CHAPTER 1: Saliva: Properties and Functions in Food Oral Processing

Saliva has an essential role in food digestion, bolus formation and sensory perception during food oral processing. Saliva is a complex heterogeneous clear fluid mainly secreted by the parotid, sublingual and submandibular glands. It consists of roughly 98% water and 2% organic and inorganic substances.¹  The presence of proteins and other large molecules means that saliva is a unique colloidal fluid with distinct rheological and lubricating properties. The origin and composition of human saliva have been studied extensively by scientists from biological, physiological and dental research backgrounds.^2–4 

Saliva is the first body fluid that contacts or interacts with ingested food, whether it is a solid or liquid. The mixture of saliva and food, which soon becomes an integrated body, ready, in most cases, to be safely swallowed, is referred to as a bolus. There has been a growth in research about saliva and food oral processing in recent years and this chapter focuses on salivary properties and functions, as well as saliva–food interactions, to present a full picture of recent research about the properties of saliva and how saliva participates in food oral processing.

The secretion of saliva differs from person to person and varies within the same individual at different times of day (a circadian rhythm). When resting with no external stimulation, the average flow rate of saliva secretion is about 0.25–0.35 mL min⁻¹. However, the secretion of saliva is greatly increased (typically 4–10 times higher) when stimulated (chemically, mechanically or aromatically), with >50% attributed to saliva secreted from the parotid gland.⁵  The inorganic components in saliva include ions and ionic groups such as Na⁺, K⁺, Cl⁻, Ca²⁺, Mg²⁺, HPO₃²⁻ and HCO₃⁻. The organic components in saliva (see Table 1.1) consist of body secretion products (urea, uric acid and creatinine), putrefaction products (putrescine and cadaverine), lipids (cholesterol and fatty acids) and >400 different types of protein. In terms of these proteins, much attention has been paid to proteins that are glandular in origin (e.g. amylase, histatins, cystatins, lactoferrins, lysozymes, mucins and proline-rich proteins) and plasma derivatives (e.g. albumin, secretory immunoglobulin A and transferrin).⁶ 

The pH of saliva is a major indicator of oral health in dentistry due to its close relation with dental caries across all age groups.^7–9  The natural pH of saliva is in the neutral range, between 5.6 and 7.6 for healthy individuals, with an average of 6.75.¹  However, this is particularly dependent on salivary calcium and phosphate concentrations, which vary between individuals.¹⁰  It has been reported that the pH of saliva has a circadian rhythm –the average intra-oral pH is about 6.7 during sleep, but this increases to 7.2 when awake.¹¹  A high salivary pH has been found to lead to better oral health and a lower incidence of dental caries.¹²  A lower pH, such as caused by tobacco consumption (containing pyridine alkaloids and aromatic hydrocarbons), may damage the oral mucosa and, furthermore, affect taste perception.¹³ 

Saliva has a buffering capacity due to the presence of bicarbonate/carbonate ions and, to a lesser extent, phosphate ions and proteins, and can neutralize acids in the oral cavity, maintaining a stable pH environment.¹⁴  Factors such as oral health, the demineralization–remineralization balance, dilution and antimicrobial activity are all important factors influencing the buffering capacity of saliva. The time it takes for the pH of saliva to return to the resting state after stimulation (e.g. drinking an acidic sparkling beverage) can be used to evaluate the buffering capacity of saliva.¹⁵  Bicarbonate is believed to be the principal buffer of saliva. Its concentration varies dramatically from about 5 mmol L⁻¹ in unstimulated human whole saliva, when it is produced at a flow rate of 0.3 mL min⁻¹, up to 24 mmol L⁻¹ in stimulated whole saliva at a flow rate >2 mL min⁻¹.¹⁴  Carbonic anhydrase VI helps to maintain a high bicarbonate level in saliva with the reversible reaction between CO₂ and HCO₃⁻. Although the optimal buffering pH for the phosphate and carbonate systems at 25 °C occurs at pH 7.2 and 6.3, respectively (the pH of the HCO₃⁻/H₂CO₃⁻ buffer system ranges from 5.1 to 8), buffering below pH 5 relies more on protein buffering.¹⁴  Protein buffer systems are mostly determined by their amino acid composition. Proteins at concentrations such as those found in human saliva (amyloglucosidase with lysozyme/α-amylase) exhibit a measurable buffering capacity.¹⁵  One study¹⁶  looked at the human salivary α-amylase subproteome and found that 67 amylase spots most frequently matched a range of isoelectric points from pH 3.5 to 7.6 with a molecular weight range of 18–59 kDa. These α-amylase variants may function like zwitterionic buffers, a buffer system operational between pH 3.5 and 5, with auxiliary buffering through anionic and cationic sites present as non-interacting carboxylate and ammonium side-chains between pH 5 and 8.¹⁷ 

