EVALUATION OF CYTOTOXIC AND ANTITUMORAL PROPERTIES OF TESSARIA ABSINTHIOIDES (HOOK & ARN) DC, “PÁJARO BOBO”, AQUEOUS EXTRACT

Along with compounds from terrestrial microorganisms, the constituents of higher plants have provided a substantial number of the natural product-derived drugs used currently in western medicine1. Elucidation of bioactive plants derived compounds is an expanding field of research, particularly of species that are used in systems of traditional medicine or are utilized as botanical dietary supplements.

Tessaria absinthioides (Hook. & Arn.) DC from the family Asteraceae, is currently named ‘pájaro bobo’, ‘suncho negro’, or ‘brea’. It is a frequent species inhabiting sandy and wet soils in Bolivia, Chile, Uruguay and Argentina, where it covers thousands of square miles in the river basins2. It is an aromatic plant commonly used in traditional medicine and as spices or seasonings through Argentina and Chile territories3. Its infusions have been employed since Aymara and Huarpe folk Medicine´s as an hypocholesterolemic, balsamic, expectorant, for hepatitis, renal insufficiency, diabetes and “empacho” (digestive disorder with a halt of undigested food and non-digestible substances, provoking a stagnation of intestinal transit), along both countries4-6.

The potential therapeutic use of this plant derivatives, have conducted to the isolation of pharmacologically active components with specific biological activities. The aqueous and organic extracts and its purified essential oils have demonstrated virucidal activity against herpes simplex virus type 1 and 2, poliovirus type 2; and Junin virus7. The hidroalcoholic extract, constituted by coumarins, flavonoids, tannins and saponins, have been related to anti-inflammatory effects by inhibition of pro-inflammatory enzymes8. Insecticidal effects have been related to isolated sesquiterpenes which induce topical toxicity, growth alteration effects and repellent activity to Tenebrio molitor and Tribolium castaneum larvae, a serious stored-product insect pest with a world–wide distribution9, 10. Furthermore, isolated tessaric acid has shown in vivo gastric cytoprotective activity11.

Despite recent scientific reports about the plant derived compounds identification and activities, in accordance to our knowledge, there is not a well conducted research about its antitumoral properties and potential toxic effects in animal models. In consequence, the present work aims were: to determine by phytochemical qualitative analysis the composition of T. absinthioides aqueous extract; to study the extract in vitro activity against tumoral and non tumoral cell lines; to discard toxicity induced by its oral administration in laboratory healthy animals; and to show the extract capability to affect in vivo tumor induction. The results of this work are compared with the chemotherapeutic agent 5-fluoracil (5-FU), which have well characterized antitumoral and toxic effects.

Materials and methods

Tessaria absinthioides (Hook & Arn) DC, was collected in Lavalle-County, Mendoza-Argentina (33°44’10”S, 68°21’30.5”W). A voucher specimen (MERL-61823) was deposited in the Mendoza Ruiz Leal herbarium. Leaves were obtained immediately. Then, 20 g of leaves were autoclaved in 200 ml of distilled water. Solids were separated by filtration and the obtained volume was lyophilized. At the moment of use, lyophilized was reconstituted with milliQ water (1 g/ml) and sterilized by passing through a 0.22 µm pore size filter.

A qualitative phytochemical screening for chemical constituents was realized on the aqueous extract. Alkaloids12, 13, flavonoids14, resins15, 16, carbohydrates17, amino acids18, sterols/terpenes19, and tannins20 were evidenced by colorimetric, and precipitation reactions. Identified chemical constituents were extracted by solid-liquid procedures from 1 g of lyophilized crude extract by selective solvents (HPLC grade – J.T Baker), dried obtained compounds were quantified.

In vitro studies were realized in the human cell lines: HeLa, uterine cervix adenocarcinoma; Gli-36, glioblastoma; HCT-116, colorectal carcinoma; MCF-7, breast adenocarcinoma; and HBL-100, non-tumoral epithelium (American Type Culture Collection, ATCC); and cultured in DMEM, supplemented with 10% fetal bovine serum, 100 IU of penicillin and 100 μg/ml streptomycin (GIBCO – USA). Culture conditions were fixed at 37 °C, in a humidified atmosphere enriched by 5% CO2.

