The gustatory and olfactory systems during infancy: Implications for development of feeding behaviors in the high risk neonate

Synopsis

This chapter reviews the development of the senses of taste and smell, which provide information on the flavor of foods, and discusses how innate predispositions interact with early-life feeding experiences to form children’s dietary preferences and habits. A basic understanding of the development and functioning of the chemical senses during early childhood may assist in forming evidence-based strategies to improve children’s diets, especially for those who experience a discontinuity or disruption in early flavor experiences.

Introduction

Our senses of taste and smell are intimately connected to nutrition and allow us to reject those foods that are harmful and to seek out those that are beneficial and pleasurable[1]. During the past several decades, researchers have begun to unravel some of the mysteries underlying the ontogeny of the function of these senses as well as the roles they play in food choice, health, and social interactions.

Building upon the scientific definition of flavor and the basic biology of taste and smell, we summarize insights gleaned from basic scientific research in the chemical senses, with a focus on the sensory capabilities of the human infant and the inherent contributions of genetic differences in taste perception and the plasticity of the chemical senses in the development of flavor and food preferences. We highlight differences between normal and high risk neonates with regard to early sensory experiences and its potential impact on learning and later feeding.

Definition of Flavor

Flavor, a powerful determinant of human ingestion throughout the life span, is a product of several sensory systems, most notably those of the chemical senses, taste and smell. The perceptions arising from these two senses are often confused and misappropriated[2]. Sensations such as garlic, chocolate, anise, and lemon are erroneously attributed to the taste system per se, when in fact only a small number of primary taste qualities can be perceived by the tongue: sweet, salty, bitter, sour, and savory. Smell sensations, on the other hand, encompass thousands of diverse qualities, including the flavors noted above. As illustrated in Figure 1, the receptors for the olfactory system, located high in the nasal chambers, are stimulated not only during inhalation (orthonasal route) but also when infants suck and when children and adults swallow, as chemical constituents in foods and beverages reach the nasal receptors by passing from the oral cavity through the nasal pharynx (retronasal route). It is this retronasal stimulation arising from the molecules of foodstuffs that leads to the predominant flavor sensations.

Figure 1.

Orthonasal (green arrows) and retronasal (purple arrow) routes of olfaction.

Basic Flavor Biology

Taste occurs when chemicals come into contact with taste receptors on the tongue, palate, throat, epiglottis, or esophagus that then send signals to the brain. Taste receptor cells are the interface between the oral environment and the nervous system (reviewed in ref[3]). These cells, arranged in groups of 50 to 100 to form taste buds, contain the proteins necessary to recognize each of the five types of taste: sweet, salty, sour, bitter, and savory. Salty and sour foods are recognized by ion channels[4]. Salty taste is most commonly imparted by sodium ions in sodium chloride, but other sodium and nonsodium salts also convey a salty characteristic. Sour taste is generated by protons in acids. Sweet, bitter, and savory tastes are translated to the brain via G-protein-coupled receptors (GPCRs): type I GPCRs (T1R1, T1R2, and T1R3) are stimulated by sweet (T1R2+T1R3) and savory (T1R1+T1R3) compounds[5,6], while bitter compounds are recognized by type II GPCRs (T2Rs)[7,8]. T2Rs recognize a variety of unpleasant-tasting compounds and may have evolved as a warning to avoid toxins[9,10].

Odors are recognized by olfactory receptors, which are located on a small patch of tissue in the nasal cavity. Olfactory receptors are GPCRs that are generated by the largest mammalian gene superfamily, with more than 400 functional genes[11,12]. The olfactory system becomes tuned to respond to stimuli in different ways based on the experience of the individual and the context in which odors are experienced[13,14]. Olfactory signals combine with taste signals to communicate flavor to the brain[15,16].

Genetics of Taste

Polymorphisms in the genes that encode taste and odorant receptors result in differential sensory patterns in humans, by altering amino acid sequences of receptors, which alters their function, or by altering gene expression[17-28]. Although these mutations are found in a variety of receptor genes, few examples have been well characterized in the literature.

