problems. Visual system anomalies similar to those that result in myopia have been obtained in animal models (Sulik and Johnston 1983), and various investigators have recently examined the pathogenesis of these ophthalmologic abnormalities (Phillips 1989; Phillips and Krueger 1990). The results suggest that an increase in retinal cell loss, along with impaired cell replication and hypomyelination of optic nerve axons, may contribute to the pathogenesis of ocular abnormalities seen in FAS. A higher incidence of hearing impairment in children with FAS and FAE has been reported (Church and Gerkin 1988). Evidence of sensorineural hearing loss in rats prenatally exposed to alcohol has been obtained from studies demonstrating abnormal electrical activity in the brain stem following presentation of highfrequency sounds and disruptions in auditory processing at the level of the cortex, as measured by the method of cortical auditory-evoked potentials (Church 1987). These results have important clinical implications, because good hearing is essential for normal speech and language development in children. Problems with balance and gait, as well as deficits in both gross and fine motor function observed in the clinical setting (Barr et al. 1990; Conry 1990; Marcus 1987; Streissguth et al. 1990), have been demonstrated in animal model systems. All of these effects resemble the motor dysfunctions that occur following disruption of cerebellar maturation, and alcohol exposure has been shown to have a deleterious effect on the developing cerebellum (e.g., West, Goodlett, Bonthius et al. 1990). Most recently, altered cerebellar morphology was identified in the same animals that exhibited behavioral motor defects (Goodlett et al. 1991; Meyer et al. 19906). Neonatal and Regulatory Behavioral Effects Fetal Movement Researchers conducting animal studies are beginning to examine how fetuses respond to the presence of alcohol in the womb. In one study, the effects of alcohol on spontaneous forelimb movements were examined in near-term fetal sheep (Leader et al. 1990). Maternal intravenous alcohol infusion that resulted in the fetal sheep's attaining blood alcohol levels of about 85 mg/dL decreased spontaneous forelimb movements. In addition, treatment of pregnant ewes with alcohol resulted in a suppression of breathing movements and brain activity in near-term fetal sheep (Smith, Brien, Carmichael et al. 1989; Smith, Brien, Homan et al. 1989). These results, along with those from earlier rodent studies (e.g., Smotherman and Robinson 1987), provide the most direct evidence that maternal ingestion of alcohol influences fetal behavior in utero. They also are consistent with a study indicating altered human fetal breathing movements monitored indirectly in mothers who had just consumed alcohol (McLeod et al. 1983). Thus, altered fetal activity as a result of in utero alcohol exposure may be a contributing factor in the expression of FAS. Human newborns prenatally exposed to alcohol may exhibit a weak sucking response and irregular sucking patterns early in life. These infants are typically described as easily distracted and fatigued when suckling. Feeding Behavior Human newborns prenatally exposed to alcohol may exhibit a weak sucking response and irregular sucking patterns early in life. These infants are typically described as easily distracted and fatigued when suckling (e.g., Van Dyke et al. 1982). Similarly, feeding abnormalities have been found in rodents prenatally exposed to alcohol. In comparison with controls, rat pups exposed to alcohol in utero took longer to attach to the nipple of a test dam, spent less time suckling, and displayed weaker suckling pressure and an altered pattern of suckling (Rockwood and Riley 1990). It has recently been hypothesized that growth retardation in rodents prenatally exposed to alcohol may be related to disturbances in feeding behavior that result from alcohol-induced alterations in developing neurochemical systems that are involved in regulating feeding (Druse et al. 1990; Middaugh and Boggan 1991; Stricker and Verbalis 1990). These human and animal data suggest that deficient feeding behavior results from in utero alcohol exposure and that such deficient behavior may contribute significantly to the postnatal growth retardation commonly seen in FAS and FAE. Regulatory Behavior Human neonates born to mothers who consumed alcohol during pregnancy have been described as having poor regulation of normal physiologic states. Sleep state disturbances, tremulousness, and