The cytotoxic activity was studied by MTT [(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrasodium bromide)] assay. This process requires active mitochondria, and the amount of MTT cleaved is directly proportional to the number of viable cells present. Cells were seeded in 96-well microplates; 24 hours later, the medium was aspirated and replaced by medium containing T. absinthioides (0 to 50 μg/ml) or 5-FU (Filaxis®, Argentina) at indicated doses (0 to 50 μg/ml). Then, cells were incubated for 72 hours; thereafter, medium was replaced by 100 μl of MTT solution (0.5 mg/ml in DMEM, without phenol red nor FBS) and incubated for an additional 4 h. MTT solution was then removed and 100 μl of DMSO added; the plates were shaken for 10 min to dissolve the formazan crystals. The optical density was measured using a ThermoScientific Multiscan Elisa reader at 570 nm. The optical density obtained in untreated cells was taken as 100% viability and the other values were calculated in accordance with controls (concentration 0 µg/ml). The assay was performed three times in triplicate. To quantify the measured cytotoxicity, the 50% of cell viability (CV50) was used.

The animals used in this work were cared in accordance with the Guiding Principles in the Care and Use of Animals of the US National Institute of Health. All procedures were approved by the Institutional Animal Care and Use Committee of School of Medical Science, Universidad Nacional de Cuyo (Protocol approval N°30/2014).

To perform the in vivo toxicity studies, Sprague Dawley rats bred in our laboratory, 8 weeks-old at the onset of treatment were used. They were kept in a controlled light (lights on 6:00 AM to 10:00 PM) and temperature (22-24 °C) room. Rats chow and tap water were available ad libitum.

The acute toxicity study procedure was performed according to the guideline for the testing of chemicals, acute oral toxicity (OECD 200121). A single dose of aqueous extract, was administered by oral gavage to 4 groups of 3 female and 3 male rats each, at doses of 0 (described as control group), 50, 300, or 2000 mg/kg, respectively. The rats were monitored for mortality, body weight changes and other clinical signs during 14 days after dosing. All rats were necropsied to detect grossly observable evidence of organ or tissue damage or dysfunction.

The 28-day repeated-dose study was performed according to the guideline for the testing of chemicals, repeated dose 28-day oral toxicity study in rodents (OECD 200822). Four groups of 3 female and 3 male rats each were administered daily, for 28 days, by oral gavage doses of aqueous extract of T. absinthioides. The doses were 0 (described as control group), 250, 500 and 1000 mg/kg/day. Body weights, water consumption, and general clinical observations were evaluated three times a week. All animals were evaluated daily for abnormal behavior and presence of clinical symptoms considered as endpoints. Observation time was extended until 1 day after the exposure period. Then, animals were sacrificed and given a gross and microscopic pathology examination. Troncal blood was collected; one aliquot was obtained with 10% EDTA and used for blood cells quantification. The other aliquot was collected to serum separation and then used for biochemical determinations.

The in vivo antitumor activity of aqueous extract was studied in adult male BALB/c mice, bred in our laboratory, 6 weeks old at the onset of treatment. They were kept in a light (lights on 6:00 AM to 10:00 PM) and temperature (22-24 °C) room. Mice chow (Cargill, Córdoba, Argentina) and tap water were available ad libitum. Four groups of 8 animals each were separated. Group 1, did not receive treatments and was considered as survival control. In the other 3 groups, colorectal tumor induction was performed by DMH, 20 mg/kg/week, for 22 weeks, S.C.22 Group 2, animals were tumor inducted and did not receive other treatment, it was considered as indicative of no treated tumor survival. Group 3, 8 weeks after tumor induction, animals receive 5-FU 30 mg/kg/week, I.P.23, until animal sacrifice. Finally (group 4), 8 weeks after tumor induction, animals were treated with a dilution of lyophilized T. absinthioides in the drinking water. Mice extract consumption was 300 mg/animal/day, until animal sacrifice. T. absinthioides dose was selected as a higher concentration of extract that not affected the animal water consumption. The human endpoints to sacrifice animals were: reduction of spontaneous activity, respiratory difficulty, not eating/drinking, increased/decreased weight more than 20%, diarrhea, no micturition, anal bleeding and anal tumor protrusion.

Along the work, data are expressed as mean ± standard error (SEM) and analyzed using GraphPad Prism 5.0 software. To assess the 50% of cell viability (CV50), a sigmoidal dose-response analysis was performed; values were considered acceptable when goodness of fit showed R2 ≥ 0.90. Comparisons between two groups were done using Student’s T test. When more than two groups were compared, one-way ANOVA followed by Dunnett’s multiple comparison test was used. Animal survival curves were compared using a Mantel Cox test. In all cases, statistical significance was considered when p < 0.05.

