Comparative analysis of fatty acid amide hydrolase and cb(1) cannabinoid receptor expression in the mouse brain: evidence of a widespread role for fatty acid amide hydrolase in regulation of endocannabinoid signaling

Fatty acid amide hydrolase (FAAH) catalyses hydrolysis of the endocannabinoid arachidonoylethanolamide ("anandamide") in vitro and regulates anandamide levels in the brain. In the cerebellar cortex, hippocampus and neocortex of the rat brain, FAAH is located in the somata and dendrites of neurons that are postsynaptic to axon fibers expressing the CB(1) cannabinoid receptor [Proc R Soc Lond B 265 (1998) 2081]. This complementary pattern of FAAH and CB(1) expression provided the basis for a hypothesis that endocannabinoids may function as retrograde signaling molecules at synapses in the brain [Proc R Soc Lond B 265 (1998) 2081; Phil Trans R Soc Lond 356 (2001) 381] and subsequent experimental studies have confirmed this [Science 296 (2002) 678]. To assess more widely the functions of FAAH in the brain and the potential impact of FAAH activity on the spatiotemporal dynamics of endocannabinoid signaling in different regions of the brain, here we have employed immunocytochemistry to compare the distribution of FAAH and CB(1) throughout the mouse brain, using FAAH(-/-) mice as negative controls to validate the specificity of FAAH-immunoreactivity observed in wild type animals. In many regions of the brain, a complementary pattern of FAAH and CB(1) expression was observed, with FAAH-immunoreactive neuronal somata and dendrites surrounded by CB(1)-immunoreactive fibers. In these regions of the brain, FAAH may regulate postsynaptic formation of anandamide, thereby influencing the spatiotemporal dynamics of retrograde endocannabinoid signaling. However, in some regions of the brain such as the globus pallidus and substantia nigra pars reticulata, CB(1) receptors are abundant but with little or no associated FAAH expression and in these brain regions the spatial impact and/or duration of endocannabinoid signaling may be less restricted than in regions enriched with FAAH. A more complex situation arises in several regions of the brain where both FAAH and CB(1) are expressed but in a non-complementary pattern, with FAAH located in neurons and/or oligodendrocytes that are proximal but not postsynaptic to CB(1)-expressing axon fibers. Here FAAH may nevertheless influence endocannabinoid signaling but more remotely. Finally, there are regions of the brain where FAAH-immunoreactive neurons and/or oligodendrocytes occur in the absence of CB(1)-immunoreactive fibers and here FAAH may be involved in regulation of signaling mediated by other endocannabinoid receptors or by receptors for other fatty acid amide signaling molecules. In conclusion, by comparing the distribution of FAAH and CB(1) in the mouse brain, we have provided a neuroanatomical framework for comparative analysis of the role of FAAH in regulation of the spatiotemporal dynamics of retrograde endocannabinoid signaling in different regions of the brain.     

CB1 and CB2 cannabinoid receptor expression during development and in epileptogenic developmental pathologies

Recent data support the involvement of the endocannabinoid signaling in early brain development, as well as a key role of cannabinoid receptors (CBR) in pathological conditions associated with unbalanced neuronal excitability and inflammation. Using immunocytochemistry, we explored the expression and cellular pattern of CBR 1 and 2 (CB1 and CB2) during prenatal human cortical development, as well as in focal malformations of cortical development associated with intractable epilepsy (focal cortical dysplasia; cortical tubers in patients with the tuberous sclerosis complex and glioneuronal tumors). Strong CB1 immunoreactivity was detected in the cortical plate in developing human brain from the earliest stages tested (gestational week 9) and it persisted throughout prenatal development. Both cannabinoid receptors were not detected in neural progenitor cells located in the ventricular zone. Only CB1 was expressed in the subventricular zone and in Cajal–Retzius cells in the molecular zone of the developing neocortex. CB2 was detected in cells of the microglia/macrophage lineage during development. In malformations of cortical development, prominent CB1 expression was demonstrated in dysplastic neurons. Both CBR were detected in balloon/giant cells, but CB2 appeared to be more frequently expressed than CB1 in these cell types. Reactive astrocytes were mainly stained with CB1, whereas cells of the microglia/macrophage lineage were stained with CB2. These findings confirm the early expression pattern of cannabinoid receptors in the developing human brain, suggesting a function for CB1 in the early stages of corticogenesis. The expression patterns in malformations of cortical development highlight the role of cannabinoid receptors as mediators of the endocannabinoid signaling and as potential pharmacological targets to modulate neuronal and glial cell function in epileptogenic developmental pathologies.     

