The Tongue Is Covered by a Series of Papillae, Some of Which Contain Taste Buds

The tongue is mostly muscle, but its surface is covered by a series of bumps and folds called papillae (Fig. 13-1). Most are small conical projections not involved in taste, but fungiform, foliate, and circumvallate papillae contain taste buds (Fig. 13-2). About 200 to 300 fungiform (“mushroom-shaped”) papillae are scattered across the surface of the anterior two thirds of the tongue, concentrated on the tip and sides. Typical fungiform papillae contain 3 to 5 taste buds. Foliate (“leaflike”) papillae are the most posterior of a series of about 20 folds on the sides of the posterior tongue; each has 100 to 150 taste buds in its walls. Finally, a series of 8 to 9 circumvallate (“surrounded by a wall”) or vallate papillae is arranged in a V-shaped line two thirds of the way back along the dorsal surface of the tongue. Each circumvallate papilla is surrounded by a deep groove in the lingual epithelium (Fig. 13-2A), with about 250 taste buds located in the walls of the groove. Hence, even though the circumvallate papillae are few in number, they contain nearly half of the 5000 taste buds found on an average tongue.

Figure 13-1 Distribution and innervation of taste buds (left), and innervation of the epithelium of the oral cavity (right). The trigeminal nerve (V) subserves general sensation (and the common chemical sense) from the anterior two thirds of the tongue, and the glossopharyngeal nerve (IX) has a similar function for the posterior third of the tongue. The innervation of the soft palate is indicated as being trigeminal, but it also receives a contribution from the facial nerve. Similarly, the innervation of the oropharynx is indicated as being glossopharyngeal, but it also receives a contribution from the vagus nerve.

Figure 13-2 Morphology of taste buds. A, Section of a single circumvallate papilla from the tongue of a cat. B, Enlargement of the area outlined in A. Individual taste buds (arrows) can be seen just beneath the surface of the papilla. C, Light micrograph of a section through the center of a single taste bud from a human fungiform papilla, showing the taste pore (arrow). D, Electron micrograph of a taste bud from a human circumvallate papilla. The section is not quite through the middle of the taste bud, but the arrow indicates the apex where the taste pore would open in a nearby section. The principal cellular elements are taste cells (TC) of various categories and basal cells (BC). Discrete synapses of the receptor cells onto afferent fibers cannot be seen easily in this micrograph, but numerous glossopharyngeal nerve processes (IX) are apparent.

(A and B, courtesy Dr. Nathaniel T. McMullen, University of Arizona College of Medicine. C and D, courtesy Pamela Eller, University of Colorado Health Sciences Center.)

Although an “average” tongue contains about 5000 taste buds, the numbers of both papillae and taste buds are surprisingly variable. For example, among normal, healthy individuals, some have as many as 100 times more taste buds in their fungiform papillae than others do. This is presumably the basis of the 100-fold variation in threshold concentration for various substances among normal individuals.

Taste buds are usually associated with the tongue, but they are also distributed widely, although in smaller numbers, over the palate and pharynx.^* The pharyngeal and palatal taste buds are probably more important for swallowing and for reflex responses to good or bad tastes than for conscious awareness of taste.

Taste Receptor Cells Are Modified Epithelial Cells with Neuron-Like Properties

Each taste bud is an ovoid structure containing about 100 spindle-shaped epithelial cells modified as taste cells. Some of these are supporting cells with glial properties, and the rest are taste receptor cells (Fig. 13-2B to D). Taste receptor cells have microvillar processes that extend through a small opening, the taste pore, where they are exposed to chemical stimuli. At the deep end of the taste bud, the receptor cells communicate with visceral sensory fibers from the facial, glossopharyngeal, and vagal nerves (Fig. 13-3). Some make typical chemical synapses, releasing adenosine triphosphate (ATP) and probably other transmitters onto these afferent endings; others use a less conventional communication method, releasing ATP directly into extracellular space, where it diffuses to the same afferent endings. Fibers from the facial nerve innervate the taste buds of fungiform and anterior foliate papillae and the palate; fibers from the glossopharyngeal nerve innervate those of circumvallate and most foliate papillae and the pharynx; and a few vagal fibers innervate those of the epiglottis and esophagus (Fig. 13-4).

Figure 13-3 Communication from taste receptor cells to peripheral endings of facial, glossopharyngeal, and vagal nerve fibers. Tastant molecules activate the transduction machinery in the apical microvilli of taste receptor cells (1) and cause the production of depolarizing receptor potentials (2). In some receptor cells, this depolarization causes entry of Ca²⁺ through voltage-gated Ca²⁺ channels (3), release of ATP onto a peripheral nerve ending (4), and increased firing of the nerve fiber (7). In others, depolarization causes the direct release of ATP onto nearby nerve endings through a Ca²⁺-independent mechanism in which gap junction channels, instead of forming a connection with another cell, connect to extracellular space and allow ATP to spill out (5). Large depolarizations are required to open these channels, and voltage-gated Na⁺ channels (6) apparently provide the needed amplification of the receptor potential.

