Comparative Anticancer Activity of Extract, Partitions, and a Two-Acetogenin Mixture from Mexican Creole Avocado Seed

TL;DR

This study shows that a compound mixture from Mexican Creole avocado seeds has strong anticancer effects on SiHa cells.

Contribution

The isolation and characterization of a two-acetogenin mixture from avocado seeds with high cytotoxic activity is novel.

Findings

The butanol partition had the highest total phenol content and antioxidant capacity.

The acetogenin mixture showed the highest cytotoxicity against SiHa cells with IC50 values between 15.37 and 28.09 µg/mL.

All samples had high hemolytic concentration values, indicating low toxicity to human erythrocytes.

Abstract

Creole avocado (Persea americana var. drymifolia) seeds are considered as biowaste; however, they constitute a rich source of bioactive compounds. The objective of this study was to evaluate the cytotoxic effect of extract, partitions, and acetogenin mixture from creole avocado seeds in SiHa cells and erythrocytes. Creole avocado seed extract was obtained using ethyl acetate (CASE), and subsequently partitioned into hexane (HP), ethyl acetate (EP), and butanol (BP). Acetogenin mixture (AM), composed of avocadene acetate and avocadyne acetate, was isolated from HP and structurally characterized. Total phenolic content, antioxidant capacity and cytotoxic effect of all samples were evaluated using SiHa cell line and human erythrocytes. BP exhibited the highest total phenol content with a value of 159.13 mg of gallic acid equivalents/g (mg GAE/g). Antioxidant capacity assessed by 2,2′-azino-bis-(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS•+) and 2,2-diphenyl-1-picrylhydrazyl (DPPH•) assays indicated that BP showed the greatest antioxidant capacity with values of 207.26 and 94.96 mg of Trolox equivalents antioxidant capacity/g (mg TEAC/g), respectively. AM demonstrated the highest cytotoxicity against SiHa cells at all exposure times, with half-maximal inhibitory concentration (IC50) values ranging from 15.37 to 28.09 µg/mL. Half-maximal hemolytic concentration (HC50) of all samples ranged from 107.39 to 160.26 µg/mL. AM, isolated from creole avocado seeds, showed the highest cytotoxic activity against SiHa cells, highlighting its potential as a promising bioactive compound for further investigation in cancer research.

Figures

Full text

1. Introduction

Cervical cancer represents a major public health challenge worldwide. According to the Global Cancer Observatory, in 2022, this malignancy was among the most common cancers affecting women globally, with 662,301 new cases and 348,874 deaths reported, as well as in Mexico, where 10,348 new cases and 4909 deaths were recorded [1]. Persistent infections with high-risk human papillomavirus (HPV) genotypes are recognized as the primary etiological factor in cervical cancer, with approximately 20 oncogenic HPV types identified. Among these, HPV-16 is the most prevalent, being detected in nearly 50% of malignant cervical tumors [2].

Despite advances in prevention and treatment, first-line therapies for cervical cancer frequently lead to the development of therapeutic resistance, thereby reducing treatment efficacy and adversely affecting patient prognosis and survival [3,4]. Consequently, there is an urgent need to explore alternative and complementary therapeutic strategies. In this context, plant-derived bioactive compounds have gained increasing attention in contemporary oncological research due to their structural diversity and potential antitumor properties [5].

Since ancient times, plants have been used in traditional medicine to treat a wide range of diseases [6], largely due to the presence of biologically active secondary metabolites in various plant tissues. These metabolites include phenolic compounds, flavonoids, tannins, terpenes, alkaloids, and other specialized metabolites with documented pharmacological activities [7]. One such example is the creole avocado, a variety native to Mexico, whose pulp, seed, and leaves have been reported to have antioxidant, antimicrobial, anticancer, anti-inflammatory, and immunomodulatory properties [8,9,10,11,12,13,14,15,16].

In creole avocado, the pulp and peel are commonly consumed, whereas the seed is generally discarded as agro-industrial waste. However, seeds from several avocado varieties are known to contain high levels of bioactive metabolites [17,18], particularly acetogenins, which are found in plants belonging to the Annonaceae and Lauraceae families, including Persea americana [19,20].

Acetogenins belong to the polyketide group, which are metabolites biosynthesized by polyketide synthases, resulting in compounds of variable polarity depending on the degree of oxidation [21]. These metabolites have been shown to exhibit antifungal, antimicrobial, cytotoxic, antiproliferative, and pro-apoptotic activity, which has stimulated growing scientific interest in recent years [22,23,24,25,26,27,28].

Notably, avocado seeds have been reported to contain a higher concentration of acetogenins than the pulp [29]. Conferring added value to this typically discarded byproduct and positioning it as a promising source of bioactive compounds for cancer prevention and treatment strategies.