Immunohistochemical distribution of the cannabinoid receptor 1 and fatty acid amide hydrolase in the dog claustrum

Cannabinoid receptor 1 (CB1R) and fatty acid amide hydrolase (FAAH) are part of the endocannabinoid system (ECB) which exerts a neuromodulatory activity on different brain functions and plays a key role in neurogenesis. Although many studies have reported FAAH and CB1R expression in the brain of different animal species, to the best of our knowledge they have never been described in the canine claustrum. Claustrum samples, obtained from necropsy of four neurologically normal dogs, were formalin fixed for paraffin embedding. Sections were either stained for morpho-histological analysis or immunostained for CB1R and FAAH. Analysis of adjacent sections incubated with the two antisera showed a complementary labeling pattern in the claustrum, with CB1R antibody staining fibers while anti-FAAH antibody stained cell bodies and the proximal portion of dendrites; this particular anatomical relationship suggests a retrograde endocannabinoid action via CB1R. CB1R and FAAH complementary immunostaining and their cellular localization reported here provide the first anatomical evidence for existence of the ECB in the dog claustrum.     

Role of water-soluble amyloid-beta in the pathogenesis of Alzheimer's disease

Water-soluble amyloid-beta (wsAbeta) is present in cerebral cortex of subjects at risk of Alzheimer's disease (AD) as well as in normal elderly subjects as a mixture of three major amyloid-beta (Abeta) species: 1-42, py3-42 and py11-42. The three wsAbeta species are nondetectable in brains of young people, free of immunohistochemically detectable amyloid plaques. In the brains of Down's syndrome and APP-mutant transgenic mice, wsAbeta appears long time before amyloid deposition, indicating that it represent the first form of Abeta aggregation and accumulation. In normal brain, wsAbeta is bound to apolipoprotein E that favours its degradation by proteases. The composition of wsAbeta, in terms of the ratio between the full-length 1-42 and the py3-42 peptides, correlates with the severity of clinical and pathological phenotype in familial early onset AD. Water-soluble Abeta is the native counterpart of the Abeta small aggregates (soluble oligomers) that show in vitro an early and high neuronal toxicity.     

Differential regulation of Abeta42-induced neuronal C1q synthesis and microglial activation

Expression of C1q, an early component of the classical complement pathway, has been shown to be induced in neurons in hippocampal slices, following accumulation of exogenous Abeta42. Microglial activation was also detected by surface marker expression and cytokine production. To determine whether C1q induction was correlated with intraneuronal Abeta and/or microglial activation, D-(-)-2-amino-5-phosphonovaleric acid (APV, an NMDA receptor antagonist) and glycine-arginine-glycine-aspartic acid-serine-proline peptide (RGD, an integrin receptor antagonist), which blocks and enhances Abeta42 uptake, respectively, were assessed for their effect on neuronal C1q synthesis and microglial activation. APV inhibited, and RGD enhanced, microglial activation and neuronal C1q expression. However, addition of Abeta10-20 to slice cultures significantly reduced Abeta42 uptake and microglial activation, but did not alter the Abeta42-induced neuronal C1q expression. Furthermore, Abeta10-20 alone triggered C1q production in neurons, demonstrating that neither neuronal Abeta42 accumulation, nor microglial activation is required for neuronal C1q upregulation. These data are compatible with the hypothesis that multiple receptors are involved in Abeta injury and signaling in neurons. Some lead to neuronal C1q induction, whereas other(s) lead to intraneuronal accumulation of Abeta and/or stimulation of microglia.     

Overexpression of monocyte chemotactic protein-1/CCL2 in beta-amyloid precursor protein transgenic mice show accelerated diffuse beta-amyloid deposition