jitteriness have been reported in infants exposed prenatally to alcohol (e.g., Scher et al. 1988; Streissguth 1986). State regulation has been studied in animal models as well. For example, several studies have demonstrated that prenatal alcohol exposure impairs the ability of rat neonates to maintain their body temperature (e.g., Zimmerberg 1989). Long-Term Behavioral One of the advantages to using animals in prenatal alcohol research is that the long-term effects of prenatal alcohol exposure can be studied in a relatively shorter period of time than would be possible in humans. Hence, animal research has addressed the persistence of behavioral and cognitive effects that follow fetal alcohol exposure. Many of the behavioral teratogenic effects of alcohol have been reported to be transient. That is, the effects appear to diminish with age, thus suggesting that prenatal alcohol exposure produces a developmental delay in the maturation of the CNS (Abel 1982). Alternatively, or in addition to a simple delay in development, some of the "normalization" that comes with age may be due to the development of independent compensatory strategies designed to cope with and overcome the disabilities. Recent data showing the reemergence of some of these deficits in older animals suggest that the compensatory systems decay with age or that the recruitment of such compensatory mechanisms compromises the ability to perform other more complex functions later in life (Riley 1990). For example, under particular testing conditions, behavioral abnormalities such as hyperactivity and impaired performance in learning and memory tasks have been observed in adult animals exposed prenatally to alcohol (e.g., Becker et al. 1989; Gianoulakis 1990; Middaugh and Ayers 1988; Reyes et al. 1989; Zimmerberg et al. 1989, 1991). In addition, behavioral deficits that are not evident under normal, unchallenged conditions may be "unmasked" when the complexity of the testing situation increases or when testing is conducted under stress and drug-challenged conditions (Riley 1990). Thus, animal research has revealed that many of the behavioral dysfunctions observed in young offspring are permanent and persist into adulthood. These findings complement those of clinical studies that have revealed persistent attentional, cognitive, and neurobehavioral deficits in adolescents and adults with FAS or FAE (e.g., Streissguth et al. 1991). Taken together, basic and clinical studies have established that prenatal alcohol exposure can potentially produce long-lasting cognitive and behavioral disabilities. Neuroanatomical Effects Adverse effects of alcohol on brain growth and development have been documented in a number of regions spanning all levels of the CNS, but the mechanisms underlying these effects remain unclear. This inability to identify precise mechanisms may be due, in part, to the various effects of alcohol on brain growth and differentiation, as well as to the complexity of the organ itself and the dynamic nature of its development. The brain is unique among organs; it is one of the first to begin to develop and the last to be completed (West 1987). For example, in the human embryo the major brain regions are recognizable at 5 weeks' gestation. Brain cells undergo 2 periods of rapid growth: at 15 to 20 weeks' gestation and again at 25 weeks' gestation to a year after birth. In addition, brain cells typically are generated in zones but migrate to their ultimate locations during brain development. Thus, the brain appears to be sensitive to the teratogenic actions of alcohol throughout its development, a period that encompasses all three trimesters (West and Goodlett 1990). In addition, although all mammals pass through the same stages of brain development, the timing of these stages relative to birth varies considerably among species (West 1987). For example, the brain is less mature at birth in rats than it is in humans. Thus, with regard to brain development, the full gestation period in the rat is equivalent to the first two trimesters in humans, and the first 10 days of postnatal life in the rat may be considered equivalent to the third trimester in humans. The brain is particularly sensitive to the adverse effects of alcohol during the period of time when it is undergoing rapid growth (the brain growth spurt). This phenomenon has been studied in an artificial-rearing procedure developed to administer alcohol to rat neonates (West et al. 1984). The model allows for external control over the dose of