Acetogenins from the Annonaceae family are the most extensively studied and have demonstrated potent anticancer activity [20,30]. Within the Lauraceae family, acetogenins have been isolated and characterized uniquely from the seeds and pulp of Hass avocado, with studies reporting significant anticarcinogenic effects [23,25,26,27,28].

In contrast, acetogenins derived from creole avocado remain largely underexplored. To date, research has been limited to evaluating the anticancer effect of lipid-rich extract from Mexican creole avocado seeds (LEAS) [9,10] and peptides isolated from the pulp of this variety, in different cancer cell lines from that evaluated in the present study [12,13,14]. Importantly, acetogenins have not yet been isolated or identified from any part of the creole avocado, and their antioxidant capacity and biological activity on malignant cells and normal erythrocytes remain unexplored.

In the present study, two acetogenins (avocadene acetate and avocadyne acetate) were isolated from creole avocado seeds. These metabolites have also been previously reported in the pulp and seeds of the Hass avocado variety [22,24,31].

Therefore, this study represents the first report on the isolation and characterization of acetogenins from creole avocado seeds and the evaluation of their biological activity in cervical cancer and erythrocytes models. This study aimed to assess, using a bioassay-guided approach, the cytotoxic effect of the crude extract, partitions, and an acetogenin mixture derived from creole avocado seeds in the SiHa cell line and human erythrocytes.

2. Results

2.1. Yields of the Creole Avocado Seed Samples

The yields obtained were 7% for CASE, 4.02% for HP, 2.10% for EP, 0.08% for BP, and 0.16% for AM.

2.2. Isolation and Identification of Acetogenin Mixture (AM)

The white solid, identified as AM, was obtained by chromatographic fractionation. Nuclear magnetic resonance (NMR) analysis of this solid showed the presence of two compounds, avocadene acetate and avocadyne acetate (Figures S1–S8). The ratio of the compounds was 41.27 and 55.37% (0.74:1), respectively (Figure S9), which was confirmed by 1D and 2D NMR mixture analysis.

Avocadene acetate (Figure 1a), ^1^H-NMR (400 MHz, CDCl_3_): δ = 5.81 (ddt, J = 16.98, 10.16, 6.80 Hz, 1H, H-16′), 4.98 (ddt, J = 16.98, 3.44, 1.42, Hz, 1H, H-17′ β), 4.92 (ddt, J = 10.16, 3.44, 1.42 Hz, 1H, H-17′ α), 4.10 (m, 2H, H-1′), 3.99 (m, 1H, H-2′), 3.89 (m, 1H, H-4′), 2.10 (s, 3H, H-2), 2.03 (m, 2H, H-15′), 1.60 (m, 2H, H-3′), 1.47 (m, 2H, H-5′), 1.37 (m, 2H, H-14′), 1.24 (m, 16H, H-6′-13′). ^13^C-NMR (101 MHz, CDCl_3_): δ = 171.38 C-1, 139.41 C-16′, 114.23 C-17′, 72.65 C-4′, 70.94 C-2, 68.69 C-1′, 39.16 C-3′, 38.32 C-5′, 33.96 C-15′, 29.69 C-10′-12′, 29.64 C-9′and 13′, 29.08 C-7′, 28.89 C-9′ and 14′, 25.43 C-6′, 21.04 C-2′.

Avocadyne acetate (Figure 1b), ^1^H-NMR (400 MHz, CDCl_3_): δ = 4.10 (m, 2H, H-1′), 3.99 (m, 1H, H-2′), 3.89 (m, 1H, H-4′), 2.17 (ddt, J = 6.80, 2.6, Hz, 2H, H-15′), 2.10 (s, 3H, H-2), 1.93 (ddt, J = 2.6 Hz, 1H, H-17′), 1.60 (m, 2H, H-3′), 1.47 (m, 2H, H-5′), 1.37 (m, 2H, H-14′), 1.24 (m, 16H, H-6′-13′). ^13^C-NMR (101 MHz, CDCl_3_): δ = 171.38 C-1, 84.96 C-17′, 72.65 C-4′, 70.94 C-2, 68.69 C-1′, 68.18 C-16′, 39.16 C-3′, 38.32 C-5′, 33.96 C-15′, 29.76 C-7′, 29.71 C-8′-9′, 29.70 C-10′, 29.61 C-11′, 29.29 C-12′, 28.89 C-13′, 28.62 C-14′, 21.04 C-2′, 18.53 C-15′.

2.3. Phenolic Compounds of Creole Avocado Seed Samples

The presence of phenols was confirmed in CASE and all partitions. As shown in Table 1, statistically significant differences (p ≤ 0.05) were observed in the concentrations among the samples analyzed. BP exhibited the highest total phenol content.