More than 1000 proteins have been reported in saliva. These have several biological and antimicrobial functions that influence numerous aspects of oral health, food digestion and taste perception.¹⁸  In dentistry, saliva is an effective medium for the monitoring and diagnosis of disease. Salivary proteins and biomarkers (immunoglobulins, lysozyme, lactoferrin, cystatins and histatins) show good diagnostic potential in monitoring and detecting periodontal disease, oral cancers and dental caries.^19–22  Table 1.1 lists the proteins widely studied in food oral processing.

As the most abundant enzyme in saliva (40% of total salivary proteins), salivary α-amylase has long been used clinically as a non-invasive biomarker for sympathetic activity.^27–30  It is a calcium-containing metalloenzyme that hydrolyses the α-1,4 linkages of starch to glucose and maltose.³¹  Adults are reported to have higher salivary α-amylase activity than children.³²  In the oral cavity, salivary α-amylase is mainly involved in the initial digestion of starch. Salivary α-amylase activity shows a distinct diurnal profile with a pronounced decrease within 60 min after awakening and a steady increase in activity during the course of the day.³³  Several studies have shown that smoking, caffeine intake and psychological stress will affect salivary α-amylase activities.^34,35 

Salivary α-amylase is also associated with obesity. Mennella et al.³⁶  compared people aged 19–54 years and found that the overweight participants had higher salivary α-amylase activity and salivary lipolysis than the normal weight participants. There may also be a genetic link between carbohydrate metabolism and obesity. A decrease in the copy number of the salivary amylase gene (AMY1) leads to a decrease in salivary α-amylase levels and a higher risk of obesity.³⁷  When the proportion of dietary carbohydrates is higher, the AMY1 copy number and salivary α-amylase activity are higher than in a low-starch diet. Mejia-Bentez et al.³⁸  evaluated the number of AMY1 copies in 597 Mexican children (293 obese and 304 normal weight) and found that the average AMY1 copy number in obese children was lower than that in normal weight children. The children with >10 copies of AMY1 were all of normal weight.

Dietary habits might also have an impact on salivary α-amylase activity.³⁹  The secretion of saliva was assessed in the two major ethnic groups of China: Chinese Han and Chinese Mongolian. People from the Chinese Han ethnic group have lived for a long time in the central regions of China, where water is readily available for crop production, and they rely heavily on carbohydrates as their staple food. By contrast, Chinese Mongolians, as a nomadic ethnic group, rely heavily on meat and dairy products. The salivary α-amylase activity of Chinese Han people is significantly higher than that of Chinese Mongolians in both stimulated and unstimulated saliva samples. This is indicative of an association between dietary intake and the biochemical properties of saliva. The concentration of α-amylase is not equal to the activity of α-amylase. α-Amylase activity is dependent on both the α-amylase concentration and protein modification (e.g. glycosylation), and even the formation of complexes with other salivary proteins, such as mucins.⁴⁰  It is therefore possible to have individuals with similar concentrations of salivary α-amylase, but different levels of salivary enzymatic activity and vice versa.

Mucin is another major category of salivary protein and concentrations as high as 16% of the total salivary protein content have reported.⁴¹  Mucins are glycoproteins consisting of a linear polypeptide core with a highly glycosylated central part accounting for up to 80% of its molecular weight, which ranges between 0.5 and 20 MDa.⁴²  Human salivary glands secrete two major types of mucin: oligomeric mucin (MG1) with a molecular weight >1 MDa (encoded by the MUC5B gene) and monomeric mucin (MG2) with a molecular weight of 200–250 kDa (encoded by the MUC7 gene).⁴³  As a result of the abundance of negatively charged groups, arising mainly from sialic acid residues and sulfated sugars, mucins generally have low isoelectric points, estimated to be between pH 2 and 3.⁴⁴  The presence of a large number of charged groups results in the pH-dependent physicochemical properties of mucins. In addition, mucin molecules have unique properties as surfactants. The naked parts of the mucin protein backbone adhere to hydrophobic materials in aqueous environments as a result of their hydrophobicity, whereas the carbohydrate side-chains are thought to orient themselves away from the surface. Mucins therefore tend to adsorb to hydrophobic surfaces via protein–surface interactions and to hold water molecules via their hydrophilic oligosaccharide clusters.