Polymorphisms in the bitter taste receptor gene TAS2R38 are the most studied of all taste receptor variants. Genetic variation in this receptor translates into individual differences in taste sensitivity for the synthetic compounds phenylthiocarbamide and propylthiouracil (PROP), as well as bitter-tasting compounds commonly found in cruciferous vegetables[29]. The polymorphisms result in changes to the amino acid sequence of the receptor from alanine-valineisoleucine (AVI) in nontasters to proline-alanine-valine (PAV) in tasters[17,30,31]. These polymorphisms allow homozygous AVI people to enjoy broccoli or turnips without perceiving the bitterness that heterozygous AVI/PAV and especially homozygous PAV people taste[32]. Studies in children and their mothers indicate that the phenotype-genotype relationship for PROP sensitivity varies with age, such that AVI/PAV heterozygous children are more sensitive to PROP than are heterozygous adults, with adolescents being intermediate[25,33]. These results imply that within the same genotype, taste sensitivity can change over the life span (from more to less bitter sensitivity).

A commonly cited example of individual variation in human olfaction is the perception of androstenone, a volatile steroid found in human perspiration, boar saliva, some pork products, truffles, and celery[34]). While some individuals describe this volatile as “sweaty and urinous,” others perceive it as smelling “sweet and floral,” or odorless[35-37]. The odorant receptor OR7D4 is activated by androstenone, and recently two polymorphisms were identified within the gene that change the amino acid sequence and impair the function of the receptor[21].

Individuals with the arginine-threonine variant smell androstenone, and those with the tryptophan-methionine variant find it to be odorless. Similar to bitterness sensitivity, the ability to detect androstenone seems to change with age[38-40].

Extraoral Taste and Nutrient Sensing

Although not much is known about their relative function, taste receptors have been found in many extraoral tissues, including the lungs, brain, gut, and reproductive system[41-44]. Sweet and bitter receptors are both found in the gut but have different functions. Sweet receptors regulate local glucose transporters to enhance glucose uptake[45], whereas one function of the bitter taste receptors is to regulate the absorption of toxic secondary plant compounds or other poisons[46]. Bitter receptors are also found in the upper and lower airways in mammals[47-50], and they are probably also present in humans. Their function in the airway is not known, but one possibility is that they sense bitter molecules secreted by bacteria and may evoke immune or other responses to clear the airway of pathogens[51]. The developmental trajectory of extraoral bitter and sweet receptors in gut, airway, and other tissues is not known, either in humans or in other species.

Pre- and Postnatal Development of Flavor

Both olfactory and taste receptors must be functional in order for a human fetus or infant to sense flavor. The primary olfactory receptors are formed by the 8th week of gestation (see [52] for a review) and are functional as early as the 24th week[53,54]. Taste cells also begin to form at 7 to 8 weeks of gestation[55,56]; by 13 to 15 weeks they look like mature receptor cells, and by around 17 weeks they are considered functionally mature. Fetal swallowing begins at approximately 12 weeks of gestation[57,58]. Around 18 weeks, gestational nonnutritive suckling begins, and the sucking and swallowing actions are coordinated by 35 to 40 weeks of gestation. Near the end of gestation the fetus swallows significant amounts of amniotic fluid. After 6 months of gestation, when the epithelial plugs no longer obstruct the air passages, amniotic fluid is also inhaled. The inhalation and swallowing of amniotic fluid are the first chemosensory experiences of the fetus and mark the beginning of flavor learning.

Amniotic fluid, the first food of infants, contains a wide range of nutrients that have particular tastes, such as glucose, fructose, lactic acid, fatty acids, and amino acids[59], as well as the flavors (for which the odors are perceived retronasally) of the foods consumed by the mother[60,61]. The fetus can detect these tastes and flavors: fetal swallowing frequency increases in response to the introduction of sweet solutions into the amniotic fluid and decreases in response to the introduction of bitter solutions[59,62]—this may be one of the first indications that our basic biology favors consumption of sweet tastes and avoidance of bitter tastes.