Results

The qualitative phytochemical analysis of the aqueous extract revealed the presence of different chemical groups. Strong positive reactions were registered for flavonoids, carbohydrates, sterols/terpenes, and tannins. Whereas, alkaloids, amino acids and resins were not detected. Present components were quantified and its dry weight, related to 1 g of aqueous extract lyophilized results as follow: flavonoids 76.5 ± 3.2 mg/g, carbohydrates 4.3 ± 0.7 mg/g, sterols/terpenes 244.4 ± 16.0 mg/g, and tannins 27.1 ± 1.3 mg/g.

We next performed the standard MTT assay to demonstrate aqueous extract cytotoxicity against human cell lines (Fig. 1). A comparative study between treatments with T. absinthioides and 5-FU was completed on tumoral cells HeLa, Gli-36, HCT-116 and MCF-7; and non-tumoral

HBL-100 cells. Induced cytotoxicity by T. absinthioides extract was strong in Hela and Gli-36 cells (CV50: 3.0 ± 1.2 and 6.1 ± 1.1 μg/ml, respectively), intermediate in HCT-116 and MCF-7 (7.6 ± 1.0 and 14.8 ± 1.3 μg/ml, respectively) and moderate on HBL-100 (29.5 ± 1.1 μg/ml). Meanwhile, a different pattern of response was observed when 5-FU was a treatment. The CV50 calculated by cell line, was: Gli-36, 2.3 ± 1.2 μg/ml; HCT-116, 3.6 ± 1.1 μg/ml; HeLa, 8.3 ± 1.2 μg/ml; HBL-100, 10.0 ± 1.4 μg/ml; and MCF-7, 18.6 ± 1.1 μg/ml.

The present data demonstrate that uterine cervix carcinoma and breast cancer cell lines were more sensitive to T. absinthioides than to 5-FU. In Hela cells the difference was significant (Student T Test, p = 0.03). On the contrary, in glioblastoma, colorectal cancer and non-tumoral cell lines 5-FU results more cytotoxic than the plant extract. Notably, when a normal cell line HBL-100 was tested, the measured CV50 induced by T. absinthiodes resulted significantly higher (3 folds) than the 5-FU obtained value (Student T Test, p = 0.0003). An increase in CV50 indicates lower cytotoxicity, when the effect is observed in non-tumoral cells (HBL-100), these results indicate a T. absinthiodes selective toxicity against cancer cells.

Then, we performed neurobehavioral, anatomical, hematological and serological studies on Sprague Dawley healthy rats to examine possible alterations induced by T. absinthioides treatment, with special attention in toxic signs similar to those reported for 5-FU treatment.

One dose by gavage of aqueous extract, at concentrations of 0, 50, 300, or 2000 mg/kg/bw did not determine acute toxicity. No females or males died as consequence of extract administration or into the period of 14 days after treatment; no animals showed neurobehavioral signs of toxicity. After necropsy, internal organs were analyzed and no-macroscopical signs of ischemia, bleeding, fibrosis or degenerative proliferation were observed. In consequence, it is possible to discard acute toxicity induced by oral administration of T. absinthioides up to 2000 mg/kg/bw.

Thereafter, we perform a dose-repeated-28-days study of toxicity. During the assay period, neuro-behavioral signs of toxicity were not registered and sensory/motor functions did not show alterations. When treatment finished, anatomo-pathological, blood and serological studies were performed to analyze whether treatment induce toxicity in detoxifying organs, such as liver, kidneys and lungs, or into a rapidly proliferating fraction of cells, such as spleen and circulating blood cells. Glucose, hepatic enzymes, urea and creatinine were determined to analyze liver and kidneys functional status. Serum values of glucose and urea were determined to evaluate metabolism integrity (Table 1). No toxic changes were registered at any used doses. Analytical results obtained at 0 and

1000 mg/kg bw/day are showed in Table 1. Final weight of females and males did not show significant changes. The organs of treated animals did not evidence differences in weight, color or structure. No macroscopical signs of ischemia, bleeding, fibrosis or degenerative proliferation were observed.