In the oral cavity, salivary mucins interact with oral bacteria and other salivary proteins to maintain oral health.^45–47  Mucins protect the oral cavity via several mechanisms influenced by their unique polymeric structures. Mucins can interact with other salivary proteins to alter their localization and retention, which could provide increased protection for the oral cavity.⁴⁸  Salivary mucins also serve as carriers for antibacterial salivary proteins within the oral cavity, increasing their retention in the dental pellicle, and/or protect proteins from proteolytic degradation through the formation of complexes. Mucins also interact with oral microorganisms to speed up their removal and/or reduce their pathogenicity. Early studies also showed that salivary mucins secreted by the submandibular/sublingual glands aggregate suspended bacteria (particularly streptococci) and induce the attachment of bacteria to mucin-coated surfaces via specific glycan anchors, thus removing them selectively.^49,50 

Albumin is the most abundant serum protein, accounting for >50% of all plasma proteins. In the oral cavity, salivary albumin is considered as a serum ultrafiltrate and it may diffuse into oral mucosal secretions.⁵¹  Patients with medically compromised conditions, such as immunosuppression, diabetes and sometimes periodontitis, typically have a higher salivary albumin content.^52,53  The ability of salivary albumin to interact with polyphenols has also been reported and, more specifically, that salivary albumin can markedly increase the antioxidant activity of polyphenols.⁵⁴ 

Lysozyme is an alkaline protein composed of a single-chain polypeptide and is an important antibacterial component in human saliva.⁵⁵  It selectively breaks down the cell walls of microorganisms without damaging other oral tissues. As a non-specific immune factor, lysozyme is involved in the regulation of homeostasis in the oral environment. The occurrence and development of caries, oral mucosal disease and periodontal disease have been reported to be related to oral lysozymes.^53,56 

The paired major salivary glands in humans – the parotid, submandibular and sublingual glands – along with hundreds of small, minor submucosal salivary glands, provide a film of mixed saliva that coats and protects the oral mucosal and tooth surfaces.⁵⁷  Salivary secretion is maintained at a default rate of 0.25–0.35 mL min⁻¹, creating a slow but mobile film and replenishing/replacing the proteins adsorbed onto the underlying soft and hard oral surfaces. Superimposed on this default secretion of unstimulated or resting saliva is the secretion of much greater volumes of saliva in response to taste, smell and chewing during periods of food intake (i.e. stimulated saliva).⁵⁸ 

Salivary glands are densely innervated by the parasympathetic and sympathetic nerves of the autonomic nervous system. Parasympathetic nerve impulses typically produce a high-flow, low-protein saliva, whereas sympathetic impulses usually produce a low-flow, high-protein saliva. Garrett⁵⁹  showed that parasympathetic stimuli are particularly important for mucin secretion. Salivary glands are composed of two main cell types: the acinar cells, which make the saliva; and the ductal cells, which modify and convey the saliva into the oral cavity.⁶⁰  The secretion of saliva involves the active secretion of salt (as sodium and chloride ions) by the acinar cells into the ductal lumen of the gland when a neural signal is received from the brain. Water, but not protein, derived from the blood system passes around, via tight junctions, and through, via aquaporin channels, the acinar cells to form saliva that is isotonic to serum. Salt is mostly recovered in the parotid and submandibular glands by the striated ducts, which are impermeable to water.⁶¹ 

The recovery of salt from saliva changes the primary isotonic saliva into hypotonic saliva. By existing in hypotonic saliva, taste buds are able to detect salt at much lower thresholds than those found in serum and this is why tears, sweat and blood taste salty. However, the reabsorption of salt is an energy-expensive process (hence the large number of mitochondria within the striated ducts) and is not upregulated during stimulated salivary secretion. The result is that stimulated saliva has higher sodium and chloride concentrations than resting saliva. By contrast, protein secretion into saliva is usually mediated by the sympathetic nerve stimulation of β-adrenergic receptors acting via intracellular changes in cyclic adenosine monophosphate to cause the fusion of secretory granules with the apical membrane of cells.⁶² 

Salivary flow is a continuous process in conscious humans and is upregulated by a reflex mostly stimulated by taste and chewing.^60,63  Resting, or unstimulated, salivary flow (0.25–0.35 mL min⁻¹) is the result of low-level autonomic stimulation, including the orbitofrontal cortex and amygdala of the brain working via the salivary centres within solitary tract nuclei in the brain stem to act on salivary glands.^64,65  These inputs from the higher centres are reduced when we are asleep. The salivary flow rate is therefore reduced (to about 0.1 mL min⁻¹) and this is why teeth are particularly susceptible to microbial attack at this time. With stress, the higher centres reduce nerve traffic to the salivary centres and then to the salivary glands, which causes a dry mouth sensation.⁵⁹ 

