Leaves of D. viscosa were collected in Chalhuanca, province of Aymaraes, in the department of Apurimac, Peru (2,880 masl) in July 2019. This plant was authenticated by Prof. Asunción Cano at the Herbarium of the Natural History Museum, Universidad Nacional Mayor de San Marcos, with a voucher specimen (125-USM-2019).

The powdered leaves (500 g) were extracted with 70% ethanol at 25°C for 7 days. Afterward, the hydroethanolic extract was filtered using a Buchner apparatus, and the filtrates was incorporated into a rota-evaporator to remove the solvent, resulting in a dark green solid extract. The hydroalcoholic extract weighed 34.02 g with a yield percentage of 6.8%

The results are presented as the mean SE (standard error) of at least two separate experiments. Utilizing one-way ANOVA and Dunnett’s post hoc test, statistical differences were evaluated. p values below 0.05 were considered significant. GraphPad Prism 7.04 for Windows was utilized for data analysis (Graph Pad Software, San Diego, California, United States).

The maceration of D. viscosa leaf powder in ethanol produced a dark green extract with a yield of 6.8% (w/w dry powder). The phytochemical constituents identified by LC-ESI-MS/MS are depicted in a positive and negative modes in Figure 1. Each retention time corresponds to a determined phytochemical constituent. The chromatographic analysis revealed 57 components, primarily phenolic compounds and diterpenes. Nine were detected in ESI (−), thirty-two were observed in ESI (+), and sixteen were observed in both modes (Table 1; Supplementary Table S1). Figure 1 depicts the positive and negative ESI chromatographic profiles for the hydroethanolic extract of D. viscosa leaves.

Some chemical constituents in other reports as Isorhamnetin (Al-Snafi, 2017), kaempferol (Omosa et al., 2010), quercetin [24], aliarin (Teffo et al., 2010), methyl dodovisate B (Niu et al., 2010), hardwickiic acid (Niu et al., 2010), fraxetin (Kaigongi et al., 2020), chlorogenic acid (Beshah et al., 2020), p-coumaric acid (Beshah et al., 2020), rutin, viscosol, catechin, and dodonic acid (Kaigongi et al., 2020) were determined by chromatographic analysis and showed antibacterial activity. Otherwise, some compounds in this study have been reported in the first time such as sucrose, quinic acid, shikimic acid, adenosine, N-(1-deoxy-1-fructosyl) isoleucine, tryptophan, 6,8-C-dihexosylnoreugenin, 6,8-hexosylnaringenin, procyanidin B2, 6-(3-benzoyloxy-2-hydroxypropoxy)-3,4,5- trihydroxyoxane-2-carboxylic acid, quercetin 3-O-[rhamnosyl-(1→2)- [rhamnosyl-(1→6)]-glucoside], 3-O-Feruloylquinic acid, kaempferol-3-O-[rhamnosyl-(1→2)- [rhamnosyl-(1→6)]-galactoside], 1-O-methyl-2-acetyl-3-p-coumaryl-myo- inositol, kaempferol-3-O-rutinoside, isorhamnetin 3-O-[α-L-Rhamnopyranosyl- (1→6)-β-D-galactopyranoside, Loliolide, 3,8,16-Trihydroxy-13-labden-15,16-olide; 5,7-dihydroxy-3′,4′,5′-trimethoxyfiavone, 7-O-[β-D-glucuronopyranosyl-(1→2)-β-Dglucopyranoside], naringenin, kauralexin B3, 3′,5,5′-Trihydroxy-7-methoxy flavanone, tamarixetin, mkapwanin, viscoflavone B, palmitamide, and 9-octadecenamide.

The cytotoxic effect of the hydroethanolic extract of D. viscosa and 5-FU (conventional treatment) was examined in colon cancer cells (SW480 and SW620) and nonmalignant cell lines (HaCaT and CHO-K1) using the sulforhodamine B test. As demonstrated in Table 2, the IC₅₀ values of the extract were equivalent to those seen for the chemotherapeutic drug at both 24 and 48 h. Furthermore, the extract exhibited slightly better selectivity than 5-FU, as shown by a decreased toxicity towards nonmalignant cells. In addition, it was discovered that cellular morphology was substantially altered when treated with the extract, demonstrating size and shape modifications, whilst the control vehicle (DMSO) seemed normal and healthy. These morphological abnormalities increased in a time- and dose-dependent way, resulting in a significant decrease in the number of cells compared to the control, indicating either a movement toward cell death or a halt in the cell cycle.

Other authors have evaluated D. viscosa using different models. Thus, Malik et al. (2022) reported that the leaf extracts of D. viscosa possess toxicity against shrimp larvae with 70% mortality. Besides, they reported that this cytotoxic activity might be due to defensive secondary metabolites as flavonoids, phenols, saponins or other compounds. In contrast, Herrera-Calderon et al. (2020) reported that D. viscosa possess a slight cytotoxic effect against colon cancer cells (HT-29) in comparison to 5-FU, besides, they found this extract does not possess any detectable cytotoxic effect on epidermal cells from mouse. These findings highlight the importance of evaluating the extract using different models to explore the real potential of D. viscosa.