2.4. Antioxidant Capacity

The antioxidant capacity of CASE, partitions, and AM was assessed using the ABTS•^+^ and DPPH• methods. As shown in Figure 2, the five samples demonstrated antioxidant capacity in both techniques, showing statistically significant differences among the samples (p ≤ 0.05). In ABTS•^+^ and DPPH• assays, BP exhibited the highest antioxidant capacity; by contrast, AM showed the lowest antioxidant capacity. All samples exhibited higher antioxidant capacity in the ABTS•^+^ assay than in the DPPH• assay.

2.5. Cytotoxic Activity

2.5.1. Malignant SiHa Cells

As shown in Figure 3, the evaluated samples exhibited cytotoxic effects against the SiHa cell line, in a dose- and time-dependent manner. When comparing the extract and the three partitions, HP demonstrated the highest biological activity, followed by CASE, EP, and BP.

HP was subjected to chromatographic fractionation, resulting in the isolation and identification of AM. In SiHa cells, AM showed the greatest cytotoxic activity from all creole avocado seed samples.

The samples showed a similar tendency at the three exposure times, with the highest biological activity at 72 h. Cisplatin (CDDP), used as a positive control, displayed greater cytotoxicity than AM at low concentrations; however, at higher concentrations, AM was more cytotoxic than CDDP.

Table 2 presents the IC_50_ values obtained in the cytotoxicity assays, which exhibit significant differences (p ≤ 0.05) among samples and exposure times. BP was the least cytotoxic in SiHa cells at all exposure times; conversely, AM was the most cytotoxic among the avocado-derived samples. Comparing CDDP and AM, at 24 h AM was significantly more cytotoxic (p ≤ 0.05), and at 48 and 72 h no difference (p ≤ 0.05) was observed between them.

2.5.2. Human Erythrocytes

The hemolysis assay in human erythrocytes revealed a dose-dependent cytotoxic effect of all samples, since increasing concentrations caused greater erythrocyte damage, with statistically significant differences (p ≤ 0.05) observed between concentrations within each sample, as shown in Figure 4. BP exhibited the lowest hemolytic activity within 25–100 µg/mL, compared to the other samples; in contrast, AM, EP, and HP demonstrated higher toxicity at the same concentrations. At 200 µg/mL, CASE induced the greatest hemolytic effect (78.4%), followed by AM, BP, EP, and HP.

HC_50_ values are shown in Table 3. BP was the lowest cytotoxic sample (p ≤ 0.05), followed by AM, HP, and EP, while CASE showed the highest cytotoxicity in human erythrocytes.

3. Discussion

The present study provides the first evidence of the isolation and purification of acetogenins from creole avocado seeds, expanding the current knowledge of bioactive acetogenins in this underexplored variety. In addition, this work represents the first evaluation of the biological activity of CASE, its partitions, and AM in a human cervical cancer cell model and erythrocytes, linking chemical characterization with anticancer effect.

Differences in phytochemical profiles were observed among CASE and its partitions, which can be attributed primarily to the polarity of the extraction solvents, as previously reported by Iloki-Assanga et al. [32]. BP exhibited the lowest yield among the samples (0.08%), but it displayed the highest phenolic content, as determined by the Folin–Ciocalteu method. This is explained by the structural characteristics of phenolic compounds, as these metabolites show greater affinity for polar solvents, such as butanol [33].

All four samples analyzed in the present study displayed higher total phenolic content than that reported by Zavala-Guerrero [34] for a hexane extract of Persea americana var. drymifolia seeds and its methanolic and chloroform partitions. In contrast, the phenolic concentration observed here were lower than those reported for Hass avocado seed extract and fractions obtained using acetone, ethyl acetate, ethanol, water, and ethanol/water mixtures [35,36,37]. These discrepancies may be attributed to the differences in avocado variety, agronomic and environmental conditions, ripening stage, post-harvest processing, and extraction solvents employed [38,39].

AM was isolated from HP using chromatographic techniques, and the structural elucidation of its constituents was achieved through comprehensive spectroscopic analyses. The chemical shifts observed in ^1^H- and ^13^C-NMR spectra were consistent with those previously reported by Domergue et al. [22]. The quantitative ^1^H NMR (qNMR-^1^H) analysis confirmed that the isolated material consisted of a 0.74:1 mixture of avocadene acetate and avocadyne acetate, both classified as lauraceous acetogenins.

Both acetogenins identified in this study share a 17-carbon aliphatic chain esterified with an acetoxy group and hydroxyl groups at positions C-2 and C-4, differing only in the degree of terminal unsaturation, with one compound containing a double bond and the other a triple bond. These two acetogenins comprising AM could not be separated using the chromatographic methods employed in this study; therefore, advanced separation techniques are required.

