Gender differences in the amount and deposition of amyloidβ in APPswe and PS1 double transgenic mice

Gender differences in the amount and deposition of amyloid� inAPPswe and PS1 double transgenic mice

Jun Wang,a Heikki Tanila,a,b Jukka Puoliva¨li,c Inga Kadish,a and Thomas van Groena,b,*a Department of Neuroscience and Neurology, University of Kuopio, Kuopio, Finland

b Department of Neurology, Kuopio University Hospital, P.O. Box 1627, FIN-70211 Kuopio, Finlandc Gladstone Institute of Neurological Disease, University of California, San Francisco, P.O. Box 419100, San Francisco, CA 94141-9100, USA

Received 22 October 2002; revised 24 February 2003; accepted 8 August 2003

Abstract

Transgenic mice carrying both the human amyloid precursor protein (APP) with the Swedish mutation and the presenilin-1 A246Emutation (APP/PS1 mice) develop Alzheimer’s disease-like amyloid� protein (A�) deposits around 9 months of age. These mice show anage-dependent increase in the level of A�40 and A�42 and in the number of amyloid plaques in the brain. A�40 and A�42 levels weremeasured, and amyloid burden and plaque number were quantified, in the hippocampus at the age of 4, 12, and 17 months in both male andfemale APP/PS1 mice. In all mice, amyloid burden and plaque number increased markedly with age, with female mice bearing a heavieramyloid burden and higher plaque number compared to male mice of the same age, both at 12 and at 17 months of age. The level of bothA�40 and A�42 significantly increased in female mice with age and was always significantly higher in female than in male mice of the sameage. Further, there were significant correlations between amyloid burden and A�42 level in female mice and between amyloid burden andplaques in both female and male mice. Together these data show that female APP/PS1 mice accumulate amyloid at an earlier age and thatthey build up more amyloid deposits in the hippocampus than age-matched male mice. Together, these results provide new insights in thepotential mechanisms of the observed gender differences in the pathogenesis of AD.© 2003 Elsevier Inc. All rights reserved.

Introduction

Alzheimer’s disease (AD) is the most common cause ofdementia in the elderly, with the neuropathological hall-marks of amyloid plaques, tangles, and a progressive loss ofneurons in the neocortex (Evans et al., 1989; Braak andBraak, 1991). At the earliest stage of AD, pathology islargely restricted to the medial temporal lobe; in later stagesit spreads to the associative temporal and parietal corticalareas and, finally, to all cortical areas (Hyman et al., 1984;Braak and Braak, 1991). Epidemiological studies haveshown higher prevalence rates of AD in women (Molsa etal., 1982; Jorm et al., 1987; Hagnell et al., 1992; Letenneuret al., 1994; Brayne et al., 1995; Fratiglioni et al., 1997,

2000; Andersen et al., 1999), but the physiological basis thatunderlies this sex difference in the prevalence of AD is stillelusive. It has been suggested that after menopause, a decliningestrogenic stimulus, either from dramatically reduced levels ofcirculating estrogen or from aromatizable androgen levels,both of ovarian origin, might make estrogen target neurons inthe brain more susceptible to age- or disease-related processes.This idea is supported by a study that showed lower serumestrogen values in women with AD than in age-matched con-trols (Honjo et al., 1989). Clinical trials with estrogen supple-mentation have suggested that the onset of AD is delayed andthat the risk of AD is reduced in the women who receiveestrogen (Honjo et al., 1989; Tang et al., 1996; Henderson etal., 2000). However, some more recent studies have suggestedthat estrogen supplementation is not beneficial for Alzheimer’sdisease, especially in patients with established AD (Cholertonet al., 2002; Fillit, 2002).

Understanding the disease mechanisms of AD has beenfacilitated by the discovery of the genetic mutations that

* Corresponding author. Department of Neuroscience and Neurology,University of Kuopio, P.O.Box 1627, Kuopio FIN-70211, Finland. Fax:�358-17-162048.

E-mail address: [email protected] (T. van Groen).

R

Available online at www.sciencedirect.com

Neurobiology of Disease 14 (2003) 318–327 www.elsevier.com/locate/ynbdi

0969-9961/$ – see front matter © 2003 Elsevier Inc. All rights reserved.doi:10.1016/j.nbd.2003.08.009

underlie inherited forms of early onset AD. Several muta-tions in the genes encoding the amyloid precursor protein(APP), presenilin-1 (PS-1), and PS-2 proteins have beenshown to lead to familial AD (Price et al., 1995; Hardy,1997). These mutations lead to the overproduction of amy-loid� (A�) which is followed by the extracellular depositionof A� in the brain (Price et al., 1995; Hardy, 1997). Atpresent several transgenic mouse models of AD (that carryAPP and/or PS1 genes with mutations) are used and most ofthese models develop progressive, age-related A� neuropa-thology with amyloid plaques and elevated levels of A�(Games et al., 1995; Hsiao et al., 1996; Borchelt et al.,1997). Our transgenic mice have elevated levels of thehighly fibrillogenic A�42 peptide and develope amyloidplaques around the age of 9 months; the amyloid plaques arefirst present in the subiculum and caudal cortex (Liu et al.,2002a), and later the amyloid plaques are present in nearlyall cortical areas (Borchelt et al., 1997).

We have noted that female mice have higher levels of A�and pathology than age-matched males (Van Groen et al.,2000). Recently this has also been shown in another aging ADmodel mouse (the Tg2576 mouse) (Callahan et al., 2001).

