Increased corticosterone secretion and early-onset of cognitive decline in female apolipoprotein E-knockout mice

Increased corticosterone secretion and early-onset of cognitive decline in female apolipoprotein E-knockout mice

Behavioural Brain Research 148 (2004) 167–177 Research report Increased corticosterone secretion and early-onset of cognitive decline in female apol...

127KB Sizes 0 Downloads 8 Views

Behavioural Brain Research 148 (2004) 167–177

Research report

Increased corticosterone secretion and early-onset of cognitive decline in female apolipoprotein E-knockout mice Jeannette Grootendorst1 , Leo Enthoven, Sergiu Dalm, E. Ron de Kloet, Melly S. Oitzl∗ Division of Medical Pharmacology, Leiden/Amsterdam Center for Drug Research, Leiden University Medical Center, Leiden University, P.O. Box 9502, 2300 RA, Leiden, The Netherlands Received 15 May 2003; accepted 16 May 2003

Abstract In the present study, the interaction of age and apolipoprotein E (apoE)-genetic background on cognitive abilities was investigated in young (5–6 months) and aged (14–16 months) female apolipoprotein E-knockout (apoE0/0) and wild-type mice. Cognitive abilities are known to be affected by the steroid hormones corticosterone and estrogen. Therefore, we measured the activity and reactivity of the hypothalamic–pituitary–adrenal (HPA) axis expressed by circadian corticosterone concentrations and responses to novelty and controlled the regularity of the estrous cycle. Young female apoE0/0 mice acquired the water maze task and showed a similar latency and search strategy to locate the platform as young female wild-type mice. Similar corticosterone responses to novelty were observed in both genotypes. Regularity of the estrous cycle was disturbed in a small percentage of the young apoE0/0 female mice. However, in aged female apoE0/0 mice water maze performance was impaired with search strategies less persistent than in aged wild-type mice. In parallel, increased corticosterone concentrations were measured in apoE0/0 mice in response to novelty and during the circadian cycle. The percentage of mice with an irregular estrous cycle increased with age, but was comparable for apoE0/0 and wild-type mice. Thus, although disruption of the apoE gene affects the regularity of the estrous cycle in young mice, it is the enhanced corticosterone secretion, which parallels the cognitive decline in the aging female apoE0/0 mice. © 2003 Elsevier B.V. All rights reserved. Keywords: Aging; Spatial learning; Estrogen; Alzheimer’s Disease; Stress; Estrous cycle

1. Introduction Besides age, inheritance of the apolipoprotein E (apoE) ε4 allele is an important risk factor for the development of sporadic Alzheimer’s disease, the most common form of this illness [3]. Using the apolipoprotein E-knockout (apoE0/0) mouse as an animal model, the potential physiological importance of murine apoE has been assessed. These studies have indicated that apoE plays an essential role in neurodegeneration and regeneration processes [15,17], and is involved in learning and memory processes

Abbreviations: apoE0/0, apolipoprotein E knockout; apoE, apolipoprotein E; vs., versus; S.E.M., standard error of the mean; HPA, hypothalamic–pituitary–adrenal ∗ Corresponding author. Tel.: +31-71-5276210; fax: +31-71-5274715. E-mail address: [email protected] (M.S. Oitzl). 1 Present address: Laboratoire de Neurosciences Comportementales et Cognitives, Universit´e Louis Pasteur, CNRS UMR 7521, 12 rue Goethe, 67000 Strasbourg, France. 0166-4328/$ – see front matter © 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0166-4328(03)00188-8

[19,26,31,40]. Despite the fact that human female carriers of apoE ε4 have a greater risk of Alzheimer’s disease than male carriers [35,41], most studies with apoE0/0 or apoE-transgenic mice were performed with male mice. It was found that male apoE0/0 mice showed spatial learning deficits [25,31,39,40,57]. Three studies with female apoE0/0 mice have been published, of which two failed to demonstrate impairment of water maze performance [1,40,54]. These results suggest that there is a gender effect which depends on the presence of apoE. Indeed, an intriguing relationship has been found between apoE and estradiol in the brain. It has been shown that estradiol induced apoE mRNA levels in the brain [47,48,50] and that synapse sprouting is stimulated by estrogen in an apoE-dependent fashion [49,52]. In the course of female life, estrogen concentrations vary during the estrous cycle and at later ages the cycle becomes irregular or stops [8,29]. A growing body of evidence suggests that estrogen replacement therapy slows down the progression and delays the onset of Alzheimer’s disease [34,46,51]. Apart from effects on synaptic plasticity in the