Based on their carbon chain length, the acetogenins identified in the present study are classified within the avocatins subgroup [19], consistent with the findings reported by Báez-Magaña et al. [15], who detected avocatins in LEAS using gas chromatography–mass spectrometry (GC–MS). Moreover, avocadene acetate has previously been identified in the pulp and seed of Hass avocado, while avocadyne acetate has been reported in the seed of the same variety [22,24,31]; however, their isolation and identification from creole avocado have not been previously described.

BP showed the highest antioxidant capacity in both assays evaluated, which may be attributed to its high concentration of total phenols. Phenolic compounds are well known for their ability to scavenge free radicals through hydrogen donation from hydroxyl groups in their structures [40]. In contrast, AM showed the lowest antioxidant capacity, likely due to the lipophilic nature of acetogenins, which generally exhibit lower radical-scavenging efficiency than hydrophilic antioxidants. It has been suggested that the antioxidant capacity of acetogenins may be associated with hydrogen atom donation mechanisms that contribute to free-radical stabilization [31].

Previous studies have reported that extracts from Hass and Fuerte avocado seeds, prepared using acetone or ethanol [36,41,42], exhibited higher antioxidant capacity than that observed in CASE, HP, and EP in the present study, but lower than that of BP. These differences may be explained by variations in avocado variety, cultivation region, extraction solvent type, and sample processing method [38,39].

The higher antioxidant capacity observed by the ABTS•^+^ assay compared to the DPPH• assay is consistent with previous reports [36,41,42] and may be attributed to the broader reactivity of the ABTS•^+^ radical, which can interact with both hydrophilic and lipophilic samples. In contrast, the DPPH• radical reacts exclusively with hydrophilic compounds [43]. Given the complexity of oxidation–antioxidation processes, no single technique can fully characterize the antioxidant profile of a sample [44], suggesting that both methods are complementary.

Cervical cancer continues to present a major public health challenge worldwide and in Mexico, underscoring the need for continued research into novel preventive and therapeutic strategies [45]. In this context, the creole avocado has gained increasing attention due to its diverse pharmacological properties, including its potential anticancer activity [46].

The cytotoxic effect observed for CASE, HP, EP, and BP against SiHa cells may be partially associated with the phenolic content, as studies have demonstrated that phenolic compounds exert anticarcinogenic effects through the modulation of signaling pathways involved in apoptosis and antiproliferative processes [47,48]. However, it is important to consider that additional classes of secondary metabolites not evaluated in the present study, such as alkaloids and terpenoids, may also contribute to the observed biological activity. These metabolites have been reported to exert anticancer effects through mechanisms involving key molecular targets, including p53, p21, BAX, and NF-kB [49].

The antioxidant capacity of creole avocado seed samples may also contribute to the cytotoxic effect observed in cancer cells. Cancer is characterized by elevated oxidative stress, which promotes carcinogenesis and disease progression through the activation of pro-oncogenic signaling pathways [50]. Antioxidant compounds have demonstrated therapeutic potential against cancer, not only by scavenging reactive oxygen species (ROS), but also by modulating gene expression, including downregulation of oncogenes, upregulation of tumor suppressor genes, and inhibition of angiogenesis. These compounds may also have a dual effect, exhibiting pro-oxidant behavior under certain conditions, thereby inducing apoptosis in cancer cells [51].

Although hydrophilic compounds are the primary contributors to antioxidant capacity, they are generally more rapidly eliminated via renal excretion. In contrast, lipophilic compounds such as acetogenins can more readily penetrate cell membranes, potentially resulting in enhanced cellular uptake and bioavailability, which may contribute to their cytotoxic efficacy [31].

When the cytotoxic effects of CASE, HP, and EP on SiHa cells were compared with those reported by Lara-Márquez et al. [9] for LEAS in Caco-2 cells after 48 h of exposure, our samples exhibited markedly higher cytotoxicity, with values approximately 3.0- to 5.7-fold greater. Similarly, comparison with the LEAS activity reported in canine osteosarcoma cells (D-17) [10] revealed that CASE, HP, and EP were approximately 1.6- to 3-fold more cytotoxic on SiHa cells. These differences are likely attributable to variations in the phytochemical composition resulting from extraction and fractionation, as well as to intrinsic biological differences among the cell lines evaluated.

LEAS induced apoptosis in Caco-2 cells through activation of caspases 8 and 9, increased mitochondrial ROS production, and inhibited fatty acid oxidation in Caco-2 cells [9]. In D-17 cells, LEAS similarly induced apoptosis via caspases 8 and 9, elevated intracellular ROS levels, and induced cell cycle arrest in the G0/G1 phase [10].

