INTRODUCTION:

Cancer is still one of the leading causes of death in the world. Its high incidence, mortality and lack of effective treatment have led to a permanent effort in bioprospecting for new chemotherapeutics. The estimated annual incidence of cancer in Chile is approximately 35,000 new cancer cases per year, with adjusted incidence rates of 226 and 180 per 100,000 male and female inhabitants, respectively (^{JIMENEZ DE LA JARA et al., 2015}). Therefore, the research and development of new, more effective and less toxic substances against cancer have become an essential. Worldwide, many plants have been tested in cancer models and found to be effective as anticancer agents. Several of them have been used routinely as anticancer drugs, such as vinblastine and paclitaxel, both compounds extracted from Catharanthus roseus and Taxus brevifolia, respectively (^{CISTERNINO et al., 2003}; ^{MAKAROV et al., 2007}). Several other drugs extracted from plants have recently proved to be useful as anti-cancer therapies, including betolunic acid (^{FULDA, 2008}), resveratrol (^{GARCÍA-ZEPEDA et al., 2013}) and homoharringtotine (^{LÜ & WANG, 2014}), among others. The north of Chile is characterized by its vegetal biodiversity due to a unique geographical and climatological conditions (e.g., altitudes from zero to 5000 m.a.s.l; temperatures of 23 °C to -5 °C and high solar radiation), offering an unexplored a unique source of naturally derived compounds. Most of these, recognized as medicinal herbs, are part of the traditional uses in folk medicine given by the local Andean communities (^{VILLAGRÁN et al., 2003}; ^{FERREIRA-MACHADO et al., 2004}). However, efforts in additional studies of these plants about new potential healing properties, not necessarily linked to its ancestral uses, have not been carried out in-depth yet. As part of a permanent effort of our laboratory, we have focused our work in bioprospecting activities. We have found two Andean species with promising anticancer activities. For example, the ethanolic extract of Senecio nutans, was able to induce cytotoxicity in breast cancer cells but not in MCF-10F, the normal counterpart, showing an interesting differential effect (^{ECHIBURÚ-CHAU, 2014}). On the other hand, the benzofurane p-coumaryloxytremetone isolated from Parastrephia lucida, showed a protective activity from reactive oxygen species (ROS) when tested in MCF10F after pyocyanin treatment (^{ECHIBURU-CHAU et al., 2017}) and recently we presented a first glance of cytotoxic activity of the Bacharis alnifolia infusion and ethanolic extract (^{SOTO et al., 2019}). The present research has considered a screening of twenty-six Andean plants, from Arica & Parinacota region, northern Chile, using an in vitro growth inhibition model in a panel of six cancer cell lines derived from different tissues (colorectal, liver, breast, lung, kidney and skin). The plants were selected by its popular use in local communities and its availability at local markets.

MATERIALS AND METHODS:

Plant material

Fresh aerial parts of the 26 plants (Table 1), from the Andean Highlands sector at the Arica and Parinacota Region, northern Chile, were obtained from the local market “Terminal Agropecuario Arica” (ASOCAPEC). Collected samples and voucher numbers were deposited in the herbarium of the Centro de Investigaciones del Hombre en el Desierto (CIHDE), Arica, Chile.

Preparation of extracts

The samples were treated as previously described (^{PARRA et al., 2018}). Dried and powdered samples (1 g of aerial parts) were macerated with absolute ethanol for 72 hours at room temperature. The extracts were filtered using a Whatman filter paper Nº 1 and concentrated on a rotary evaporator under reduced pressure at 40 °C. The residues were re-dissolved in EtOH 97% to yield a final concentration of 1 mg mL^-1. The samples were refrigerated at 20 °C until its use.

Cell culture

Six cells lines were used, A549 (Human lung carcinoma), B-16 (Murine melanoma), Caco-2 (Human colorectal adenocarcinoma), HEK-293 (Human embryo kidney), HepG2 (Human liver carcinoma) and MCF7 (Human breast adenocarcinoma). All cell lines were cultured in DMEM medium supplemented with 10% of fetal bovine serum (Hyclone, South Logan, Utah, U.S), only when it was necessary the medium was supplemented with 100 mg mL^-1 streptomycin, 2.5 mg mL^-1 amphotericin B (all from Corning, Tewksbury, MA, USA), The incubation condition was established at 37 ̊C, humid atmosphere and 5% CO₂. (^{ECHIBURÚ-CHAU, 2014}).

In vitro growth inhibition screening

To measure growth inhibition effects, cell viability was assessed using MTT assay. Cells were seeded for 1 day prior to exposure in a 96-well format in sextuplicate, for A549, Caco-2, MCF7 and HepG2 10,000 cells per well were seeded, and 20,000 cells/well for HEK-293 and B-16 cell lines. After this incubation period, culture medium was discarded and cells were washed with sterile DPBS, and replaced with two concentrations of ethanolic extracts (10 and 100 µg mL^-1) in DMEM medium without FBS for a period of 24 hours at same incubation conditions established before.