A similar response pattern is seen shortly after birth—within hours and days of being born, young infants react as would be expected to pleasurable and aversive taste stimuli[63-72]: provision of sweet or savory solutions to neonates elicits rhythmic tongue protrusions, lip smacks, lip and finger sucking, and elevation of the corners of the mouth, all of which have been interpreted as a positive or hedonic response[71,72]. In contrast, neonates gape, wrinkle their noses, shake their heads, flail their arms, and frown in response to a bitter solution[63,72]. Concentrated sour solutions elicit lip pursing and, to a certain extent, gaping, nose wrinkling, and arm flailing, as well as tongue protrusions and lip smacking[63,72,73]. Unlike the other basic tastes, salt taste receives a neutral reaction from neonates—the taste for salt does not emerge until later in infancy and then remains throughout childhood and adolescence[74].

These specific affective reactions to differing taste stimuli are strikingly similar across cultures[68,73,75] and species[72,76-79], suggesting a basic biological underpinning for the flavors and foods youngsters prefer and avoid. The convergence of research findings supports the conclusion that the innate preference for sweets and rejection of bitter tastes in humans are consequences of selection, favoring consumption of high-energy, vitamin-rich fruit and vegetable diets and avoidance of bitter, poisonous fruits and plants. Thus, when we examine children’s dietary patterns from the perspective of the development of taste, the foods children naturally prefer (e.g., sweet snacks) and those they dislike (e.g., bitter-tasting green vegetables) are not surprising and reflect their basic biology.

In addition to containing chemicals with distinct taste properties, amniotic fluid contains volatile chemicals (flavors) transmitted from the maternal diet[60,61,80], which, by at least the second trimester, appear to be detected by the fetus. Shortly after birth, infants will respond differently to flavors experienced in amniotic fluid, indicating that memories are formed from these early sensory experiences. For example, neonates whose mothers consumed an anise-flavored beverage or ate garlic-containing foods throughout pregnancy were more accepting of and interested in (as measured by mouthing and orienting) anise and garlic odors[80,81]. Similar findings were observed with alcohol odors[82].

Learning about the dietary choices of the mother continues when infants experience the flavors of the mother’s diet transmitted in breast milk. Young mammals first learn about the dietary choices of their mothers through transmitted flavor cues, a type of learning documented in a wide variety of species (see[83] for review). Following from this, researchers determined that many flavors (e.g., anise, garlic, ethanol, carrot, mint, vanilla, bleu cheese) pass from mother to offspring through breast milk[60,61,83-86]. Human infants detect the flavors in mother’s milk, as evidenced by changes in their suckling rate, patterning and duration of feeding and intake[60,85,86], and differential acceptance of similarly flavored foods at weaning and beyond[60,87-89]. Similarly, breastfed infants were more accepting of fruits and vegetables than were formula-fed infants, but only if their mothers regularly ate these foods themselves[87].

That these early flavor experiences can influence the acceptance of foods was first demonstrated in a randomized, controlled study of mothers who consumed carrot juice for several days each week during the last trimester of pregnancy or for a similar period during the first 3 months of lactation[88]. The control group drank water and avoided carrots and carrot juice during both pregnancy and lactation. When mothers weaned their infants around 6 months of age, the children were tested for acceptance of plain cereal on one day and carrot-flavored cereal on another. Infants who experienced the flavor of carrots in either amniotic fluid or mother’s milk responded more favorably (e.g., ate more, made fewer faces of distaste) to carrot-flavored cereal than did nonexposed control infants. Thus, as with many other mammals, human infants’ pre- and postnatal experiences with food flavors transmitted from the mother’s diet lead to greater acceptance and enjoyment of these foods during weaning.

