Aging, gender and APOE isotype modulate ... - Wiley Online Library

113 downloads 43924 Views 144KB Size Report
never develop histological amyloidosis [i.e. APOE knockout. (KO) mice with or ... apolipoprotein E; APP, amyloid b precursor protein; F2-isoPs,. F2-isoprostane ...... Reich E. E., Markesbery W. R., Roberts L. J., 2nd Swift L. L., Morrow. J. D. and ...
Journal of Neurochemistry, 2004, 90, 1011–1018

doi:10.1111/j.1471-4159.2004.02532.x

Aging, gender and APOE isotype modulate metabolism of Alzheimer’s Ab peptides and F2-isoprostanes in the absence of detectable amyloid deposits Jun Yao,* Suzana S. Petanceska,* Thomas J. Montine,  David M. Holtzman,à Stephen D. Schmidt,* Carolyn A. Parker,§ Michael J. Callahan,¶ William J. Lipinski,¶ Charles L. Bisgaier,¶ Brian A. Turner,** Ralph A. Nixon,* Ralph N. Martins,** Charles Ouimet,§ Jonathan D. Smith,   Peter Davies,àà Eugene Laska,* Michelle E. Ehrlich,§§ Lary C. Walker,¶¶ Paul M. Mathews* and Sam Gandy§§ *Department of Psychiatry, New York University, The Nathan S. Kline Institute for Psychiatric Research, Orangeburg, New York, USA  Department of Pathology, University of Washington, Seattle, Washington, USA àDepartments of Neurology, Molecular Biology and Pharmacology, Washington University School of Medicine, St Louis, Missouri, USA §Department of Biomedical Sciences, College of Medicine, Florida State University, Tallahassee, Florida, USA ¶Neuroscience Discovery, Pfizer Worldwide Research, Ann Arbor, Michigan, USA **Psychiatry and Clinical Neurosciences, University of Western Australia, and School of Biomedical and Sports Science, Edith Cowan University, Perth, Western Australia   Department of Cell Biology, Cleveland Clinic Foundation, Cleveland, Ohio, USA ààAlbert Einstein College of Medicine, Bronx, New York, New York, USA §§Farber Institute for Neurosciences of Thomas Jefferson University, Philadelphia, Pennsylvania, USA ¶¶Yerkes Primate Center, Emory University, Atlanta, Georgia, USA

Abstract Aging and apolipoprotein E (APOE) isoform are among the most consistent risks for the development of Alzheimer’s disease (AD). Metabolic factors that modulate risk have been elusive, though oxidative reactions and their by-products have been implicated in human AD and in transgenic mice with overt histological amyloidosis. We investigated the relationship between the levels of endogenous murine amyloid b (Ab) peptides and the levels of a marker of oxidation in mice that never develop histological amyloidosis [i.e. APOE knockout (KO) mice with or without transgenic human APOE e3 or human APOE e4 alleles]. Aging-, gender-, and APOE-genotypedependent changes were observed for endogenous mouse brain Ab40 and Ab42 peptides. Levels of the oxidized lipid F2-isoprostane (F2-isoPs) in the brains of the same animals as those used for the Ab analyses revealed aging- and genderdependent changes in APOE KO and in human APOE e4 transgenic KO mice. Human APOE e3 transgenic KO mice did not exhibit aging- or gender-dependent increases in F2-isoPs. In general, the changes in the levels of brain F2-isoPs in mice according to age, gender, and APOE genotype mirrored the

changes in brain Ab levels, which, in turn, paralleled known trends in the risk for human AD. These data indicate that there exists an aging-dependent, APOE-genotype-sensitive rise in murine brain Ab levels despite the apparent inability of the peptide to form histologically detectable amyloid. Human APOE e3, but not human APOE e4, can apparently prevent the aging-dependent rise in murine brain Ab levels, consistent with the relative risk for AD associated with these genotypes. The fidelity of the brain Ab/F2-isoP relationship across multiple relevant variables supports the hypothesis that oxidized lipids

Received January 18, 2004; revised manuscript received March 12, 2004; accepted April 1, 2004. Address correspondence and reprint requests to Sam Gandy, Farber Institute for Neurosciences of Thomas Jefferson University, 900 Walnut St., Suite 467, Philadelphia, PA 19107–5587, USA. E-mail: [email protected] Abbreviations used: Ab, amyloid b; AD, Alzheimer’s disease; APOE, apolipoprotein E; APP, amyloid b precursor protein; F2-isoPs, F2-isoprostane; GC, gas chromatography; KO, knockout; PS1, PS2, presenilins/c-secretases; SDS/PBS, sodium dodecyl sulfate/phosphatebuffered saline.

 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 1011–1018

1011

1012 J. Yao et al.

play a role in AD pathogenesis, as has been suggested by recent evidence that F2-isoPs can stimulate Ab generation and aggregation. Keywords: Ab40 and Ab42 peptides, APOE isoform, APOE knockout mice, human APOE e3 transgenic knockout

mice (apoE3-expressing mice or apoE3 mice), human APOE e4 transgenic knockout mice (apoE4-expressing mice or apoE4 mice), isoprostane, oxidation. J. Neurochem. (2004) 90, 1011–1018.