Salivary secretion is upregulated above the resting rate by taste and chewing stimulation and, to a lesser degree, by aroma (smelling) stimulation. Chewing foods stimulates the receptors in the periodontal ligament⁶²  sandwiched between the tooth and the alveolar process of the jawbone. However, these receptors are not stimulated by empty chewing, such as teeth grinding.⁶⁶  Citric acid is the most commonly reported taste stimulant. It generates by far the highest salivary flows, often at one-tenth of the concentration of other tastants such as sweet, salty, bitter and umami.⁶⁷  It is difficult to collect truly pure saliva samples from single salivary glands (e.g. submandibular or sublingual). Anatomically speaking, the submandibular gland duct passes through the sublingual gland and can sometimes connect into the sublingual duct before entering the mouth.⁶⁸ 

Of the major salivary glands, the parotid glands are the largest and contribute the greatest flow (as much as 60% of the total) when stimulated by taste or chewing⁶⁴  (Table 1.2). The parotid gland secretes a serous saliva that contains no mucin, but is rich in α-amylase and proline-rich proteins. It is the gland most responsive to changes in diet, such that the gland shrinks in size and becomes atrophic during nutrient starvation, but can regenerate with the resumption of feeding.⁶⁹  The submandibular and sublingual glands are less responsive to changes in diet and contribute more to the resting salivary flow rate.

There are also hundreds of minor salivary glands located in the submucosa covering the oral cavity. These glands secrete small volumes of mucin-rich saliva and are generally considered not to secrete reflexively – that is, there is no increase in salivary flow in response to food intake or stimulation. Although only contributing about 10% of salivary flow, the minor glands are important in maintaining a mucin-rich layer adjacent to the mucosa.⁶⁶  A small subset of these minor glands, the von Ebner's glands, is of particular interest to scientists. The von Ebner's glands are located at the base of crypts that surround the foliate and circumvallate papillae on the tongue, where the majority of the taste buds are located. Secretions from these glands tend to be more serous than mucous. They also contain proteins important in food processing – lipocalin and lingual lipase – both of which have been speculated to have roles in the detection of dietary fat in the oral cavity.⁵⁷ 

Once a food is placed inside the mouth, the mechanical stimulation from chewing and the chemical stimulation from food components will lead to an increase in the secretion of saliva via neural reflexes.⁷⁰  Under the action of teeth grinding and tongue manipulation, the food and saliva are gradually mixed to form a bolus conducive to swallowing. Saliva coats the surfaces of food particles, destabilizes emulsions and other colloidal systems, participates in the formation of soluble and insoluble aggregates and clusters, breaks down compounds through enzymatic action, dissolves tastants and binds aroma molecules. The perception of food attributes will be affected through the deformation, destabilization and size reduction of food structures.

Size reduction and mechanical clustering are facilitated by the teeth and tongue, but the cohesion of food particles during bolus formation is mainly supported by salivary enzymes, especially α-amylase and lipase as the major oral digestive enzymes. The α-amylase present in saliva is mainly secreted by the parotid glands and breaks down α-1,4 glycosidic bonds. When eating starchy products, the enzymatic interactions of α-amylase with the starch components help to break down food structures and/or reduce the viscosity (for fluid foods) on the timescale of oral processing.⁷¹  Correspondingly, different sensory attributes are generated during mastication and size reduction, such as melting, vanilla flavour, slippery lip–tooth feel, thickness, creamy after-feel, creaminess, fattiness, roughness and stickiness.^72–74 

The structure of the food product itself determines the degree of hydrolysis and bolus formation during oral processing. Hoebler et al.⁷⁵  studied the oral digestion of bread and spaghetti by 12 French participants. With a pre-determined swallowing point, they found a significant difference in the amount of saliva added to the food products: 220 ± 12 g kg⁻¹ fresh matter for bread and 39 ± 6 g kg⁻¹ for spaghetti. This difference in the degree of moisture (the incorporation of the amount of saliva) led directly to different bolus structures. They found highly degraded structures in the bread boluses, but a relatively intact microstructure of the food product in the spaghetti boluses. The oral hydrolysis of starch in bread was twice as high as that seen in spaghetti. Gao et al.⁷⁶  conducted a similar study comparing steamed bread with baguettes. They found that the addition of saliva to the baguettes was much higher than the addition of saliva to the steamed bread, but they also pointed out that no significant difference was seen for the bolus hardness of these two products. However, the difference in reducing sugar (an indicator of how much starch is hydrolysed) was not studied.