Hematocrit and the number of circulating blood cells were similar after treatment. Extract toxicity on rapidly proliferating cells or organs related to the hematological system can therefore be excluded. None of the assessed biochemical parameters showed statistical differences between groups.

All the parameters had no significant differences when 0 and 1000 mg/kg dose/bw/day were compared (Student´s T test, p ≤ 0.05). Moreover, determined parameters are considered normal to the specie24.

Concluding, this section demonstrates that oral administration of T. absinthioides aqueous extract, for 28 days, it is not toxic at doses up to 1000 mg/kg bw/day.

After acute and 28 repeated-dose toxicity was discarded, to evaluate the capability of aqueous extract to exert in vivo antitumoral activity, we induced colorectal tumors in BALB/c mice by DMH administration. In the control group (animals without DMH tumoral induction) animals presented 100% survival. No spontaneous animal deaths were registered, and no animals evidence the humane endpoints criteria during the experimental time (52 weeks after tumor induction start).

All animals treated with DMH developed colorectal cancer and were sacrificed as consequence of tumor disease. Necropsy of animals evidenced 10 to 24 colorectal polyps (Fig. 2, right lower panel), without significant differences between groups; neither difference in polyps´ volume were found. Histologic analysis of tumoral pieces confirms the diagnosis of colorectal adenocarcinoma (Fig. 2, right upper panel).

Only the median survival was different between the groups in which tumors were induced (Fig. 2, left panel). While no treated group median survival was 24 weeks after treatment starts, in 5-FU and T. absinthioides groups were 27 and 30, respectively. Differences between no treated group and T. absinthioides were significant (Mantel Cox test, p = 0.0071). The presented data clearly indicate the antitumoral effects of T. absinthioides when is orally administrated. When T. absinthioides group is compared with 5-FU group, no significant differences were found (Mantel Cox test, p = 0.1179).

Discussion

In accordance with our results, T. absinthioides aqueous extract has a selective cytotoxicity against different human tumoral cell lines showing a reduced activity in non-tumoral cells. In addition, oral administration of aqueous extract does not determine acute or subchronic toxic effects in laboratory healthy animals; whereas, the extract demonstrates in vivo antitumoral activity by increasing the median survival of BALB/c mice with colorectal cancer.

Both, cytotoxic in vitro and antitumoral in vivo results, were compared with the worldwide distributed chemotherapeutic, 5-FU. The antimetabolite 5-FU acts by inhibition of thymidylate synthase enzyme, thus preventing DNA synthesis25. It is widely used in the treatment of various

malignancies like gastrointestinal, breast, colorectal, liver, skin, head and neck cancers26. Although 5-FU is generally well tolerated in vivo, it also possesses severe side effects such as: nephrotoxicity, hepatotoxicity, ototoxicity, and neurotoxicity27. T. absinthioides demonstrated activity, allowed us to compare its efficiency against cancer cells with 5-FU, with a reduced or absent secondary undesirably effects against non tumoral and healthy tissues.

Because T. absinthioides evidenced anticancer activity, a new window of cancer research was opened for this plant. It is remarkable that 61% of anticancer agents approved are natural derived products28. Some phytochemical components of the T. absinthioides aqueous extract are widely reported as active compounds against cancer cells. As examples, flavonoids actions are related to direct free radical scavenger, cytotoxicity and protection against pro-carcinogenic compounds29, 30; terpenes and sterols were described as cytotoxic compounds against several tumoral cells and potentiating pro-apoptotic agents31-33; while, vegetal derived tannins were demonstrated to act as cytotoxic by induction of antiproliferative and pro-apoptotic actions34, 35. In accordance to the exposed antecedents, reported chemical groups derived from T. absinthioides aqueous extract, represent a promissory field of study for plant-derived anti-cancer products.