In the present study, we have used transgenic mice ex-pressing both the human APP695swe and the PS1-A246Emutation (Borchelt et al., 1997) to investigate more closelythe gender difference in A� levels and amyloid burdenbetween female and male mice and the development of thisrelationship with age.

Material and methods

Animals

Transgenic mice expressing either the human PS1(A246E mutation) or a chimeric mouse/human APP695(K595N, M596L, Swedish mutation; APPswe) (Borchelt etal., 1997) were back-crossed to C57BL/6J mice for sixgenerations, and then the lines were crossed together togenerate double transgenic mice, i.e., expressing both trans-genes. Male and female double transgenic mice were usedin this study. The housing conditions were controlled (tem-perature, 21 � 1°C; light from 7:00 to 19:00; humidity50–60%), and food and water were freely available. Theexperiments were conducted according to the Council ofEurope (Directive 86/609) and Finnish guidelines and ap-proved by the State Provincial Office of Eastern Finland.

Four groups of transgenic mice were used, at three ages:4 months (male � 17, female � 14), 12 months (male � 11,female � 8), and 17 months (male � 15, female � 22) forcomparison of the A�40 and A�42 levels. For the 12- and17-month-old mice we also analyzed the correlations be-tween amyloid burden and A� level and amyloid burdenand plaque number (the 4-month-old mice do not have anyplaques or amyloid deposits and were therefore not includedin this analysis). However, we measured the amyloid burden

only in a subgroup of the 17-month-old mice (male � 7,female � 6).

For biochemical measurements, the hippocampus fromone hemisphere was dissected, frozen in liquid nitrogen, andstored at �70°C until the A� assays were performed. Theother hemisphere was immersed in 4% paraformaldehydeand used for histology (see below; amyloid plaques andamyloid burden).

A�40 and A�42 ELISAs

To analyze total A�, the hippocampus was homogenizedin guanidine buffer (5.0 M guanidine–HCl/50 mM Tris–HCl, pH 8.0) in proportion to weight. The samples andA�-peptides used as standards were prepared to contain 0.5M guanidine–0.5% BSA–1 mM AEBSF in the final com-position. The levels of A�40 and A�42 were quantifiedusing Signal Select Beta Amyloid ELISA kits (BioSourceInternational, Inc.) according to the manufacturer’s proto-col. The level of total A�40 and A�42 was standardized tobrain tissue weight and expressed as nanograms of A� pergram (brain tissue).

Histology

The other hemisphere of the brain was immersed in 4%paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) andthen transferred to a 30% sucrose solution for cryoprotec-tion. The half-brains were cut in coronal sections (six 1 in 6series of 30 �m) using a freezing, sliding microtome. Thefirst series of sections was mounted on subbed slides andstained with cresyl violet, the second series was stained forA� (W0-2 antibody, mouse anti-human A�1-16; K.Beyreuther) using a standard protocol (see below), and thethird series was silver-stained using the Garvey method(Garvey et al., 1991). In short, the second series of sectionswas rinsed overnight in phosphate buffer. The next morn-ing, following a 30-min pretreatment with a Nacitrate solu-tion at 85°C, the sections were incubated in the primaryantibody solution (at 1:20,000) for 18 h on a shaker table atroom temperature (20°C) in the dark, and then the sectionswere rinsed and transferred to the solution containing thesecondary antibody (Goat anti-mouse biotin at 1:400;Sigma). After 2 h the sections were rinsed and transferred toa solution containing ExtrAvidin (1:1000) and, followingrinsing, the sections were incubated with Ni-enhanced DAB(Liu et al., 2002b). The stained sections were mounted onslides and coverslipped.

Measurements

The appropriate sections (i.e., from the hippocampal forma-tion) were digitized using a Nikon Coolpix 990 camera, andthe images were converted to grayscale using the Paint ShopPro 7 program. To avoid changes in lighting, which mightaffect measurements, all images were acquired in one session.

319J. Wang et al. / Neurobiology of Disease 14 (2003) 318–327

To analyze the amyloid burden, the area in the hippocampuscovered by A� (i.e., the amyloid burden) was measured inthree sections through the hippocampus in the W0-2-stainedmaterial using the ScionImage (NIH) program, with the A�burden being reported as stained area fraction. For counting thenumber of amyloid plaques, all W0-2-stained sections thatcontained the hippocampal formation were used. However, forthe 17-month-old animals, using a similar procedure, plaqueswere counted on the sections that were silver-stained, since thedense amyloid load (i.e., from plaques and diffuse deposits)prevented proper counting of the plaques in the A�-stainedmaterial. The plaque number is reported as the total number ofplaques counted.

Statistics

The effects of mouse age, A� levels, amyloid plaquedensity, and A� burden and their interactions were analyzedby ANOVA.

Results

A� levels, age

In male mice from the age of 4 to 12 months there wasa significant increase in the level of A�40 (Table 1; Figs. 1Aand 2A), but from the age of 12 to 17 months there was anonsignificant increase in the level of A�40 (Table 1; Figs.1A and 2A). On the other hand, the level of A�42 increasedsignificantly from 4 to 12 and from 12 to 17 months (Table1; Figs. 1A and 2B). In the female mice, however, there wasa significant increase in both the A�40 (Table 1; Figs. 1Aand 2A) and the A�42 (Table 1; Figs. 1B and 2B) level from4 to 12 and from 12 to 17 months.

