Cytotoxicity of plants from the Brazilian semi-arid region: A comparison of different selection approaches

•We study the role of different plant selection criteria to find extracts with cytotoxicity activity.•Three approaches were used: ethnobotanical, chemical-ecological and random.•Contrary to our expectations, neither approach were especially useful to find cytotoxic agents in the local flora.

This study aimed to answer the following questions: (1) Are plants that are selected using the ethnobotanical or chemical-ecological approach more cytotoxic than those selected using a random approach? (2) Are trees and shrubs more cytotoxic than herbs? (3) For a given ethnobotanical approach, are the cytotoxicity results in cancer cell lines using medicinal plants commonly selected for cancer treatment different from the results in the same cell lines using plants selected using the syndromic importance value (SIV) method? We selected 18 plant species using the ethnobotanical approach, 20 using the random approach and 20 using the chemical-ecological approach. After acquiring 50 μg/mL of hexanic and methanolic extracts, the cytotoxic activity of the samples was tested in HEp-2 (larynx cancer) and NCI-H292 (lung cancer) cell lines using the MTT method (3-[4.5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide). No significant differences were observed between the ethnobotanical, chemical-ecological and random selection approaches. Additionally, no significant differences were observed when the cytotoxicity of trees was compared with that of the shrubs and herbs. Although the percentage of active plant species used for cancer treatment was higher than those selected using the SIV, 50% and 16.67%, respectively, these differences were not significant. Neither approach provided meaningful insight into the search for cytotoxic agents in the local flora.