About 5% of familial Alzheimer disease (AD) is caused by missense mutations in genes for either the Alzheimer amyloid b precursor protein (APP) or the amyloid-metabolizing enzymes, known as presenilins/c-secretases (PS1, PS2) (for review, see Gandy et al. 2004). All pathogenic mutations in APP, PS1 or PS2 facilitate the process of cerebral and/or cerebrovascular amyloidogenesis. Another genetic factor, the e4 allele of the cholesterol transport protein apolipoprotein E (APOE e2, e3, e4 genetic polymorphism; apoE2, apoE3, apoE4 protein isoforms) can increase the relative risk for AD, but the e4 allele is not a deterministic mutation (for review, see Holtzman 2001). The cause(s) of most sporadic AD remain(s) obscure, although it has long been proposed that abnormal oxidation of brain proteins might be a key feature (Smith et al. 1991). Aberrant oxidation could explain cerebral and cerebrovascular amyloidogenesis since the reactive by-products of aging-related oxidative reactions can accelerate Ab aggregation (Dyrks et al. 1992). While no pathogenic mutations in oxidation-related enzymes have been identified in AD, a suggestive clue from Alzheimer genetics that potentially implicates oxidative metabolism comes from the observation that APOE isotypes exert their protective or deleterious effects according to their abilities to act as antioxidants (apoEe 2 > apoEe 3 > apoEe 4) (Miyata and Smith 1996). Some, but not all, clinical data support the notion that antioxidant drugs might slow the progression of AD (Mayeux and Sano 1999). Further assessment of the protective capacities of antioxidant compounds is underway, based on the reasoning that toxic oxidation reaction byproducts might exert their effects at a very early stage of disease pathogenesis, possibly as amyloidosis initiation factors. Pinpointing metabolic factors that modulate risk for AD has proven difficult. One area of interest has involved measurement of peroxidized membrane prostaglandins known as isoprostanes. The levels of oxidized lipids and their products (especially those of the isoprostane class) have been determined in tissues and body fluids of normal humans as well as those suffering from AD (Pratico et al. 1998, 1999, 2000, 2001; Montine et al. 1999a,b, 2001, 2002; Pratico 1999, 2000; Waddington et al. 1999; Reich et al. 2001a,b; Arlt et al. 2002). F2-isoPs are also of interest as they have been directly demonstrated to accelerate Ab generation and aggregation (Boutaud et al. 2002; Qin et al. 2003) so could plausibly lie upstream of amyloidogenesis.

Studies of Tg2576 plaque-forming transgenic mice showed that brain, plasma, and urine levels of the isoprostane, 8,12-iso-iP F2a, rise during aging, beginning at a point when cerebral amyloidosis of transgenic human Ab is incipient (Pratico et al. 2001). This pattern raises the question of whether F2-isoPs rise first or whether amyloidosis precedes the F2-isoP rise. We have extended that work by measuring brain F2-isoPs and endogenous mouse Ab. As mouse Ab does not form histologically detectable fibrillar amyloid (Selkoe et al. 1987), this system enables us to determine whether gross amyloidosis is required for the isoprostane levels to rise. Furthermore, we have used either mice that are genetically deficient in apoE (mu APOE –/–), or mice that are both deficient in endogenous apoE and transgenic for human apoE isoforms, using the glial fibrillary acidic protein promoter (mu APOE –/–X hu APOEe3 +/+ and mu APOE –/–X hu APOEe4 +/+). Since APOE genotype is one of the best studied genetic risks for sporadic AD, and since apoE has been implicated in the regulation of oxidation reactions (Miyata and Smith 1996), we wanted to investigate the possibility that APOE genotype modulates isoprostane levels. Here we report the results of simultaneous measurement of Ab peptides and the oxidation marker F2-isoprostane (F2-isoP) in order to refine our understanding of the relationship between the two.

Materials and methods Transgenic mice Mu APOE –/– mice were generated as described (Plump et al. 1992; Zhang et al. 1992). (mu APOE –/–) X (GFAP-hu APOE e3 +/+) and (mu APOE –/–) X (GFAP-hu APOE e4 +/+) were prepared as described (Holtzman et al. 2000). All were maintained on the C57BL/6 background. All mice were housed at a constant temperature with 12-h light/dark cycles and fed regular mouse chow until sacrifice. Mice from 1 to 16 months of age were sacrificed in a carbon dioxide-charged chamber. The brains were immediately removed and dissected into two hemispheres. One hemisphere was processed for quantitation of Ab species, and the other hemisphere was immediately snap-frozen and stored at ) 80C until used for quantitation of F2-isoPs. Antibodies JRF/cAb40/10 and JRF/cAb42/26 monoclonal antibodies were generated against the C-terminal five or 10 residues of Ab 1–40