In male mice, the A�42/40 ratio changed significantlyfrom a ratio of 0.2 (i.e., more A�40 then A�42) to a ratio ofapproximately 3 (i.e., more A�42 then A�40) from 4 till 12months of age, but after that age the A�42/40 ratio de-creased slightly (nonsignificant) between the age of 12 and17 months (Fig. 1C and D). Similarly, in the female mice,the A�42/40 ratio changed significantly from 0.3 to a ratioof approximately 2.2 from the age of 4 till 12 months (Fig.1C and D), and again there was a nonsignificant decrease in

the A�42/40 ratio in female mice between the ages of 12and 17 months (Fig. 1C and D).

A� levels, gender

Comparing male and female mice, there was a significanthigher level of A�40 in the female animals at all ages (i.e.,at 4, 12, and 17 months of age; Table 1, Fig. 2A). Further,at all ages there was a significant higher A�42 level infemale mice compared to male mice (Table 1, Fig. 2B). TheA�42/40 ratio was significantly higher in the young femalemice (at 4 months of age; Table 1, Fig. 1D); however, atlater ages no significant differences were present betweenthe two genders in the A�42/40 ratio.

A� levels, correlations

For both A�40 and A�42 levels, the ANOVA revealed ahighly significant effect of age (F(2,87) � 38.9 [A�40] and84.7 [A�42], P � 0.001) and gender (F(1,87) � 23 [A�40]and 53.6 [A�42], P � 0.001). The levels significantly in-creased with age (Fig. 1) and were higher in females (Fig. 2).In addition, the age � gender interaction was significant(F(2,87) � 11.2 [A�40] and 20.7 [A�42], P � 0.001), suchthat the increase with age was higher in females (Fig. 2A andB). A significant effect of age (F(2,87) � 69.5, P � 0.001) buta nonsignificant effect of gender (F(1,87) � 2.8, P � 0.05)was found for the A�42/40 ratio. The ratio increased from 4 till12 months of age, but then decreased with further aging (Fig.1C) and was lower in females (Fig. 1C). No age � genderinteraction was found for the A�42/40 ratio.

A�, histology

The number of amyloid plaques was measured using theW0-2-immunostained material only in the 4- and 12-month-old male and female mice. At 4 months of age no amyloiddeposits were present in either the silver- or the immuno-stained material (not illustrated), and at 12 months of agethe female mice had significantly more plaques in the hip-pocampus than male mice (268 � 81 and 92 � 23, respec-tively; Figs. 2 and 3). In the 17-month-old mice the plaquecounts were based on the silver-stained material, and cor-respondingly a separate statistical analysis was performed

Table 1Comparison of hippocampal amyloid� levels between male and female APP/PS1 transgenic mice at the age of 4, 12, and 17 months (mean � SEM)

4 months 12 months 17 months

M F M F M F

A�40 (ng/g) 71.1 � 2.5 89.8 � 3.2 1432 � 512 4585 � 1451 3465 � 868 11374 � 1173F (1,30) � 21.8; P � 0.001 F (1,18) � 5.3; P � 0.05 F (1,36) � 24.6; P � 0.001

A�42 (ng/g) 15.4 � 1.8 27.5 � 2.2 3564 � 993 9566 � 2249 5455 � 655 16151 � 974F (1,30) � 18.2; P � 0.001 F (1,18) � 7.2; P � 0.05 F (1,36) � 67.5; P � 0.01

A�42/A�40 ratio 0.20 � 0.02 0.30 � 0.02 3.05 � 0.22 2.25 � 0.35 2.30 � 0.29 1.80 � 0.23F (1,30) � 9.7; P � 0.004 F (1,18) � 2.4; P � 0.05 F (1,36) � 1.8; P � 0.05

320 J. Wang et al. / Neurobiology of Disease 14 (2003) 318–327

for each age group. Female mice again had significantlymore plaques than male mice (153 � 19 and 39 � 7,respectively; Figs. 2 and 3); the plaque load increased dra-matically with age and was higher in females (Fig. 2C). TheANOVA analysis revealed a significant effect of age(F(1,34) � 299.8, P � 0.001) and gender (F(1,34) � 12.5,P � 0.001) in the plaque load; the age � gender interactionwas nonsignificant (F(1,34) � 1.8, P � 0.05).

Amyloid plaques, age and gender

Amyloid plaques were measured using the W0-2-immuno-stained material in the 12-month-old male and female mice.The female mice had significantly (F(1,19) � 5.2, P � 0.05)more plaques than male mice in the hippocampus (268 � 81and 92 � 23, respectively; Figs. 2D and 3). At 17 months ofage the amyloid plaque number was measured using the silver-stained material. Female mice again had significantly (F(1,13)� 38.3, P � 0.001) more plaques than male mice (153 � 19and 39 � 7, respectively; Figs. 2D and 3).

A� burden, age and gender

The A� burden was measured in the hippocampal for-mation in all transgenic mice, i.e., in the 4-, 12-, and 17-month-old female and male mice. The amyloid burden sig-nificantly increased from 4 to 12 till 17 months of age inboth sexes (Figs. 2C and 3). Further, there was a significant(F(1,19) � 5.2, P � 0.05) higher A� burden in 12-month-old females compared to males (1.71 � 0.55 and 0.56 �0.99%, respectively); similarly, at 17 months of age thefemale mice had a significantly (F(1,13) � 6.1, P � 0.05)higher A� burden compared to the male mice (11.52 � 0.82and 8.95 � 0.66%, respectively; Fig. 2C).

