Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Neurobehavioral Time Bombs: Their Nature and Their Mechanisms Roger W. Russell Basic to a complete discipline of neurobehavioral toxicology is the recognition that behavior is but one of the properties of living organ- isms that are affected by the chemical environments in which they live. Behavior does not exist independently of dynamic biochemical and electrophysiological processes taking place constantly in various structural (morphological) sites within the body. The major objective of this chapter is to place behavior in its proper perspective within the "integrated organism" (Russell, 1979~. This is done by discussing the nature and mechanisms involved in three examples of what are generally referred to as "progressive degenerative dementias" (PDDs). It is well to begin with consideration of a general framework within which the trilogy may be analyzed. Relations Between Behavior and Chemical Environment It is well recognized that the biological effects of chemicals in the external physical environment can only be a result of physiochemical interactions between molecules of the agent and receptor sites on particular molecules present in the body (Doull, 1980~. Technological advances have now made it possible to study some of the "cascade" of biological events that follows such an interaction, although there is still much more to learn. The events may be viewed as progressing from the molecular level through morphological sites synapses, neurons, nerve networks, nuclei, and systems eventually to exert an effect on 206
NEUROBEHAVIORAL TIME BOMBS BEHAVIORS ~ 1 | SYSTEMS l NUCLEI a 1 NETWORKS NEURONS SYNAPSES [ MOLECULES 207 FIGURE 1 Sites of action of neurod~emical events in the nervous system. The dia- gram represents levels of increasing complexity extending from molecules to endpoints measurable as changes in behavior. behavior (Figure 1~. Although the matter is not pursued here, it should be remembered that research in psychosomatics has shown that behavior may produce consequent changes even at the molecu- lar level. The sequence of the neurochemical events taking place in these sites is shown diagrammatically in Figure 2. Events al to an preceding the formation of the chemical-receptor complex, AR, are involved in the processes by which the chemical, A, reaches its site of action. The transport of A to its receptor may involve progress through several different membranes and chemical milieu, during which A may undergo biotransformation. In some cases the resulting molecule may be much more potent in terms of its biological effects than the parent substance. Conversion of the organophosphorus pesticide parathion to paraoxon, the active form, is an example. Binding of A to its receptor site may be reversible or irreversible. In the latter instance, receptor molecules must be synthesized de nova; hence, recovery from the effects produced by A may be delayed. Effects e, to en following formation of the AR complex are indepen- dent of those preceding it, but are influenced by the state of the organ-
208 EXTERNAL I A ROGER W. RUSSELL INTERNAL TO ORGANISM PHARMACOKINETIC I PHARMACODYNAMIC a1 a2 a 3 a n A Rt e1 e2 e 3 e n 1 FIGURE 2 Sequence of neurochemical events taking place between the entry of an exogenous chemical into the body and its effects on behavior. The chemical, A, is transported to its site of action, a,-an, where it binds to its specific receptors R. initiat- ing an extensive series of events, even, leading to effects on behavior. . . . ~ ~ ism and by other processes that interact with those stimulated by AR. It is logical that the further an endpoint is from its receptor activa- tion, the greater is the possibility that other events may influence the nature of the effect. Some forms of behavior are linked more directly to their biochemical correlates than others. Where the linkage is direct (e.g., in sensory-reflexive responses), changes in biochemical events are reflected in specific changes in behavior, but where the linkage is diffuse, as it is in cognitive behaviors, changes in bio- chemical events may affect a variety of behavior patterns. The nature of the relation between the magnitude of exposure to an environmental chemical and its effects on behavior is familiar to psychologists, as well as to pharmacologists and toxicologists. It takes the general form of an ogival or cumulative normal population curve. Very low levels of exposure produce no effects. As exposure increases, a level is reached where behavioral effects begin to appear, the "basal threshold." Further increases induce proportional changes in behavior until a level is reached, the "terminal threshold," at which behavioral malfunctions begin to occur. Activity prior to the termi- nal threshold is characteristic of self-regulator~y, self-correcting biological processes, to which the term "homeostasis" has been applied. How- ever, there are limits to this plasticity. A premise on which the con- cepts of terminal threshold and of behavioral plasticity depend is that some low magnitude of exposure exists for all chemical sub- stances which will not produce an effect no matter how long the exposure. Coupled with this is the corollary that all substances will produce an effect at some higher level of exposure. It follows from this that any chemical introduced into the physical environment may set an "ecological trap" for the behavior of living organisms. An example is presented next which illustrates major points drawn _ _ . . · . .
NEUROBEHAVIORAL TIME BOMBS 209 from studies of both human and animal models designed to provide information about chemicals affecting the cholinergic neurotransmit- ter system. These chemicals are involved in what may well be the widest diversity of purposes of any substances known today. They are applied therapeutically, a new anticholinesterase (antiChE) pres- ently undergoing mass clinical trials for potential treatment of Alzhe~mer's disease. They appear in both indoor and outdoor environments as pesticides. Some were developed but not used during World War II as the so-called nerve gases. They have a basic neurochem~cal mechanism in common. CHOLINESTERASE IN PROGRESSIVE DEGENERATIVE DEMENTIAS The role of acetylcholinesterase (ACHE) is normally associated with the inactivation (hydrolysis) of the neurotransmitter, acetylcholine (ACh) once the transmitter has been released into the synaptic cleft and, therefore, beyond the presynaptic side of the cholinergic syn- apse (Figure 3~. However, evidence has been accumulating that AChE also appears to be involved presynaptically. Results of recent experi- ments have demonstrated that presynaptic AChE and the high-affin- - Glucose Glycolysis Phospholipid Synthesis and Degradation k ~ Embden- | Meyerhof ~ Pathway J Choline ~ / ~ Pyruvate / Plasma / Choline ~ ~ / *Choline Acet' rI-CoA Ax/ Acetyltransferase - Acetylcholine Tricarboxylic \\ / I Acid Cycle ~ / Cholinesterase (TCA) J ~ ~ ~ Acetate Choline FIGURE 3 Metabolic pathways involved in the normal biosynthesis and hydrolysis of the neurotransmitter acetylcholine.