 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 1011–1018

Parallel trends in Ab and isoprostanes

or of Ab 1–42, respectively, and JRF/rAb1–15/2 monoclonal antibody was raised against a synthetic peptide including residues 1–15 of murine Ab beta (see Mathews et al. 2002 for details of epitope specificities); all three antibodies were kindly provided by Dr Marc Mercken (Janssen Pharmaceutica/Johnson and Johnson Pharmaceutical Research and Development). Antibodies 6E10(antiAb1-16) and 4G8 (anti-Ab17–24) were purchased from Signet Laboratories (Dedham, MA, USA). Preparation of brain extracts for detection of Ab Brain supernatants were processed for sandwich ELISA as described (Petanceska et al. 2000). Briefly, mouse cerebral hemispheres were Dounce-homogenized in 0.2% diethylamine (DEA)/50 mM NaCl at 1 : 10 w/v. Brain homogenates were centrifuged for 1 h at 100 000 · g at 4C, and supernatants were neutralized to pH 8.0 with 1 : 10 v/v of 0.5 M Tris-HCl/pH 6.8. Protein concentrations in DEA extracts were determined using the BCA Protein Assay Kit (Pierce, Rockford, IL, USA). Ab sandwich ELISAs Endogenous murine brain Ab40 and Ab42 levels were quantitated using an established sandwich ELISA (Mathews et al. 2002; Rozmahel et al. 2002a,b). Ab signals were calculated based upon standard curves run on every ELISA plate, and Ab levels reflect the averages of duplicate measurements of each sample. Isoprostane measurements F2-isoprostanes (e-iso-Ps) were measured in one cerebral hemisphere, as described (Reich et al. 2001b). Briefly, each hemisphere was homogenized in Folch solution, and esterified lipids were converted to O-methyloxime derivatives in Folch solution. Saponification was used to hydrolyze the compounds. F-Ring isoPs were analyzed using gas chromatography (GC) with negative ion chemical ionization mass spectrometry and selective ion monitoring as described (Reich et al. 2001b). Tissue marker studies ApoE-deficient mice from two sources were analyzed. Ten ‘aged’ (14–20 month), five ‘middle-aged’ (8–9 month) and nine ‘young’ (2–4 month) homozygous mu APOE –/– knockout mice from Jackson Laboratories were studied. These mice were originally produced at the University of North Carolina (Zhang et al. 1992); the embryonic stem cell line was a 129-derived E14Tg2a cell, and the mutation was maintained on a C57BL/6 background. For comparison, nine wild-type, C57BL/6 mice obtained from the National Institute on Aging were studied. Among this group, three mice were 2 months old, three were 8 months old, and three were 20 months old. In addition, seven aged mu APOE –/– mice were obtained from a colony at The Rockefeller University (Plump et al. 1992), also derived from E14 embryonic stem cells and maintained on a C57BL/ 6 inbred background. Seven age-matched, wild-type C57BL/6 J littermate controls from the Rockefeller colony were also studied. Tissue preparation and immunohistochemical analysis Animals were deeply anesthetized with an overdose of sodium pentobarbital and perfused with cold, phosphate-buffered saline, pH 7.4, for 90 s. Brains were removed and subdissected for analyses. Right hemispheres were immersion-fixed in 4% parafor-

1013

maldehyde, cryoprotected in buffered sucrose, frozen and sectioned at a thickness of 19 lm with a Microm 500 cryostat (Carl Zeiss, Inc.; Thornwood, NY, USA). On some slides, astrocytic inclusions were visualized using a standard Bielschowsky stain. In most mice, the fresh unfixed hippocampus and cerebral cortex were removed from the left hemisphere, weighed, and frozen in isopentane cooled in dry ice. These tissue samples were homogenized individually with a Tissumizer (Tekmar, IKA, Labortechnik) in 500 lL of cold brain homogenization buffer, pH 8.0 (10.25 mM phosphate, 150 mM NaCl, 5.0 benzamidine, 3.0 mM EDTA, 1.0 mM magnesium sulfate plus a protease inhibitor cocktail containing phenylmethylsulfonyl fluoride, aprotinin, leupeptin, pepstatin A, and antipain). Samples were centrifuged at 100 000 g for 8 min at 4C. P2 fractions were resuspended in 200 lL homogenization buffer. Samples from all fractions were assayed for protein content (Bio-Rad) and applied to polyacrylamide gels (see below). Quantitative immunoblotting Aliquots of homogenates (50 lg per lane) were separated in SDS– polyacrylamide gel electrophoresis gradient gels (4–12%) and then electroblotted onto nitrocellulose paper. Blots were blocked in 0.1% Tween 20 in PBS, pH 7.4 for 2 h at 4C and incubated overnight at 4C with a primary antibody against an antigen of interest: synaptophysin (Boehringer Mannheim, Indianapolis, IN, USA; 1: 100), microtubule associated protein-2 (MAP-2; 5 lg/mL; Boehringer Mannheim, Indianapolis, IN, USA), b-amyloid precursor protein (b-APP) (Boehringer Mannheim, 22C11, 1: 200), and Ab antibody 4G8 (K.S. Kim, Staten Island, NY, USA). Standard curves were included on each blot, and samples were read from the linear portion of the curve. Choline acetyltransferase assays Brain homogenate was diluted 1 : 1 with a solution containing 269 mM NaCI, 90 mM NaH2PO4, 0.45% Triton-X100, 0.9 mM EGTA, 0.9 mM Na2EDTA and 179 lM physostigmine. 10 lL of this mixture (all samples in quadruplicates) were added to 5 lL of choline bromide (32 mM). The incubation was started by the addition of 5 lL of [14C]Acetyl coenzyme-A (50 nCi/assay; 0.242 mM final concentration). After 15 min at 37C, the tip of the incubation tube was cut with a razor blade and put into a mixture of 5 mL sodium phosphate buffer (10 mM, pH 7.4) with 2 mL of sodium tetraphenylborate in acetonitrile (5 mg/mL). From this mixture the newly formed [14C]acetylcholine was extracted by careful shaking with 10 mL of toluene scintillator. Following separation of the aqueous phase from the organic phase the samples were directly counted by LSC. Method of statistical analysis The statistical analysis was based on a three way factorial model with an unequal number of observations in each cell. The three factors are age (three ordered categories), gender and APOE genotype (e3,e4,KO). This was carried out to determine whether age, gender and APOE genotype individually and together influence brain Ab accumulation. The following factorial model Y ¼ intercept + age + gender þ APOE genotype þ age*gender þ age*APOE genotype gender*APOE genotype þ age*gender*APOE genotype + error