alcohol administered: It involves infusing a milk formula containing alcohol directly into the neonates stomach. Such neonatal alcohol exposure in rats (during the brain growth spurt) results in reduced brain growth (Bonthius and West 1988; Goodlett et al. 1990), brain cell loss (West, Goodlett, Bonthius, et al. 1990), alterations in brain microvasculature (Kelly et al. 1990), and behavioral abnormalities (Barron and Riley 1990; Kelly et al. 1988; Meyer et al. 1990a, 19906). Furthermore, several studies have demonstrated that brain growth and development are vulnerable to the teratogenic actions of both acute and chronic alcohol exposure (Goodlett et al. 1990; Kotkoskie and Norton 1988, 1990). Prenatal alcohol exposure alters the generation, proliferation, and migration of cerebral cortical neurons in rats (e.g., Miller 1988, 1989; Miller and Nowakowski 1991; Miller and Potempa 1990) and results in an abnormal pattern of distribution and organization of these neurons in the cortex (Miller 1988; Miller and Potempa 1990). Miller et al. (1990) also found that the projections and density of corticospinal neurons were abnormal in adult rats exposed prenatally to alcohol. Alcohol treatment also results in neuronal cell death among more mature populations of brain cells. For example, alcohol exposure restricted to the early postnatal days in rats reduces the number of neurons in the hippocampus (West and Pierce 1986) and cerebellum (Pierce et al. 1989; West, Goodlett, Bonthius et al. 1990). Thus, alcohol exposure may be damaging to the developing brain because it results in cell death, because it affects neuronal generation and migration, or for both reasons. Further, various brain regions have been shown to be differentially sensitive to alcoholinduced damage (Bonthius and West 1990; West, Goodlett, Bonthius et al. 1990). Alcohol also inhibits the action of nerve growth factor, which is essential for the survival of some nerve cells (Dow and Riopelle 1985). The implication of these findings is that different regions of the developing brain may have different thresholds and perhaps different critical periods of susceptibility to alcohol injury. These differences further complicate the question of how much alcohol exposure is harmful to the developing brain. The answer appears to be that it depends not only on the dose, pattern, and timing of the alcohol exposure, but on the brain region as well (West and Goodlett 1990). Finally, a great deal of attention has been focused on examining how alcohol-induced brain injury corresponds to behavioral dysfunctions. For example, Riley (1990; Riley and Barron 1989) has suggested that many of the behavioral abnormalities that result from alcohol exposure during development are remarkably similar to effects observed following damage to the hippocampal region of the brain, such as impaired learning (West, Goodlett, Bonthius, et al. 1990; Wigal and Amsel 1990). The hippocampus is important in the process of leaming and memory consolidation. Similarly, motor deficits have been linked to structural defects in the cerebellum. Adult animals that were exposed to alcohol during the neonatal brain growth spurt and that have exhibited deficits in performance on several tasks requiring balance and motor coordination have also exhibited retarded cerebellar growth (Goodlett et al. 1991; Meyer et al. 1990b). New associations between structure and function may be uncovered by research applying more sophisticated neuropsychological procedures with children exposed prenatally to alcohol (West, Goodlett, and Brandt 1990). In addition to morphologic alterations, alcohol exposure during CNS development produces alterations in brain microcircuitry, including neurochemical aberrations, as well as disturbances in the integrity of the synapse (the junction between two neurons). Neurochemical Effects In addition to morphologic alterations, alcohol exposure during CNS development produces alterations in brain microcircuitry, including neurochemical aberrations, as well as disturbances in the integrity of the synapse (the junction between two neurons). For example, neurochemical and electrophysiological abnormalities were identified in hippocampal tissue of 45-day-old rats that had been exposed to relatively low levels of alcohol in utero (maternal blood alcohol levels were in the range of 30 to 40 mg/dL) (Farr et al. 1988; Savage et al. 1989). Further, a decreased capacity to generate long-term potentiation of electric impulses, as measured electrophysiologically