It is widely accepted that humans can taste mono- and disaccharides as sweet substances, but they cannot taste longer chain oligo- and polysaccharides. From the evolutionary standpoint, the ability to taste starch or its oligomeric hydrolysis products would be highly adaptive, given their nutritional value. Here, we report that humans can taste glucose oligomer preparations (average degree of polymerization 7 and 14) without any other sensorial cues. The same human subjects could not taste the corresponding glucose polymer preparation (average degree of polymerization 44). When the sweet taste receptor was blocked by lactisole, a known sweet inhibitor, subjects could not detect sweet substances (glucose, maltose, and sucralose), but they could still detect the glucose oligomers. This suggests that glucose oligomer detection is independent of the hT1R2/hT1R3 sweet taste receptor. Human subjects described the taste of glucose oligomers as "starchy," while they describe sugars as "sweet." The dose-response function of glucose oligomer was also found to be indistinguishable from that of glucose on a molar basis.

The sense of taste is essential for identifying potential nutrients and poisons. Accordingly, specialized taste receptor cells are activated by food-derived chemicals. Because of its importance in the human diet, oral detection of starch, or its degradation products, would presumably be highly beneficial. Yet, it has long been assumed that simple sugars are the only class of carbohydrates that humans can taste. There is, however, considerable evidence that rodents can taste starch degradation products (i.e., glucose polymers composed of maltooligosaccharides with 3-10 glucose units and maltopolysaccharides with >10 glucose units) and that their detection is independent of the sweet taste receptor, T1R2/T1R3. The present study was designed 1) to measure individual differences in human taste perception of glucose polymers, 2) to understand individual differences in the activity of salivary α-amylase, and 3) to investigate the role that salivary α-amylase may play in the taste perception of glucose polymers. In the first experiment, subjects rated taste intensity of glucose, sucrose, NaCl, and glucose polymers of various chain lengths, while their noses were clamped. Saliva samples from the subjects were also collected and their salivary α-amylase activity was assayed. Results showed that the perceived intensities of glucose, sucrose, and NaCl were significantly correlated (r = 0.75-0.85, P < 0.001), but not with the longer chain glucose polymers, whereas intensity ratings of all glucose polymers were highly correlated with one another (r = 0.69-0.82, P < 0.001). Importantly, despite large individual differences in α-amylase activity among subjects, responsiveness to glucose polymers did not significantly differ between individuals with high and low α-amylase activity. A follow up experiment was conducted to quantify the concentrations of glucose and maltose that were inherently present in the glucose polymer stimuli and to determine whether the amounts were within a perceptually detectable range. Results revealed that the amounts of simple sugars present in the test stimuli were trivial and were mostly at an undetectable level. These data together provide strong evidence that humans can taste glucose polymers and that the responsiveness to glucose polymers is independent of that to simple sugars.

Since starch is a significant part of human diet, its oral detection would be highly beneficial. This study was designed to determine whether starch or its degradation products can be tasted and what factors influence its perception. Subjects were asked 1) to taste 8% raw and cooked starch samples for 5, 15, and 35 s and rate perceived intensities of sweetness and "other" taste (i.e., other than sweet), 2) to donate saliva to obtain salivary flow rate (mg/s) and salivary α-amylase activity (per mg saliva), and 3) to fill out a carbohydrate consumption survey. Subsequently, in vitro hydrolysis of starch was performed; saliva was collected from 5 subjects with low and high amylase activities and reacted with 8% raw and cooked starch at 2, 15, and 30 s. Hydrolysis products were then quantified using a High performance liquid chromatography. The results showed cooking increased the digestibility of starch such that the amount of hydrolysis products increased with reaction time. However, cooking did not influence taste ratings, nor were they influenced by tasting time. Subjects' salivary amylase activities were associated with the efficacy of their saliva to degrade starch, in particular cooked starch, and thus the amount of maltooligosaccharide products generated. Effective α-amylase activity [i.e. α-amylase activity (per mg saliva) × salivary flow rate (mg/s)] and carbohydrate consumption score (i.e. consumption frequency × number of servings) were also independently associated with sensory taste ratings. Human perception of starch is undoubtedly complex as shown in this study; the data herein point to the potential roles of salivary α-amylase activity and carbohydrate consumption in the perception of cooked starch.

Considerable research is focused on understanding the functionality of starch hydrolysis products (SHP) consisting of glucose, maltose, maltooligosaccharides (MOS), and maltopolysaccharides (MPS). A confounding factor in this research is the high molecular dispersity of commercially available SHP. The study presented herein characterizes a flexible fractionation approach for lowering the dispersity of such products. This was accomplished by fractionating a corn syrup solids (CSS) preparation based on the differential solubility of its component saccharides in aqueous-ethanol solutions. Products obtained from selected fractionations were characterized with respect to degree of polymerization (DP; liquid chromatography), dextrose equivalency (reducing sugar assays), and prevalence of branching (NMR). Glucose and maltose were preferentially removed from CSS using high (⩾90%) ethanol extractants. Preparations with relatively narrow ranges of MOS, lower DP MPS, and higher DP MPS were obtained through repetitive 70%-ethanol extractions. Linear, as opposed to branched, MOS and MPS were preferentially extracted under all conditions tested.

No abstract