 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 1011–1018

1014 J. Yao et al.

was fit using SAS’s General Linear Model (GLM) program (SAS Institute Inc., Cary, NC, USA). The model includes main effects for the factors age, gender, and APOE genotype, as well as all two way interactions and the three way interaction. The same model was used to separately analyze the dependent variables Ab40, Ab42, and Total Ab ¼ Ab40 + Ab42. Whenever there were significant (p < 0.05) main effects and/or interaction effects, pairwise contrasts (using the least-square means) between the levels of the factors and the levels of the interactions were computed. The method of TukeyKramer was used to control the familywise error rate at 0.05. That is, the chance that there is an error among the contrasts that were declared statistically significant is one in twenty.

Results and discussion

Aging, gender, and APOE genotype modulate brain levels of endogenous murine Ab in the absence of histologically detectable amyloidosis The analysis revealed that for Ab40, there is a significant main effect of age such that the levels of Ab40 at 5 and 10 months of age were lower than at 16 months of age (p < 0.001 for each comparison). For Ab40, gender and APOE genotype were not found to be statistically significant factors. We observed a significant main effect of age on the accumulation of Ab42; similar to Ab40, the levels of Ab42 in brain at 16 months of age were significantly higher than at 5 or 10 months of age (p < 0.001 for each comparison). Moreover, the levels at 10 months tended to be higher than at 5 months of age, but this difference failed to reach statistical significance (p ¼ 0.07). We also observed a similar trend in main effect of gender on the accumulation of Ab42 such that males tended to accumulate higher levels of the peptide than females, but again this was not significant (p ¼ 0.07). In contrast to Ab40, APOE genotype significantly influenced Ab42 accumulation. The presence of human e4 was associated with significantly higher Ab42 levels in brain than the presence of APOE e3 or the absence of APOE (p < 0.001 and p < 0.05, respectively). The main effect of age on total brain Ab accumulation was such that there was significant increase with increasing age (5 months < 10 months < 16 months; p < 0.05 for 5 months versus 10 months, and p < 0.001 for 10 months versus 16 months and 5 months versus 16 months) (Fig. 1). Total brain Ab was significantly influenced by gender, the accumulation being higher in males when compared to females (p < 0.05) (Fig. 1). A significant main effect of APOE genotype was also observed such that APOE e4 carriers had higher levels than either APOE e3 mice or APOE KO mice. Total brain Ab accumulation was not significantly different between the APOE e3 mice and mice deficient in apoE. When the impact of all three factors on brain Ab accumulation was determined, the highest average values

Fig. 1 Main effects of age, gender and APOE genotype on total brain Ab levels. 16 M male E4–16 M female E4–16 M male KO > rest; 16 M male KO > 16 M female KO, p < 0.05; 5 M < 10 M, p < 0.05; 5 M < 16 M, p < 0.001; 10 M < 16 M, p < 0.001; Female < Male; p < 0.05; e3 KO; p ¼ 0.36; e3 < e4; p < 0.001; KO < e4, p < 0.05. *highest level. Numbers represent group means by the method of least squares.

for Ab40, Ab42 and total Ab were observed in the aged male APOE e4 mice. However for all three variables, similar levels of Ab were also observed in aged male APOE knockout mice and aged female APOE e4 mice. In addition, the mean value for brain Ab40 levels in the aged male APOE e4 mice were not significantly different from the means for 10-month-old APOE e4 female mice and aged (16-month-old) female APOE e3 mice. The average brain Ab42 levels in the aged male APOE e4 mice were significantly higher than in all of the APOE e3 experimental cells but were similar to the mean values for the 10-monthold female APOE e4 mice. In spite of the gender effect observed for total Ab, aged male APOE e4 mice accumulated similar levels of Ab to aged female APOE e4 mice due to the strong influences of age and APOE e4 on Ab accumulation (Fig. 1). It is also of note that even though the main effect of APOE genotype was such that higher mean values for Ab (Ab42 and total Ab) were observed in APOE e4 mice compared to apoE-deficient mice, there were no significant differences in Ab accumulation between the aged male APOE e4 mice and the aged male APOE KO mice, due to the very strong influence of age on Ab accumulation (Fig. 1). Gender differences in Ab accumulation in the aged experimental cells were observed only for total Ab in the apoE-deficient mice.

