Human APOE protein localized in brains of transgenic mice

Transgenic mice carrying the three common human apolipoprotein E (APOE) alleles have been developed. In this study, brains of the transgenic mice have been analyzed by in situ histohybridization, immunohistochemistry, and immunoblots to determine sites of gene expression, to identify specific brain cells associated with human apoE protein, and to determine the relative concentrations of the human apoE. Results indicate that (1) human APOE mRNA and apoE protein occur in the gray and white matter of transgenic mouse brains; (2) in the hippocampus of transgenic brains, human apoE protein reacts immunologically within the same cells as the glial fibrillary acidic protein (GFAP), a specific marker for astrocytes; and (3) concentrations of the apoE isoforms determined in three heterozygous transgenic brains range from 22 to 250 pmol/g wet weight of brain.     

Apolipoprotein E isoform-specific regulation of dendritic spine morphology in apolipoprotein E transgenic mice and Alzheimer's disease patients

Dendritic spines are postsynaptic sites of excitatory input in the mammalian nervous system. Apolipoprotein (apo) E participates in the transport of plasma lipids and in the redistribution of lipids among cells. A role for apoE is implicated in regeneration of synaptic circuitry after neural injury. The apoE4 allele is a major risk factor for late-onset familial and sporadic Alzheimer's disease (AD) and is associated with a poor outcome after brain injury. ApoE isoforms are suggested to have differential effects on neuronal repair mechanisms. In vitro studies have demonstrated the neurotrophic properties of apoE3 on neurite outgrowth. We have investigated the influence of apoE genotype on neuronal cell dendritic spine density in mice and in human postmortem tissue. In order to compare the morphology of neurons developing under different apoE conditions, gene gun labeling studies of dendritic spines of dentate gyrus (DG) granule cells of the hippocampus were carried out in wild-type (WT), human apoE3, human apoE4 expressing transgenic mice and apoE knockout (KO) mice; the same dendritic spine parameters were also assessed in human postmortem DG from individuals with and without the apoE4 gene. Quantitative analysis of dendritic spine length, morphology, and number was carried out on these mice at 3 weeks, 1 and 2 years of age. Human apoE3 and WT mice had a higher density of dendritic spines than human E4 and apoE KO mice in the 1 and 2 year age groups (P<0.0001), while at 3 weeks there were no differences between the groups. These age dependent differences in the effects of apoE isoforms on neuronal integrity may relate to the increased risk of dementia in aged individuals with the apoE4 allele. Significantly in human brain, apoE4 dose correlated inversely with dendritic spine density of DG neurons cell in the hippocampus of both AD (P=0.0008) and aged normal controls (P=0.0015). Our findings provide one potential explanation for the increased cognitive decline seen in aged and AD patients expressing apoE4.     

Cellular distribution of insulin-like growth factor-II/mannose-6-phosphate receptor in normal human brain and its alteration in Alzheimer's disease pathology

The insulin-like growth factor-II/mannose-6-phosphate (IGF-II/M6P) receptor is a multifunctional membrane glycoprotein, which binds different classes of ligands including IGF-II and M6P-bearing lysosomal enzymes. Besides participating in the process of endocytosis this receptor functions in the trafficking of lysosomal enzymes from the trans-Glogi network (TGN) or the cell surface to lysosomes. In Alzheimer's disease (AD) brain, marked overexpression of certain lysosomal enzymes in vulnerable neuronal populations and their association to beta-amyloid (Abeta) containing neuritic plaques has been correlated to altered metabolic functions. In the present study, we measured the levels of IGF-II/M6P receptor and characterized its distribution profile in selected regions of AD and age-matched normal postmortem brains. Western blot analysis revealed no significant alteration in the levels of IGF-II/M6P receptor either in the hippocampus, frontal cortex or cerebellum between AD and age-matched control brains. However, a significant gene dose effect of apolipoprotein E (APOE) epsilon4 allele on IGF-II/M6P receptor levels was evident in the hippocampus of the AD brain. At the cellular level, immunoreactive IGF-II/M6P receptors were localized in the neurons of the frontal cortex, hippocampus and cerebellum of control brains. In AD brains, the labeling of the neurons was less intense in the frontal cortex and hippocampus than in the age-matched control brains. Additionally, IGF-II/M6P receptor immunoreactivity was observed in association with a subpopulation of Abeta-containing neuritic plaques as well as tau-positive neurofibrillary tangles both in the frontal cortex and the hippocampus. Reactive glial cells localized adjacent to the plaques also occasionally exhibited IGF-II/M6P receptor immunoreactivity. These results, when analyzed in context of the established role of the IGF-II/M6P receptor in the regulation of the intracellular trafficking of lysosomal enzymes, suggest that alterations in IGF-II/M6P receptor levels/distribution are possibly associated with altered functioning of the lysosomal enzymes and/or loss of neurons observed in AD brains, especially in patients carrying APOE epsilon4 alleles.     