Lipase is another salivary enzyme that is believed to play a part in food digestion in the mouth. Human salivary lipase, mainly secreted from the von Ebner's glands, is highly hydrophobic. It can trap fat globules and hydrolyse medium- to long-chain triglycerides into free fatty acids. The fatty acids not only enhance the perception of fat, but also increase the lubricity of the bolus, thus reducing the time of oral processing.⁷⁷  However, the general consensus is that the lipase content of human saliva is very low, making it unlikely that fat digestion is influenced by the secretion of the von Ebner's and other minor salivary glands.⁷⁸ 

Bolus properties and the oral processing time depend on the nature of the food itself, the strength of mastication and the salivary flow rate⁷⁹  (see also Chapters 3 and 4). The amount of saliva added to the food or comminuted particles influences aggregation, dispersion and further bolus formation. The initial moisture content of a food also affects the number of chews and the amount of saliva required for a safe swallow.⁸⁰  In the case of shortbread biscuits, for example, a paste-like structure is typically obtained at a moisture content of ∼45% and they are safe to swallow at ∼55%. Eck et al.⁸¹  suggested that the addition of toppings to bread and crackers assists bolus formation. Topping with cheese spread or mayonnaise resulted in a shorter chewing time with fewer chews than when bread and crackers were consumed alone. These toppings contributed to faster bolus formation by providing moisture, so that less saliva needed to be incorporated into the bolus during mastication.

Flavour perception during food oral processing, discussed in detail in Chapter 5, is more than the properties of the food itself – it is the perception of flavour in a mixture of food with saliva.⁸²  The concentrations of taste compounds dissolved in saliva are more closely related to the taste perception actually felt than the concentrations printed on the package label.⁸³  The individual variability (in saliva) explains the variabilities in perception and acceptability of the same food products by different individuals across different dietary backgrounds and cultures.⁸⁴  In addition to the dilution effect caused by the flow of saliva, the interaction of saliva constituents with food components also affect the perception of flavours,⁸⁵  including ions and ionic groups, salivary enzymes (e.g. carbonic anhydrase VI, α-amylases and lipases) and salivary proteins (e.g. cystatins, histatins, albumins and mucins).^82,86,87  Salty, sweet, sour, bitter and umami are recognized as the five basic tastes and, recently, fatty has also been argued to be a basic taste.⁸⁸ 

Sodium chloride is important in maintaining food texture as well as preserving food quality and safety. Sodium is perceived through epithelial Na⁺ channels, allowing the diffusion of sodium through the cell membrane⁸⁹  (see also Chapter 5). This diffusion depends on a concentration gradient between the extra- and intracellular fluids. Elevated salivary flow rates are correlated with the lower release of sodium and saltiness perception.⁹⁰  The structure and breakdown of food also affect the transportation of salty tastes to the taste receptors during the continuous mixing of food with saliva. With starch-thickened foods, Ferry et al.⁸⁵  found that individuals with higher α-amylase activities showed a lower perception of saltiness, in conjunction with a lowered thickness perception. The disruption of the granular structure of gelatinized starch caused by amylase decreases the mixing efficiency of starch with water and leads to the reduced transport of sodium to the taste buds. The preference for saltiness is associated with the biochemistry of saliva. A positive correlation with salivary proteolysis ability and a negative correlation with salivary carbonic anhydrase VI concentration were observed in >200 French participants.³⁴ 

The effect of salivary flow on sweetness is less clear than the effect on saltiness. Sweet taste receptors (see also Chapter 5) are a heterodimer composed of taste type 1 receptor 2 (T1R2) and taste type 1 receptor 3 (T1R3), which sense sweet tastes in the taste buds,⁹¹  a process that, unlike saltiness, does not require a concentration gradient.^92,93  Aoyama et al.⁹⁴  developed mathematical regression models for the relation between sweetness sensitivity and salivary biochemical composition. They showed that the pH of unstimulated saliva has the strongest effect on sweetness sensitivity in Japanese young adults. Other correlations were also found between sweet sensitivity and α-amylase levels (negatively, in 159 Portuguese students)⁹⁵  and leptin levels (negatively, in 32 Australian adults).⁹⁶  The perception of simple sugars with T1R2 and T1R3 receptors might not be the same as the perception of glucose polymers. Lapis et al.⁹⁷  showed that humans are capable of tasting glucose polymers with various chain lengths and that the responsiveness is different from that of simple sugars. It also has little correlation with individual salivary α-amylase concentrations. Low et al.⁹⁸  recruited 34 participants who were able to sense complex carbohydrates (maltodextrin and oligofructose) in the oral cavity and found that the detection threshold for complex carbohydrates was not directly correlated with the sweetener detection threshold.