BP had the highest total phenol content and antioxidant capacity; however, it showed the lowest cytotoxic activity against SiHa cells, as well as the lowest yield. For these reasons, and considering the bio-guided nature of the study, this fraction was considered the least interesting for further analysis.

The findings of Velderrain-Rodríguez et al. [42], who reported that an 80% ethanolic extract of avocado seed with high phenolic content and antioxidant capacity, exhibited moderate cytotoxicity against Caco-2 cells (IC_50_ of 200 µg/mL at 72 h), which is comparable to the cytotoxic effect for BP at the same exposure time in the present study.

Cancer cells are characterized by elevated basal ROS levels compared to normal cells; however, excessive ROS accumulation can trigger apoptosis through enhanced oxidative DNA damage [52]. Therefore, the high antioxidant content of BP may exert a protective effect by scavenging ROS and preventing oxidative stress-mediated cell death, thereby attenuating its cytotoxic potential. This observation is consistent with previous studies indicating that a high phenolic content and strong antioxidant capacity do not necessarily correlate with cytotoxicity, as phenolic compounds may maintain redox homeostasis rather than inducing cell death, highlighting that antioxidant activity alone is not a reliable predictor of anticancer activity [53,54].

In contrast, isolated AM, a 0.74:1 mixture of avocadene acetate and avocadyne acetate, exhibited the highest cytotoxic effect against SiHa cells. These findings are consistent with those reported by D’Ambrosio et al. [23], who demonstrated a synergistic cytotoxic effect of this acetogenin mixture isolated from Hass avocado pulp in 83-01-82CA oral cancer cells, particularly when avocadyne acetate was present in higher proportions. Notably, when each acetogenin was evaluated individually, a reduced cytotoxic effect was observed, highlighting the importance of metabolite ratios in determining biological activity. Moreover, these authors reported that the mixture exerted antiproliferative effects via modulation of the EGFR/RAS/RAF/MEK/ERK1/2 signaling pathway

Previous studies have investigated the anticancer effects of acetogenins from Hass avocado seeds and pulp. Avocadyne has shown to inhibit fatty acid oxidation by targeting very long-chain acyl-CoA dehydrogenase in acute myeloid leukemia cells (TEX and OCI-AML2) [26,28]. Similarly, avocatin B, a 1:1 mixture of avocadyne and avocadene, was reported to accumulate in mitochondria, inhibit fatty acid oxidation, reduce NADPH levels, and induce apoptosis through ROS generation in the same leukemia models mentioned before [25]. Persin induced Bim-dependent apoptosis and cell cycle arrest at the G2/M phase in MCF-7 breast cancer cells [27]. Collectively, these findings underscore the anticancer potential of avocado acetogenins and support their relevance as candidates for further therapeutic exploration.

The antitumor activity of annonaceous acetogenins has been largely documented in the literature. For instance, annonacin and annonamuricins A, B, C, and D, isolated from Annona muricata, have been evaluated in PC-3 prostate cancer cells, where they exhibited important cytotoxic effects [55]. In a related study, a mixture of five annonaceous acetogenins, with bullatacin as the major constituent, demonstrated cytotoxic and pro-apoptotic activities across multiple gastric cancer cell lines, and induced cell cycle arrest at the G0/G1 phase [56]. Likewise, annonacin and muricin P were shown to reduce intracellular ATP levels, to trigger apoptotic pathways, and to exert synergistic cytotoxic effects when combined with the chemotherapeutic agent sorafenib in hepatocellular carcinoma cells [57]. Furthermore, squamocin P and annosquatin III, isolated from the seeds of Annona squamosa, were assessed in multidrug-resistant cancer cell lines derived from liver (SMMC-7721/T), breast (MCF-7/ADR), and lung (A549/T), where they displayed significant cytotoxic activity [58].

It is important to emphasize that annonaceous acetogenins are structurally distinct from lauraceous acetogenins. Differences in chain length, functional groups, presence of lactones, and degrees of unsaturation result in distinct biological activities and mechanisms of action [59], and therefore, findings from annonaceous acetogenins cannot be directly extrapolated to avocado-derived acetogenins.

AM was isolated from HP, which in turn was obtained from CASE; therefore, CASE and HP likely contain avocadene acetate and avocadyne acetate at lower concentrations, which may explain their reduced cytotoxic effect compared to purified AM. Additionally, other acetogenins may be present in CASE, HP, and EP, contributing to their biological activity, as Persea americana seeds are known to be rich sources of acetogenins [29], which are extractable with non-polar and moderately polar solvents such as hexane and ethyl acetate [60].