Although, some in vitro studies were published using T. absinthioides natural derivatives, there were not previous reports of significant cytotoxicity against tumoral cell lines. Describing virucidal activity, García et al.9, 10 reported lack cytotoxicity for Vero cells of the essential oils extracted from T. absinthioides. Other studies demonstrate the antiproliferative activity of semi-synthetic molecules derived from tessaric acid obtained from aerial parts of the plant; sensitive cell lines were A2780 (ovarian cancer), HBL-100 (non-tumoral epithelium), HeLa (uterine cervix adenocarcinoma), SW1573 (lung alveolar carcinoma), T-47D (breast adenocarcinoma) and WiDr (colorectal adenocarcinoma); curiously, in the same work, the natural derivative tessaric acid was reported as inactive36. Other study performed by Torres Carro et al.8 reported nontoxic activity of hydro-alcoholic extract of T. absinthioides on RAW 264.7 (murine macrophages) at doses up to 200 μg/ml; interestingly, phytochemical identification of this extract reported flavonoids, tannins, but not sterol/terpenes contents. In consequence, the cytotoxic results presented in the current work represent a valuable evidence of T. absinthioides natural extract activity on cancer cells. In relation to the cytotoxicity, it is important to remark, the capability of the aqueous extract to exert a selective cytotoxicity on tumoral cells. While the 5-FU shows an intermediate cytotoxicity on HBL-100 cell line, similar to the measured for HeLa uterine cervix cancer cells, the CV50 of T. absinthioides extract was the highest on non tumoral cells, been the double than values obtained for the extract on MCF-7 cells; and the triple of the value presented by 5-FU on the HBL-100. Altogether, the evidence show in this work make the aqueous extract an important material to perform bio-guided fractionation procedures destined to isolate and identify specific chemical molecules with oncologic relevance.

On the other side, the present results show that acute and repeated-doses of T. absinthioides do not exert undesirable effects; while it is widely accepted that, in laboratory animals, the common clinical side effects of 5-FU include myelosuppression, diarrhea, vomiting, mucositis, cardiotoxicity and neurotoxicity37. The obtained data about non-toxic consumption of T. absinthioides support their ethnopharmacological, culinary and biomedical uses. Moreover, they validate the in vivo experimentation of previous reported antiviral, anti-inflammatory and gastric cytoprotective agent properties7, 8, 11, 38. About its use as insecticidal9, 10, the results shown herein provide experimental data that exclude induction of toxic effects on vertebrate animals. Overall, the safety oral administration, analyzed together with the cytotoxicity against cancer cells and extract antitumoral effect, constitute strong evidence to promote investigation of T. absinthioides plant derivatives and its biomedical and biological properties.

When the antitumoral effects are considered, the median survival increase obtained by oral administration of T. absinthioides, indirectly indicates bioavailability, stability and activity of natural plant derivative/s on in vivo models. In the context of cancer research, the in vivo confirmation of the hypothesized antitumoral properties represents a critical result to ensure compound efficacy.

Despite its in vivo results, and the cytotoxic effects described against the human colorectal cancer cell line (HCT-116), it is not possible to stablish a clear correlation between the effective doses used in both models. In fact, the efficiency of in vitro cytotoxic assays to predict animal toxicity remains controversial39-42. In accordance with Huntjens et al.43, we consider that pharmacokinetic and pharmacodynamic data form the basis for scaling and predicting drug effects in vivo. Although, much more information needs to be explored about T. absinthioides antitumoral actions, the present work represents a well demonstrated report of plant value into the cancer research.

Altogether, the results presented in the current work open a new window about this plant research into the oncology field. Deep studies are required to elucidate the chemical structure of bioactive molecules, its pharmacological properties, the specific cellular responses, its related molecular pathways, and the capability of the compound to potentiate others current used chemotherapeutic agents.

In conclusion, the present work is a relevant report of T. absinthioides aqueous extract activity related to cancer. The natural compound evidence cytotoxic activity against different human tumoral cell lines, representative of glioblastoma, uterine cervix, colorectal and mammary adenocarcinomas; with a selective and significant diminished effects on non tumoral cells. Acute and dose-repeated in vivo administration of aqueous extract it is not toxic; in healthy animals, administration of plant derivatives does not induce neurobehavioral, anatomical, hematological, serological nor lethal effects. In mice induced colorectal cancer, oral administration of extract increases significantly the median survival. Altogether, the current work make the T. absinthioides derived products promising for cancer research and treatment.

Acknowledgements: This work was supported by 06/J473 grant, from Secretaría de Ciencia, Técnica y Postgrado (SeCTyP) – Universidad Nacional of Cuyo – Argentina; PICT2014-1877 grant, from Agencia Nacional de Promoción Científica y Tecnológica (ANPCyT) and Instituto Nacional del Cáncer (INC) grant, both from Ministerio de Ciencia, Tecnología e Innovación Productiva – Argentina; and grant 11220150100579CO, from Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET).

Conflict of interests: None to declare.