210 ROGER W. RUSSELL ity transport (HAChT) of the acetylcholine precursor, choline, are localized very close to each other on the cholinergic terminal membrane, suggesting some functional relationship between the two mechanisms (Raiteri et al., 1986~. Furthermore, the existence of molecular hetero- geneity (stack et al., 1983; Michalek et al., 1981) and the different cellular distributions of AChE support the view that the enzyme may have more functions than hydrolyzing ACh postsynaptically (Greenfield, 1984~. Preclinical Studies of Acety~cholinesterase and Behavior The possibility that manipulation of presynaptic AChE indepen- dently from the postsynaptic enzyme might produce differential effects on behavior awaits the invention of techniques for varying the one without the other in the intact organism. Meanwhile, it is relevant to the development of the present theme to summarize briefly the nature of behavioral effects when available antiChEs are employed as phar- macological tools. Early experiments using animal models concluded that exposure to antiChEs produced differential effects on behavior, some behavior patterns being affected and others not. Behaviors affected involved the extinction of old responses that were no longer appropriate in coping with new environmental demands (Russell, 1958~. More re- cent research has made it quite clear that cognitive behaviors (learn- ing and memory) are particularly sensitive to manipulation of AChE activity. Dose-effect relations indicate that, behaviorally, the cholin- ergic system is capable of adaptive changes only within limits. Be- havioral subsensitivity characterizes levels of AChE activity below this "normal" range and supersensitivity, levels above it. Activity at both extremes is associated initially with nonadaptive responses. However, prolonged changes in AChE activity at these extremes may initiate compensatory mechanisms within the cholinergic system [e.g., downregulation of muscarinic receptors (mAChRs), differential recovery of AChE isoenzymes] that are paralleled by the return of behavior to normal. Such "tolerance development" is an important form of behavioral homeostasis and also has significant implications for the use of antiChEs (e.g., physostigmine) as therapeutic agents in neurodegenerative disorders involving hypofunctioning of the cholinergic system [e.g., Alzheimer's disease (DATA. Acety~cholinesterase in Behavioral Disorders Major neurochemical, morphological, and behavioral symptoms of DAT are summarized in Figure 4. Available evidence indicates that
NEUROBEHAVIORAL TIME BOMBS ChAT ,> ChE >l Cholinergic cell bodies Symptoms of DAT 211 Reactivity Attention span Learning Memory Problem solving ~ FIGURE 4 Major symptoms of Alzheimer's disease (DAT). Levels of activity of the synthesizing choline acetyltransferase (ChAT) and inactivating cholinesterase (ChE) enzymes are decreased. Relatively selective loss of cholinergic neurons occurs in cer- tain regions of the brain, particularly the projections from large cholinergic cell bodies in the basal forebrain (nucleus basalis of Meynert) to the neocortex and projections from the medial septum to the hippocampus. Behavioral effects are characterized by hyperreactivity and by decreases in attention span, learning, memory, and other cog- nitive functions. AChE activity is decreased in DAT: ". . . the first report of an altered distribution of acety~cholinesterase molecular forms in a disease of the central nervous system" distinguished three such forms in post- mortem tissues from both the normal and the DAT neocortex (stack et al., 1983~. Losses in activity levels appeared selectively in the intermediate form assayed in DAT samples. Involvement of presyn- aptic AChE in human behavioral disorders is further suggested by the fact that deteriorative neuronal changes found in DAT are rich in AChE. As these changes increase, the AChE activity decreases. The results of such investigations are interpreted as indicating that changes in cortical cholinergic innervation are an important feature in patho- genesis and progressive development (PDD) (Struble et al., 1982~. The possibility that antiChEs, alone or in conjunction with other means for manipulating the cholinergic system when it is hypofunctional, might serve a therapeutic purpose has been under consideration for several years. Indeed, physostigmine continues to be a therapeutic strategy by which the half-life of ACh in the synaptic cleft is pro- longed by decreasing its hydrolysis (inactivation). The varied success obtained, as well as the difficulties (i.e., short half-life, peripheral side effects, very narrow therapeutic window) involved in the clinical application of this particular compound, are reflected in a number of reports during the past decade (Davis and Mobs, 1982~. The PDDs considered in this example have characteristic behav- ioral sequelae that include significant deterioration of memory and other cognitive functions such as language, spatial or temporal orien-
212 ROGER W. RUSSELL ration, judgment, and abstract thought. These behavioral changes are not readily discernible during early stages of the disorders and, even later, are difficult to differentiate from senile dementia accom- panying the aging processes. It is generally accepted that final con- firmation of DAT depends upon evidence of cellular degeneration in a particular area of the brain. The possibility that at least some PDD may result from exposures to chemicals in the external environment has been recognized in a recent publication on cementing diseases by the U.S. National Institutes of Health (1987~. The major points emphasized here are that detailed analyses of neurochemical mechanisms of action underlying behavior can pro- vide (1) knowledge about differential behavioral effects of different toxicants upon which differential diagnostic criteria may be established; (2) information necessary for regulating exposures; (3) rational bases for patient management in neurodegenerative disorders, thereby eliminating procedures that have high probabilities of being unsuccessful. THE TRUTH ABOUT A FALSE TRANSMITTER Some half-century ago the possibility that a false precursor lead- ing to the synthesis of a false transmitter might serve as a means for examining neurotransmitter systems at a molecular level began to receive attention. In