Correlations among A�42 level, amyloid burden, andamyloid plaque number in male and female mice at theage of 12 months

There was a significant correlation between the A�42level and amyloid burden (r � 0.76, P � 0.05; Fig. 4A) and

Fig. 1. Hippocampal A�40 (A) and A�42 (B) levels (ng/g) in male and female APP/PS1 double transgenic mice at the age of 4, 12, and 17 months. C displaysthe A� burden in the hippocampus in male and female APP/PS1 double transgenic mice at the age of 4, 12, and 17 months (A� burden is defined as thepercentage of area in the measurement field occupied by A�), and D shows the number of plaques at 12 and 17 months of age. The number of amyloid plaquesis counted using the W0–2-immunostained material at the age of 4 and 12 months and the silver-stained material at the age of 17 months in the male andfemale APP/PS1 double transgenic mice. Asterisks denote significant differences between the sexes; *P � 0.05; **P � 0.01; ***P � 0.001.

321J. Wang et al. / Neurobiology of Disease 14 (2003) 318–327

between amyloid plaque number and amyloid burden (r �0.72, P � 0.05; Fig. 4C) in 12-month-old female mice.There was no significant correlation between A�42 leveland amyloid burden (r � 0.44, P � 0.05; Fig. 4B), but therewas a significant correlation between amyloid plaque num-ber and amyloid burden (r � 0.60, P � 0.05; Fig. 4D) in12-month-old male mice.

Correlations between the amyloid burden and amyloidplaque number in male and female mice at the age of 17months

There was a significant correlation between the amyloidplaque number and amyloid burden (r � 0.89, P � 0.05) in17-month-old female mice. Similarly, there was a signifi-cant correlation between amyloid plaque number and amy-loid burden (r � 0.88, P � 0.004) in 17-month-old malemice.

Discussion

Our data show that gender plays an important role in thepathogenesis of AD, i.e., sex-related differences are presentin A� levels and in amyloid deposits in the hippocampus inour AD model mice. First, the A�40 level and the A�42level are significantly increased in female mice compared tomale mice, at all ages measured (i.e., at 4, 12, and 17months of age). Second, at both 12 and 17 months of age theamyloid burden is substantially higher in female mice com-pared to male mice (at 4 months of age there are no depos-its).

Gender differences in the prevalence rate of AD havebeen shown by some epidemiological studies (Molsa et al.,1982; Jorm et al., 1987; Hagnell et al., 1992; Letenneur etal., 1994; Brayne et al., 1995; Fratiglioni et al., 1997);however, some more recent studies have suggested that

Fig. 2. Hippocampal A�40 and A�42 levels (ng/g) comparing male (A) and female (B) APP/PS1 double transgenic mice at the ages of 4, 12, and 17 months.C demonstrates the comparison of hippocampal A�42/40 ratio in between male and female transgenic mice at the age of 4, 12, and 17 months, and D showsthe A�42/40 ratio development over the three ages measured. The asterisks denote significant differences between A�40 and A�42 levels and A�42/40 ratioin mice of the same age. ***P � 0.001; # denotes significant differences between A�42/40 ratio in different ages of mice of the same sex, #P � 0.004.

322 J. Wang et al. / Neurobiology of Disease 14 (2003) 318–327

gender does not play a role in AD prevalence (Ganguli etal., 2000; Ruitenberg et al., 2001). The mechanisms thatunderlie this putative gender distinction in AD are not clear,but the disparity in estrogen level between males and fe-males, together with the dramatic change in estrogen levelwith menopause, has been hypothesized to cause these gen-der differences. Clinical trials with estrogen supplementa-tion in women have suggested positive effects on the pro-gression and incidence of AD (Fillit et al., 1986; Honjo etal., 1989, 2001; Henderson et al., 1994, 2000; Asthana et al.,

1999). Contrastingly, results from more recent papers sug-gest that estrogen supplementation is not beneficial forslowing the progression of Alzheimer’s disease (Broe et al.,1990; Mortel and Meyer, 1995). Further, Petanceska et al.(2000) have shown that in very young, normal guinea pigsovariectomy causes an increase of 50% in the level of A�.Recently two groups have studied the effects of ovariec-tomy on A� levels in young APP transgenic mice. Levin-Allerhand and Smith (2002) demonstrated in APPswe micethat the level of A� in the brain increased nonsignificantly,

Fig. 3. Eight low-power photomicrographs demonstrating amyloid deposits and plaques in cortex and hippocampus in male and female APP � PS1 doubletransgenic mice at ages of 12 and 17 months. The top four photographs show the hippocampus and dorsal cortex in material that is stained with the W0–2antibody. The four pictures at the bottom are of the adjacent silver-stained sections (Garvey’s silver stain).