Sensitive Period for Flavor Learning

Although the types of flavors that breastfed infants experience before their first taste of solid foods reflect the culinary practices of their mothers, which varies from infant to infant[60,87], formula-fed infants are usually exposed to constant flavors after birth and prior to weaning, since most formula-fed infants experience a single type of formula[90]. The absence of a robust experimental paradigm, like that employed for other sensory systems (e.g., vision, audition/language) and other animals, has inhibited progress in understanding whether human flavor programming exhibits age-related changes in functional plasticity, commonly referred to as sensitive periods. To address this gap, a model system was used that exploits the naturally occurring flavor variation in infant formulas[91,92].

In the United States, formulas are available for healthy term infants and for special medical purposes (such as preterm infants or infants with inborn errors of metabolism). Among the formulas for healthy term infants, one of the main distinctions is their protein source and/or degree of protein hydrolysis. Cow milk formula (CMF) is the most common formula consumed by infants, accounting for 76% of all U.S. infant formula sales in 2000[93]. Its protein usually includes combinations of intact casein and whey proteins[94,95]. Extensive protein hydrolysate formula (ePHF), a type of formula typically fed to infants who have cow milk protein allergy or intolerance to intact protein, is less prevalent in use than is CMF[93]. The milk proteins (i.e., whey, casein) in ePHF are treated with enzymes to break down the protein structure to reduce allergenicity; these formulas contain low-molecular-weight peptides and free amino acids. Partial protein hydrolysate formulas (pPHF) contain whey or casein milk proteins that are enzymatically treated but to a lesser extent than for ePHF. The varying composition and degree of hydrolysis among hydrolysate formulas affect formula flavor profiles[96]. To adults who were not fed ePHF during infancy, it is extremely unpalatable compared with CMF because of PHF’s distinctive, unpleasant flavors, including both volatile (odors) and nonvolatile (e.g., bitter and sour tastes) components.

Using the flavor differences between CMF and ePHF as a model system, a “window” of acceptance was identified during which young infants readily accept ePHF. Beginning around 4 months of age and continuing through adulthood, its flavor is rejected unless the individual was exposed to ePHF earlier in life (consistent with anecdotal pediatrician reports that it is difficult to begin feeding ePHF to infants 4 or more months of age). Thus, depending on an individual’s exposure to its flavor during the first few months of life, ePHF acquires a completely different “hedonic tone,” or perceived pleasantness[91].

A randomized trial was conducted to begin to characterize the effects of the timing and duration of early-life exposure when hedonic responses to PHF flavors are established. Infants were randomized to be fed ePHF for 1 month beginning at 1.5 months, 2.5 months, or 3.5 months or for 3 months beginning at 1.5 months[92]. All groups were then compared with control groups that had either no ePHF exposure or 7 months of ePHF exposure. At 7.5 months, infant acceptance of ePHF was tested with complete “meals” of both formulas. Among infants who began feeding ePHF at 1.5 months, those fed for 1 month were as accepting of ePHF as those fed for 3 months. That is, flavor experience of a relatively brief occurrence (1 month) before the baby is 3.5 months of age is sufficient to maintain acceptance. However, infants fed ePHF for 1 month were less accepting than infants exposed to ePHF for the entire 7 months. Early exposure is also important: infants exposed to ePHF for 1 month starting at 3.5 months were less accepting of ePHF at 7.5 months of age than were infants exposed at an earlier age. Maternal perceptions of infants’ enjoyment of the formulas and the frequency of facial expressions of distaste were consistent with both the exposure- and timing-related differences in intake. Early exposure eliminated the age-related rejection seen in unexposed infants and resulted in a complete shift in hedonic tone[92].

The effects of early exposure to ePHF are persistent, leading to heightened preferences for the taste and aroma of ePHF and foods containing similar volatiles or tastes (e.g., bitter, sour, savory) at weaning and several years after children’s last exposure to the formula. Children fed ePHF had an increased preference for sour-flavored apple juice[75,97] and savory-, bitter-, and sour-tasting and plain cereals compared with other children[98]. The mothers of these children were also more likely to list broccoli as one of their child’s preferred vegetables than were mothers of infants fed CMF[97].