viva studies of choline analogues began to ap- pear in scientific journals. Criteria that must be satisfied if a compound is to be accepted as a false transmitter were established. The first direct demonstration that a choline analogue, triethylcholine, could be acetylated and released in cholinergic synthesis of false transmitters, might provide valid animal models for studying clinical states involving dysfunctions in the central nervous system (CNS). A few experiments were designed to use such behaviors. The relevance of the early studies to PDD was recognized when it became apparent that the lack of natural precursor chemicals in the diet or the presence of a false precursor could significantly alter normal functioning of the cholinergic system and thus affect behavior (lender et al., 1987~. A major difficulty faced in the earlier studies was to demonstrate that chronic dietary administration of a choline analogue did in fact result in functionally significant replacement of choline in the syn- thesis of an endogenous analogue of ACh. Ways in which the availability of quantitative analytical techniques eventually solved this problem are illustrated in a series of experiments in our laboratories reported during the past five years, involving the false precursor N-amino-N,N- dimethylaminoethanol (N-aminodeanol, NADe) (Newton and lender, 1985~: NADe is taken up by the choline transport system in competi-
NEUROBEHAVIORAL TIME BOMBS PHOSPHOLIPIDS a\ a/ BRAIN CHOLINE i' BLOOD ACETYLCHOLINE /? 213 FIGURE 5 Competition for avail- able choline. One hypothesis for the selective vulnerability of chol~nergic neurons in Alzheimer's disease as- serts that a competition for available choline occurs between biochemical pathways involved in maintaining the membrane integrity of cholin- ergic neurons (phospholipids) and in synthesizing the neurotransmit- ter (acetylcholine). tion with choline. It is acetylated by choline acetyltransferase (ChAT), stored as O-acetyl-N-aminodeanol (ANADe) in vesicles, and released on stimulation. Stores of ACh are depleted as they are replaced with NADe. Upon release, ANADe interacts with both muscarinic and nicotinic receptors and is hydrolyzed by AChE. Because the potency of ANADe at these receptors is only 4 and 17 percent that of ACh, respectively, there occurs a profound interference with cholinergic transmission, particularly at muscarinic sites. Replacement of choline with NADe in the diet of weaning rats for periods of 60-120 days results in the replacement of 85-95 percent of free choline by free NADe in brain, plasma, and peripheral tissues; ChAT is reduced in the cortex, hippocampus, striatum, and ileum, suggesting the loss of cholinergic neurons. This evidence shows that NADe satisfies the neurochemical requirement of a false precursor, leading to the syn- thesis of a false cholinergic transmitter, and enables the study of these behavioral and physiological consequences. We had two major objectives in mind. The first was to test a hypothesis about the etiology of DAT, and the second, to develop a useful animal model for the disorder. According to the hypothesis, competition for available choline is the central element (lender, 1986~. Cholinergic neurons use choline both for the synthesis of the neuro- transmitter ACh and, via phospholipid metabolism, as a structural component in cell membranes (Figure 5~. The hypothesis states that when there exists a deficiency in the normal supply of choline for use in the two biochemical pathways involved, cholinergic neurons may break down membrane phosphatidylcholine to maintain the choline concentration required for ACh synthesis. This could lead to what has been imaginatively termed "autocannibalism" of the neuron (Wurtman et al., 1985~. Conditions under which such degeneration could occur
214 ROGER W. RUSSELL include (1) a relatively short supply of choline and (2) the overuse of choline for ACh synthesis. It has been suggested that "this competi- tion could be precipitated by a number of inherited or acquired char- acteristics, is likely to be age-related and could result in failure of synaptic transmission, of mechanisms for cell membrane renewal or both" (lender, 1986). Although this hypothesis has not yet been put to an adequate test, it is consistent with a number of facts already available. Our approach to an experimental test involves the replace- ment of choline with its analogue NADe. General observations during our early experiments failed to show any gross neurobehavioral toxicity, although the animals showed a discernible hypertonic when handled. This did not, of course, mean that the replacement of choline by NADe had no concomitant behav- ioral or physiological effects. For this reason an extensive series of experiments has now been carried out using quantitative assays of variables known to be cholinergically "coded." Initially these variables were measured only between 25 and 32 days on the NADe diet, yet even after this relatively short time, they showed some significant effects (Newton et al., 1986). A more extensive series of experiments (as yet unpublished) has now been completed, measuring many more variables periodically as the level of replacement of choline by NADe increased progressively over a much longer time. The results provide the data necessary to relate magnitudes of behavioral and physiological changes to levels of the false transmitter. There were no significant effects on total caloric intake, on the maintenance of body fluid balance, or on core body temperature, results which strongly suggest that the behavioral effects described below are unlikely to be attributable to imbalances in basic homeostatic mechanisms. The behavioral changes concomitant with increasing replacement of choline by NADe may be summarized in three major categories. The first includes measures that are primarily sensory-r`fexive in nature, i.e., innate, appearing without the necessity for learning (e.g., reflexive responses to electric shock and acoustic stimuli). Such behaviors were affected significantly, be- ing evidenced in sensory hypersensitivity and motoric hyperreactiv- ity. Clearly apparent in DAT and other degenerative disorders of aging are symptoms of a sensory-perceplual nature, e.g., failure to "un- derstand" events in the physical and psychosocial environments, spatial disorientation, and eventually a general loss of response to most stimuli. The effects were reminiscent of those observed in earlier experiments in which the transport of choline to the site of ACh synthesis in nerve endings was impaired by a specific inhibitor, hemicholinium-3 (HC- 3), administered intracerebroventricularly (Russell and Macri, 1978), and of those found in septally lesioned animals with reduced cholin-