323J. Wang et al. / Neurobiology of Disease 14 (2003) 318–327

but that the level of APP decreased significantly; however,they noted that very young ovariectomy (4 weeks of age)also led to early death in the transgenic mice. In anotherrecent study they demonstrated that ovariectomy in APPswemice does not change the level of A� (Levin-Allerhand etal., 2002); however, treatment with estrogen decreased theamount of A� in the brains of these animals. The levels ofAPP protein, however, were not changed in these animals,but the A�/sAPP� ratio significantly increased in the ovari-ectomized mice, together suggesting that processing of APPby the �-secretase is increased by estrogen (Levin-Aller-hand et al., 2002). Surprisingly, however, this study showedthat treatment with 17�-estradiol was more effective thanwith 17�-estradiol (Levin-Allerhand et al., 2002). In con-trast, Zheng et al. (2002) showed that ovariectomy in APPtransgenic mice (Tg2576) did increase the levels of A�40and A�42 in the brain and that estrogen replacement ther-apy reduced this increase in A� levels. However, followingthe same experiment in the Tg2576�PS1 double mutantmice this effect was much reduced. Together these datasuggest a relation between a decline in estrogen levels andan increase the generation of A�, but the physiological basisfor this effect is still unclear. It must be noted that prelim-inary studies from our lab have shown that long-term ovari-

ectomy (i.e., a period of 9 months) in our AD model micedoes not change the amount of deposited amyloid� at 12months of age. Additionally, it should be noted that in ourmice the sex differences are present at all ages, from beforemenopause is present (i.e., at 4 and 12 months of age).Further, at 17 months of age most female mice are noncyc-ling (e.g., Frick et al., 2000), but this does not significantlyincrease the amyloid burden.

However, estrogen can also have some indirect effects onA� generation, for example, by direct trophic effects onneurons (Luine, 1985; Keefe et al., 1994; Lustig, 1994;McEwen and Woolley, 1994), and it may act as an antiox-idant (Behl et al., 1995; Gridley et al., 1997). A recent studyby Lee et al. (Lee et al., 2002) has suggested that a sexdifference in levels of (synaptic) zinc contributes to thegender-disparate plaque formation in Tg2576 mice (APP-swe mutation). They have shown that females have muchhigher plaque numbers compared to age-matched males,that old females have higher levels of zinc than male mice,and that animals with a lower amount of the zinc transporterexpression also have lower numbers of plaques. However,the gender differences in A� deposits disappear in theanimals lacking the zinc transporter. Further, they demon-strate that higher levels of zinc significantly correlate with

Fig. 4. Four line graphs demonstrating the correlations between the amyloid burden and A�42 level (A, male, B, female) and amyloid burden and numberof plaques in 12-month-old male (C) and female (D) APP/PS1 double transgenic mice. The asterisk denotes a significant difference between male and femalemice, *P � 0.05.

324 J. Wang et al. / Neurobiology of Disease 14 (2003) 318–327

both insoluble A�40 and A�42 levels. Several studies havesuggested that AD plaques have high levels of zinc and thatzinc contributes to A� aggregation (Bush and Tanzi, 2002).It must be noted, however, that the analysis of our Timm’sstained material does not suggest different amounts of zincbetween the sexes in our mice (unpublished observations).

Comparing our results with the data of Callahan et al.(Callahan et al., 2001), although two different strains oftransgenic mice have been used (that present slightly dif-ferent phenotypes of A� accumulation), the A� burden isincreased similarly in aged mice, with a significantly higherburden in female compared to male mice. However, in theCallahan et al. study, which uses the Tg2576 mice carryingonly the APPswe mutation, a higher A�40 level comparedto the A�42 level at 15 months of age was shown, theopposite of our mice which show a much higher A�42 level.Further, they (Callahan et al., 2001) have shown that in15-month-old female mice there are significantly higherlevels of A�40 compared to male mice of the same age,whereas there is no difference in the A�42 level betweenfemale and male mice. In our study (which uses transgenicmice carrying both APPswe and PS1 mutations) we dem-onstrate a higher A�42 than A�40 level at 12 and 17months of age; likely this is caused by the combination ofmutations, which has been shown to specifically increasethe production of A�42 compared to the production ofA�40 (Borchelt et al., 1996). Our study shows that there isno significant difference in the ratio of A�42 to A�40between female and male mice at these ages (12 and 17months of age), even when there are clear differences in theamount of amyloid� deposits. One possible explanation forthis are the time points used in this study. We have dem-onstrated that till the age of 12 months there is a steadyincrease in the A�42/40 ratio in male mice (4 months � 0.2,8 months � 1.5, and 12 months � 3.0). Further, there is adramatic increase in the levels of both A�40 (from 155 to1432 ng/g) and A�42 (from 242 to 3564 ng/g) from 8 to 12months of age (vide infra and Liu et al., 2002b, 2000c).After 12 months of age there is a relatively stable increasein both A�40 and A�42 levels to 17 months of age. Itshould be noted that males do not yet have any pathology at8 months of age, but do show deposits at 12 months of age.The relatively unchanged ratio during the period of 12 to 17months (both in female and in male mice) indicates thatgender likely affects only the overproduction of APP, butnot the generation of its breakdown product, i.e., A�. Re-cently, the group of Duff has shown that A�42 levels aresignificantly higher in estrogen-deprived mice than intactmice, and this effect could be reversed through the admin-istration of estradiol. It must be noted, however, that theseresults have been obtained in young Tg2576 mice, i.e., at anage before plaque deposition is present (Zheng et al., 2002).Further, they have demonstrated a similar result in doubletransgenic mice (i.e., Tg2576 and PS1) at the age when theyare starting to form amyloid deposits, but they have notmeasured any estrogenic effects on plaque deposition