Adults with high salivary flows can neutralize the acidity of sour solutions more efficiently than those with low salivary flows.⁹⁹  This is probably due to the higher dilution effect or buffering capacity of the high salivary flow rate. However, the evidence is still controversial for the perception of sourness. Bonnans and Noble⁹³  could not find any significant difference between sourness perception and salivary flow rate, whereas Heinzerling et al.⁹²  used an imposed alteration of the salivary flow rate and showed that an increased salivary flow rate lowered the perception of sourness. This imposition of an alteration of the salivary flow rate also affected the perception of saltiness, but not bitterness nor sweetness. Building on previous work demonstrating a number of genes expressed in sour-sensing receptors, Zhang et al.¹⁰⁰  found that a proton-selective ion channel, Otopetrin-1, is essential in sensing sour. They also showed that sourness is represented by its own dedicated population of neurons in the brain system.

The most widely studied bitter taste compounds are those with thiocyanate moieties, such as phenylthiocarbamide and 6-n-propylthiouracil (PROP).¹⁰¹  There are some arguments about the relationship between salivary protein levels and the bitter taste response. Cabras et al.¹⁰²  showed that the basal levels of two basic proline-rich proteins, II-2 and Ps-1, were significantly higher in PROP super-tasters than in unstimulated saliva samples from non-tasters. This increased rapidly with PROP stimulation, but only in PROP super-tasters. Higher salivary ionic zinc concentrations and carbonic anhydrase VI were also related to a lower responsiveness to PROP in adults¹⁰³  and urea in infants.¹⁰⁴  Martin et al.¹⁰⁵  showed that model animals (Long Evans rats) on a protein-inducing diet (14–37 kDa) had a higher bitterness (quinine) detection threshold than control animals. Using an artificial bitter stimulant (sucrose octa-acetate) as the control, they showed that salivary protein–bitter stimulant interactions were stimulus-specific rather than task-specific.

Umami is a pleasant feeling originally associated with monosodium l-glutamate (MSG), but repeatedly reported with other substances, including amino acids, bi-functional acids, Maillard reaction products and peptides.^106–108  The perception of umami is substance-dependent and differs with individual salivary biochemistry and oral physiology. Scinska-Bienkowska et al.¹⁰⁹  divided participants based on their salivary endogenous glutamate levels and found that, although not significantly different in rating the intensity and pleasantness of samples with lower MSG concentrations (0.03–1.0%), the group with low salivary glutamate levels rated the samples with higher MSG concentrations (3.0 and 10.0%) as significantly more unpleasant. This difference in hedonic response, originating from variations in the basal salivary biochemistry, needs further investigation for the taste and recognition mechanisms of taste receptors to umami compounds.^110,111  The perception of umami, like other taste perceptions, decreased with age in white European¹¹²  and Japanese participants.¹¹³  However, the stimulation of MSG could induce a longer secretion of saliva, as demonstrated by Sasano et al.,¹¹⁴  in addition to significant reductions in several other symptoms associated with decreased saliva secretion (dry mouth, difficulty speaking and swallowing, masticatory disturbance, mucosal pain and taste disorder). This makes umami a potential candidate for the treatment of clinical xerostomia.

Fat perception is a complex sensation dependent on different sensory cues, such as texture, olfaction and taste.¹¹⁵  Many researchers believe that oral lipase hydrolyses triacyl glycerides release free fatty acids and result in fat perception¹¹⁶  via mechanisms such as delayed-rectifying potassium channels, G protein-coupled receptor 120 and CD36 glycoprotein receptors.¹¹⁷  However, the amount of lipase secreted from minor salivary glands is very small and fatty acids are poorly soluble in aqueous solvents, which might preclude their access to sensory receptors. Poette et al.¹¹⁸  showed that participants with a higher salivary lipolytic activity had a higher detection threshold for an oleic acid oil-in-water emulsion when considering only the role of lingual lipase (with a nose clip). However, when the nose clip was taken away, thus adding in multimodal olfactive and trigeminal stimuli, salivary protein appeared to contribute more negatively to the detection of fatty acids. This was also suspected by Gasymov et al.¹¹⁹  of a hydrophobic binding pocket at the centre of lipocalin-1, which functions as a carrier of fatty acids. This hypothesis was evidenced by a study conducted in the authors' research group, in which a depletion of certain proteins during the oral processing of rapeseed oil and pork fat and the formation of a stable saliva emulsion was observed.¹²⁰  More details can be found in Chapters 7 and 9.