NEUROBEHAVIORAL TIME BOMBS 215 ergic activity. The third category of behavioral effects included learning, memory, and other cognitive processes. Because these are among the early signs of dysfunction in degenerative disorders involving the cholinergic system, they received particular attention and were stud- ied in several different test situations. Measures that are intrinsically dependent upon the functioning of the cholinergic system showed a consistent pattern of effects: NADe animals took more trials to learn, made more errors, were slower in their response times, and had poorer memories. Furthermore, these effects increased progressively as the available supplies of choline decreased. Earlier in this chapter the hypothesis of autocannibalism was in- troduced as a possible mechanism underlying behavioral disorders associated with hypofunctioning of the cholinergic system, a condi- tion that might be a consequence of competition for available choline supplies between the needs for it to maintain the membrane integrity of cholinergic neurons and to synthesize ACh. In the present experiments the supply of choline was progressively replaced by the much less efficient NADe. In the continuing competition, both free and lipid- bound choline were very significantly reduced. Paralleling these re- ductions were highly significant impairments of behavioral variables, the changes being analogous to those characteristic of such primary degenerative disorders as DAT. Research in our laboratory is now underway to study still further the neurochemical, behavioral, and histopathological properties of the model to determine whether it can be influenced by drugs intended to enhance cholinergic function and to discover the extent to which the various effects may be reversed when the diet is returned to normal. At the present time we believe that it is a potentially useful model for research on DAT and similar human disorders. We also believe that it may have significant implications for their pathogenesis. The example discussed in this section of my trilogy illustrates mechanisms of action by which chemicals included (or not included) in the daily diet may induce malfunctions of normal behaviors. It also shows how the effects produced may be considerably delayed in their appearance. Earlier in this volume, Dr. Spencer provides some very striking examples of "long-latency neurotoxic disorders" as evi- denced in humans and in animal models (see Spencer, 1987; Spencer et al., 1987~. Since 1930 an increasing concern has been developing about such delayed neuropathies and their behavioral components. Particular attention has been directed to populations at special risk (i.e., pregnant women, young children, and workers exposed occupa- tionally). However, concern generalized to populations at large has become evident in such regulations as those requiring the labeling of
216 ROGER W. RUSSELL foodstuffs for their chemical compositions. Knowledge about the neurochemical modes of action underlying behavioral effects can contribute not only to regulation, but also to the invention of procedures for early detection of delayed neurotoxicities and to treatment of them once they are recognized. NEUROTROPHIC FACTORS AND BEHAVIOR1 The third part of this trilogy involves hypotheses about interac- tions of brain "transplants," neurotrophic factors, and behavior. For reasons that will become apparent, this relatively new area of study is already at one of the more exciting frontiers in the biomedical sciences. In 1981, in a paper "A Unifying Hypothesis for the Cause of Amyotrophic Lateral Sclerosis, Parkinsonism, and Alzheimer Dis- ease," Appel presented the thesis ". . . that each of these disorders is due to lack of a disorder-specific neurotrophic hormone" (Appel, 1981~. The hypothesis postulated that the failure of target tissues to supply necessary neurotrophic factors is the primary manifestation of disor- ders: ". . . in each system, the lack of an appropriate hormone re- leased from postsynaptic cells would impair the viability of the presynap~c cells" (Appel, 1981~. More specifically, in DAT the failure would be in the production of nerve growth factor (NGF) by hippocampal and cortical cells, resulting in a gradual deterioration of septal and basal nuclei and associated decreases in ChAT activity and ACh synthesis (Appel et al., 1986; Hefti, 1983~. Because knowledge of such pro- cesses derived primarily from research on animal models, a basic step toward testing this hypothesis was to establish the presence of NGF in the human brain. The evidence now indicates ". . . that NGF acts as a trophic factor for cholinergic neurons in the human brain in a similar way as has been established in recent years for the rat brain" (Heft) et al., 1986~. Furthermore, "the similarity between the response in rodents to entorhinal cell loss and that in AD [Alzheimer's disease] patients indicates that studies using the rodent model may be di- rectly applicable to AD" (Colman et al., 1986~. Neurotrophic Factors Neurotrophic Factors and Their Environments The term "trophic" has been defined as ". . . any relatively long term influence that passes from one cell or tissue to another either during development or in the mature state" (Varon and Bunge, 1978~. Interest in the remarkable ways in which neurons form synapses and