(Zheng et al., 2002). These data suggest that, in vivo, es-trogen depletion can lead to the accumulation of A� in thebrain, which can be reversed through replacement of estra-diol (Zheng et al., 2002). However, a recent paper by Levin-Allerhand and Smith (2002) indicates that ovariectomy in-creases A� levels (nonsignificantly) by approximately 50%at 12 months of age in APPswe mice, but decreases APPlevels and the level of secreted APP (sAPP). In a furtherstudy, they (Levin-Allerhand et al., 2002) have demon-strated that ovariectomy did not change brain A� levels, butestrogen treatment did reduce the A� level, and no changewas found in the amount of APP in the brain. In contrast, thelevel of sAPP� was decreased in the ovariectomized mice,but treatment with estrogen increased the level. It should benoted, however, that 17�-estradiol was more effective than17�-estradiol (Levin-Allerhand et al., 2002). Together thissuggests that estrogen levels possibly do influence the pro-cessing of APP, and thus the levels of A�, and A� deposi-tion, but on the other hand our data indicate that estrogen isnot the most likely candidate for the gender difference weobserve. It should be noted that our mice have an overpro-duction of APP and that the double mutation causes anincrease in the levels of A�42 (Borchelt et al., 1997).

A possible explanation for the observed gender differ-ence in amyloid deposition could be ApoE. It has beendemonstrated that the ApoE isoform is a risk factor inAlzheimer’s disease (Soininen and Riekkinen, 1996) andthat the different isoforms have distinct pathologies in trans-genic mice expressing both AD genes and human ApoEisoforms (Raber et al., 1998). Importantly, it has beenshown that the effects of human ApoE isoforms are genderspecific, with females being more sensitive than male miceto the effects of the ApoE isoform (Raber et al., 1998),suggesting that ApoE could differently influence the rate ofamyloid deposition in the two sexes.

The current description of differences between the twosexes in A� accumulation in AD model mice (carrying bothAPP and PS1 mutations), together with the gender differ-ence in elevated A�40 and A�42 levels, strongly suggeststhat gender has an effect on the pathogenesis of AD. Com-bining the data from our animal model with the data fromhuman studies may provide useful information for the fur-ther study of gender differences in AD, especially in thecontext of the amyloid hypothesis.

Acknowledgments

This work was supported by the Academy of Finland(Grant 46000) and EVO-Grant (Grant 5510) of KuopioUniversity Hospital. We thank Dr. David Borchelt (JohnsHopkins University, Baltimore, MD, USA) for providingthe APP and PS1 mouse lines and Dr. Tobias Hartmann(Heidelberg University, Germany) for providing the W0-2antibody for the study. The authors also thank Pasi Mietti-nen and Paivi Rasanen for their skillful technical assistance.

325J. Wang et al. / Neurobiology of Disease 14 (2003) 318–327

References

Andersen, K., Launer, L.J., Dewey, M.E., Letenneur, L., Ott, A., Copeland,J.R., Dartigues, J.F., Kragh-Sorensen, P., Baldereschi, M., Brayne, C.,Lobo, A., Martinez-Lage, J.M., Stijnen, T., Hofman, A., 1999. Genderdifferences in the incidence of AD and vascular dementia: theEURODEM Studies. EURODEM Incidence Research Group. Neurol-ogy 53, 1992–1997.

Asthana, S., Craft, S., Baker, L.D., Raskind, M.A., Birnbaum, R.S.,Lofgreen, C.P., Veith, R.C., Plymate, S.R., 1999. Cognitive and neu-roendocrine response to transdermal estrogen in postmenopausalwomen with Alzheimer’s disease: results of a placebo-controlled, dou-ble-blind, pilot study. Psychoneuroendocrinology 24, 657–677.

Behl, C., Widmann, M., Trapp, T., Holsboer, F., 1995. 17-beta Estradiolprotects neurons from oxidative stress-induced cell death in vitro.Biochem. Biophys. Res. Commun. 216, 473–482.

Borchelt, D.R., Ratovitski, T., van Lare, J., Lee, M.K., Gonzales, V.,Jenkins, N.A., Copeland, N.G., Price, D.L., Sisodia, S.S., 1997. Accel-erated amyloid deposition in the brains of transgenic mice coexpressingmutant presenilin 1 and amyloid precursor proteins. Neuron 19, 939–945.

Borchelt, D.R., Thinakaran, G., Eckman, C.B., Lee, M.K., Davenport, F.,Ratovitsky, T., Prada, C.M., Kim, G., Seekins, S., Yager, D., Slunt,H.H., Wang, R., Seeger, M., Levey, A.I., Gandy, S.E., Copeland, N.G.,Jenkins, N.A., Price, D.L., Younkin, S.G., Sisodia, S.S., 1996. FamilialAlzheimer’s disease-linked presenilin 1 variants elevate Abeta1-42/1-40 ratio in vitro and in vivo. Neuron 17, 1005–1013.

Braak, H., Braak, E., 1991. Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol. (Berl.) 82, 239–259.

Brayne, C., Gill, C., Paykel, E.S., Huppert, F., O’Connor, D.W., 1995.Cognitive decline in an elderly population—a two wave study ofchange. Psychol. Med. 25, 673–683.

Broe, G.A., Henderson, A.S., Creasey, H., McCusker, E., Korten, A.E.,Jorm, A.F., Longley, W., Anthony, J.C., 1990. A case–control study ofAlzheimer’s disease in Australia. Neurology 40, 1698–1707.

Bush, A.I., Tanzi, R.E., 2002. The galvanization of beta-amyloid in Alz-heimer’s disease. Proc. Natl. Acad. Sci. USA 99, 7317–7319.

Callahan, M.J., Lipinski, W.J., Bian, F., Durham, R.A., Pack, A., Walker,L.C., 2001. Augmented senile plaque load in aged female beta-amyloidprecursor protein-transgenic mice. Am. J. Pathol. 158, 1173–1177.