Regional and neuronal reductions of polyadenylated messenger RNA in Alzheimer's disease.

Messenger RNA (mRNA) is the key intermediate in the gene expression pathway. The amount of mRNA in Alzheimer's disease (AD) brains has been determined using in situ hybridization histochemistry (ISHH) to detect the poly(A) tails of polyadenylated mRNA (poly(A) + mRNA). On a regional basis, AD cases had significantly less poly(A) + mRNA than controls in hippocampus (field CA3) and cerebellum (granule cell layer). Analysis of constituent pyramidal neurons showed mean reductions per cell within AD hippocampus (field CA3) and temporal cortex, but not in visual cortex. Similar changes were seen in a small group of non-AD dementias. The finding of reduced poly(A) + mRNA content is another indication of the altered brain gene expression occurring in AD. It is proposed that measurement of poly(A) + mRNA may be valuable in identifying functionally impaired neuronal populations. The methodology also provides a means by which changes in the quantitative distribution of individual mRNAs can be determined relative to that of poly(A) + mRNA as a whole.     

Apolipoprotein E receptors and amyloid expression are modulated in an apolipoprotein E-dependent fashion in response to hippocampal deafferentation in rodent

The entorhinal cortex lesion paradigm is a widely accepted and efficient method to provoke reactive synaptogenesis and terminal remodeling in the adult CNS. This approach has been used successfully to contrast the profile of reactivity from various proteins associated with Alzheimer's disease pathophysiology in wild-type and apolipoprotein E (apoE)-deficient (APOE ko) mice. Results indicate that the production of the beta-amyloid 1-40 peptide (A beta 40) is increased in response to neuronal injury, with a timing that is different between wild-type and APOE ko animals. Moreover, we report that baseline levels of the A beta 40 peptide are significantly higher in the APOE ko mice. The expression of the apolipoprotein E receptor type 2 (apoER2) is also modulated by the deafferentation process in the hippocampus, but only in APOE ko mice. These results provide novel insights as to the molecular mechanisms responsible for the poor plastic response reported in apoE4-expressing and apoE deficient mice in response to hippocampal injury.     

The role of apolipoprotein E in Alzheimer's disease, acute brain injury and cerebrovascular disease: evidence of common mechanisms and utility of animal models

The epsilon 4 allele of apolipoprotein E (APOE denotes gene; apoE denotes protein) is a major risk factor for Alzheimer's disease (AD). More recent evidence indicates an association with a poor outcome after acute brain injury including that due to head trauma and intracerebral hemorrhage. APOE gene polymorphism also influences the risk of hemorrhage in cerebral amyloid angiopathy. These diverse brain disorders seem to have some mechanisms in common. The multiplicity of the roles of apoE within the central nervous system is currently being unraveled. For example, apoE can interact with amyloid beta-protein and tau, proteins central to the pathogenesis of AD. In addition to these effects, it is proposed that one of the major functions of apoE is to mediate neuronal protection, repair and remodeling. In all of the different roles proposed, there are marked apoE-isoform specific differences. Although it remains to be clarified which is the most important mechanism(s) in each disorder in which apoE is involved, these isoform specific differences seem to underly a genetically determined susceptibility to outcome from acute brain injury and to AD with APOE epsilon 4 conferring relative vulnerability. This review focuses on apoE research, from clinical studies to animal models, in AD, acute brain injury and cerebrovascular disease and explores the common mechanisms that may explain some of the complex underlying neurobiology.     