Microglia accumulation at the site of amyloid plaques is a strong indication that microglia play a major role in Alzheimer's disease pathogenesis. However, how microglia affect amyloid-beta peptide (Abeta) deposition remains poorly understood. To address this question, we developed a novel bigenic mouse that overexpresses both amyloid precursor protein (APP) and monocyte chemotactic protein-1 (MCP-1; CCL2 in systematic nomenclature). CCL2 expression, driven by the glial fibrillary acidic protein promoter, induced mononuclear phagocyte (MP; monocyte-derived macrophage and microglial) accumulation in the brain. When APP/CCL2 transgenic mice were compared to APP mice, a fivefold increase in Abeta deposition was present despite increased MP accumulation around hippocampal and cortical amyloid plaques. Levels of full-length APP, its C-terminal fragment, and Abeta-degrading enzymes (insulin-degrading enzyme and neprilysin) in APP/CCL2 and APP mice were indistinguishable. Sodium dodecyl sulfate-insoluble Abeta (an indicator of fibrillar Abeta) was increased in APP/CCL2 mice at 5 months of age. Apolipoprotein E, which enhances Abeta deposition, was also increased (2.2-fold) in aged APP/CCL2 as compared to APP mice. We propose that although CCL2 stimulates MP accumulation, it increases Abeta deposition by reducing Abeta clearance through increased apolipoprotein E expression. Understanding the mechanisms underlying these events could be used to modulate microglial function in Alzheimer's disease and positively affect disease outcomes.     

Comparative analysis of cortical gene expression in mouse models of Alzheimer's disease

Three mouse models of Alzheimer's disease (AD) were used to assess changes in gene expression potentially critical to amyloid beta-peptide (Abeta)-induced neuronal dysfunction. One mouse model harbored homozygous familial AD (FAD) knock-in mutations in both, amyloid precursor protein (APP) and presenilin 1 (PS-1) genes (APP(NLh/NLh)/PS-1(P264L/P264L)), the other two models harbored APP over-expression of FAD mutations (Tg2576) with the PS-1 knock-in mutation at either one or two alleles. These mouse models of AD had varying levels of Abeta40 and Abeta42 and different latencies and rates of Abeta deposition in brain. To assess changes in gene expression associated with Abeta accumulation, the Affymetrix murine genome array U74A was used to survey gene expression in the cortex of these three models both prior to and following Abeta deposition. Altered genes were identified by comparing the AD models with age-matched control littermates. Thirty-four gene changes were identified in common among the three models in mice with Abeta deposition. Among the up-regulated genes, three major classes were identified that encoded for proteins involved in immune responses, carbohydrate metabolism, and proteolysis. Down-regulated genes of note included pituitary adenylate cyclase-activating peptide (PACAP), brain-derived neurotrophic factor (BDNF), and insulin-like growth factor I receptor (IGF-IR). In young mice without detectable Abeta deposition, there were no regulated genes common among the three models, although 40 genes were similarly altered between the two Tg2576 models with the PS-1 FAD knock-in. Finally, changes in gene expression among the three mouse models of AD were compared with those reported in human AD samples. Sixty-nine up-regulated and 147 down-regulated genes were found in common with human AD brain. These comparisons across different genetic mouse models of AD and human AD brain provide greater support for the involvement of identified gene expression changes in the neuronal dysfunction and cognitive deficits accompanying amyloid deposition in mammalian brain.     

Soluble Abeta homeostasis in AD and DS: impairment of anti-amyloidogenic protection by lipoproteins

In order to assess whether lipoproteins are physiologically able to balance and modulate the sAbeta homeostasis in vivo, soluble Abeta levels in lipoprotein-depleted plasma were measured as a function of age in normal controls, Alzheimer's disease (AD) patients, and Down's syndrome (DS) cases. The reshaping of sAbeta homeostasis, in particular the sAbeta42-lipoprotein interaction, takes place over normal-60's, whereas mild AD patients appear to have impaired this anti-amyloidogenic mechanism resulting in a significant increase of lipoprotein-free sAbeta42. Similar loss of function takes place in Down's syndrome patients. Lipoprotein-free sAbeta remains significantly elevated from the pre-symptomatic through the symptomatic stages of the disease, and declines with the progression of the AD-like pathology. The dissociation of sAbeta from lipoprotein-particles also occurs in brain parenchyma and the presence of soluble dimeric lipoprotein-free Abeta prior to its parenchymal deposition in AD brains would support the hypothesis that functionally declined lipoproteins may be major determinants in the production of metabolic conditions leading to higher levels of sAbeta species and cerebral amyloidosis.     

Expression of the cannabinoid CB2 receptor in the rat cerebellum: an immunohistochemical study

Reports of cannabinoid CB2 receptor protein in the brain have been ambiguous. We therefore tested for CB2 immunoreactivity in the rat brain using immunofluorescence. We detected CB2 labeling in fine fibers in the granule layer. This CB2 labeling did not co-localise with the astrocyte marker glial fibrillary acidic protein (GFAP) and, therefore, the CB2-positive fibers were not astrocytes and were possibly microglial or neuronal. Additionally, strong CB2 labeling was detected in capillary endothelia in the granule, Purkinje cell, and molecular layers. Our results suggest that the role of CB2 receptors in the brain may have been previously underestimated.     