Cholerton, B., Gleason, C.E., Baker, L.D., Asthana, S., 2002. Estrogen andAlzheimer’s disease: the story so far. Drugs Aging 19, 405–427.

Evans, D.A., Funkenstein, H.H., Albert, M.S., Scherr, P.A., Cook, N.R.,Chown, M.J., Hebert, L.E., Hennekens, C.H., Taylor, J.O., 1989. Prev-alence of Alzheimer’s disease in a community population of olderpersons. Higher than previously reported. J. Am. Med. Assoc. 262,2551–2556.

Fillit, H., Weinreb, H., Cholst, I., Luine, V., McEwen, B., Amador, R.,Zabriskie, J., 1986. Observations in a preliminary open trial of estradioltherapy for senile dementia–Alzheimer’s type. Psychoneuroendocrinol-ogy 11, 337–345.

Fillit, H.M., 2002. The role of hormone replacement therapy in the pre-vention of Alzheimer disease. Arch. Intern. Med. 162, 1934–1942.

Fratiglioni, L., Launer, L.J., Andersen, K., Breteler, M.M., Copeland, J.R.,Dartigues, J.F., Lobo, A., Martinez-Lage, J., Soininen, H., Hofman, A.,2000. Incidence of dementia and major subtypes in Europe: a collab-orative study of population-based cohorts. Neurologic Diseases in theElderly Research Group. Neurology 54, S10–S15.

Fratiglioni, L., Viitanen, M., von Strauss, E., Tontodonati, V., Herlitz, A.,Winblad, B., 1997. Very old women at highest risk of dementia andAlzheimer’s disease: incidence data from the Kungsholmen Project,Stockholm. Neurology 48, 132–138.

Frick, K.M., Burlingame, L.A., Arters, J.A., Berger-Sweeney, J., 2000.Reference memory, anxiety and estrous cyclicity in C57BL/6NIA miceare affected by age and sex. Neuroscience 95, 293–307.

Games, D., Adams, D., Alessandrini, R., Barbour, R., Berthelette, P.,Blackwell, C., Carr, T., Clemens, J., Donaldson, T., Gillespie, F., et al.,1995. Alzheimer-type neuropathology in transgenic mice overexpress-ing V717F beta-amyloid precursor protein. Nature 373, 523–527.

Ganguli, M., Dodge, H.H., Chen, P., Belle, S., DeKosky, S.T., 2000.Ten-year incidence of dementia in a rural elderly US communitypopulation: the MoVIES Project. Neurology 54, 1109–1116.

Garvey, W., Fathi, A., Bigelow, F., Jiminez, C.L., Carpenter, B.F., 1991.Rapid, reliable and economical silver stain for neurofibrillary tanglesand senile plaques. J. Histotechnol. 14, 39–42.

Gridley, K.E., Green, P.S., Simpkins, J.W., 1997. Low concentrations ofestradiol reduce beta-amyloid (25–35)-induced toxicity, lipid peroxi-dation and glucose utilization in human SK-N-SH neuroblastoma cells.Brain Res. 778, 158–165.

Hagnell, O., Ojesjo, L., Rorsman, B., 1992. Incidence of dementia in theLundby Study. Neuroepidemiology 11, 61–66.

Hardy, J., 1997. The Alzheimer family of diseases: many etiologies, onepathogenesis? Proc. Natl. Acad. Sci. USA 94, 2095–2097.

Henderson, V.W., Paganini-Hill, A., Emanuel, C.K., Dunn, M.E., Buck-walter, J.G., 1994. Estrogen replacement therapy in older women.Comparisons between Alzheimer’s disease cases and nondementedcontrol subjects. Arch. Neurol. 51, 896–900.

Henderson, V.W., Paganini-Hill, A., Miller, B.L., Elble, R.J., Reyes, P.F.,Shoupe, D., McCleary, C.A., Klein, R.A., Hake, A.M., Farlow, M.R.,2000. Estrogen for Alzheimer’s disease in women: randomized, dou-ble-blind, placebo-controlled trial. Neurology 54, 295–301.

Honjo, H., Kikuchi, N., Hosoda, T., Kariya, K., Kinoshita, Y., Iwasa, K.,Ohkubo, T., Tanaka, K., Tamura, T., Urabe, M., Kawata, M., 2001.Alzheimer’s disease and estrogen. J. Steroid Biochem. Mol. Biol. 76,227–230.

Honjo, H., Ogino, Y., Naitoh, K., Urabe, M., Kitawaki, J., Yasuda, J.,Yamamoto, T., Ishihara, S., Okada, H., Yonezawa, T., et al., 1989. Invivo effects by estrone sulfate on the central nervous system—seniledementia (Alzheimer’s type). J. Steroid Biochem. 34, 521–525.

Hsiao, K., Chapman, P., Nilsen, S., Eckman, C., Harigaya, Y., Younkin, S.,Yang, F., Cole, G., 1996. Correlative memory deficits, Abeta elevation,and amyloid plaques in transgenic mice. Science 274, 99–102.

Hyman, B.T., Van Horsen, G.W., Damasio, A.R., Barnes, C.L., 1984.Alzheimer’s disease: cell-specific pathology isolates the hippocampalformation. Science 225, 1168–1170.

Jorm, A.F., Korten, A.E., Henderson, A.S., 1987. The prevalence of de-mentia: a quantitative integration of the literature. Acta Psychiatr.Scand. 76, 465–479.

