Based on recent studies in adult subjects, saliva composition is increasingly considered as a physiological factor contributing to taste sensitivity or acceptance. In order to evaluate a possible link between salivary protein composition and taste acceptance in infants, 73 infants participated longitudinally in taste acceptance tests and donated saliva at the age at 3 and 6 months. Intake ratios, reflecting acceptance of a taste solution relative to water were calculated for the five basic tastes. Salivary proteins were separated by one-dimensional electrophoresis and bands were semi-quantified by image analysis. Partial least square (PLS) regression analyses were performed for each taste at both ages to explain intake ratios by band intensities. Bitterness acceptance in the younger infants was unique in the sense that salivary protein profiles could partly predict bitter taste acceptance. At that age, infants were on average indifferent to the 0.18-M urea solution, but great variability in acceptance was observed. The six bands considered as the best predictors for bitterness acceptance were identified by MALDI-TOF mass spectrometry. Higher abundance of bands containing secretory component, zinc-α-2-glycoprotein and carbonic anhydrase 6 was associated to a lower bitterness acceptance, while higher abundance of bands containing lactoperoxidase, prolactin-inducible protein and S-type cystatins was associated to a higher bitterness acceptance. In a second stage, S-type cystatin abundance was measured by Western blotting in order to tentatively confirm this particular finding in an independent group of 22 infants. Although not reaching statistical significance, probably due to a relatively small sample size, it was again observed that cystatin abundance was higher in infants accepting more readily the bitter solution over water. Conclusion: saliva protein composition may contribute to bitter taste acceptance in the younger infants.