Figure 13-4 Innervation of taste buds in different parts of the oral cavity by the facial (VII), glossopharyngeal (IX), and vagus (X) nerves. The central processes of all three terminate in rostral parts of the nucleus of the solitary tract. CT, chorda tympani nerve; GG, geniculate ganglion; GP, greater petrosal nerve; IG IX, inferior ganglion of the glossopharyngeal nerve (petrosal ganglion); IG X, inferior ganglion of the vagus nerve (nodose ganglion).

(Drawing of hemisected head, modified from Parker CA: A guide to diseases of the nose and throat and their treatment, New York, 1906, Longmans, Green.)

Taste receptor cells differentiate from the surrounding lingual epithelium and subsequently depend on chemical interactions with the gustatory nerves for their continued existence; denervation of an area of tongue causes degeneration of its taste buds. Despite this epithelial origin, taste receptor cells have some very neuron-like properties: they contain transduction machinery in their apical membranes and produce receptor potentials in response to appropriate taste stimuli; some make typical chemical synapses on the peripheral endings of gustatory nerves, and others even produce action potentials when sufficiently depolarized by a receptor potential.^* However, unlike almost all neurons, taste receptor cells have a limited life span. Each lives only a week or two before being replaced by differentiation of basal cells, which migrate in from the surrounding epithelium and wait, as their name implies, near the base of the taste bud.

Taste Receptor Cells Utilize a Variety of Transduction Mechanisms to Detect Sweet, Salty, Sour, and Bitter Stimuli

The four basic taste qualities traditionally recognized are sweet, salty, sour, and bitter, but there are others. For example, the flavor-enhancing effect of glutamate (as in monosodium glutamate, or MSG) is a separate taste in its own right. This taste is often referred to as umami, which is Japanese for “delicious,” and it is based on taste receptor cells specifically sensitive to glutamate. Other receptor cells may be selectively sensitive to fats and other components of tastants. Although some areas of the tongue are somewhat more sensitive to certain taste qualities—the tip to sweet, the sides to salt and sour, and posterior portions to bitter—all parts of the tongue are sensitive to all kinds of tastants. Corresponding to these multiple taste qualities, taste receptor cells utilize multiple methods to transduce chemical stimuli into electrical signals (Fig. 13-5). Some are epithelial mechanisms adapted for use in taste transduction, and others are familiar ligand-gated or G protein–coupled mechanisms. Transduction processes include the following:

1. The transduction process for sodium chloride (NaCl), the prototypical salty stimulus, is ultimately simple. No receptor molecules are involved: cation channels in the apical membranes of taste receptor cells allow inward movement of Na⁺ ions, depolarizing the cell (Fig. 13-5A). Some of these channels are selective for Na⁺, reflecting the dietary importance of NaCl; others are nonselective, accounting for the salty taste of compounds such as potassium chloride (KCl).

2. Acids taste sour, but not in a way that is related in a simple way to the pH of saliva. Weak organic acids (e.g., acetic acid in vinegar) taste more sour than a hydrochloric acid (HCl) solution of the same pH. Corresponding to this, there are at least two different transduction mechanisms for acidic solutes (Fig. 13-5B). Weak organic acids diffuse across the receptor cell apical membrane, dissociate, acidify the cytoplasm, and initiate the opening of cation channels. Stronger acids depolarize receptor cells by acting as the ligand that opens pH-sensitive cation channels.

3. Sweet compounds and glutamate, in contrast, bind to G protein–coupled receptor molecules that, through a second messenger cascade, cause cation channels to open in other parts of the cell (Fig. 13-5C).

4. Bitter tastes, typical of many toxic substances, serve a protective function and are usually avoided by humans and other animals.^* Bitter-sensitive taste receptor cells use another set of about 30 different G protein–coupled receptors, allowing them to be depolarized by a broad range of chemicals. Using the same second messenger cascade as sweet- and umami-sensitive receptors, they too release ATP onto afferent endings (Figs. 13-3 and 13-5C).

Figure 13-5 Transduction mechanisms used by taste receptor cells. A, Increased extracellular Na⁺ concentration moves the Na⁺ equilibrium potential in a positive direction, and Na⁺ ions (salty taste) flow directly through open Na⁺ channels; increased extracellular concentrations of other cations also cause depolarization via open nonselective cation channels. B, Weak organic acids (sour taste) diffuse across receptor cell membranes in an undissociated state (1), then acidify the cytoplasm (2). Intracellular acidity causes cation channels to open (3). The extracellular protons from stronger acids cause pH-sensitive cation channels to open (4). C, The transduction process for sweet and bitter substances and glutamate (umami) is more convoluted. All bind to different G protein–coupled receptors (1) on different taste receptor cells. Dissociation of the G protein from one of these receptors activates an enzyme, phospholipase Cβ2 (2), whose products (inositol trisphosphate and diacylglycerol) lead to the release of Ca²⁺ from internal stores (3) and the opening of Ca²⁺-gated cation channels (4).