168

J. Grootendorst et al. / Behavioural Brain Research 148 (2004) 167–177

hippocampus [56], several studies in mice and rats demonstrated that estrogen is implicated in spatial learning abilities [10,20,24,33,42,55]. In view of these findings, it is conceivable that gender effects may occur in aging of apoE0/0 mice. Cognitive decline during the aging process is often associated with hypercorticism and a flattened circadian rhythm in circulating corticosteroids, at least in a subgroup of individuals [21,22,27,28]. Interestingly, Peskind et al. [36] found that higher cortisol concentrations in cerebro-spinal fluid were associated with increased frequency of the apoE ε4 allele. We demonstrated the association of corticosterone and the lack of apoE in young adult male apoE0/0 mice which showed impaired water maze performance in parallel with elevated corticosterone secretion under basal conditions as well as in response to a novelty stressor [12]. Recently, we showed that the effect of corticosterone on cognitive performance depends on the apoE-genetic background [14]. These observations suggest that apoE, corticosterone, and cognitive performance are linked. Since estrogens are known to improve cognitive performance and to attenuate age-induced corticosterone responses to stress [2,7,18], we were interested if apoE is implicated in age-related changes in spatial learning and stress responsiveness in female mice. In the present study, we tested the hypothesis that in the aging female apoE0/0 mouse, the activity of the hypothalamic–pituitary–adrenal (HPA) axis, reflected in the pattern of corticosterone secretion, is associated with cognitive performance. For this purpose young (5–6 months) and aged (14–16 months) female mice were examined for the relationship between apoE, corticosterone, and performance in the Morris water maze. The activity of the HPA axis was monitored at two levels. First, basal activity was determined by measuring corticosterone concentrations in the morning and evening, representative for the basal trough and peak of the circadian rhythm. Second, the reactivity of the HPA axis was estimated by monitoring the corticosterone secretion during exposure to a novel clean cage. Furthermore, we determined if the regularity of the estrous cycle was dependent on apoE and/or age. Therefore, the estrous cycle of all mice was controlled before exposing them to the behavioral task. We expected that during aging, female apoE0/0 mice will have elevated corticosterone concentrations compared to wild-type mice and that in apoE0/0 mice, this increased HPA axis activity coincides with impaired water maze performance.

2. Materials and methods 2.1. Animals ApoE0/0 mice were generated from C57BL/6J and 129/Sv-Ev mice as described previously by interbreeding heterozygous mutants to obtain mice homozygous for the disrupted apoE allele [53]. ApoE0/0 mice, backcrossed to C57BL/6J for nine generations, and wild-type controls

(siblings or C57BL/6J) were bred and housed under SPF conditions in the transgenic animal facilities of TNO, Leiden, The Netherlands. When mice reached 4–5 months (young) or 13–15 months (aged) of age, they were transported to our department and allowed to acclimatize to their new environment for at least 14 days (food and water ad libitum; lights on from 07.00 to 19.00 h in a temperature (21 ± 1 ◦ C) and humidity-controlled room). Since survival rate of apoE0/0 mice decreased from 18 months onwards (observations at TNO breeding facilities), we took 14–16 months old mice for the aged experimental groups. Testing took place between 08.30 and 14.00 h. Experiments were approved by the Local Committee for Animal Health, Ethics and Research of the University of Leiden and carried out in accordance with the European Communities Council Directive of November 1986 (86/609/EEC). 2.2. Experimental design To determine the regularity of the estrous cycle, vaginal smears were collected daily of all mice over a period of three weeks (see Section 2.5). Mice were housed singly and left undisturbed for one week before behavioral and endocrine testing started. Blood samples were collected in the morning and in the evening, i.e. around the time of the circadian trough and peak of corticosterone secretion. Subsequently, mice were divided over two groups. One week later, one group was trained in the water maze while the other group was subjected to novelty stress, i.e. a novel clean cage without sawdust. During the whole experimental period, bodyweights were monitored weekly. Mice were tested at 5–6 months of age (young adult female mice; water maze: apoE0/0, n = 12; wild-type, n = 13; novelty stress: apoE0/0, n = 14; wild-type, n = 12) and at 14–16 months of age (aged female mice; water maze: apoE0/0, n = 16; wild-type, n = 10; novelty stress: apoE0/0, n = 21; wild-type, n = 20). 2.3. Water maze schedule and procedure 2.3.1. Handling of the animals Particular attention was paid to handle the mice gently and quietly. From their cages, mice were picked up at the base of their tail and placed in the water maze. When search latencies exceeded 60 s, a metal grid (5 cm×20 cm) was used to guide the animals to the platform of the water maze. The same grid was used to remove the animals from the water maze. Upon presentation of the grid, animals climbed onto it and could easily be transported to their home cage. Any unwanted punishment for finding the platform or chasing the mouse through the pool was hence avoided by this procedure. 2.3.2. Water maze procedure Mice were tested in the water maze for their spatial learning abilities. A pool (white, diameter 140 cm) was filled with warm water (26 ± 1 ◦ C) and made opaque by the addition

