Discrimination of Hypotrigona species (Apidae: Meliponinae) in Kenya using nest architecture, cephalic secretions and molecular tools

Stingless bees are important pollinators contributing significantly to biodiversity and food security. Stingless bees produce honey that has high medicinal value that fetches higher prices compared to the honey produced by honey bees (Apis mellifera). However,identification and classification of Afrotropical stingless bees, which is key and important for their domestication, still remains ambiguous and solely reliant on the use of morphological features. In this study, an integrative taxonomy approach was applied to identify and differentiate three Hypotrigona species: Hypotrigona gribodoi, H.ruspolii and H. araujoi that are found in Kenya. Nesting sites, nest architecture,morphometrics, DNA barcoding and hemotaxonomy using whole head extracts were employed as complementary tools to identify and differentiate the Hypotrigona species.Colonies of the three Hypotrigona species from Kakamega forest and Mwingi, Kenya were nested at meliponary on the campus of icipe in Nairobi, Kenya. Nest sites, nest entrance (colour and sizes) and nest architecture (brood cells arrangement and sizes,honey and pollen pots sizes, presence or absence of involucrum and colour of propolis)were recorded. It was found that nest sites are specific with Hypotrigona gribodoi nesting mostly in revices of mud walls in homesteads, while H. ruspolii and H. araujoi nest in specific ndigenous tree species found in Kakamega forest. The colour of external nest entrances varies between species. Those in H. araujoi were yellowishbrown; white or cream in H. gribodoi while H. ruspolii’s were dark brown. There is an internal nest entrance in H. gribodoi, which is absent in the other species. Brood cells are clustered in H. gribodoi’s and H. ruspolii’s nests, whereas H. araujoi’s form vertical semicomb-like layers. The surface area of the apical opening of the entrance tube and volumes of brood cells, honey and pollen pots differ significantly between the three Hypotrigona species. Using veins on the right forewing and hind leg for morphometrics analysis, H. gribodoi and H. ruspolii were separated from H. araujoi. However, there is an overlap between H. gribodoi and H. ruspolii. On the other hand, using mitochondrial DNA, COI gene, the three Hypotrigona species were clearly separated. A lower genetic distance exists between H. araujoi and H. gribodoi from Kakamega (1.4%) than between H. gribodoi collected from Kakamega and H. gribodoi from Mwingi (4.3%).Using gas chromatography and mass spectrometry, analysis of extracts from the head of workers, 50 components belonging to six chemical classes; hydrocarbons, aldehydes,alcohols, terpenoids, steroids and fatty acids were identified. Twenty-nine compounds were found in the cephalic extracts of H. araujoi, 26 in H. gribodoi and 33 in H.ruspolii. Workers were successfully grouped into their respective species and colonies using sixteen components among which: heptacosene, heptacosanol and octadecanol contributed most to the separation into species.In conclusion, nest entrance and nest architecture show variation between the three Hypotrigona species. Use of morphometrics and molecular taxonomic approaches (DNA barcoding) provides a convenient, robust and reliable way to identify Hypotrigona species. The large genetic distance between H. gribodoi collected from Kakamega and Mwingi suggests that H. gribodoi ex-Mwingi may be a new undescribed putative species. Gas chromatographic and mass spectrometric analysis of head extract reveals differences in secretions among the Hypotrigona species. Thus, integrative taxonomy tools applied in this study provides a valuable alternative to identify Hypotrigona species. In addition, this study indicates the need for a thorough revision of Hypotrigona species.