Second-Order Gustatory Neurons Are Located in the Nucleus of the Solitary Tract

Chemosensory information reaches consciousness as the perception of flavor, but in its role in autonomic responses and the acquisition of food, it has much closer ties to the hypothalamus and limbic system than do other senses. The second-order neurons that mediate the involvement of taste in all these connections are located in the nucleus of the solitary tract (see Fig. 12-11). The nucleus of the solitary tract, the principal visceral sensory nucleus of the brainstem, receives (via the solitary tract) gustatory afferents as well as the other visceral sensory fibers mentioned in Chapters 12 and 23. However, the gustatory fibers and chemosensitive trigeminal fibers end separately in lateral and rostral portions of the solitary nucleus (Fig. 13-4).

Second-order taste fibers do two things (Fig. 13-6). Some participate in reflex activities, such as salivation, swallowing,^* and coughing, by way of cranial nerve motor nuclei. Others, like fibers in most sensory systems, project to the cerebral cortex by way of the thalamus. In this case, however, the projection is uncrossed. Fibers travel through the ipsilateral central tegmental tract to the most medial part of the ventral posteromedial (VPM) nucleus, where they end adjacent to the uncrossed fibers of the dorsal trigeminal tract (see Fig. 12-16). This medial part of VPM then projects to gustatory cortex, which is located in the insula and the medial surface of the frontal operculum, near the base of the central sulcus. Gustatory cortex projects in turn to orbital cortex of the frontal lobe (see Fig. 13-21), where taste information is integrated with olfactory and other information, and to the amygdala, through which taste information reaches the hypothalamus and the limbic system. (In most mammals, taste information reaches the hypothalamus and amygdala more directly, through a projection from the parabrachial nuclei^* of the pontine reticular formation, which also distribute nociceptive information and information from visceral structures. The gustatory portion of this set of functions appears to have been lost in primates.)

Figure 13-6 Taste pathways in the CNS. Second-order neurons feed into reflexes both by direct projections (e.g., to the nearby dorsal motor nucleus of the vagus [DMN X]) and by connections with the reticular formation (RF). The projection from the parabrachial nucleus to the hypothalamus (H) and amygdala (A) is dashed because, although it is present in most mammals, its existence in primates is doubtful.

Information About Taste Is Coded, in Part, by the Pattern of Activity in Populations of Neurons

Sensory systems such as somatic sensation and vision depend on the receipt of information about specific aspects of a stimulus, highly organized in space and time (see Figs. 3-30, 10-20, and 17-35). Activity in many of the neurons in these systems signals the occurrence of a particular kind of event—particularly its location, as is evident in the detailed topographic maps in places such as somatosensory, visual, and motor cortex (see Figs. 3-30 and 17-29). In contrast, the nervous system also uses combinations of inputs to interpret stimuli, as in the case of Ruffini endings that are involved in the sense of touch but produce no sensation when stimulated in isolation (see Chapter 9). Taste and olfaction also utilize this combinatorial type of processing. Taste stimuli are widely dispersed in the mouth during chewing, and odorants fill the nose during breathing, so spatial localization is less important and we actually use the touching of the tongue by a piece of food to localize it (see Box 9-2). Instead, the gustatory and olfactory systems together are faced with the task of coding the identities of many thousands of different chemicals, both singly and in mixtures. In principle, this could be accomplished by having many thousands of different receptor types, each particularly sensitive to one chemical (“labeled-line” coding). An alternative strategy is to code the identity of a stimulus in terms of the pattern of activity in a large population of neurons with multiple sensitivities (“across-fiber” coding). Although this might seem like a cumbersome way to do things, it is actually an efficient system for encoding information about a large variety of potential stimuli using a relatively small number of neurons, similar in principle to using just three kinds of cones as the basis for perceiving hundreds of hues (see Chapter 17).

The gustatory system uses a combination of labeled-line and across-fiber coding; the relative importance of the two strategies is still debated. Taste receptor cells apparently contain in their apical membranes only one of the transduction mechanisms shown in Figure 13-5, but at every subsequent level of the gustatory system individual cells respond to more than one kind of tastant. Gustatory nerve fibers branch and innervate multiple taste buds that may even be located in multiple papillae, and ATP released by an assortment of taste receptor cells reaches each afferent ending. As a result, each responds to more than one tastant, although one kind of sensitivity usually predominates for any given nerve fiber. This “tuning” corresponds to the relative sensitivities of different areas of the tongue: most facial nerve (chorda tympani) fibers are maximally sensitive to sweet or salty stimuli, and most glossopharyngeal fibers to sour or bitter stimuli. Second-order neurons in the nucleus of the solitary tract are more broadly tuned (Fig. 13-7A), although there is still some selectivity. By the time gustatory information reaches higher cortical levels, specific features of a taste stimulus have been extracted, and neurons with particular chemical sensitivities are found (Fig. 13-7B).

Figure 13-7 Responses of single gustatory neurons in the nucleus of the solitary tract (A) and the orbital cortex (B) of monkeys to tastants applied at the time indicated by the vertical dashed lines. The brainstem neuron responds to multiple tastants (but not all of them), whereas the cortical neuron is more selective.

(A, from Scott TR et al: J Neurophysiol 55:182, 1986. B, from Rolls ET, Yaxley S, Sienkiewicz ZJ: J Neurophysiol 64:1055, 1990.)