NEUROBEHAVIORAL TIME BOMBS 217 become affilitated with their eventual target cells and tissues during embryonic, fetal, and neonatal growth raised questions about the dy- namic processes involved. In the early 1950s, Nobel Laureate R.W. Sperry proposed that during maturation of the nervous system, genetic ". . . specification of the neurons makes possible the formation of selective synaptic linkages on the basis of a chemoaffinity" (Sperry, 1951~. Although this concept was on a fruitful path, it was not until somewhat later that the current model took a definite shape, with the hypothesis that neurotrophic factors might be produced by other "target" tissues to provide a favorable cellular environment for axonal generation and regeneration. Initially, the physiological role and the distribution of NGF were well characterized peripherally as synthesized by target tissues innervated by sympathetic and certain sensory neurons and taken up by those neurons to be transported retrogradely to their cell bodies. More recent evidence, derived from studies of both human material and animal models, supports the involvement of NGF in an analogous role in the CNS. In the brain, as in the periphery, NGF is elaborated by target cells and binds to specific receptors on the in- nervating neurons. Behavioral adjustments (including cognitive processes) to ever-chang~ng physical and psychosocial environments require a multitude of dy- namic biochemical events that take place in defined morphological sites within the CNS. There is a rapidly growing body of knowledge describing mechanisms by which these sites are themselves influenced by biochemical processes. It appears that neurotrophic factors determine which neurons will survive during ontogeny and thereby regulate the development of neural pathways. Present results also indicate that (1) there may be a large class of neurotrophic factors, each specific to particular neuronal populations; (2) they regulate neuronal survival during adulthood as well as during earlier development; and (3) inadequate neurotrophic activity may lead to neurological malfunc- tions arising from nerve cell death. It has been proposed that, phylo- genetically, ". . . specific heritable, trophic interactions during devel- opment, which determine cell survival and pathway size, form a substrate for neural evolution" (Black, 1986~. Clearly the growth of neuronal cell structures to synapse on other "target" tissues is a process of great importance for the normal functioning of the nervous system and hence of the ability of individuals to cope with the demands of their environments. Award of the 1986 Nobel Prize for Physiology or Medicine to Rita Levi-Montalcini and Stanley Cohen for their discoveries of neurotrophic factors called special attention to-the potential significance of these agents not only neurologically, but also behaviorally. Results of Levi-
218 ROGER W. RUSSELL Montalcini's early research (1964) had suggested the possibility that interactions might occur between behavior and biochemical events involved in the synthesis or release of nerve growth factors. "It seems a reasonable hypothesis that there may exist a mechanism by which during the coding of new behavioral patterns, the effects of informa- tion input on protein synthesis increase the production of protein molecules capable of modifying the structure of nerve or glial cells. The structural modifications would then serve as engrams for long- term memory storage" (Russell, 1966~. This prediction came to have special meaning when it was established that deficiencies in memory were about the most prominent characteristics of DAT. More recent research makes it clear that some special relationship does in fact exist between NGF and the cholinergic neurotransmitter system, with behavior as the third among three partners in the relationship. Roles During Neuronal Development and Beyond Early development of the nervous system is characterized by a very significant overproduction of cells. During a circumscribed phase of embryonic life, neuronal degeneration occurs by which a high per- centage of these cells die, despite the fact that they develop proper- ties of mature neurons. The degeneration generally coincides with the arrival of surviving cells in their target area, indicating that de- velopmental survival depends upon NGF derived from target cells. The development of neural pathways and connections is similarly dependent, with neurons whose axons have extended to an inappro- priate target area being eliminated. It has been suggested that during early development of the mammalian brain the "transient cells" " function as neurons in a synaptic circuitry that disappears by adulthood" (shun et al., 1987~. In addition to its role during early development, NGF may be involved in at least two series of events vital to the behavior of living organisms: (1) the maintenance and viability of a normally function- ing mature nervous system, and (2) the modification of neural struc- tures during behavioral adjustments to changing physical and psychosocial environments (e.g., memory). Indeed, it has been suggested ". . . that NGF may function not only as a trophic agent, but also as a modulator of neurotransmission in the CNS" (Rennert and Heinrich, 1986~. Recovery from Morphological Lesions Evidence that central neurons exhibit a capacity for axonal sprout- ing and generation of new connections in response to injury, not only
NEUROBEHAVIORAL TIME BOMBS 219 in the developing but in the mature CNS, has been accumulating for some time. In peripheral models, axotomy is followed by a dramatic increase in the density of NGF receptors in Schwann cells through which axonal regeneration must occur. In the CNS, lesions of the septohippocampal pathway are followed by an increase in NGF con- tent in the hippocampus and septum. Of major interest is the fact that, both centrally and peripherally, NGF-supported regeneration occurs following injury to the cholinergic system. That the regenera- tion results in "correct wiring" (i.e., in functional connections) is evi- denced by electrophysiological activity and the recovery of behaviors disrupted by the original lesion. The use of immunohistochemical techniques has made it possible to examine the morphological effects of NGF on axotomized septal neurons (Gage et al., 1988~: infusion of NGF protected most of the immunoreactive neurons from degeneration and prevented the appearance of plaque-like neurons. Findings that suggest the occurrence of neuronal death following brain lesions are based primarily on the detectability of neurotrans- mitter-related enzymes. Very recently results of experiments using special staining techniques have demonstrated ". . . that the initial loss of ChAT-positive neurons following fimbria-fornix transection is due mainly to a reduction of the ChAT-stainability rather than actual neuronal death" (Hagg et al., 1988~. Apparently, during the