 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 1011–1018

Parallel trends in Ab and isoprostanes

APOE genotype-dependent and gender-dependent changes in brain Ab levels in aged mice paralleled alterations in levels of a biochemical marker of oxidative injury There is evidence that oxidative damage in brain coincides with increased brain Ab levels and precedes apparent Ab deposition, suggesting an instrumental role for oxidative damage in AD pathogenesis (Dyrks et al. 1992). To determine whether APOE-genotype, aging, or gender-related increases in brain Ab levels were associated with increased oxidative damage, we measured by mass spectrometry the levels of a quantitative in vivo marker of oxidative injury, F2-isoPs, in contralateral brain tissue from the same animals used for Ab measurements. F2-isoPs are generated in vivo by nonenzymatic, free-radical-mediated peroxidation of arachidonic acid, through a series of endoperoxide intermediates which are eventually reduced to form PGF2-like compounds (Montine et al. 1999a). Quantification of F2-isoPs has emerged as one of the most popular approaches to assessing oxidative injury in vivo. APOE genotype was found to influence F2-isoPs in male mice. Aged male APOE KO and human APOE e4 transgenic knockout mice had significantly higher levels of brain F2-isoPs when compared to aged male APOE e3-expressing mice (1.5- to 1.6-fold increases) (Fig. 2a). Female mice, however, did not exhibit aging- or APOE-genotype-associated changes in brain F2-isoP levels at 16-months of age (Fig. 2b). Aged male APOE KO and human APOE e4 transgenic knockout mice had significantly higher levels of F2-isoPs when compared to female mice matched for age and genotype (1.3- to 1.4-fold) (Fig. 2c). There were no gender-related differences in F2-isoPs in aged APOE e3expressing mice. This pattern of changes in brain F2-isoPs levels closely paralleled the APOE -genotype- and genderdependent changes in brain Ab40 and Ab42 levels. APOE-deficient mice have no obvious decrement in markers of neuronal or synaptic density In order to investigate the possibility that neurodegeneration might be responsible for the changes in Ab and F2-isoPs in APOE KO mice, we examined various neuronal and synaptic markers in wild-type and APOE KO mice. Quantitative immunoblotting for Alz50, PHF-1, MAPs 1b and 2, synaptophysin, synapsin, or SNAP-25 (not shown) confirmed that no significant variations were detectable in the levels of these proteins in the hippocampi or cerebral cortices of 2- or 20month-old APOE-deficient mice when compared to the cortices of age-matched, wild-type mice. Similar results have been reported by Hartman et al. (2001). There was also no detectable alteration of neuronal density in the CA1 region of the hippocampi of aged APOE-deficient mice when assessed using the Gunderson unbiased stereological technique to compare with the neuronal density of the CA1 region of the hippocampi of age-matched wild-type

1015

(a)

(b)

(c)

Fig. 2 APOE-genotype and gender-dependent differences in F2-isoP levels in the brains of aged mice. The levels of F2-isoPs in aged mice were compared based on APOE genotype in males (a) and females (b). Comparison of F2-isoP levels between genders in aged mice of each APOE genotype (c). Statistical significance was determined using a two-way ANOVA (**p < 0.005).

mice (Table 1). Morris water maze performances of APOEdeficient and wild-type mice were indistinguishable (not shown). Choline acetyltransferase (ChAT) Whole hemisphere brain homogenates were analyzed for choline acetyltransferase activity. No differences in ChAT activity were observed when levels from young and aged APOE-deficient mice were compared to levels from young and aged wild-type mice, respectively (not shown), confirming the results of Fagan et al. (1998). Astrocytic inclusions Periodic acid-Schiff reagent histochemistry identified granular clusters in hippocampal astrocytes in both wildtype- and APOE-deficient mice with no grossly obvious relationship between genotype and cluster density. These structures have been previously described and termed ‘agingrelated astrocytic inclusions’, (Jucker et al. 1992) and an earlier study proposed that the density of these clusters might be increased in APOE deficient mice (Robertson et al. 1998). The clusters were initially misidentified as Ab plaques by Wirak et al. (1991) because of the tendency of the clusters to

 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 1011–1018

1016 J. Yao et al.

APOE KO-1 APOE KO-2 APOE KO-3 wildtype-1 wildtype-2 wildtype-3 wildtype-4

Vref (lm3)