Keefe, D., Garcia-Segura, L.M., Naftolin, F., 1994. New insights intoestrogen action on the brain. Neurobiol. Aging 15, 495–497.

Lee, J.Y., Cole, T.B., Palmiter, R.D., Suh, S.W., Koh, J.Y., 2002. Contri-bution by synaptic zinc to the gender-disparate plaque formation inhuman Swedish mutant APP transgenic mice. Proc. Natl. Acad. Sci.USA 99, 7705–7710.

Letenneur, L., Commenges, D., Dartigues, J.F., Barberger-Gateau, P.,1994. Incidence of dementia and Alzheimer’s disease in elderly com-munity residents of south-western France. Int. J. Epidemiol. 23, 1256–1261.

Levin-Allerhand, J.A., Smith, J.D., 2002. Ovariectomy of young mutantamyloid precursor protein transgenic mice leads to increased mortality.J. Mol. Neurosci. 19, 163–166.

Levin-Allerhand, J.A., Lominska, C.E., Wang, J., Smith, J.D., 2002. 17�-Estradiol and 17�-estradiol treatments are effective in lowering cere-bral amyloid-� levels in A�PPSWE transgenic mice. J. AlzheimersDis. 4, 449–457.

Liu, L., Ikonen, S., Heikkinen, T., Heikkila, M., Puolivali, J., van Groen,T., Tanila, H., 2002a. Effects of fimbria–fornix lesion and amyloidpathology on spatial learning and memory in transgenic APP � PS1mice. Behav. Brain Res. 134, 433–445.

Liu, L., Ikonen, S., Heikkinen, T., Tapiola, T., van Groen, T., Tanila, H.,2002b. The effects of long-term treatment with metrifonate, a cholines-terase inhibitor, on cholinergic activity, amyloid pathology, and cog-

326 J. Wang et al. / Neurobiology of Disease 14 (2003) 318–327

nitive function in APP and PS1 doubly transgenic mice. Exp. Neurol.173, 196–204.

Liu, L., Ikonen, S., Tapiola, T., Tanila, H., van Groen, T., 2002c. Fimbria-fornix lesion does not affect APP levels and amyloid deposition in thehippocampus of APP � PS1 double transgenic mice. Exp. Neurol. 177,565–574.

Luine, V.N., 1985. Estradiol increases choline acetyltransferase activity inspecific basal forebrain nuclei and projection areas of female rats. Exp.Neurol. 89, 484–490.

Lustig, R.H., 1994. Sex hormone modulation of neural development invitro. Horm. Behav. 28, 383–395.

McEwen, B.S., Woolley, C.S., 1994. Estradiol and progesterone regulateneuronal structure and synaptic connectivity in adult as well as devel-oping brain. Exp. Gerontol. 29, 431–436.

Molsa, P.K., Marttila, R.J., Rinne, U.K., 1982. Epidemiology of dementiain a Finnish population. Acta Neurol. Scand. 65, 541–552.

Mortel, K.F., Meyer, J.S., 1995. Lack of postmenopausal estrogen replace-ment therapy and the risk of dementia. J. Neuropsychiatry Clin. Neu-rosci. 7, 334–337.

Petanceska, S.S., Nagy, V., Frail, D., Gandy, S., 2000. Ovariectomy and17beta-estradiol modulate the levels of Alzheimer’s amyloid beta pep-tides in brain. Exp. Gerontol. 35, 1317–1325.

Price, D.L., Sisodia, S.S., Gandy, S.E., 1995. Amyloid beta amyloidosis inAlzheimer’s disease. Curr. Opin. Neurol. 8, 268–274.

Raber, J., Wong, D., Buttini, M., Orth, M., Bellosta, S., Pitas, R.E.,Mahley, R.W., Mucke, L., 1998. Isoform-specific effects of humanapolipoprotein E on brain function revealed in ApoE knockout mice:increased susceptibility of females. Proc. Natl. Acad. Sci. USA 95,10914–10919.

Ruitenberg, A., Ott, A., van Swieten, J.C., Hofman, A., Breteler, M.M.,2001. Incidence of dementia: does gender make a difference? Neuro-biol. Aging 22, 575–580.

Soininen, H.S., Riekkinen Sr., P.J., 1996. Apolipoprotein E, memory andAlzheimer’s disease. Trends Neurosci. 19, 224–228.

Tang, M.X., Jacobs, D., Stern, Y., Marder, K., Schofield, P., Gurland,B., Andrews, H., Mayeux, R., 1996. Effect of oestrogen duringmenopause on risk and age at onset of Alzheimer’ s disease. Lancet348, 429 –432.

Van Groen, T., Puolivali, J., Tapiola, T., Iivonen, S., Miettinen, P., Kadish,I., 2000. Differences in amyloid-� levels and in development of pa-thology during aging between male and female APP/PS1 mutant mice.Six International Stockholm/Springfield Symposium on Advances inAlzheimer Therapy. April 5–8, 2000, Stockholm. (abstract 230).

Zheng, H., Xu, H., Uljon, S.N., Gross, R., Hardy, K., Gaynor, J., Lafran-cois, J., Simpkins, J., Refolo, L.M., Petanceska, S., Wang, R., Duff, K.,2002. Modulation of A� peptides by estrogen in mouse models. J. Neu-rochem. 80, 191–196.

327J. Wang et al. / Neurobiology of Disease 14 (2003) 318–327