The antiproliferative activity of D. viscosa was assessed against SW480 and SW620 colon cancer cells using the colorimetric assay sulforhodamine B. Based on the data depicted in Figure 2, it was determined that D. viscosa exhibits strong antiproliferative action in both cell lines beginning on day 2 at concentrations as low as 5 g/mL. In addition, when cells were treated with the extract, the assessment under an optical microscope revealed that the cells underwent morphological changes, including alterations in size and form, indicating that they were in the process of dying. Cao et al. (2009) evaluated different extracts of the same plant in a different cancer model. They reported that triterpenoid saponins of D. viscosa possess antiproliferative activity against the A2780 human ovarian cancer cell line (Cao et al., 2009). So, considering that chemoprevention involves the reversion, suppression, or prevention either the initial phases of carcinogenesis or the progression of premalignant cells to invasive disease using natural, synthetic, or biological agents (Steward and Brown, 2013), our findings revealed that the natural extract of D. viscosa could exert chemopreventive potential inducing antiproliferative activity in colorectal cancer, suggesting it could also be an important candidate for future investigations in this matter.

Mitochondrial membrane potential (ΔΨm) plays a significant role in apoptosis (Ly et al., 2003). Changes in ΔΨm have been shown to precede caspase activation and cell death. Because of this, the hydroethanolic extract of D. viscosa was examined 48 h after treatment to determine alterations in mitochondrial membrane potential. Using a lipophilic dye (Dioc6(3)) that is specific for the mitochondria of living cells, flow cytometry was performed. This accumulates in the mitochondrial matrix and is discharged into the cytosol following the decrease in ΔΨm. In addition, propidium iodide was utilized concurrently to assess alterations in the integrity of the cell membrane (Herrera-R et al., 2019). According to the results shown in Figure 3, it was observed that the extract of D. viscosa did not alter the mitochondrial membrane potential of SW480 cells, nevertheless, it was observed a significant mitochondrial depolarization in SW620 cells. A similar study was reported by Mossa and Al-Shawi (2015) who evaluated the extract of D. viscosa in human breast MDA-MB231 cancer cells and observed that membrane potential decreased with higher doses of the extract (Mossa and Al-Shawi, 2015). Considering that those changes in mitochondrial membrane potential are associated with apoptosis and cell death, our findings suggest this could be a possible mechanism related to the activity of the extract of D. viscosa in SW620 colon cancer cells.

The impact of D. viscosa on SW480 and SW620 cell cycle distribution was studied using propidium iodide (PI) by flow cytometry. This fluorescent dye intercalates into the main groove of double-stranded DNA to provide a highly fluorescent signal that identifies the percentage of cells in one of the three interphase phases. Cells were treated with either 1% DMSO alone or the IC₅₀-determined extract (the same concentration was used in all experiments). Following 48 h of treatment, cells were stained with PI and examined using DNA flow cytometry. Figure 4 depicts an example histogram for each cell line. In SW480 cells (Figures 4A, C), there were no significant differences between the control and the cell cycle distribution. In contrast, in SW620 cells (Figures 4B, D), a large number of cells were undergoing apoptosis, as shown by a high accumulation of cells in sub-G0/G1 and a minor drop in the other phases. These results are comparable with those published by Mossa and Al-Shawi (2015), who discovered that the hydroethanolic extract of D. viscosa promotes cell cycle arrest at the S phase in MDA-MB231 human breast cancer cells.

On the other hand, different authors have reported that the activity of D. viscosa is attributed to different polyphenols detected in the plant, including apigenin, rutin, quercetin and flavonoids, among others, who have demonstrated a plethora of biological activities (Lawal and Yunusa, 2013; Malik et al., 2022). These findings agree with those reported by Cai et al. (2011) which found that the apigenin might be responsible for the observed activity in human hepatoma Huh7 cells, through arresting in the cell cycle at the G2/M phase (Cai et al., 2011). Likewise, Weiqun Wang and others (Wang et al., 2004) indicated that treatment with different analogs of apigenin (chrysin, acacetin, kaempferol, luteolin, or quercetin) resulted in the cell-cycle arrest at the G2/M phase in a dose-dependent manner in SW480 cells. Considering the urgent need for the development of new treatment strategies with the ability to target cancer cells without harming normal cells, the antiproliferative activity of D. viscosa observed in this study highlights the importance of evaluating natural sources as potential treatments or adjuvants even in those cases associated with chemoresistant phenotype.