Aroma perception is an important factor driving food preference and acceptance. Saliva plays an important part in aroma release and perception. The release of aroma from fluid foods was different when the fluid foods were added with various solvents – water, artificial saliva or human whole saliva.^121,122  With soft solid foods, the flow rate of stimulated saliva also affected aroma release as a result of differences in chewing performance.¹²³  Salivary proteins function via hydrophobic effects, electrostatic interactions and other mechanisms to regulate and facilitate aroma release and perception.¹²⁴  Salivary mucins have a high molecular weight and several hydrophobic domains. The adsorption of mucins for aromatic molecules such as ketones and esters in the presence of hydrophobic effects therefore increases with the length of the aliphatic chain.¹²⁵  α-amylase, although globular, also exhibits hydrophobic effects and decreases the release of aroma compounds.^125,126  Despite efforts to reproduce human saliva, artificial saliva still lacks similarity when it comes to interactions with aromatic compounds with different aroma release behaviours, mainly because of a lack of salivary enzyme function in ester hydrolysis,¹²⁵  aldehyde degradation¹²⁷  and the oxidation of thiols.¹²⁸  More details are discussed in Chapter 5.

Texture and rheology are considered as the two major physical properties in relation to food oral processing and affect the sensory perception of consumed foods. The texture properties of food include mechanical strength (force–deformation relationships), tactile sensations, as well as visual and auditory responses.¹²⁹  The specific textural properties of certain foods can be evaluated by trained panellists and are more relevant to material property measurements.¹³⁰  Texture-analysing devices have been used to study the material properties of foods since the 1960s, typically a universal testing machine^131,132  and, later, a texture analyser. Textural profile analysis, currently the most widely used methodology, is still applied, although its correlation with sensory properties and terminologies is questioned by the scientific community.^133–135 

Rheological properties focus on the deformation and flow of foods. By accurate control of the contact surface area, strain rate and temperature, a rheometer can provide an idea of how thick a fluid food might be perceived in the mouth,^136–139  or the chewiness of a gel.^140,141  There is currently no single evaluation methodology that can simultaneously reflect all the dynamic changes during food oral processing. There are some specific reviews on the textural considerations and perceptions of solid foods,¹⁴²  hard and soft solid foods,¹⁴³  composite foods¹⁴⁴  and food bolus.¹⁴⁵  Chapters 4, 5, 6, 10 and 12 also discuss texture perception.

Tribology is the study of friction and lubrication between interacting surfaces in relative motion. There are plenty of interacting surfaces in the mouth during food consumption, such as teeth–teeth, tongue–palate, tongue–teeth, teeth–food, tongue–food, tongue–bolus, lips, lips–food, bolus–palate, food particles–oral surfaces.¹⁴⁶  The main function of saliva is to lubricate oral surfaces and protect them from damage and to ensure that food moves easily around the oral cavity and is then comfortably swallowed. A lack of saliva will make eating, drinking and talking a very unpleasant experience.¹⁴⁷ 

Conventional rheological studies applied in food science assume a physical perception of food texture, determined by the mechanical properties of the food material, especially food deformation, flow and fracture. However, when it comes to fluids and, more often, surface properties, the limitations of considering only food rheology (and texture) becomes clear. Sensory descriptions such as creaminess, slipperiness and astringency cannot be fully explained by a rheological test. Perceived thickness in the oral cavity does not always relate to the viscosity measured at a shear rate of 50 s⁻¹ in a rheometer.¹⁴⁸  Studies have shown that during food oral processing, sliding and pressing of the tongue against the palate could result in a force of between 0.01 and 90 N and a sliding speed of up to 200 mm s⁻¹.^149,150  When bulk properties dominate the system and the thickness of the film of fluid-like foods or beverages between oral surfaces is >100 µm, rheology and texture analyses can explain sensory attributes such as thickness, hardness or melting. However, when the surface properties start to dominate at smaller length scales (microns to nanometres), tribology has shown itself to be a better option in explaining sensory properties such as creaminess, smoothness, astringency and roughness, or after-feel.¹⁵¹  Chapters 2, 7, 9 and 10 discuss tribology and the friction properties of certain foods.