After this period, culture medium with extract treatments was discarded and replaced with 200 µL MTT medium (1,2 mM of 3-(4,5-dimethylthiazol-2- yl)-2,5-diphenyltetrazolium bromide in DMEM medium without serum), and incubated for 2.5 h at 37 °C and 5% CO₂ (^{KUMAR et al., 2018}). An amount of 200 µL of DMSO was used for solubilizing the formazan crystals and incubate cells for 15 minutes at 37 °C.. The absorbance was measured with a TECAN Infinite pro 200 plate reader at 560 nm. The viability percentage was calculated against a non-drug treated control (designated 100% viability) considered as a control vehicle (≤ 0.5% DMSO). For background absorption, some wells remained cell-free with DMSO as a blank control.

Statistical analysis

Data were analyzed using GraphPad 6.0 software for Windows. The experimental data were expressed as means ± SD; the comparisons were made between controls and treated cultures using a one-way ANOVA followed by Dunnett’s post-hoc test for multiple comparisons. p<0.05 was considered to indicate a statistically significant difference between values.

RESULTS AND DISCUSSION:

Five plant extracts (Rosmarinus officinalis, Cynodon dactylon, Dunalia spinosa, Psoralea glandulosa and Azorella compacta) exhibited the highest cytotoxicity (cell viability ≤ 50%), at 10 µg mL^-1 at least in one cell line, representing the 18.5% of the total plants (Table 2). A. compacta shows to be active against MCF7 (50.0 ± 9.0), HEK-293 (43.6 ± 14.2) and B16 (11.4 ± 7.1), is effective in a major range of cell lines; R. officinalis was active for HEK-293 (59.0 ± 11.7) and B16 (30.0 ± 8.3). C. dactylon shows a cytotoxic effect only in HepG2 (31.8 ± 5.1); Finally, D. spinosa and P. glandulosa show activity only against B16 cell lines (38.0 ± 6.6 and 44.5 ± 6.2, respectively). Previously, we isolated an azorellane diterpene from A. compacta with cytotoxic activity, showing to be effective on breast cancer cells (MCF7) after 24 h exposure (^{BÓRQUEZ et al., 2016}).

Now the spectrum of potential studies expands for new cancer types. D. spinosa was only active against melanoma B16 cell line. Supporting these results, these species have reports of an isolated active metabolite, withaferin A (^{ERAZO et al., 2008}). A steroidal lactone that has been reported as an inhibitor of cell proliferation of uveal melanoma cells with an IC₅₀ value of 2,42 µM, also shifts G₂/M cell cycle arrest, and induces apoptosis in multiple cell lines in vitro and decreases melanoma tumor growth in vivo (^{SAMADI et al., 2012}; ^{SAMADI, 2015}). Furthermore, the present findings show that ethanolic extract was able to induce strong activity on the human lung carcinoma cell line A549 at 100 µg mL^-1 (0.4 ± 1.3 of viability) and 10 µg mL^-1 (70.0 ± 5.5), opening new possibilities for new anticancer applications. The R. officinalis extract showed good activity at 10 µg mL^-1, decreasing the cell viability (30,0 ± 8,3) of melanoma B16 and kidney HEK-293 cell lines. This effect is supported by previous results where the extract shows complementary activity as an agent in anticancer therapy (^{GONZÁLEZ-VALLINAS et al., 2015}), specifically against two Human melanoma cell lines, M14 and A375.

Moreover, the R. officinalis extract has shown a protective effect on plasmid DNA damage (^{RUSSO et al., 2009}). Also, studies of the whole extract suggest possessing antiproliferative activity, not attributed to the main chemical constituents isolated such as carnosic acid, carnosol, ursolic acid and rosmarinic acid (^{HUANG et al., 1994}; ^{CATTANEO et al., 2015}; ^{ANDRADE et al., 2018}; ^{BOURHIA et al., 2019}). Other authors have demonstrated a differential cytotoxic effect of C. dactylon, being effective on laryngeal HEP-2, cervical HELA and MCF7 cancer cell lines but with minimum effect on monkey kidney Vero cells (^{KANIMOZHI, 2012}; ^{AL-SNAFI, 2016}). The extract showed strong activity on MCF7, recording a 50% of anticancer activity at the concentration of 0.625 µg mL^-1, in comparison to the present work where no activity was detected. Other interesting cases are the extracts of P. glandulosa that in the present work induced cytotoxic activity on B16 melanoma cell line. Authors have reported activity of the aerial parts exudate from such plant able to inhibit the growth of A2058 melanoma cells with an IC₅₀ value of 10.5 µg mL^-1 after 48 hours of treatment (^{MADRID et al., 2015}). Moreover, bakuchiol acetate, an isolated meroterpene from this plant was tested, exhibiting a major cytotoxic effect against melanoma cells. The present work reveals additional potential as a natural source to be studied on HEK-293 kidney cells showing strong activity at 100 µg mL^-1 (5.3 ± 2.2 of viability) linked to the decrease on viability up to 68.4 ± 10.9.

Our results indicated that five of the 26 studied ethanolic extracts from highland species in northern Chile shows strong activity at least in one cell line tested at 10 g mL^-1 in a first stage (Table 2). In a second line, our findings present five plants worthy to be evaluated as promissory ethanolic extracts with cytotoxic activity, in spite not to be considerably strong at 10 µg mL^-1. Further efforts will be made to analyze the IC₅₀ values and also to extend the treatment times from 24 to 48 and 72h to obtain more reliable results that can lead us to find biologically active extracts to perform fractioning and isolation of the active compounds. This work is the first glance and part of a systematic attempt to screen the cytotoxic activity of Andean plants looking for new candidates to be used in cancer therapies.