As mentioned above for isolated acetogenins from Hass avocado and LEAS, AM could act through antiproliferative signaling pathways, caspase-mediated apoptosis, ROS modulation, and cycle arrest in SiHa cells. However, further studies are required to evaluate its mechanism of action.

CDDP exerts its anticancer effect through the formation of DNA adducts, interfering with replication and triggering apoptosis [61]. Although the mechanism of action of AM is currently unknown and is likely distinct from CDDP, its use as a positive control is justified, as CDDP remains a first-line chemotherapeutic agent for cervical cancer treatment [62].

Given the adverse side effects associated with chemotherapy and radiotherapy [63], there is a pressing need to find alternative therapeutic agents that are effective and safer for patients [64]. Evaluation of hemolytic activity represents a useful preliminary approach for assessing cytotoxicity in non-cancerous cells, as erythrocytes are readily available, do not require cell culture, and allow rapid spectrophotometric quantification of hemoglobin release [65].

The hemolytic activity observed for CASE, HP, EP, and BP may be partially attributed to their phenolic content, as these compounds possess amphiphilic structures capable of interacting with the erythrocyte membrane. The hydrophobic moieties of phenolic compounds can insert into the lipid bilayer, altering its integrity, fluidity, and permeability, as well as inducing changes in cytoskeletal architecture [66].

According to Pagano and Faggio [67], a sample is considered non-toxic if it induces less than 9% hemolysis, mildly toxic between 10–49%, toxic between 50–89%, and highly toxic when it exceeds 90%. Therefore, CASE at 25 µg/mL and BP up to 50 µg/mL are considered non-toxic, whereas all samples at 200 µg/mL would be classified as toxic.

In relation to the HC_50_ values obtained in the present study, CASE and its partitions exhibited greater toxicity compared to those reported by other authors for LEAS and the ethanolic extract of Quintal avocado peel, evaluated on human and sheep erythrocytes [9,68], likely reflecting differences in phytochemical composition.

Based on the hemolytic activity of AM, it would be classified as mildly toxic at concentrations between 25 and 100 µg/mL, according to the previously described classification. These findings indicate that additional studies are required to further assess its safety. In contrast, Montalvo-González et al. [69] reported that annonaceous acetogenins did not induce hemolysis at concentrations of 250 and 500 µg/mL. This discrepancy may be attributed not only to differences in experimental conditions, but also to the substantial structural and physicochemical differences between annonaceous acetogenins and those derived from Persea americana, which exhibit different lipid chain composition, polarity, and functional groups.

The hemolytic activity observed by AM could be associated with its lipophilic nature, which could facilitate its interaction with the erythrocyte membrane and compromise its integrity; however, further mechanistic studies are required to confirm this.

This study focused on the isolation of AM and the evaluation of its cytotoxic activity in a single malignant cervical cancer cell line, as well as its hemolytic effects in human erythrocytes in vitro. These findings should be interpreted considering certain limitations, including the evaluation of a limited number of biological models and the absence of in vivo validation. Therefore, future studies should aim to isolate the two acetogenins from AM, other acetogenins, and bioactive metabolites from creole avocado seeds and to assess their biological activity in cancer cell lines, as well as in non-malignant cells. Furthermore, comprehensive mechanistic investigations and in vivo studies will be essential to elucidate the therapeutic potential, selectivity, and safety profile of AM and related compounds.

4. Materials and Methods

4.1. Plant Material

Creole avocados were obtained in Michoacán, Mexico. Seeds were manually separated, washed, and dried in an oven with air circulation at 40 °C. Dried seeds were then finely ground using a manual mill to obtain the seed powder.

4.2. Obtaining CASE

The seed powder (1500 g) was macerated in 2.5 L ethyl acetate for two weeks in darkness with sporadic stirring at room temperature. The resulting liquid was filtered and rotary evaporated to dryness at 40 °C (Büchi R-114, Flawil, Switzerland), obtaining CASE.

4.3. Liquid–Liquid Partition of CASE

Three partitions were obtained by liquid–liquid partition. Briefly, in a separation funnel, 40 g of CASE were dispersed in distilled water (200 mL), followed by adding hexane (250 mL). The mixture was stirred, and the organic phase corresponding to HP was collected. Subsequently, 250 mL of ethyl acetate was added to the aqueous phase to obtain EP. Finally, the procedure was repeated, adding 250 mL of butanol, obtaining BP. Partitions were dehydrated using anhydrous sodium sulfate and rotary evaporated at 40 °C (Büchi R-114, Flawil, Switzerland).

