, demonstrating that these combinations can be promising for increasing the efficacy of colorectal cancer therapy.
Furthermore, these results are in line with those obtained through the in silico simulations, where we predicted that 5-FU would have an enhanced ability to enter cancer cells and to exert more cytotoxic effects than DOXO and PTX, as well as higher bioavailability, making it more available to interact with peptides.  combined effect was better than only 5-FU alone ( Figure 20D). Combinations of 5-FU with 25 and 50 μM of P1 ( Figure 20A), P2 ( Figure 20B), and P3 ( Figure 20C) resulted in a higher reduction of HT-29 cell viability than both the corresponding peptides and 5-FU alone, demonstrating these peptides can be used to enhance the delivery of 5-FU into cancer cells when presented at higher concentrations. The results of the analysis of cell morphology ( Figure 21) support these results and demonstrate a reduction in the number and size of cell aggregates after all treatments, compared to vehicle. . Cell viability was obtained using the MTT assay, and the results are given as the mean ± SEM (n = 3). * Statistically significant vs. drug alone at p < 0.05, ** statistically significant vs. drug alone at p < 0.01, *** statistically significant vs. drug alone at p < 0.001, **** statistically significant vs. drug alone at p < 0.0001. . Cell viability was obtained using the MTT assay, and the results are given as the mean ± SEM (n = 3). * Statistically significant vs. drug alone at p < 0.05, ** statistically significant vs. drug alone at p < 0.01, *** statistically significant vs. drug alone at p < 0.001, **** statistically significant vs. drug alone at p < 0.0001.

Combination Index Evaluation in the Combination of Peptides P1-P4 with 5-FU in HT-29 Colorectal Cancer Cells
After finding the most promising drug combinations based on the cell viability results and the software CompuSyn, we finally assessed the combination synergism. This was only performed for the combinations of 5-FU plus each peptide, since the combinations tested in MCF-7 and PC-3 cancer cells did not present significant combined anticancer effects.
The combination index (CI) plot for the drug combinations of 5-FU plus peptides P1-P4 can be seen in Figure 23; in this plot are represented the individual values of CI vs the fractional effect (Fa). CI values higher than one indicate antagonism, equal to one indicate additivity, and lower than one indicate synergism. An Fa value equal to zero indicates no cell death whereas one indicates complete cell death. These results are also summarized in Table 4. Table 4. Fractional effect (Fa) and CI values of peptides and 5-FU (3 µM) combinations for 48 h in HT-29 cells. C1 < 1 represents synergism, CI = 1 represents additivity, and C1 > 1 represents antagonism. The fractional effect shows the degree of cell death, with zero being no cellular death and one being total cellular death. CI in bold indicates drug pairs that are synergic.     Figure 22A). Combining the different peptides with 5-FU showed similar effects on cell viability between themselves in HT-29 cells ( Figure 22B). Comparing the combinations with each other in PC-3 cells shows a relatively uniform effect among them, with the DOXO and P2 combination being marginally stronger for higher concentrations of P2 ( Figure 22C). Taken together, the results previously obtained in MCF-7 breast cancer cells and PC-3 prostate cancer cells demonstrate that the combination of peptides P1-P4 is not advantageous over the use of the chemotherapeutic agents alone. Nevertheless, in the case of HT-29 colorectal cancer cells, it was possible to verify that the combination of P1-P3 with 5-FU at higher concentrations (25 and 50 µM) induces higher cytotoxicity than 5-FU and each peptide alone, demonstrating that these combinations can be promising for increasing the efficacy of colorectal cancer therapy.