initial period, neurons may be in a state of reversible trauma, before irreversible damage sets in. The behavioral characteristics of tissue transplants may be consequences of the ". . . activation of 'silent' pathways by neurotrophic factors. . . " (Stein and Mufson, 1987~. Brain Transplants Early conceptualization of the roles of NGF suggested that, if the deficiencies underlying behavioral changes and neuronal degenera- tion in disorders such as DAT are insufficient concentrations of specific neurotrophic factors, it should be possible to manipulate these fac- tors by administering them directly or by transplanting cells that are rich trophic sources. Research could lead to the chemical isolation, purification, and eventually, synthesis of the trophic molecule. Im- portant steps have already been taken to implement these actions. Information about the survival of brain transplants has come from studies of their effects, as well as from evidence of their continued existence. The capacity to survive within a host has been demon- strated repeatedly. Neurochemical and histological evidence, supple- mented by the fact that neural transplants can reverse behavioral impairments induced by brain lesions, supports the conclusion that transplants not only survive but also influence the functional capac-
220 ROGER W. RUSSELL ity of newly regenerated neural connections. The evidence also indi- cates the importance of the relationship between transplants and NGF: survival of a large number of cells and a considerably larger amount of nerve fiber formation occur in the presence of NGF. The question of whether the probability of survival may be affected by untoward immunological accidents has been investigated. A review of the present state of affairs has come to the conclusion that the CNS is a site where transplants may enjoy a ". . . prolonged, but not always indefinite, survival" (Mason et al., 1985~. Nerve Growth Factor, Brain Implants, and Behavior It is now time to relate the information about neurochemical and morphological factors to behavior, the property heavily involved in the diagnoses of PDDs, in monitoring their progression, and in evaluating treatment outcomes. Again, DAT will be used as the main example, with special attention given to memory because of the central role it plays in normal coping behavior as well as in disease states. Behavioral Recovery from Brain Lesions During the past few years, relations between neurochemical events and changes in behavior have encouraged a spate of experimentation designed to manipulate NGF and neuronal regeneration as means of compensating for adverse behavioral effects of experimental brain lesions. Two experimental approaches have been used: (1) repeated (chronic) injections or infusion of NGF and (2) transplantation of neuronal or target tissue. Both of these are reported to have produced at least partial compensation for the effects of brain damage, restoring the pattern of cholinergic innervation and producing concomitant im- provements in some but not all behavioral patterns. A broad range of behaviors have served as dependent variables in the search for evidence that cellular changes following neuronal re- generation stimulated by implants or by cell suspensions are func- tional. Interest in motoric abnormalities in Huntington's disease has led to the development of animal models involving lesions produced by neurotoxins. Bilateral lesions in the striatum followed by bilateral fetal striatal implants reversed the spontaneous motor abnormalities induced by the lesions. Severe striatal neuronal cell loss and shrink- age following ibotenic acid lesions of the caudate-putamen produced hyperactivity that was completely compensated by "neural grafting," i.e., implantation of a dissociated cell suspension from fetal rat striatum into the lesioned sites. Cognitive deficits have been reduced in ani-
NEUROBEHAVIORAL TIME BOMBS 221 mats with frontal cortex implants after bilateral damage to the me- dial frontal cortex. Particular attention has been given to learning and memory, by using a variety of different assays. Retention has been found to be very significantly improved in lesioned animals following implantation of cholinergically rich cells when measured by such well-established techniques as spatial alternation, inhibited (passive) avoidance, and learning and memory in water or radial maze situations. These examples establish that exogenous NGF and tissue implants may produce neural regeneration and concomitant full or partial re- covery of behaviors impaired by brain lesions. The behavioral effects appear to be differential in the sense that some behaviors may be affected and others, not. Recovery may be transient or long range depending upon the procedure used to induce increases in NGF activity. As knowledge about the effects of NGF and of transplanting tis- sues into brain grew at the basic bioscience level, there began to appear an interest in its implications for human therapy. In 1985, reports appeared describing the "first clinical trials," apparently begun some three years earlier, involving transplantation of adrenal medullary tissue into the striatum of patients with severe Parkinson's disease. The transplantations were autologous ("autografts"), i.e., involving tissues within the same individual. The results of a series of "continuing clinical experiments" by this group of investigators are summarized in the statement: "The brief and limited response in our patients after the transplantation seems to indicate a limited survival of the graft" (Backlund et al., 1987~. Greater success has been reported by other investigators: "Our results suggest that grafting chromaffin cells in direct contact with both the cerebrospinal fluid and the caudate nucleus produced excellent amelioration of most of the clinical signs of Parkinson's disease in our two patients" (Madrazo et al., 1987~. Results of research on animal models had indicated greater suc- cess with transplants of fetal tissue than with tissue taken later in development. In 1988, results were reported of transplants in two patients with Parkinson's disease of tissue from a spontaneously aborted fetus of 13 weeks. One patient received transplants from the fetal substantia nigra; the other, from the fetal adrenal medulla. The grafts were placed within a cavity of the right caudate nucleus and in con- tact with CSF. Significant improvement was recorded in both cases. The investigators concluded that". . . the use of fetal tissue as donor grafts may prove superior to autografting to treat Parkinson's dis- ease" (Madrazo et al., 1988~. As happens with most "breakthroughs" at a frontier of science, this picture is not as clear as it must eventually become. Disparities