in hippocampal slices, was demonstrated in these animals (Swartzwelder et al. 1988). Long-term potentiation is thought to represent the major biological process underlying learning and memory consolidation. In another recent electrophysiological study of adult rats exposed prenatally to alcohol, the results indicated that prenatal alcohol exposure can result in long-term subtle abnormalities in hippocampal function (Tan et al. 1990). In recent years, neurochemistry studies have focused on the monoamines (dopamine and serotonin), although effects on other systems, such as acetylcholine (Brodie and Vernadakis 1990; Kelly et al. 1989; Light et al. 1989a, 1989b, Okonmah et al. 1989), y-aminobutyric acid (GABA) (Zhulin and Zabludovskii 1989; Zimmerman et al. 1990), and glutamate (Farr et al. 1988; Morrisett et al. 1989; Noble and Richie 1989), have been reported. Prenatal exposure to alcohol markedly affects the development of the dopamine system. A 30- to 40-percent reduction in striatal and cortical dopamine levels, along with a reduction in dopamine uptake sites and a reduction in the number of some receptor sites, was found in rats exposed to alcohol in utero (Druse et al. 1990). Given the role of dopamine in a variety of neurobehavioral events, such as feeding, motor activity, and arousal, these effects on the developing dopamine system may be related to feeding, motor, and attention deficits observed in children born to mothers who consumed alcohol during pregnancy. In addition, such perturbations of the dopamine system may underlie altered responsiveness to drugs acting on the dopamine system (Hannigan 1990; Hannigan et al. 1990) as well as suppressed lesioninduced plasticity of dopaminergic neurons (Gottesfeld et al. 1989) in alcohol-exposed offspring. Decreased levels of serotonin (5-HT) and its major metabolite (5-HIAA) have also been detected in some brain regions (cortex, cerebellum, and brain stem), but not in others (hippocampus, hypothalamus, and striatum). Moreover, the deficiency of 5-HT and 5-HIAA in the brainstem and cortex was found as early as gestation days 15 and 19 (Druse et al. 1991). These effects on the dopamine and serotonin systems are particularly interesting because dopamine and 5-HT are not only implicated in a variety of neurobehavioral functions, but are also thought to play an important role in neuronal maturation and differentiation during embryonic development (Lankford et al. 1988; Lauder et al. 1983). In summary, animal research has revealed that prenatal alcohol exposure alters the cytoar chitectural structure of numerous brain regions, disturbs the integrity of synaptic neurotransmission, and perturbs a variety of neurotransmitter systems. These effects have been identified in animals that exhibit no external physical abnormalities; thus, they may represent the neurobiological basis for the behavioral teratogenic actions of alcohol. This contention has gained support from recent studies demonstrating neuroanatomical and neurochemical alterations in animals exhibiting behavioral deficits. Effects on Neuroendocrine Systems Alcohol consumption during pregnancy disturbs the normal functioning of several neuroendocrine systems, including prolactin secretion, which is necessary for milk production; thyroid function; and gonadal and adrenocortical hormonal control. For example, maternal alcohol consumption both prior to and following birth produces alterations in the structure as well as the function of mammary gland tissue (Steven et al. 1989; Vilaro et al. 1989). Moreover, although basal prolactin levels in plasma are not perturbed by alcohol, maternal infusion of alcohol that resulted in blood alcohol levels higher than 65 mg/dL inhibited suckling-induced prolactin release and milk yield (Subramanian and Abel 1988; Subramanian et al. 1990). Prolactin is essential for the structural development of mammary glands during pregnancy and for the initiation and maintenance of adequate milk delivery following parturition (Whitworth 1988). The mechanism by which alcohol inhibits sucklinginduced prolactin release is not known. Hypothalamic-Pituitary-Thyroid Axis A remarkable similarity exists between the physical, sensorimotor, and cognitive-behavioral abnormalities associated with prenatal alcohol exposure and those produced by perinatal hypothyroidism. Fetal exposure to alcohol can result in abnormal thyroid function (Portoles et al. 1988). Recently, a 10- to 15-percent reduction in serum thyroxine (T4) concentrations was reported in alcohol-exposed