Why should there be a sensitive period in the early acceptance of the flavor of ePHF? First, presuming there is an adaptive reason, it clearly has nothing directly to do with hydrolyzed protein formulas, which were introduced only a half-century ago. Indeed, these observations with formulas may conveniently expose a much more fundamental aspect of early mammalian flavor learning. We hypothesize that it is important for the human infant to accept and be particularly (but not exclusively) attracted to the flavors that are consumed by the culture and, more specifically, by the mother. All else being equal, these are the flavors that are associated with nutritious foods or, at the very least, foods the mother has access to—and the foods and flavors that the infant will experience at weaning and probably thereafter. Under this hypothesis, much of the normal exposure would occur in utero and during breastfeeding, as flavors mothers consume are transferred to these chemosensory environments. Additional research is needed to determine the extent to which early exposure (and the lack of early exposure) to these flavors, perhaps during sensitive periods of development, helps establish enduring preferences for foods and flavors.

Challenges for the High Risk Neonate

The first few months of life are an essential part of the flavor learning process for humans, and during this period the sensory experiences of the high risk neonate are drastically different from those of a typical infant, lacking continuity with prenatal sensory experiences. Preterm infants are often unable to coordinate sucking, swallowing, and breathing, so nasogastric or orogastric tube feeding is used to provide adequate nutrition[99]. When fed by a tube, infants likely have a relatively constrained olfactory and flavor experience in the context of feeding because their nutrition bypasses the oral and nasal cavities. Even those who are tube fed human milk may not have the opportunity to experience retronasally the flavors present in milk. Furthermore, it is unknown how the body responds when nutrients are sensed in the gut but have not been sensed in the oral cavity.

Tube-fed infants increase nonnutritive sucking when exposed to the smell of mother’s milk through an infant olfactometer, suggesting that exposure to maternal nutrient odor may assist in transition to oral feeding[100]. However, in the neonatal intensive care unit and normal infant wards in hospitals, infants are also exposed to (and learn about) unpleasant or noxious odors, including disinfectants, antibacterial compounds, and cleaning solutions[101]. The long-term consequences of this altered sensory environment remain unknown.

Tube feeding is generally done using either preterm formula or fortified human milk[102-104]. Greater efforts are being made to increase the amount of human milk given to preterm infants because human milk provides many benefits such as improved immune status and increased cognitive development, while successful expression of breast milk allows for more maternal involvement in feeding and increases maternal confidence[105-119]. We hypothesize that, given the presence of functional taste receptors in both the oral cavity and the gut[45], increasing intake of human milk is beneficial for future feeding behavior because the extraoral stimulation of milk on gut taste receptors may aid in the transition from tube feeding to breastfeeding. Encouraging the mother to pump breast milk also increases the likelihood that the child will eventually be able to transition to the breast and experience the flavors in mother’s milk within the sensitive window for flavor learning[106].

In addition to altered oral sensory exposure, infants who are tube fed do not have early experience with traditional feeding behaviors (sucking, swallowing, and chewing). However, we emphasize that there has been a paucity of experimental research in this area on how such altered sensory experiences affect later behaviors associated with feeding. Several case studies from the 1960s revealed that if children were not introduced to solid foods at the time when they are first able to chew, acceptance of these foods became very difficult[120]. In three of the cases, the children had esophageal atresia and were tube fed beginning days after birth. Repair of the esophagus did not happen until 16 to 22 months later, at which point two of the three patients readily accepted fluids, but all three had significant difficulty transitioning to solid food. Long-term tube feeding may affect the physical development of feeding behaviors, with consequences lasting into childhood.