The plasma membrane’s direct barrier to the extracellular environment is essential for the maintenance of tissue homeostasis. Loss of membrane integrity can be associated with apoptosis, post-apoptotic secondary necrosis, necroptosis, and other kinds of programmed cell death, which terminates cellular life. Using a double staining with annexin-FITC and propidium iodide, it was determined if the extract of D. viscosa produces plasma membrane damage and likely cell death, which was confirmed by prior findings for mitochondrial membrane potential and cell cycle distribution. Figure 5 depicts the observed results. After 48 h of exposure, it was discovered that the extract stimulates plasma membrane breakdown in SW480 (Figure 4A) and SW620 (Figure 5B) cells, as shown by the movement of the cells onto the top quadrant with positive propidium iodide staining. These results were statistically significant in SW480 cells (Figures 5A–C), confirming that the extract triggers cell death independent of mitochondria in SW480 cells.

According to (Mossa and Al-Shawi, 2015), increasing doses of the hydroethanolic extract of D. viscosa elicit apoptosis and a decrease in mitochondrial membrane potential in human breast cancer MDA-MB231 cells. In addition, it has been postulated, as in the case of cell cycle analysis, that the activity might be linked to the presence of various polyphenols (Malik et al., 2022). Cai et al. (2011) discovered that apigenin, one of the chemicals identified in the extract of D. viscosa, significantly enhances apoptosis in human hepatocellular carcinoma Huh7 cells. These findings suggest that D. viscosa could be a promising candidate for the chemoprevention of various cancers, and it is essential to continue researching this natural source to identify both the molecules responsible for its chemopreventive activity and the signaling pathway associated with the mechanism of action of these molecules.

To complement the previous findings and get closer to the possible mechanism associated with the hydroethanolic extract of D. viscosa, we evaluated different apoptotic biomarkers. Firstly, considering that apoptosis is a highly regulated process, we evaluated caspases 3 and −7 which are the hallmarks of the degradation phase of apoptosis and are responsible for initiating cell shrinkage, membrane blebbing, and DNA fragmentation. According to the results, it was found that the hydroethanolic extract was only active on SW620 cells (Figure 6A), causing an important increase in the concentration of the active form of this protease, suggesting that one of the possible mechanisms of the extract on SW620 cells could be related to apoptosis mediated by caspase 3. On the contrary, in SW480 cells, it was not observed an important change caused by the extract (data not shown). We also evaluated caspase 8 to understand if the extract could be responsible for death receptor-induced cell death, however, we did not observe any changes on this protease (data not shown) which suggests that the apoptotic process in SW620 cells is mediated by mitochondria, confirming our previous findings on mitochondrial membrane potential (ΔΨm). In addition, considering the pivotal role of the tumor-suppressor protein p53 in the regulation of different cellular processes such as apoptosis (Xie et al., 2001; Mantovani et al., 2019), we evaluated if the extract of D. viscosa could modulate the expression of this protein. According to the results, this extract caused an important increased in the concentration levels of this protein in SW620 cells (Figure 6B) but not in SW480 cells (data not shown). This finding is very important because the tumor suppressor protein p53 is mutated in SW620 cells, and our experimental results suggest that the extract could activate the tumor suppressor protein in SW620 cells. These results are supported by those investigations reported by other authors who have demonstrated that p53 retains some of the functions and maintains residual DNA-binding ability (Bykov et al., 2014), being possible to activate it both in vitro and in vivo through different mechanisms (Selivanova and Wiman, 2007; Bykov et al., 2014), complementing our previous findings and suggesting that the apoptotic process induced by this extract could be mediated by the intrinsic pathway in response to the activation of p53. All our findings together with the previous investigations in the same matter suggest that D. viscosa could potentially be considered as chemopreventive agent or adjuvant in the treatment of colorectal cancer, even in those cases with resistance to conventional chemotherapy because of the lack of p53 expression or function, and thus, it is necessary to carry out further investigations with this natural source to discover new molecules against this disease.

Although in this study, it was used the hydroalcoholic extract, some compounds found in the chromatographic analysis, specially flavonoids and their derivates might be responsible for the anticancer activity like quercetin through downregulation, this flavonoid primarily targets the pro-survival Bcl-2 component of the p53 pathway, the members of the PI3K/AKT/mTOR, Wnt/-catenin, NF-B, and MAPK signaling pathways, the MMPs, the anabolism with AMPK as a marker, and the stress response to reactive oxygen species (ROS) (Neamtu et al., 2022). Other flavonoids like rutin can also trigger apoptosis in cancer cells by activating p53. In colon cancer cells like HCT cell line, rutin treatment activated caspase-3. In addition, it has been demonstrated that rutin activates both the intrinsic and extrinsic apoptotic pathways in colon cancer (HT-29) cells by upregulating caspases. This evidence significantly supports rutin’s ability to induce apoptosis by activating both the intrinsic (mitochondria-mediated) and extrinsic (death receptor-mediated) apoptotic pathways (Pandey et al., 2021).