rats from 8 to 24 days of age (Hannigan and Bellisario 1990). Alcohol may act directly on the fetus, where it has been shown to have profound effects on the thyroid axis (Yamamoto et al. 1989), or indirectly via the mother, inasmuch as T4 is known to cross the placenta (Vulsma et al. 1989). Since thyroid hormones are critical trophic factors for normal somatic and neural maturation and since fetal alcohol exposure reduces serum T4 levels, impaired trophic regulation of growth, differentiation, and general metabolic activity by thyroid hormones may represent a potential mechanism by which alcohol produces its teratogenic actions. Further support for this hypothesis was provided by a recent study showing that early thyroid hormone treatment was effective in reversing some of the developmental delays resulting from prenatal alcohol exposure (Gottesfeld and Silverman 1990). Hypothalamic-Pituitary-Gonadal Axis Prenatal alcohol exposure influences the hypothalamic-pituitary-gonadal (HPG) axis at all levels in males and females. For example, hypothalamic luteinizing hormone-releasing hormone and plasma luteinizing hormone (LH) are significantly reduced in 30- to 40-day-old female rats exposed prenatally to alcohol (Morris et al. 1989). Alcohol also decreases in vitro LHinduced production of estrogens and progesterone in cultured human ovarian granulosa cells (Saxena et al. 1990). These changes may be related to delayed sexual maturation, reproductive dysfunction, and altered sexual behavior observed in female offspring exposed prenatally to alcohol (e.g., Becker and Randall 1987; Creighton-Taylor and Rudeen 1991). Hormonal perturbations produced by perinatal alcohol exposure may underlie alterations of the normal gender differences in brain structure and function (behavior) observed in alcoholexposed offspring. The development of male and female neuroanatomical and behavioral patterns depends on the androgenic milieu of the fetus during the critical period of brain sexual differentiation. Adult male rats exposed prenatally to alcohol exhibit demasculinized sexual behavior (Barron et al. 1988; Rudeen et al. 1986) and a feminized pattern of maze learning (McGivern et al. 1984), saccharin preference (McGivern et al. 1984, 1987), and rough-andtumble play (Meyer and Riley 1986). Female rats exposed prenatally to alcohol display a masculine pattern of saccharin consumption, maze learning performance, and play behavior, as well as a marked reduction in maternal behavior (Barron and Riley 1985; McGivern et al. 1984; Meyer and Riley 1986). Sex differences in sensitivity to effects of prenatal alcohol exposure on immune function and stress responsiveness have been noted as well (Weinberg 1988; Weinberg and Jerrells 1991). Furthermore, these effects on sexual dimorphic behavioral patterns have been extended to CNS anatomical alterations. For example, male offspring exposed prenatally to alcohol exhibit demasculinization of certain brain structures (Barron et al. 1988; Rudeen 1986; Rudeen et al. 1986; Zimmerberg and Mickus 1990; Zimmerberg and Reuter 1989). Thus, although alcohol-induced changes in sexual hormones may be transient, the ultimate effects may be permanent because in utero alcohol exposure occurred during the critical period for brain sexual differentiation. A remarkable similarity exists between the physical, sensorimotor, and cognitive-behavioral abnormalities associated with prenatal alcohol exposure and those produced by perinatal hypothyroidism. Hypothalamic-Pituitary- Following exposure to a stressful situation, a number of physiological and behavioral events that ready the body for a response are set in motion. For example, the hypothalamus and pituitary gland activate the adrenal cortex to release corticosteroid hormones. Alcohol consumption during pregnancy has been shown to have profound effects on both maternal and fetal hypothalamic-pituitary-adrenocortical (HPA) axes (Weinberg et al. 1986). These effects have been demonstrated under both nonstressful and stressful conditions. (See Chapter 8, Effects of Alcohol on Health and Body Systems.) Maternal alcohol intake may directly or indirectly influence the HPA axis in offspring after birth and later in adulthood. For example, plasma, brain, and adrenal corticosteroid levels, along with hypothalamic corticotropin-releasing factor content and pituitary adrenocorticotropin hormone, are significantly higher at birth in rodents exposed prenatally to alcohol than in control offspring (e.g., Redei et al. 1989; Weinberg |