Many studies of the oral lubrication of foods have taken an empirical approach to characterizing the differences between samples and comparing these differences with sensory studies. Correlations have been found between the friction coefficient and certain texture descriptions, such as smoothness, fattiness and creaminess.^152–154  Emulsions have been studied extensively as a model food system to investigate the influence of oils and fats on texture and mouthfeel; see also Chapter 9 on emulsions in oral processing and Chapter 7. Several studies have indicated that the localized coalescence of emulsions in oral tribological contacts may be the driver for fat-related textural attributes such as creaminess. The emulsions that were more sensitive to in-mouth coalescence led to a higher creamy mouthfeel and fatty/oily sensations.¹⁵³  Upadhyay et al.¹⁵⁵  summarized a perception mechanism for gel and liquid emulsions, emphasizing the importance of oral coating, saliva viscosity and droplet size in the thick, rich, smooth and coating attributes of creaminess (see Figure 1.1). Droplets that are small in size in relation to surface roughness do not usually contribute significantly to the lubrication properties of the emulsion.¹⁵⁶  There have been extensive studies on creaminess perception using instruments such as the mini-traction machine, ball-on-plates and ring-on-plates devices, and an optical tribological configuration.^{152,157–162} 

Most current commercially available tribological devices have contact surfaces that are planar (plates and discs) or curved (balls or rings). Typical stainless-steel surfaces cannot really reflect biological tissues in the oral cavity and therefore soft surfaces, such as rubbers and silicones that more closely resemble oral surfaces, have been widely applied in food oral processing studies.^163,164  Animal tongue tissues have also been used as substrates for tribological studies, but they usually lack repeatability and vary biologically as a result of the different stages of decay of the tissues.¹⁵³ 

Chojnicka-Paszun and de Jongh¹⁶⁵  argued that good lubrication is more important than the mechanical properties (hardness or roughness) of the contacting surface during oral tribology. In soft tribological contacts, the hydrodynamic lubrication properties of oils have been shown to be the same as those of an aqueous fluid of the same viscosity.¹⁶⁶  This highlights the fact that oil does not necessarily have unique lubricating properties in a soft tribological contact because it is a Newtonian fluid, whereas many aqueous polymers used for thickening are highly shear thinning and hence their viscosity at high shear rates is not necessarily much more than that of water. This low viscosity at high shear rates may explain why they are not very effective as replacers for fat-related mouthfeel properties because friction is dominated by the high shear viscosity of the material in the contact zone. As stated by Sethupathy et al.,¹⁶⁷  tribology is a system property and therefore possible correlations between lubrication and food texture are dependent on the chosen testing conditions (such as the tribopair, sliding speed, applied load, surface roughness, hydrophobicity and modulus) and food samples (amount of saliva and transitional properties).

Although mainly comprised of water, proteins and ions change saliva into a unique colloidal fluid vital to humans. Building on the progress made so far, further research is required in the following areas. A full understanding of saliva proteomics is needed to reveal the functionalities of saliva proteins. Over the past decade, advances in technology have led to a greater understanding of some protein structures and how these might affect protein–protein interactions. However, we only know about a small portion of the entire salivary protein bank and this hinders our exploration of salivary proteins. A better understanding of salivary proteomics will also fill the knowledge gap in saliva–food interactions. How food is broken down and how food particles stick/adhere together in the presence of saliva is a major focus of research in the food oral processing community. This needs to be considered from physical, physicochemical, physiological and biological perspectives and needs to take into consideration both saliva itself and the interactions between saliva composition and food components.

Understanding and predicting saliva–food interactions are both important and urgent given the rapid increase in the elderly population worldwide. Safe eating and swallowing are now a major concern in almost all countries and cultures. The appropriate design of food for elderly people needs a better understanding of the procedures undergone by food before reaching the swallowing point and how these procedures could be optimized to reduce difficulty in swallowing and the occurrence of dysphagia.

We need smarter, high-precision and easily applied evaluation methodologies. We still need an exact correlation between food sensory properties and the output of assessment devices. This requires not only improving existing technologies, but also integrative inputs from disciplines other than food science, such as neuroscience (about how the brain responds to stimuli), mechanics and engineering, and how to mimic and simulate orofacial and oral muscle behaviours. Computer science can help us to evaluate response output (such as language and facial expression) using non-invasive approaches and to improve prediction accuracy by the use of computational algorithms and virtual reality. The advancement of sensory evaluation methodologies will also support measurement and device development. All these call for global cooperation between academic and industrial professionals to push forward the discipline of oral food processing.

This work was supported financially by the National Natural Science Funding Council of China (Grant No. 31801637) and the National Key Research and Development Plan (2017YFD0400101) from the Ministry of Science and Technology of China.