Cannabinoid CB1 receptor protein expression in the rat choroid plexus: a possible involvement of cannabinoids in the regulation of cerebrospinal fluid

Cannabinoid CB1 receptors in the brain are expressed on axon terminals presynaptic to neurons that express fatty acid amide hydrolase (FAAH). Postsynaptic FAAH catabolizes endocannabinoids which act as short-range transmitters. It has been previously shown that FAAH is also expressed in the epithelial cells of the choroid plexus. Using immunohistochemisty, we found that CB1 receptor protein is also expressed in choroid plexus epithelia. This is consistent with the hypothesis that FAAH in choroid plexus epithelial cells catabolizes endocannabinoids close to their site of action. Cannabinoids may then act directly on choroid plexus cells, and thereby contribute to the regulation of the composition of the CSF.     

Neuronal or glial expression of human apolipoprotein e4 affects parenchymal and vascular amyloid pathology differentially in different brain regions of double- and triple-transgenic mice

Apolipoprotein E4 (ApoE4) is associated with Alzheimer's disease by unknown mechanisms. We generated six transgenic mice strains expressing human ApoE4 in combination with mutant amyloid precursor protein (APP) and mutant presenilin-1 (PS1) in single-, double-, or triple-transgenic combinations. Diffuse, but not dense, amyloid plaque-load in subiculum and cortex was increased by neuronal but not glial ApoE4 in old (15 months) double-transgenic mice, whereas both diffuse and dense plaques formed in thalamus in both genotypes. Neuronal and glial ApoE4 promoted cerebral amyloid angiopathy as extensively as mutant PS1 but with pronounced regional differences: cortical angiopathy was induced by neuronal ApoE4 while thalamic angiopathy was again independent of ApoE4 source. Angiopathy correlated more strongly with soluble Abeta40 and Abeta42 levels in cortex than in thalamus throughout the six genotypes. Neither neuronal nor glial ApoE4 affected APP proteolytic processing, as opposed to mutant PS1. Neuronal ApoE4 increased soluble amyloid levels more than glial ApoE4, but the Abeta42/40 ratios were similar, although significantly higher than in single APP transgenic mice. We conclude that although the cellular origin of ApoE4 differentially affects regional amyloid pathology, ApoE4 acts on the disposition of amyloid peptides downstream from their excision from APP but without induction of tauopathy.     

Slow cortical potential consequences of elctroconvulsive shock in rats

Slow potential (SP) changes were recorded from the dura of rats following the administration of transpinnate or transcortical electroconvulsive shock (ECS). A relatively brief negative wave of several mV was followed by a prolonged positive SP shift when seizures were induced. Several quantitative characteristics of the pattern of shifts could be related to the intensity of the ECS even at suprathreshold levels for induction of a full behavioral convulsion. Low intensity ECS sometimes resulted in spreading depression waves without a behavioral convulsion. It was conjectured that the pattern of shifts might be related to the severity of retrograde amnesia reported by others to be induced by ECS.     

Metallothionein-I and -III expression in animal models of Alzheimer disease

Previous studies have described altered expression of metallothioneins (MTs) in neurodegenerative diseases like multiple sclerosis (MS), Down syndrome, and Alzheimer's disease (AD). In order to gain insight into the possible role of MTs in neurodegenerative processes and especially in human diseases, the use of animal models is a valuable tool. Several transgenic mouse models of AD amyloid deposits are currently available. These models express human beta-amyloid precursor protein (AbetaPP) carrying different mutations that subsequently result in a varied pattern of beta-amyloid (Abeta) deposition within the brain. We have evaluated the expression of MT-I and MT-III mRNA by in situ hybridization in three different transgenic mice models of AD: Tg2576 (carrying AbetaPP harboring the Swedish K670N/M671L mutations), TgCRND8 (Swedish and the Indiana V717F mutations), and Tg-SwDI (Swedish and Dutch/Iowa E693Q/D694N mutations). MT-I mRNA levels were induced in all transgenic lines studied, although the pattern of induction differed between the models. In the Tg2576 mice MT-I was weakly upregulated in cells surrounding Congo Red-positive plaques in the cortex and hippocampus. A more potent induction of MT-I was observed in the cortex and hippocampus of the TgCRND8 mice, likely reflecting their higher amyloid plaques content. MT-I upregulation was also more significant in Tg-SwDI mice, especially in the subiculum and hippocampus CA1 area. Immunofluorescence stainings demonstrate that astrocytes and microglia/macrophages surrounding the plaques express MT-I&II. In general, MT-I regulation follows a similar but less potent response than glial fibrillary acidic protein (GFAP) expression. In contrast to MT-I, MT-III mRNA expression was not significantly altered in any of the models examined suggesting that the various MT isoforms may have different roles in these experimental systems, and perhaps also in human AD.     