4.4. Isolation of AM

AM (avocadene acetate and avocadyne acetate) was isolated from 5.2 g of HP, separated by silica gel column chromatography (15 × 5.5 cm, Merck, Boston, MA, USA), eluted sequentially with 250 mL of hexane, followed by 250 mL each of the mixtures hexane/ethyl acetate (9:1), (8:2), (65:35), (1:1), and (3:7). Thirty fractions of 55 mL each were collected and monitored by thin-layer chromatography (TLC; Merck 60 F254, Boston, MA, USA), fractions showing similar chromatographic profiles were combined and rotary evaporated (Büchi R-114, Flawil, Switzerland). A majority fraction was observed (R_f_ = 0.4285) in 17–18 fractions, which yielded 211 mg of a white solid. This was further purified by a second silica gel column chromatography (10 × 2.5 cm) using 60 mL of hexane, followed by 60 mL each of hexane/ethyl acetate (7:3) and (1:1). Thirteen fractions of 15 mL were obtained and analyzed by TLC, and AM (205 mg) was isolated from fraction 11. The established separation conditions yielded a white solid. Normal-phase TLC (eluent: hexane/acetone 7:3) showed two distinct components (R_f_ = 0.4047 and 0.4523). These compounds were only detectable by TLC and revealed using I_2_, phosphomolybdic acid, KMnO_4_, 5% H_2_SO_4_, and vanillin/H_2_SO_4_ (Figure S10), as they showed no absorption under UV radiation at 254 and 365 nm.

4.5. NMR of AM

AM was dissolved in 99.8% deuterium atom CDCl_3_ (Sigma Aldrich, St. Louis, MO, USA). The composition of AM was assessed through NMR mixture analysis, incorporating both 1D and 2D methods: ^1^H-, and ^13^C-NMR experiments were carried out at 400 and 100 MHz; correlation spectroscopy (COSY), heteronuclear single quantum correlation (HSQC), and heteronuclear multiple-bond correlation (HMBC) experiments, using a Bruker^®^ 400 MHz spectrometer (Bruker, Karlsruhe, Germany) at 298 K. The free induction decay (FID) was processed using MestReNova version 2.0 software, and the chloroform solvent signal was used as reference: 7.26 and 77.160 ppm for ^1^H and ^13^C spectra, respectively. The measurement of the ratio of avocadene acetate and avocadyne acetate was performed by qNMR-^1^H, using 98% 1,4-dinitrobenzene (Sigma-Aldrich, St. Louis, MO, USA), following the methodology of Pauli et al. [70]. Parameters qNMR-^1^H: single pulse, without carbon decoupling (zg 90° pulse), 298 K, relaxation delay (D1) 60 s, acquisition time 4 s, spectral window 30 ppm, and 64 scans.

4.6. Yields of the Obtained Samples

The yields were calculated based on the initial weight of the seed powder (1.5 kg).

4.7. Total Phenolic Content

It was determined by the Folin–Ciocalteau method [71], and a calibration curve was prepared with gallic acid (Sigma-Aldrich, St. Louis, MO, USA). Phenols were extracted from samples CASE, HP, EP, and BP with ethanol (1.5 mg/mL) for 1 h. For the analysis, aliquots of 50 µL of each sample were reacted with 250 µL of Folin’s solution (1:1; Sigma-Aldrich, St. Louis, MO, USA) and 200 µL of Na_2_CO_3_ (7.5%; Sigma-Aldrich, St. Louis, MO, USA). The reaction mixture was allowed to react in darkness for 30 min and then measured at 765 nm using a microplate reader (BioTek Epoch, Winooski, VT, USA). The results were reported as mg GAE/g.

4.8. Antioxidant Capacity

ABTS•^+^: It was performed using the ABTS•^+^ (Sigma-Aldrich, St. Louis, MO, USA) method described by Re et al. [72]. ABTS•^+^ (3.6 mg/mL) was prepared 24 h before use at room temperature in darkness, in the presence of potassium persulfate (0.6 mg/mL; Sigma-Aldrich, St. Louis, MO, USA) in deionized water and then diluted in ethanol to obtain an absorbance of 0.70 ± 0.02 at 734 nm. Antioxidant extraction was performed in ethanol from CASE, HP, EP, BP, and AM (1 mg/mL) for 1 h. Subsequently, 50 µL aliquots of each sample were mixed with 450 µL of the ABTS•^+^ solution, incubated in darkness for 5 min, and the absorbance was recorded at 734 nm (BioTek Epoch, Winooski, VT, USA). A calibration curve was prepared using 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox; Sigma-Aldrich, St. Louis, MO, USA), and the results were expressed as mg TEAC/g.