222 ROGER W. RUSSELL between the findings reported by various investigators have cast a "cloud over Parkinson's therapy" (Lewin, 1988~. A final evaluation of the effectiveness of procedures designed to use endogenous chemicals (e.g., NGF) in the treatment of neurodegenerative disorders awaits much fuller information than has yet appeared publicly. The ratio- nale is clear and research on animal models has made it persuasive. Even if its validity is established, ethical and legal questions must be answered. CONTRIBUTIONS IN THE FUTURE About a decade ago a committee of the National Research Council, investigating "Decision Making for Regulating Chemicals in the En- vironment" reported: "All difficult decisions are characterized by in- adequate information.... Problems of regulating chemicals in the environment are particularly beset with information characterized by a high degree of uncertainty. For some aspects of these problems there exists no information at all" (National Research Council, 1975~. One of the major objectives of this trilogy has been to indicate how such uncertainties may be narrowed by basic and clinical research designed to understand the neurochemical mechanisms by which ex- posures to toxicants affect behavioral and other biological indicators. Procedures may be developed by which such processes can be ob- served and measured. The procedures can provide more precise in- formation about the relation between the amount of a toxicant reaching receptors in its target tissues ("effective dose") and the consequent effects on biological endpoints. The study of neurodegenerative dis- eases can also provide opportunities to follow the progression of chemically induced malfunctions in the nervous system to which be- havior has been shown to be especially sensitive. It can be hoped that by relating behavior to other biological properties of living organisms, those responsible for regulation can be convinced that behavioral endpoints should be introduced more fully into the regulatory arena (Buckholtz and Panem, 1986~. NOTE 1. A more extensive discussion of this subject, including a more complete list of references, appeard in a recent review (Russell, 1988). REFERENCES Appel, S. H. 1981. A unifying hypothesis for the cause of amyotrophic lateral sclero- sis, Parkinsonism and Alzheimer disease. Annals of Neurology 10:449-505.
NEUROBEHAVIORAL TIME BOMBS 223 Appel, S. H., Y. Tomozawa, and R. Bostwick. 1986. Trophic factors and neurologic disease. Pp. 75-85 in Alzheimer's and Parkinson's Diseases, A. Fisher, I. Hanin, and C. Lackman, eds. New York: Plenum. Atack, J. R., E. K. Perry, J. R. Bonham, R. H. Perry, B. E. Tomlinson, G. Blessed, and A. Fairbairn. 1983. Molecular forms of acetylcholinesterase in senile dementia of the Alzheimer type: Selective loss of intermediate (10s) form. Neuroscience Letters 40:199. Backlund, E. O., L. Olson, A. Seiger, and O. Lindvall. 1987. Toward a transplantation therapy in Parkinson's disease: A progress report from continuing clinical experi- ments. Annals of the New York Academy of Sciences 495:658-673. Black, I. B. 1986. Trophic molecules and evolution of the nervous system. . Proceed- ings of the National Academy of Sciences 83:8249-8252. Buckholtz, N. S., and S. Panem. 1986. Regulation and evolving science: Neurobehavioral toxicology. Neurobehavioral Toxicology and Teratology 8:89-96. Chun, J. J. M., M. J. Nakamura, and C. J. Shatz. 1987. Transient cells of the developing mammalian telencephalon are peptide-immunoreactive neurons. Nature 325:617- 620. Cotman, C. W., M. Nieto-Sampedro, and J. W. Geddes. 1986. Synaptic plasticity in the hippocampus: implications for Alzheimer's disease. Pp. 99-117 in Treatment Devel- opment Strategies for Alzheimer's Disease, T. Crook, R. T. Bartus, S. Ferris, and S. Gershon, eds. Madison, Conn.: Mark Powley Associates. Davis, K. L., and R. C. Mohs. 1982. Enhancement of memory processes in Alzheimer's disease with multiple-dose intravenous physostigmine. American Journal of Psy- chiatry 139:1421-1424. Doull, J. 1980. Factors influencing toxicology. Pp. 70-83 in Toxicology: The Basic Science of Poisons, J. Doull, C. D. Klaasen, and M. O. Amour, eds. New York: Macmillan. Gage, F. H., D. M. Armstrong, L. R. Williams, and S. Varon. 1988. Morphological response of axotomized septal neurons to nerve growth factor. Journal of Com- parative Neurology 269:147-155. Greenfield, S. 1984. Acetylcholinesterase may have novel functions in the brain. TINS 7:364-368. Hagg, T., M. Manthrope, H. L. Vahlsing, and S. Varon. 1988. Delayed treatment with nerve growth factor reverses the apparent loss of cholinergic neurons after acute brain damage. Experimental Neurology 101:303-312. Hefti, F. 1983. Alzheimer's disease caused by a lack of nerve growth factor? Annals of Neurology 13:109-110. Hefti F., J. Hartkka, A. Salvatierra, W. J. Weiner, and D. C. Mash. 1986. Localization of nerve growth factor receptors in cholinergic neurons of the human basal fore- brain. Neuroscience Letters 69:37-41. Jenden, D. J. 1986. The pharmacology of cholinergic mechanisms and senile brain disease. Pp. 205-215 in The Bioloqical Substrates of Alzheimer's Disease, A. B. Scheibel and A. P. Wechsler, eds. New York: Academic Press. Jenden, D. J., R. W. Russell, R. A. Booth, S. D. Lauretz, B. J. Knusel, M. Roch, K. M. Rice, R. George, and J. J. Waite. 1987. A model hypocholinergic syndrome pro- duced by a false choline analog, N-aminodeanol. Journal of Neural Transmission (Suppl.) 24:325-329. Levi-Montalcini, R. 1964. Growth control of nerve cells by a protein factor and its anti-serum. Science 143:105-110. Lewin, R. 1988. Cloud over Parkinson's therapy. Science 240:390-392. Madrazo, I., R. Drucker-Colin, V. Diaz, J. Martinez-Mata, G. Torres, and J. J. Becerril. 1987. Open microsurgical autograft of adrenal medulla to the right caudate nucleus in two patients with intractable Parkinson's disease. New England Journal of Medicine 316:831-834.