Time-dependent increase in basic fibroblast growth factor protein in limbic regions following electroshock seizures

Brief experimentally induced seizures have been shown to increase the expression of mRNA encoding basic fibroblast growth factor (FGF-2) in specific brain regions. However, the extent to which this change in mRNA affects the expression of FGF-2 protein in these brain regions has not been examined. In the present study, we exposed rats to brief non-injurious seizures to determine whether this treatment would lead to an increase in FGF-2 protein expression in selected brain regions. Because initial results indicated that the elevation of FGF-2 protein was not significant following acute seizure exposure, we examined both acute and chronic seizure treatment to determine whether FGF-2 protein expression could be increased under conditions of repeated seizures. Brief limbic seizures were induced by minimal electroconvulsive shock (ECS) given as daily treatments for 1 (acute) or 7 (chronic) days. FGF-2 protein was measured in hippocampus, rhinal cortex, frontal cortex, and olfactory bulb at 20, 48, and 72 h following the last seizure.No significant increases in FGF-2 protein were observed in any region following acute ECS. In the chronic ECS-treated groups, significantly elevated FGF-2-like immunoreactivity was found in the frontal and rhinal cortex as compared with the same regions from both control and acute ECS animals. Increases after chronic ECS were maximal at 20 h, and remained significantly elevated as long as 72 h. These increases were predominantly observed for the 24-kDa and 22/22.5-kDa FGF-2 isoforms.Because chronic ECS, which has been shown to be protective against neuronal cell death, induced significantly more FGF-2 immunoreactivity than did acute ECS, we suggest that FGF-2 expression may be an important substrate for the neuroprotective action of non-injurious seizures. A prolonged induction of the high molecular weight isoforms of FGF-2, as occurs after chronic ECS, may selectively reduce the vulnerability of certain brain regions to a variety of neurodegenerative insults.     

Interferon-gamma and tumor necrosis factor-alpha regulate amyloid-beta plaque deposition and beta-secretase expression in Swedish mutant APP transgenic mice

Reactive astrocytes and microglia in Alzheimer's disease surround amyloid plaques and secrete proinflammatory cytokines that affect neuronal function. Relationship between cytokine signaling and amyloid-beta peptide (Abeta) accumulation is poorly understood. Thus, we generated a novel Swedish beta-amyloid precursor protein mutant (APP) transgenic mouse in which the interferon (IFN)-gamma receptor type I was knocked out (APP/GRKO). IFN-gamma signaling loss in the APP/GRKO mice reduced gliosis and amyloid plaques at 14 months of age. Aggregated Abeta induced IFN-gamma production from co-culture of astrocytes and microglia, and IFN-gamma elicited tumor necrosis factor (TNF)-alpha secretion in wild type (WT) but not GRKO microglia co-cultured with astrocytes. Both IFN-gamma and TNF-alpha enhanced Abeta production from APP-expressing astrocytes and cortical neurons. TNF-alpha directly stimulated beta-site APP-cleaving enzyme (BACE1) expression and enhanced beta-processing of APP in astrocytes. The numbers of reactive astrocytes expressing BACE1 were increased in APP compared with APP/GRKO mice in both cortex and hippocampus. IFN-gamma and TNF-alpha activation of WT microglia suppressed Abeta degradation, whereas GRKO microglia had no changes. These results support the idea that glial IFN-gamma and TNF-alpha enhance Abeta deposition through BACE1 expression and suppression of Abeta clearance. Taken together, these observations suggest that proinflammatory cytokines are directly linked to Alzheimer's disease pathogenesis.     