Nv

N

101,795,455 102,140,375 95,697,865 115,586,905 103,294,120 92,501,350 89,356,955

0.013545 0.021010 0.018060 0.014859 0.017157 0.013299 0.014859

1.379 2.146 1.728 1.718 1.772 1.230 1.328

Table 1 Unbiased estimates of reference volume (Vref) in lm3 x x x x x x x

106 106 106 106 106 106 106

Numbers of cells per unit volume (Nv) and neuronal density (N) in hippocampal CA1 regions of wildtype mice 1-4 and APOE-deficient mice 1-3.

bind IgG non-specifically. The precise protein composition of the clusters is unknown, and we cannot completely exclude the possibility that some mouse Ab might be incorporated into them, though this seems highly unlikely. Conclusions

In the current study, we observed that the levels of endogenous mouse brain Ab and F2-isoPs were significantly modulated by aging, APOE genotype, and gender. We observed that the aging-dependent increase in brain Ab levels was paralleled by a rise in F2-isoPs. Among the aged mice, the levels of Ab40 and F2-isoPs appeared to change in parallel as a function of gender and APOE genotype. Since either rising Ab levels or rising isoprostane levels could be a consequence of neurodegeneration, we investigated the possibility that neurons or synapses might show signs of damage in association with the changes in Ab levels or isoprostane levels, but, in contrast to some earlier reports (Masliah et al. 1995; Veinbergs and Masliah 1999; Veinbergs et al. 1999), we could find no alteration in neuronal numbers or markers or in synaptic markers. We also found no change in the levels of choline acetytransferase, again in contrast to previous reports (Chapman and Michaelson 1998). These data suggest that overt structural neuropathological features of AD, including cholinergic deficiency and detectable fibrillar amyloidosis, are not a prerequisite for elevations in levels of Ab levels or in markers of lipid peroxidation. Amyloidosis and cholinergic deficiency were also not required for the modulation of either Ab levels or isoprostane levels by aging, APOE genotype, or gender. Faithful maintenance of the pattern of parallel changes in Ab levels and isoprostane levels across three relevant variables strengthens the likelihood that F2-isoPs may prove to be causally related to alterations in Ab metabolism. Instead, or in addition, isoprostane levels might be useful surrogates for Ab levels. The equilibrium of brain, cerebrospinal fluid, and plasma isoprostane concentration may make this a more reliable marker in AD where Ab equilibria have been complicated to interpret and difficult to apply clinically.

Our data also suggest interactions across other variables. Aging-dependent rises in murine Ab levels can be differentially modulated by human APOE isoforms, indicating that at least some component of the modulation of Ab metabolism by apoE isoforms is probably upstream of senile plaque formation. Human APOE e3, but not human APOE e4, apparently prevented the aging-dependent elevation of murine Ab levels. This phenomenon suggests that the effect of APOE e4 on some aspects of Ab homeostasis more closely resembles a ‘loss-of-function’ phenotype, rather than the ‘gain-of-function’ phenotype that is believed to operate for most dominant neurodegenerative alleles, including mutant APP. These relationships may prove useful in determining the precise molecular underpinnings of the aging- and APOE-genotype-dependent effects that appear to control cerebral Ab amyloid accumulation and, consequently, the risk for developing Alzheimer’s disease. Indeed, the phenomena described here have all been associated with the risk for Alzheimer’s, although the associations have not been universal. As various epidemiological studies may legitimately differ across ethnically diverse groups, the use of inbred, genetically homogeneous mice provides a means of assessing these variables rigorously in the near-absence of environmental influences. In addition, these data provided an important background upon which studies of effects on Ab and oxidized lipids due to metabolic derangements such as diabetes and hypercholesterolemia may be conducted. At the molecular level, plausible mechanisms have been proposed for the role of APOE genotype and hormonal status in controlling resistance to oxidative stress and Alzheimer’s disease. Major questions continue to surround the basic nature of the aging effect as well as ‘chicken-or-egg’ questions linking oxidation and amyloidosis. Our data provide good reason to postulate that oxidation may play a major role in normal and pathological aging of the brain, as is true of other tissues, and that oxidation may initiate amyloidogenesis in a process that becomes self-propagating, both by ‘seeding’ and by positive feedback from oxygen radicals generated by aggregated amyloid (Hensley et al. 1994). Such a model suggests that antioxidants, anti-

 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 1011–1018

Parallel trends in Ab and isoprostanes

amyloid-aggregation drugs and vaccines, and metal chelators should be useful in the treatment of Alzheimer’s disease. Evidence suggests that each of these is beneficial (for review, see Gandy et al. 2004), raising the possibility that combination therapy with a cocktail of such agents may provide the best strategy for Alzheimer’s treatment and prevention. Acknowledgements This research was supported by USPHS grants AG10491 and NS45913 and the NH and MRC of Australia.