When tube feeding occurs for a short period of time (15-20 days), in combination with nonnutritive sucking, infants generally transition well to oral feeding[121-124]. However, if the tube feeding lasts for a longer period of time (>45 days), it becomes much more difficult for the child to make the transition. A study of 9 infants who were tube fed for at least 2 months starting from birth, the infants refused all attempts at oral feeding and reacted with agitation, arching, tongue thrusting, gagging, and vomiting[124]. The infants also had an absent or deficient sucking reflex and a gag reflex that was triggered by any foreign object. To help wean the children, during tube feedings at regular intervals the infants were provided with stimulation to reproduce normal feeding as closely as possible. They were cradled in their mother’s arms, the gums and palate were massaged, and the tongue was stimulated with breast milk from the mother’s finger to stimulate the sucking reflex. Eight of the children eventually weaned from tube to oral feeding, with those who were tube fed the longest requiring the most time to establish normal eating behavior. The authors of the study hypothesize that stimulation during tube feeding helps inhibit the gag reflex, creates an association among tactile, olfactory, and taste sensations and the mechanical replenishing of the stomach, and establishes normal circadian rhythms[124].

High risk infants are also faced with a wide array of medical conditions that contribute to temporary or permanent alteration of taste and smell as adults. Many medications, including antibiotics and anti-inflammatory agents, have been shown to alter taste and smell, and these are commonly given to high risk neonates[125-127]. Feeding problems can lead to vitamin deficiencies, which has also been linked to altered taste and smell[128]. Gastroesophageal reflux disease is another common problem in preterm infants and results in a sour or bitter taste in the mouth from reflux of stomach acid up the esophagus and into the throat[129]. The long-term effects of these alterations on the development of flavor preferences in the child are not known.

Concluding Remarks

Every culture differs in the flavor principles that characterize its cuisine and the types of foods preferred by the families who identify with its traditions. Thus, cultural traditions guide the types of food individuals eat on a daily basis. Although many would argue that learning about these flavor principles, food preferences, and cultural traditions begins when parents serve their children cultural meals during family dinners, research demonstrates that this learning begins long before a child ever consumes solid food. Flavors of the mother’s diet are transmitted to the offspring through the amniotic fluid and breast milk, and infants more readily accept flavors they’ve already experienced through these two mediums when fed as solid foods at weaning. The recent discoveries of taste receptors in tissues outside of the oral cavity only add to the complexity of this system. Infants may also be sensing bitter stimuli in the airways or sweet stimuli in the gut, and the development of these sensory systems is not yet understood.

Because the senses of taste and smell are the major determinants of whether young children will accept a food (e.g., they eat only what they like), these senses take on greater significance in understanding the biological basis for children’s food choices. Not being exposed to the flavors of healthy foods early in life can have detrimental consequences. While there are innate responses to the basic tastes, and some individuals may be more sensitive to some tastes due to genotype, the development of these chemical senses has inherent plasticity that interacts with early-life experiences to shape and modify flavor and food preferences. Such functional plasticity, one of the main characteristics of the brain, highlights the ability to change behavior based on experience. In other words, our biology is not necessarily our destiny.

Although we are beginning to learn how the chemical senses develop during infancy and their impact on food choices and other behaviors, there are many gaps in our knowledge. In particular, we know little about the contingencies for early learning and how the absence of early postnatal chemosensory experience (e.g., absence of breastfeeding), disruptions in mother-infant attachment (e.g., tube feeding of high risk infants), or negative associations with early feeding (e.g., chemical smells in hospital settings) interfere with the acquisition of feeding skills. The increasing awareness of the importance of infant feeding behavior makes it imperative to determine the extent to which restoration of normal oral motor and sensory experiences affect feeding behavior and nutrition.

Clearly, more research is needed to develop evidence-based practices aimed at infant feeding difficulties, which constitute a medically and economically important complication for some neonatal diseases. Applying the knowledge gleaned from such research and clinical practice, which takes into account the developing sensory world of the child, could have long-term consequences in preventing eating disorders in early infancy. Moreover, understanding the development and functioning of these senses may assist in the development of evidence-based strategies to improve children’s diets, since many of the illnesses that plague modern society (e.g., obesity, diabetes, and hypertension) are often the consequence of poor food choices that start in childhood.