Peptide
Furthermore, these results are in line with those obtained through the in silico simulations, where we predicted that 5-FU would have an enhanced ability to enter cancer cells and to exert more cytotoxic effects than DOXO and PTX, as well as higher bioavailability, making it more available to interact with peptides.  After finding the most promising drug combinations based on the cell viability results and the software CompuSyn, we finally assessed the combination synergism. This was only performed for the combinations of 5-FU plus each peptide, since the combinations tested in MCF-7 and PC-3 cancer cells did not present significant combined anticancer effects.
The combination index (CI) plot for the drug combinations of 5-FU plus peptides P1-P4 can be seen in Figure 23; in this plot are represented the individual values of CI vs the fractional effect (Fa). CI values higher than one indicate antagonism, equal to one indicate additivity, and lower than one indicate synergism. An Fa value equal to zero indicates no cell death whereas one indicates complete cell death. These results are also summarized in Table 4.
The combination index plot obtained for the combinations in HT-29 cells demonstrates that most CI values obtained for the combination of 5-FU with each peptide are below one, which indicates that most of the combined pairs present synergic interactions. Nevertheless, in the combination of 5-FU with P4, some CI values are closer to additivity or even antagonism, which is in accordance with the previous MTT results obtained in Figure 9.
Analyzing Table 4, no CI values were possible to calculate for the combination of P1 + 5-FU. All the combinations of 5-FU with P2-P4 demonstrate CI values of synergism except for the combination of 5-FU with 50 μM of P4, which was antagonistic. In fact, the CI values for the combination of 5-FU and P4 tend to increase with the increaseinf P4 concentration, which indicates that this combination gets worse with the increasing addition of peptide.  The combination index plot obtained for the combinations in HT-29 cells demonstrates that most CI values obtained for the combination of 5-FU with each peptide are below one, which indicates that most of the combined pairs present synergic interactions. Nevertheless, in the combination of 5-FU with P4, some CI values are closer to additivity or even antagonism, which is in accordance with the previous MTT results obtained in Figure 9.
Analyzing Table 4, no CI values were possible to calculate for the combination of P1 + 5-FU. All the combinations of 5-FU with P2-P4 demonstrate CI values of synergism except for the combination of 5-FU with 50 µM of P4, which was antagonistic. In fact, the CI values for the combination of 5-FU and P4 tend to increase with the increaseinf P4 concentration, which indicates that this combination gets worse with the increasing addition of peptide.
Taken together, these results are in line with the in silico predictions where it was found that P2, P3, and P4 had greater bioavailability compared to P1. Therefore, we believe that these peptides are more likely to interact with 5-FU and further increase the delivery of this drug into cells. Moreover, as we demonstrated that these peptides have a lower T max than P1, we also believe that these peptides interact faster and produce a rapid therapeutic effect when combined with 5-FU in these cells, resulting in synergistic interactions.
These results demonstrate that in silico studies may be a viable complement to predict the cytotoxic activity of chemotherapeutic agents and peptides, both alone and in combination, for in vitro studies using cancer cells.

Cell Morphology Visualization
Cell morphological analysis was assessed after each treatment and before the MTT assays using a Leica DMI 6000B microscope equipped with a Leica DFC350 FX camera. Images were then analyzed with the Leica LAS X imaging software (v3.7.4).

Synergistic Effect Analysis
To quantify the drug interaction between the chemotherapeutic drugs and each peptide, the Chou-Talalay equation and CompuSyn software (version 1.0; ComboSyn, Paramus, NJ, USA) were used to estimate the Combination Index (CI) [25]. The CI was used to determine the types of drug interactions, where CI < 1 indicates synergism, CI = 1 indicates additivity, and CI > 1 represents antagonism. The simulations were performed using a non-fixed ratio of doses with a fixed concentration of chemotherapeutic drug and variable concentrations of each peptide.

Statistical Analysis
All assays were performed in triplicate with at least three independent experiments. All data were reported as mean ± standard error of the mean (SEM). Statistical analysis was carried out using GraphPad Prism, version 9.0 (San Diego, CA, USA). Statistical analysis was performed using an ordinary one-way ANOVA, with Dunnett's multiple coparisons test. P-values < 0.05 were considered statistically significant.

Conclusions
Although in silico approaches represent a relatively new avenue of inquiry, these studies are starting to be used widely in studies to predict how drugs may interact and act with the organism and against cancer cells. In this study, we hypothesized that peptides P1-P4 could be used in combination with chemotherapeutic agents to enhance their delivery into different cancer cells. We performed in silico simulations to determine the PK profile of each peptide and chemotherapeutic agent, and based on these results, we made assumptions on the possible anticancer effect of these compounds in combination. We found 5-FU and P2-P4 to have the most promising PK profiles among all compounds simulated. We next evaluated each drug combination using in vitro assays and, in line with the in silico results, have found that higher concentrations of P2-P4 combined with 5-FU in HT-29 colorectal cancer cells resulted in enhanced cytotoxic effects than each molecule alone, characterized by synergism. Taken together, these results demonstrate that in silico approaches can be a promising auxiliary tool for the prediction of the interaction between drugs in combination and that these peptides can be used in combination with 5-FU to enhance the delivery of this drug into cancer cells for colorectal cancer therapy. In the future, these peptides should be further investigated for the determination of their exact mechanism of action in cancer cells and how they act synergistically with chemotherapeutic agents. Complementary assays should also be performed to confirm the efficacy of these peptides in the delivery of chemotherapeutic agents into cancer cells and to determine their safety against normal tissues. Moreover, since these peptides are biodegradable, their effect must be carefully studied in vivo to determine if they are still effective in more advanced cancer models in the delivery of anticancer agents.