Reduced levels of human apoE4 protein in an animal model of cognitive impairment

The APOE4 allele is the most common genetic determinant for Alzheimer's disease (AD) in the developed world. APOE genotype specific differences in brain apolipoprotein E protein levels have been observed in numerous studies since the discovery of APOE4's link to AD. Since the human apoE4 targeted replacement mice display characteristics of cognitive impairment we sought to determine if reduced levels of apoE might provide one explanation for this impairment. We developed a novel mass spectrometry method to measure apoE protein levels in plasma. Additionally, we developed an ELISA that replicates the mass spectrometry data and enables the rapid quantitation of apoE in plasma, brain and cerebrospinal fluid. We detected a significant decrease in plasma, brain and cerebrospinal fluid apoE levels in the apoE4 mice compared to apoE2 and E3 mice. We also measured a small (∼19%) decrease in brain apoE levels from aged, non-demented APOE4 carriers. Our findings suggest that a fraction of APOE4-linked AD may be due to insufficient levels of functional apoE required to maintain neuronal health.     

APOE and the regulation of microglial nitric oxide production: a link between genetic risk and oxidative stress

The mechanism linking the APOE4 gene with increased susceptibility for Alzheimer's disease (AD) and poorer outcomes following closed head injury and stroke is unknown. One potential link is activation of the innate immune system in the CNS. Our previously published data demonstrated that apolipoprotein E regulates production of nitric oxide, a critical cytoactive factor released by immune active macrophages. To determine if immune regulation is different in the presence of apolipoprotein E4 compared to apolipoprotein E3, we have measured NO production by peritoneal and CNS macrophages (microglia) cultured from transgenic mice that only express the human apoE4 or apoE3 protein isoform. Significantly more NO was produced in APOE4 mice compared to APOE3 transgenic mice that only express human apoE3 protein. Similarly, monocyte derived macrophages from humans carrying APOE4 gene alleles also produce significantly greater NO than those individuals with APOE3. The mechanism for this isoform-specific difference in NO production is not known and multiple sites in the NO production pathway may be affected. Expression of inducible nitric oxide synthase (iNOS) mRNA and protein are not significantly different between the APOE3 and APOE4 mice, suggesting that induction of iNOS is not a primary cause of the increased NO production in APOE4 animals. One alternative regulatory mechanism that demonstrates isoform specificity is arginine transport, which is greater in microglia from APOE4 transgenic mice compared to microglia from APOE3 mice. Increased transport is consistent with an increased production of NO and may reflect a direct or indirect effect of the APOE genotype on microglial arginine uptake and microglial activation in general. Overall, greater NO production in APOE4 carriers where characteristically high levels of oxidative/nitrosative stress are found in diseases such as AD provides a mechanism that potentially explains the genetic association between APOE4 and human diseases.     

Marked regional differences of brain human apolipoprotein E expression in targeted replacement mice

We used three human apolipoprotein (apo) E targeted replacement mouse lines, each expressing one of the three common human apoE isoforms to study the pattern of apoE expression in the central nervous system (CNS). Immunocytochemistry on brain sections from all three lines of targeted replacement mice, wild type mice, African green monkeys, and humans show a predominantly glial pattern of apoE expression. The levels of human apoE protein in hippocampus and frontal cortex were similar between targeted replacement mice and non-demented human tissue. Within a given brain region, the levels of apoE were very similar amongst all three isoforms, which contrasts sharply with plasma, where apoE2 levels are 16-fold higher than apoE3 and E4 levels. Across brain regions, cerebellar apoE levels were significantly higher than cerebral apoE levels. In conclusion, we provide detailed analysis of a human apoE animal model system that recapitulates both the pattern and level of apoE expression in non-demented humans. The neurobiology of human apoE isoforms can now be studied in both the normal and post-injury state, since all apoE regulatory sequences are intact. Finally, the differences in apoE levels we observed may explain the regional vulnerability of neuronal degeneration in Alzheimer's disease.     

Comparison of neuronal inositol 1,4,5-trisphosphate 3-kinase and receptor mRNA distributions in the adult rat brain using in situ hybridization histochemistry.