References Arlt S., Beisiegel U. and Kontush A. (2002) Lipid peroxidation in neurodegeneration: new insights into Alzheimer’s disease. Curr. Opin. Lipidol. 13, 289–294. Boutaud O., Ou J. J., Chaurand P., Caprioli R. M., Montine T. J. and Oates J. A. (2002) Prostaglandin H2 (PGH2) accelerates formation of amyloid beta1–42 oligomers. J. Neurochem. 82, 1003–1006. Chapman S. and Michaelson D. M. (1998) Specific neurochemical derangements of brain projecting neurons in apolipoprotein E-deficient mice. J. Neurochem. 70, 708–714. Dyrks T., Dyrks E., Hartmann T., Masters C. and Beyreuther K (1992) Amyloidogenicity of beta A4 and beta A4-bearing amyloid protein precursor fragments by metal-catalyzed oxidation. J. Biol. Chem. 267 (25). Fagan A. M., Murphy B. A., Patel S. N., Kilbridge J. F., Mobley W. C., Bu G. and Holtzman D. M. (1998) Evidence for normal aging of the septo-hippocampal cholinergic system in apoE (–/–) mice but impaired clearance of axonal degeneration products following injury. Exp. Neurol. 151, 314–325. Gandy S., Martins R. and Buxbaum J. (2004) Molecular basis for antiamyloid therapy in the prevention and treatment of Alzheimer’s disease. Alzheimer’s Dis. 17, 259–266. Hartman R. E., Wozniak D. F., Nardi A., Olney J. W., Sartorius L. and Holtzman D. M. (2001) Behavioral phenotyping of GFAP-apoE3 and GFAP-apoE4 transgenic mice: apoE4 mice show profound working memory impairments in the absence of Alzheimer’s-like neuropathology. Exp. Neurol. 179, 326–344. Hensley K., Carney J. M., Mattson M. P., Aksenova M., Harris M., Wu J. F., Floyd R. A. and Butterfield D. A. (1994) A model for beta-amyloid aggregation and neurotoxicity based on free radical generation by the peptide: relevance to Alzheimer disease. Proc. Natl Acad. Sci. USA 91, 3270–3274. Holtzman D. M. (2001) Role of apoE/Abeta interactions in the pathogenesis of Alzheimer’s disease and cerebral amyloid angiopathy. J. Mol. Neurosci. 2, 147–155. Holtzman D. M., Bales K. R., Tenkova T. et al. (2000) Apolipoprotein E isoform-dependent amyloid deposition and neuritic degeneration in a mouse model of Alzheimer’s disease. Proc. Natl Acad. Sci. USA 97, 2892–2897. Jucker M., Walker L. C., Martin L. J., Kitt C. A., Kleinman H. K., Ingram D. K. and Price D. L. (1992) Age-associated inclusions in normal and transgenic mouse brain. Science 255, 1443–1445. Masliah E., Mallory M., Ge N., Alford M., Veinergs I. and Roses A. D. (1995) Neurodegeneration in the central nervous system of apoEdeficient mice. Exp. Neurol. 136, 107–122. Mathews P. M., Jiang Y., Schmidt S. D., Grbovic O. M., Mercken M. and Nixon R. A. (2002) Calpain activity regulates the cell surface distribution of amyloid precursor protein: inhibition of calpains