Fatty acid amide hydrolase is located preferentially in large neurons in the rat central nervous system as revealed by immunohistochemistry

The distribution in the rat brain of fatty acid amide hydrolase (FAAH) an enzyme that catalyzes the hydrolysis of the endogenous cannabinoid anandamide was studied by immunohistochemistry. An immunopurified, polyclonal antibody to the C terminal region of FAAH was used in these studies. The large principal neurons, such as pyramidal cells in the cerebral cortex, the pyramidal cells the hippocampus, Purkinje cells in the cerebellar cortex and the mitral cells in the olfactory bulb, showed the strongest FAAH immunoreactivity. These FAAH-containing principal neurons except the mitral cells in the olfactory bulb are in close proximity with cannabinoid CB1 receptors as revealed by our previous immunohistochemical study. Moderately or lightly stained FAAH-containing neurons were also found in the amygdala, the basal ganglia, the deep cerebellar nuclei, the ventral posterior nuclei of the thalamus, the optic layer and the intermediate white layer of the superior colliculus and the red nucleus in the midbrain, and motor neurons of the spinal cord. These data demonstrate that FAAH is heterogeneously distributed and this distribution exhibits considerable, although not complete, overlap with the distribution of cannabinoid CB1 receptors in rat brain.     

Compartment-specific localization of cannabinoid 1 (CB1) and mu-opioid receptors in rat nucleus accumbens

Interactions between cannabinoid and opioid systems have been implicated in reward and drug seeking behaviors involving neuronal circuitry in the nucleus accumbens (Acb) shell and core. To determine the relevant sites, we examined the electron microscopic localization of cannabinoid type-1 (CB1) receptors and mu-opioid receptors in each Acb compartment in rat brain. CB1 receptor immunogold labeling was seen on the plasma membrane and within the cytoplasm of neuronal and glial profiles throughout the Acb. These neuronal profiles included somata and dendrites as well as axon terminals, many of which formed excitatory-type, asymmetric synapses with notable perforations that are often associated with synaptic plasticity. The number of CB1-labeled terminals within the neuropil of the Acb shell was significantly greater than in the core. Mu-opioid receptors were also detected in axonal and dendritic profiles. These dendrites were most prevalent in the Acb shell, where mu-receptors also were located in 21% of the dendritic profiles and 3% of the axon terminals containing CB1 receptors. More of the CB1-labeled terminals contacted dendrites expressing mu-opioid receptors in the shell (19%) compared with the core (13%). Conversely, of the synaptic mu-labeled terminals, 20% in the shell and 10% in the core contacted dendrites containing CB1 receptors. These findings provide ultrastructural evidence that cannabinoid-opioid interactions are mediated by activation of CB1 and mu-opioid receptors within the same or synaptically linked neurons in the Acb shell and core. They also suggest a particularly important role for presynaptic CB1 receptors in the reward circuit of the Acb shell.     

Amyloid A4 protein and its precursor in Down's syndrome and Alzheimer's disease

In patients with Alzheimer's disease, amyloid fibrils that are aggregates of A4 protein subunits are deposited in the brain. A similar process occurs at an earlier age in persons with Down's syndrome. To investigate the deposition of amyloid in these diseases, we used a radioimmunoassay to measure levels of the amyloid precursor (PreA4) in the serum of 17 patients with Down's syndrome, 15 patients with Alzheimer's disease, and 33 normal elderly controls. The mean (+/- SD) concentration of serum PreA4 was increased 1.5-fold in patients with Down's syndrome (2.49 +/- 1.13 nmol per liter) as compared with that in controls (1.68 +/- 0.49 nmol per liter; P less than 0.007); the levels in patients with Alzheimer's disease were similar to those in controls (1.83 +/- 0.78; P less than 0.98). We also found that the concentration of PreA4 in the brain tissue of two adults with Down's syndrome (100 and 190 pmol per gram) was higher than that in the brain tissue of either 26 patients with Alzheimer's disease (64.4 +/- 17.3 pmol per gram) or 17 elderly controls with neurologic disease (68.5 +/- 26.3 pmol per gram). Immunocytochemical studies of brain tissue from 26 patients with Down's syndrome showed that the deposition of A4 protein amyloid began in these patients approximately 50 years earlier than it began in 127 normal aging subjects studied previously, although the rate of deposition was the same. We conclude that, since the gene for PreA4 is on the long arm of chromosome 21, which is present in triplicate in Down's syndrome, overexpression of this gene may lead to increased levels of PreA4 and amyloid deposition in Down's syndrome. However, since increased levels of PreA4 are not present in Alzheimer's disease, additional factors must account for the amyloid deposition in that disorder.