As a result of its interaction with a specific receptor, inositol 1,4,5-trisphosphate mobilizes intracellular calcium. The metabolism of inositol 1,4,5-trisphosphate is rather complex: inositol 1,4,5-trisphosphate 3-kinase produces inositol 1,3,4,5-tetrakisphosphate, a putative second messenger. In order to elucidate inositol 1,3,4,5-tetrakisphosphate function, a comparative in situ hybridization study of the distributions of inositol 1,4,5-trisphosphate 3-kinase and receptor mRNAs was performed in the adult rat brain using oligonucleotides derived from their cDNA sequences. The neuronal distributions of the mRNA for the receptor were larger than for the kinase. Highest levels of both mRNAs were found in the cerebellar Purkinje cells, where they were enriched in their neuronal perikarya and to a lesser extent in their dendrites. In addition to the cerebellum, mRNAs were mainly detected in the hippocampal pyramidal cells of the CA1 sector of the Ammon's horn and in the granule cells of the dentate gyrus, and also in a majority of the neurons in the cortical layers II-III and V, especially in the frontal cortex and cingulate cortex; caudate-putamen, accumbens, olfactory tubercle and Calleja islets; claustrum; anterior olfactory nucleus; taenia tecta; piriform cortex; dorsolateral septum; bed nucleus stria terminalis; amygdala; hippocampal CA2-4 sectors and subiculum. The inositol 1,4,5-trisphosphate receptor mRNA but not kinase mRNA was found in a majority of the neurons in the thalamus, especially in the parafascicular nucleus; hypothalamus, especially the medial hypothalamus; substantia nigra pars compacta and ventral tegmental area; superior colliculus; lateral interpeduncular nucleus and central gray. Taking into account the limitation in sensitivity of the technique, both mRNAs were not detected in glial cells and in the olfactory bulb; basal nucleus of Meynert, diagonal band nuclei; medial septal nucleus; substantia innominata; globus pallidus; entopeduncular nucleus; substantia nigra pars reticulata; ventral pallidum; subthalamic nucleus; spinal cord and dorsal root ganglia. In conclusion, cerebellum and hippocampus appear to contain almost similar levels of kinase mRNA. This is in contrast to receptor mRNA levels which were at much higher levels in the cerebellum when compared with the hippocampus. For this reason, we have chosen hippocampal CA1 pyramidal cells and dentate gyrus granule cells for studying inositol 1,4,5-trisphosphate 3-kinase function.     

An apolipoprotein E-based therapeutic improves outcome and reduces Alzheimer's disease pathology following closed head injury: evidence of pharmacogenomic interaction

Apolipoprotein E (apoE) modifies glial activation and the CNS inflammatory response in an isoform-specific manner. Peptides derived from the receptor-binding region of apoE have been demonstrated to maintain the functional activity of the intact protein, and to improve histological and functional deficits after closed head injury. In the current study, APOE2, APOE3, and APOE4 targeted replacement (TR) mice expressing the human apoE protein isoforms (apoE2, apoE3 and apoE4) were used in a clinically relevant model of closed head injury to assess the interaction between the humanized apoE background and the therapeutic apoE mimetic peptide, apoE(133-149). Treatment with the apoE-mimetic peptide reduced microglial activation and early inflammatory events in all of the targeted replacement animals and was associated with histological and functional improvement in the APOE2TR and APOE3TR animals. Similarly, brain beta amyloid protein (Abeta)(1-42) levels were increased as a function of head injury in all of the targeted replacement mice, while treatment with apoE peptide suppressed Abeta(1-42) levels in the APOE2TR and APOE3TR animals. These results suggest a pharmacogenomic interaction between the therapeutic effects of the apoE mimetic peptide and the human apoE protein isoforms. Furthermore, they suggest that administration of apoE-mimetic peptides may serve as a novel therapeutic strategy for the treatment of acute and chronic neurological disease.     

c-fos protein-like immunoreactivity: distribution in the human brain and over-expression in the hippocampus of patients with Alzheimer's disease.

c-fos protein-like immunoreactivity was investigated in the human brain post mortem, using a polyclonal antiserum raised against the N-terminal conserved peptide of c-fos protein. Immunostaining was found in the cerebral cortex, hippocampus, striatum, thalamus and cerebellum but not in the upper brainstem and the adrenal gland. c-fos-like immunoreactivity predominated in neuronal elements, but was also observed in neuropil and glial cells. In addition to a nuclear localization, the staining could be seen in neuronal dendrites (i.e. in the pyramidal cells of hippocampus or in some cortical areas). In order to analyse the effect of brain injury on c-fos expression, the characteristics of the immunostaining were analysed in the hippocampus of patients deceased with Alzheimer's disease known to be associated with a preferential vulnerability of the pyramidal neurons. No staining was observed in the senile plaques or in neurofibrillary tangles, the histopathological stigmata of the disease. Densitometric measurement of the intensity of c-fos-like staining revealed a significant increase in the hilus, the fimbria and the CA1 field of the pyramidal layer in brains of patients with Alzheimer's disease compared to controls. These modifications may result from a suffering stage of hippocampal cells or from a compensatory mechanism in the still surviving neurons not yet affected by the pathological process.     