1017

enhances endosomal generation of beta-cleaved C-terminal APP fragments. J. Biol. Chem. 277, 36415–36424. Mayeux R. and Sano M. (1999) Treatment of Alzheimer’s disease. N. Engl. J. Med. 341, 1670–1679. Miyata M. and Smith J. D. (1996) Apolipoprotein E allele-specific antioxidant activity and effects on cytotoxicity by oxidative insults and beta-amyloid peptides. Nat. Genet. 271, 30647–30655. Montine T. J., Beal M. F., Cudkowicz M. E., O’Donnell H., Margolin R. A., McFarland L., Bachrach A. F., Zackert W. E., Roberts L. J. and Morrow J. D. (1999a) Increased CSF F2-isoprostane concentration in probable AD. Neurology 52, 562–565. Montine T. J., Sidell K. R., Crews B. C., Markesbery W. R., Marnett L. J., Roberts L. J. and 2nd Morrow J. D. (1999b) Elevated CSF prostaglandin E2 levels in patients with probable AD. Neurology 53, 1495–1498. Montine T. J., Kaye J. A., Montine K. S., McFarland L., Morrow J. D. and Quinn J. F. (2001) Cerebrospinal fluid Ab42, tau, F2-isoprostane concentrations in patients with Alzheimer disease, other dementias, and in age-matched controls. Arch. Pathol. Lab. Med. 125, 510–512. Montine T. J., Quinn J. F., Milatovic D., Silbert L. C., Dang T., Sanchez S., Terry E., Roberts L. J., Kaye J. A. and Morrow J. D. (2002) Peripheral F2-isoprostanes and F4-neuroprostanes are not increased in Alzheimer’s disease. Ann. Neurol. 52, 175–179. Petanceska S. S., Nagy V., Frail D. and Gandy S. (2000) Ovariectomy and 17beta-estradiol modulate the levels of Alzheimer’s amyloid beta peptides in brain. Neurology 54, 2212–2217. Plump A. S., Smith J. D., Hayek T., Aalto-Setala K., Walsh A., Verstuyft J. G., Rubin E. M. and Breslow J. L. (1992) Severe hypercholesterolemia and atherosclerosis in apolipoprotein E-deficient mice created by homologous recombination in ES cells. Cell 2, 343– 353. Pratico D. (1999) F2-isoprostanes: sensitive and specific non-invasive indices of lipid peroxidation. Atherosclerosis 147, 1–10. Pratico D. (2000) Oxidative in vivo injury in diseases of the central nervous system: focus on Alzheimer’s disease. Am. J. Med. 109, 577–585. Pratico D., Lee V. M.-Y., Trojanowski J. Q., Rokach J. and FitzGerald G. A. (1998) Increased F2-isoprostanes in Alzheimer’s disease: evidence of enhanced lipid peroxidation in vivo. FASEB J. 12, 1777–1783. Pratico D., Rokach J. and Tangirala R. K. (1999) Brains of aged apolipoprotein E-deficient mice have increased levels of F2-isoprostanes in vivo markers of lipid peroxidation. J. Neurochem. 73, 734–741. Pratico D., Clark C. M., Lee V. M.-Y., Trojanowski J. Q., Rokach J. and FitzGerald G. A. (2000) Increased 8,12-iso-iPF2a-VI in Alzheimer’s disease: correlation of a non-invasive index of lipid peroxidation with disease severity. Ann. Neurol. 48, 809–812. Pratico D., Uryu K., Leight S., Trojanowski J. Q. and Lee V. M.-Y. (2001) Increased lipid peroxidation precedes amyloid plaque formation in an animal model of Alzheimer amyloidosis. J. Neurosci. 21, 4183–4187. Qin W., Ho L., Pompl P. N., Peng Y., Zhao Z., Xiang Z., Robakis N. K., Shioi J., Suh J. and Pasinetti G. M. (2003) Cyclooxygenase (COX)-2 and COX-1 potentiate beta-amyloid peptide generation through mechanisms that involve gamma-secretase activity. J. Biol. Chem. 278, 50970–50977. Reich E. E., Montine K. S., Gross M. D., Roberts L. J., Swift L. L., Morrow J. D. and Montine T. J. (2001a) Interactions between apolipoprotein E gene and dietary a-tocopherol influence cerebral oxidative damage in aged mice. J. Neurosci. 211, 5993–5999. Reich E. E., Markesbery W. R., Roberts L. J., 2nd Swift L. L., Morrow J. D. and Montine T. J. (2001b) Quantification of F-ring and D-/

 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 1011–1018

1018 J. Yao et al.

E-ring isoprostanes and neuroprostanes in Alzheimer’s disease. Adv. Exp Med. Biol. 500, 253–256. Robertson T. A., Dutton N. S., Martins R. N., Roses A. D., Kakulas B. A. and Papadimitriou J. M. (1998) Age-related congophilic inclusions in the brains of apolipoprotein E-deficient mice. Neuroscience 82, 171–180. Rozmahel R., Huang J., Chen F. et al. (2002a) Normal brain development in PS1 hypomorphic mice with markedly reduced gammasecretase cleavage of betaAPP. Neurobiol. Aging 23, 187–194. Rozmahel R., Mount H. T., Chen F. et al. (2002b) Alleles at the Nicastrin locus modify presenilin 1-deficiency phenotype. Proc. Natl Acad. Sci. USA 99, 14452–14457. Selkoe D. J., Bell D. S., Podlisny M. B., Price D. L. and Cork L. C. (1987) Conservation of brain amyloid proteins in aged mammals and humans with Alzheimer’s disease. Science 235, 873–877. Smith C. D., Carney J. M., Starke-Reed P. E., Oliver C. N., Stadtman E. R., Floyd R. A. and Markesbery W. R. (1991) Excess brain protein oxidation and enzyme dysfunction in normal aging and

in Alzheimer disease. Proc. Natl Acad. Sci. USA 88, 10540– 10543. Veinbergs I. and Masliah E. (1999) Synaptic alterations in apolipoprotein E knockout mice. Neuroscience 91, 401–403. Veinbergs I., Mante M., Jung M. W., Van Uden E. and Masliah E. (1999) Synaptotagmin and synaptic transmission alterations in apolipoprotein E-deficient mice. Neuropsychopharmacol. Biol. Psychiatry. 23, 519–531. Waddington E., Croft K., Clarnette R., Mori T. and Martins R. (1999) Plasma F2-isoprostane levels are increased in Alzheimer’s disease: Evidence of increased oxidative stress in vivo. Alzheimer’s Reports 2, 277–282. Wirak D. O., Bayney R., Ramabhadran T. V. et al. (1991) Deposits of amyloid beta protein in the central nervous system of transgenic mice. Science 253, 323–325. Zhang S. H., Reddick R. L., Piedrahita J. A. and Maeda N. (1992) Spontaneous hypercholesterolemia and arterial lesions in mice lacking apolipoprotein E. Science 258, 468–471.

 2004 International Society for Neurochemistry, J. Neurochem. (2004) 90, 1011–1018