224 ROGER W. RUSSELL Madrazo I., V. Leon, C. Torres, C. Aguilera, G. Varela, F. Alvarez, A. Fraga, R. Drucker- Colin, F. Ostrosky, M. Skurovich, and R. Franco. 1988. Transplantation of fetal substantia nigra and adrenal medulla to the caudate nucleus in two patients with Parkinson's disease. New England Journal of Medicine 318:51. Mason, D. W., H. M. Charlton, A. Jones, D. M. Perry, and S. J. Simmons. 1985. Immu- nology of allograft rejection in mammals. Pp. 91-98 in Neural Grafting in the Mammalian CNS, A. Bjorklund and U. Stenevi, eds. Amsterdam: Elsevier. Michalek, H., G. M. Bisso, and A. Meneguza. 1981. Comparative studies on rat brain acetylcholinesterase and its molecular forms during intoxication by DEP and paraoxon. Pp. 847-852 in Cholinergic Mechanisms: Phylogaretic Aspects, Central and Periph- eral Synapses and Clinical Significance, G. Papeu and H. Ladinsky, eds. New York: Plenum. National Research Council. 1975. Decision Making for Regulating Chemicals in the Environment. Washington, D.C.: National Academy Press. Newton, M. W., and D. J. Jenden. 1985. Mechanism and subcellular distribution of N- amino-N,N-dimethylaminoethanol (N-aminodeanol) in rat striatal synaptosomes. Journal of Pharmacology and Experimental Therapeutics 235:135-146. Newton, M. W., R. D. Crosland, and D. J. Jenden. 1986. Effects of chronic dietary administration of the cholinergic false precursor N-amino-N,N-dimethylaminoethanol on behavior and cholinergic parameters in rats. Brain Research 373:197-204. Raiteri, M., M. Marchi, and A. M. Caviglia. 1986. Studies on a possible functional coupling between presynaptic acetylcholinesterase and high-affinity choline uptake in the rat brain. Journal of Neurochemistry 47:1696-1699. Rennert, P. D., and G. Heinrich. 1986. Nerve growth factor mRNA in brain: Localiza- tion by in situ hybridization. Biochemistry and Biophysics Research Communica- tions 138:813-818. Russell, R. W. 1958. Effects of "biochemical lesions" on behavior. Acta Psychologica 14:281-294. Russell, R. W. 1966. Biochemical substrates of behavior. Pp. 185-246 in Frontiers in Physiological Psychology, R.W. Russell, ed. New York: Academic Press. Russell, R. W. 1979. Neurotoxins: A systems approach. Pp. 1-7 in Neurotoxins: Fundamental and Clinical Advantages, I. Chubb and L.B. Geffen, eds. Adelaide, South Australia: University of Adelaide Press. Russell, R. W. 1988. Brain "transplants," neurotrophic factors and behavior. Alzheimer Disease and Associated Disorders 2:77-95. Russell, R. W., and J. S. Macri. 1978. Some behavioral effects of suppressing choline transport by cerebroventricular injection of hemicholinium-3. Biochemistry and Pharmacology Behavior 8:399~03. Spencer, P. S. 1987. Guam ALS/Parkinsonism-dementia: A long-latency neurotoxic disorder caused by "slow toxin(s)" in food? Canadian Journal of Neurological Science 14:347-357. Spencer P. S., P. B. Nunn, J. Hugon, A. C. Ludolph, S. M. Ross, D. N. Roy, and R.C. Robertson. 1987. Guam amyotrophic lateral sclerosis-parkinsonism-dementia linked to a plant excitant neurotoxin. Science 237:517-522. Sperry, R. W. 1951. Mechanisms of neural maturation. Pp. 236-280 in Handbook of Experimental Psychology, S. S. Stevens, ed. New York: John Wiley & Sons. Stein, D. G., and E. J. Mufson. 1987. Morphological and behavioral characteristics of embryonic brain tissue transplants in adult, brain damaged subjects. Annals of the New York Academy of Sciences 495:444~71. Struble, R. G., L. C. Cork, P. J. Whitehouse, and D. L. Price. 1982. Cholinergic inner- vation in neuritic plaques. Science 216:413~15.
NEUROBEHAVIORAL TIME BOMBS 225 U.S. National Institutes of Health. 1987. Differential diagnosis of cementing diseases. Washington, D.C.: National Institutes of Health. Varon, S. S., and R. P. Bunge. 1978. Trophic mechanisms in the peripheral nervous system. Annual Review of Neuroscience 1:327-361. Wurtman, R. J., J. K. Blusztajn, and J. C. Maire. 1985. "Autocannibalism" of choline- containing membrane phospholipids in the pathogenesis of Alzheimer's disease- A hypothesis. Neurochemistry International 7:369-372.