Permanent focal and transient global cerebral ischemia increase glial and neuronal expression of heme oxygenase-1, but not heme oxygenase-2, protein in rat brain

Two heme oxygenase (HO) proteins have been identified to date; HO-1, a stress-induced protein, and HO-2, a constitutively expressed isoform. Recently, it was demonstrated that HO-1 mRNA expression is increased following transient global ischemia. The present study examined the effects of global and focal ischemia on HO-1 and HO-2 protein, using immunocytochemistry. Following 20 min of ischemia (rat 4 vessel occlusion model with hypotension) and 6 h of recirculation, increased HO-1 immunoreactivity was evident in hippocampal neurons. After 24 h of recirculation, HO-1 was observed in both hippocampal neurons and astroglial cells. By 72 h, expression was primarily glial and restricted to CA1 and CA3c. In addition to hippocampus, HO-1 was also evident in both neurons and glia in cerebral cortex and thalamus, and in striatal glial cells. Twenty-four hours following permanent focal ischemia, HO-1 immunoreactivity was observed in astroglial cells in the penumbra region surrounding the infarct. In contrast to HO-1, the pattern of HO-2 immunoreactivity was not altered following transient global or permanent focal ischemia. The increased expression of HO-1 following ischemia may confer protection against oxidative stress, but might also contribute to the subsequent neuronal degeneration.     

Expression pattern of growth/differentiation factor 3 in human and murine cerebral cortex, hippocampus as well as cerebellum

Growth/differentiation factor 3 is a member of GDF/BMP subfamily of the TGF-beta superfamily, which has been reported to be implicated in testis carcinoma and deposition of adipose tissue. Interestingly, present work indicated that GDF3/Gdf3 genes were expressed in cerebral cortex, hippocampus as well as in cerebellum, as revealed by RT-PCR, in situ hybridization and immunostaining. Results of RT-PCR in 10 human tissues and 12 rat tissues indicated that GDF3/Gdf3 genes were abundantly transcribed in both human and murine brain, including cerebral cortex, hippocampus and cerebellum. In situ hybridization and immunohistochemistry results revealed that in cerebral cortex, GDF3 was evenly distributed. In hippocampus, it was expressed in most of the neurons in CA2 and DG region, especially only in a restricted number of neurons in the regions of CA1 and CA3 and in Purkinje cells in cerebellum. Present data suggested that GDF3 might play important roles in the central nervous system (CNS), especially in cerebral cortex, hippocampus and cerebellum, and it shed new light on further research of GDF3 in the central nervous system.     

Occurrence of complement protein C3 in dying pyramidal neurons in rat hippocampus after systemic administration of kainic acid

To evaluate the roles of complement in kainic acid (KA)-induced neuronal damages, the immunohistochemical localization of the complement protein C3 was examined in rat hippocampus after systemic KA injection. The immunoreactivity for C3 was found in glial cells in control rats, and such glial cells were increased in number after KA injection. Our confocal study showed that C3-positive glial cells were microglia. Three to seven days after KA, C3 immunoreactivity appeared in CA1 and CA3 pyramidal neurons. Double staining for C3 combined with the terminal deoxynucleotidyl transferase-mediated dUTP nick end labelling showed that occurrence of C3 immunoreactivity in neurons coincided well with that of DNA fragmentation. Western blot analysis and RT-PCR experiments suggested local synthesis of C3 by brain cells. Our results suggest that C3 contributes greatly to neuronal death after systemic KA administration, and that microglia and neurons are the local source of C3 in KA-induced brain injury.     

Radial glial cells and the lamination of the cerebellar cortex

Radial glial cells are stem cells that play an important role in neuronal migration and proliferation in the developing brain. However, how radial glial cells contribute to the lamination of the cerebellar cortex is not well understood. We therefore used immunohistochemistry and BrdU labeling to follow radial glial cell differentiation, cell migration and cerebellar cortex development in mice from embryonic day 8 to postnatal day 180. We report that radial glial cells represent the stem cell population of the neuroepithelium of the neural tube, and act as progenitors for both neurons and neuroglia. In addition, radial glial cells not only give rise to the principal cells of the cerebellar cortex, the Purkinje and granule cells, but they also provide a scaffold for the migration of these cells. We conclude that radial glial cells play a pivotal role in establishing the laminar structure of the cerebellar cortex.     

