Skip to main content
Dryad logo

Lowered sensitivity of bitter taste receptors to β-glucosides in bamboo lemurs: An instance of parallel and adaptive functional decline in TAS2R16?


Imai, Hiroo et al. (2021), Lowered sensitivity of bitter taste receptors to β-glucosides in bamboo lemurs: An instance of parallel and adaptive functional decline in TAS2R16?, Dryad, Dataset,


Bitter taste facilitates the detection of potentially harmful substances and is perceived via bitter taste receptors (TAS2Rs) expressed on the tongue and oral cavity in vertebrates. In primates, TAS2R16 specifically recognizes β-glucosides, which are important in cyanogenic plants’ use of cyanide as a feeding deterrent. In this study, we performed cell-based functional assays for investigating the sensitivity of TAS2R16 to β-glucosides in three species of bamboo lemurs (Prolemur simus, Hapalemur aureus, and H. griseus), which primarily consume high-cyanide bamboo. TAS2R16 receptors from bamboo lemurs had lower sensitivity to β-glucosides, including cyanogenic glucosides, than that of the closely related ring-tailed lemur (Lemur catta). Ancestral reconstructions of TAS2R16 for the bamboo-lemur last common ancestor (LCA) and that of the Hapalemur LCA showed an intermediate sensitivity to β-glucosides between that of the ring-tailed lemurs and bamboo lemurs. Mutagenetic analyses revealed that P. simus and H. griseus had separate species-specific substitutions that led to reduced sensitivity. These results indicate that low sensitivity to β-glucosides at the cellular level – a potentially adaptive trait for feeding on cyanogenic bamboo – evolved independently after the Prolemur-Hapalemur split in each species.


Calcium assays were performed as previously described (Itoigawa et al., 2019). Calcium 4 and Calcium 5 (Molecular Devices, Sunnyvale, CA) were used as calcium indicators. Final concentrations of salicin were set to 0, 0.25, 1.0, 2.5, 5.0, 10, 20, and 40 mM; those of arbutin were set to 0, 0.25, 1.0, 2.5, 5.0, and 10 mM; and those of linamarin was set to 0, 0.25, 1.0, 2.5, 5.0, 10, and 20 mM. Data were collected from 3–5 independent experiments. The calcium response is expressed as the normalized peak response (F) relative to background fluorescence (F0): ΔF/F (= [FF0]/F0). The response of cells transfected with the empty pEAK10 vector (no insert) and Gα16/gust44 was defined as the TAS2R-independent response and was subtracted from all responses. To calculate dose-response relationships, ΔF/F values were fitted to the nonlinear regression model f(x) = min + [(max − min)/(1 + x/EC50)h], where x is the test compound concentration and h is the Hill coefficient, using the drc package in R (Ritz et al., 2015). Threshold concentrations were defined as the lowest substance concentration where the normalized fluorescence (ΔF/F) was higher than that in 0 mM (Dunnett's test, p < 0.05). Maximal signal amplitudes (Amax) were defined as the maximum normalized fluorescence (ΔF/F) within the tested substance concentrations. Statistical comparisons of results for wildtype, reconstructed ancestral, and point-mutant receptors were performed by Welch’s t test with Benjamini–Hochberg (BH) correction.


Japan Society for the Promotion of Science, Award: Nos. 18J22288 to AI, 16K18630 to TH, and 18H04005 and 19K21586 to HI