DPPH•: It was performed following the method described by Schenk and Brown [73] using the DPPH• radical (Sigma-Aldrich, St. Louis, MO, USA). Antioxidants from CASE, HP, EP, BP, and AM were extracted in ethanol (1 mg/mL) for 1 h. Subsequently, 450 µL of DPPH• solution (0.0075%) was added to 50 µL of each sample, and the mixtures were incubated in darkness for 1 h. The absorbance was then measured at 540 nm using a BioTek Epoch (Winooski, VT, USA). The results were expressed as mg TEAC/g, based on a calibration curve prepared with Trolox as the standard.

4.9. Cytotoxic Activity

4.9.1. SiHa Cervical Cancer Cells

SiHa cells (ATCC HTB-35, Rockville, MD, USA) were employed in this study, which contain an integrated HPV-16 genome (1–2 copies per cell). Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Gibco, Grand Island, NY, USA) supplemented with 5% fetal bovine serum (FBS; Gibco, Grand Island, NY, USA) and 0.1% penicillin/streptomycin (Sigma-Aldrich, St. Louis, MO, USA). Cultures were maintained at 37 °C in a humidified atmosphere containing 5% CO_2_ (Sanyo, Osaka, Japan).

A cytotoxicity assay was performed using the colorimetric method described by Mosmann [74], employing 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich, St. Louis, MO, USA). Cells were seeded in flat-bottom 96-well microplates; for the 24 h assay, 10,000 cells/well were plated, while for the 48- and 72 h assays, 5000 cells/well were seeded. Cells were exposed to CASE, HP, EP, BP, AM, and CDDP at concentrations ranging from 0 to 250 µg/mL. Following the exposure period, MTT (5 mg/mL) was added and incubated for 3 h. The medium was then removed, and the formazan crystals produced by viable cells were solubilized with dimethyl sulfoxide (DMSO). Absorbance was measured at 540 nm using a BioTek Epoch microplate reader (Winooski, VT, USA). Cell viability was calculated by considering the untreated control as 100% viability, and IC_50_ were subsequently determined.

4.9.2. Normal Human Erythrocytes

Hemolysis assay was conducted using human erythrocytes, following the method described by Sæbø et al. [75] with modifications. A+ blood was obtained from a healthy donor who provided written informed consent. Erythrocytes were isolated by centrifugation at 1500 rpm for 10 min, washed, and subsequently diluted to 4% in NaCl (0.9%). Aliquots of 15 µL of the erythrocyte suspension were mixed with 500 µL of CASE, HP, EP, BP, and AM (0–200 µg/mL in 0.9% NaCl), and incubated for 30 min at 37 °C. The mixtures were centrifuged at 500 rpm for 5 min, and the supernatants containing released hemoglobin were collected. Absorbance was measured at 405 nm using a BioTek Epoch microplate reader (Winooski, VT, USA). The percentage of hemolysis was determined using 10% Triton X-100 as the reference standard (100% hemolysis), and HC_50_ were subsequently determined.

4.10. Statistical Analysis

Data were analyzed using one-way ANOVA followed by Tukey’s post hoc test (p ≤ 0.05) to determine significant differences among extracts in antioxidant capacity, total phenols content, and hemolytic activity. IC_50_ values obtained from cytotoxicity assays were analyzed using independent one-way ANOVAs, followed by Tukey’s post hoc tests (p ≤ 0.05); comparisons were performed among samples at each exposure time and, subsequently, among exposure times within each sample.

All statistical analyses were performed using StatGraphics Centurion Version 19 (StatGraphics, The Plains, VA, USA).

5. Conclusions

CASE and its partitions obtained from Persea americana var. drymifolia seeds contained phenolic compounds, exhibited antioxidant capacity, and demonstrated cytotoxic effects against SiHa cervical cancer cells, as well as hemolytic activity in human erythrocytes. Among the evaluated partitions, HP exhibited the highest cytotoxic activity and was therefore selected for the chromatographic separation of AM, whose components were identified as avocadene acetate and avocadyne acetate through spectroscopic analyses.

Among all evaluated samples, AM showed the highest cytotoxic activity against SiHa cells, highlighting its pharmacological relevance and potential as a promising bioactive compound or structure for cancer research. However, AM was classified as mildly toxic to toxic in erythrocytes, indicating that further studies in non-cancerous cells are necessary to assess its safety profile.

Considering the limited information available on the metabolite composition and biological effects of Persea americana var. drymifolia, these findings provide novel insights into the bioactive potential of creole avocado seed metabolites. Nevertheless, the individual metabolites comprising AM should be separated using advanced chromatographic techniques, and additional metabolites present in CASE should also be isolated and characterized. Furthermore, comprehensive in vitro, in vivo, and mechanistic studies are required to elucidate the molecular pathways involved, assess selectivity and safety, and determine the therapeutic applicability of these compounds.