Presenilin-1-immunoreactive neurons are preserved in late-onset Alzheimer's disease.

Recent studies have suggested that missense mutations in the presenilin-1 gene are causally related to the majority of familial early-onset Alzheimer's disease (AD). To examine the possible involvement of presenilin-1 in late-onset sporadic AD, a quantitative analysis of its distribution in the cerebral cortex of nondemented and AD patients was performed using immunocytochemistry. Stereological analyses revealed that AD brains showed a marked neuronal loss in the CA1 field of the hippocampus and hilus of the dentate gyrus, subiculum, and entorhinal cortex. In these areas, however, the fraction of neurofibrillary tangle (NFT)-free neurons showing presenilin-1 immunoreactivity was increased compared with nondemented controls. In contrast, cortical areas, which displayed no neuronal loss, did not show any significant increase in the fraction of presenilin-1-positive neurons. Moreover, presenilin-1 immunoreactivity was reduced in NFT-containing neurons. Thus, in AD, the fraction of NFT-free neurons that contained presenilin-1 varied from 0.48 to 0.77, whereas the fraction of NFT-containing neurons that were presenilin-1 positive varied from 0.1 to 0.24. Together, these observations indicate that presenilin-1 may have a neuroprotective role and that in AD low cellular expression of this protein may be associated with increased neuronal loss and NFT formation.     

Reduction of thyroid hormone receptor c-ERB A alpha mRNA levels in the hippocampus of Alzheimer as compared to Huntington brain.

A history of thyroid dysfunction has been cited as a possible risk factor for Alzheimer's disease (AD). Neurologic symptoms displayed by hypothyroid patients resemble, in part, those manifested by Alzheimer patients. To determine if a relationship exists between thyroid hormone receptor message levels and AD, in situ hybridization with tritiated antisense RNA probes for thyroid hormone receptors was used to examine the expression of these genes in Alzheimer and Huntington brain tissue. Message levels for a thyroid hormone receptor highly expressed in brain (c-ERB A alpha) was reduced by 52% in CA1 and 43% in CA2 in Alzheimer hippocampus as compared to Huntington controls. In contrast, message levels for another form of thyroid hormone receptor (c-ERB A beta 1) in Alzheimer hippocampus were not significantly different from Huntington controls. Temporal and cerebellar levels of c-ERB A alpha were elevated by 1.6-fold whereas temporal but not cerebellar levels of c-ERB A beta 1 were elevated 2.0-fold in Alzheimer brain. There was no correlation between thyroid hormone receptor levels and brain weight, autopsy interval, patient age, or the extent of neurofibrillary degeneration. Instead, decreased thyroid hormone receptor mRNA levels in Alzheimer-affected hippocampus were due to an increase in the percentage of neurons expressing lower message levels for these proteins.     

Effects of high cholesterol diet on gliosis in apolipoprotein E knockout mice. Implications for Alzheimer's disease and stroke

Hypercholesterolemia has been suggested as a risk factor for Alzheimer's disease (AD). A genetic risk factor for AD is the E4 allele of apolipoprotein E (apoE). ApoE is the major lipoprotein transporter in the brain, and is mainly produced by glial cells. The present study is focussed on analysing the effects of high cholesterol (HC) diet, duration 9 months, on glial activation in the brain, both in wild type (WT) mice and in mice with a null mutation in the apoE gene (knock-out, KO) mice. The activation of astrocytes and microglia was analysed after immunohistochemical labelling of glial fibrillary acidic protein (GFAP), and F4/80, respectively. In addition, the expression of the antioxidant enzyme NAD(P)H:quinone oxidoreductase (NQO1) was analysed. There was a marked stimulation of astrocyte and microglial activation as well as induced expression of NQO1 in the hippocampus and cerebral cortex upon HC diet. Furthermore, there was significant astrocyte activation in the apoE KO mice, as compared to the WT mice, on ND. The long time exposure to HC diet combined with apoE deficiency resulted in a synergistic effect on the expression of NQO1 in the brain.