Aparatrigona). Melissopalynological analyses of the present study identified pollen grains of these same floral sources in the analysed honeys. However, even these data showed a diversity of pollen collected by Meliponini. Stingless bees may change their trophic niche throughout the year due to several factors. The availability of floral resources (pollen, nectar

and resin), climatic oscillations, distance between the colony and the flowering plant species, and competition exerted by exotic and other native bee species represent some factors that contribute to oscillations. The total phenolic selleck screening library content of the methanol extracts of the honey samples ranged from 17 to 66 mg GAE/g of extract (Table 2). These figures are related to the honey signaling pathway floral source, because the phenolic

compounds are related to the botanical origin of the nectar and pollen and to the species of the honey-producing bees (Gheldof & Engeseth, 2002). The samples with predominance of a single pollen type, CAD4, SAD3 and CAD2, presented the highest quantities of total phenolic contents, and the lowest total phenolic contents were observed for the honeys SAD2 and SAD1. In relation to the ABTS + cation radical-scavenging activity, the methanol extracts of the honeys showed activities in which the CE50 varied from 210 ± 0.25 to 337 ± 3.17 mg mL−1 ( Table 2). Among the honeys analysed, the samples collected at CAD4, SAD3 and CAD2 showed the highest antioxidant capacity, most likely as a consequence of their higher total phenolic content compared with the remaining samples, because antioxidant Ribose-5-phosphate isomerase activity can be increased by the synergetic interaction between compounds that have the capacity to scavenge free radicals, such as phenolic compounds. SAD1 and SAD2 showed the smallest total phenolic contents and displayed the smallest antioxidant activities. The results of the present paper agree with previous works that report the correlation between total phenolic contents and antioxidant activity (samples displaying smaller total phenolic contents also showed smaller antioxidant responses) (Aljadi and Kamaruddin, 2004, Alvarez-Suarez et al., 2012 and Bertoncelj et al., 2007). The similarity between the

honeys considering the results of total phenolic content and antioxidant activity showed a dendrogram with four clusters (A, B, C and D). The cluster A included the related samples CAD1, CAD3 and CAD2, which showed intermediary values for phenolic content and radical scavenging activity (Table 2). The cluster A had weakly correlation with cluster C, formed by SAD1 and SAD2; these two samples showed the lowest antioxidant activity, probably as a consequence of the lower phenolic compounds content when compared to the others samples (Table 2). The clusters A, C and B had no correlation with the cluster D. This cluster (D) included the samples SAD3 and CAD4, which possessed the highest antioxidant activity and the highest phenolic content (Table 2).