J. Grootendorst et al. / Behavioural Brain Research 148 (2004) 167–177

of chalk. A platform (8 cm in diameter) was situated 1 cm below the surface of the water, invisible for the animal (spatial condition) or 1 cm above the water level (dark-colored rim; visible condition). In the training trials the pool was divided into four quadrants with the platform in the middle of one of the quadrants. For each trial, the mouse was placed in the water at one of the four possible starting positions. A maximum of 60 s was allowed, during which the mouse had to find the platform and climb onto it, where it remained for 10 s. If the animal did not find the platform, it was guided there with a grid and was allowed to stay for 10 s on the platform. Animals were run sequentially with an inter-trial interval of approximately 5 min. After each trial, mice were placed under a red-light warming lamp to dry. Free swim trials (no platform present) were run before and during training. The mouse was placed into the water, opposite to the former location of the platform, and allowed to swim for 60 s. 2.3.3. Training schedule Three days before spatial training in the water maze started, the pool was filled with 2 cm of warm water and a large flat object to climb on. This was the mouse’s first contact with water and each mouse was allowed to move around for 120 s (water adaptation trial). Water maze training on day 1 started with a 120 s free swim in absence of the platform (before training). This allowed estimation of the ability of the mice to swim and to determine the pre-training swim pattern of the animals (i.e. their exploratory strategy), indicative for any preferences for a certain part of the pool that may be present already. One hour later, spatial training started which consisted of 23 trials over 4 consecutive days with a submerged platform (see Fig. 1 for details). Within a training day, inter-trial intervals were 5 min, but 60 min between trials 7–8 (day 2), trials 15–16 (day 3), and trials 20–21 (day 4). These variable intervals between trials (5 and 60 min, 24 h) served to challenge possible differences in performance. On day 3, after 15 trials, the platform was removed, and the search strategy used during this second free swim trial was assessed. On day 5, motor and sensory abilities were observed during four visible platform trials. The platform was positioned in one of the other three quadrants with respect to the position of the submerged platform and for each trial, the position of the platform was changed. For all platform training trials, we assessed the time needed (seconds), distance swum (centimeters) to find and climb on the platform. We analyzed the first 60 s of the free swim trials. Total travelled distance indicates the level of general activity. The swim pattern was quantified by travelled distance, swim velocity (cm/s), the standard measures of latency to and crossings of the platform, time spent in platform quadrant and the more advanced measure cumulative distance to the platform location. Cumulative distance is the calculated position of the animal with respect to platform locations, 5 times per second. Originally, Gallagher et al. presented cumulative distance to zone (e.g. platform)

169

as a measure to quantify and compare search patterns for the platform of old and young rats [11]. We showed that the cumulative distance to zone, combined with the time spent in the platform quadrant, allowed to identify specific and characteristic swim patterns of the mice which were defined as persistent, concentric, or random search strategies [4]. This characterization of the swim patterns [12] further improves the analysis of the free swim trials. We subsequently calculated the percentage of animals showing either of these three search strategies. Behavior was recorded on videotape and analyzed by EthoVision 1.95 (Noldus Information Technology BV, Wageningen, The Netherlands). 2.4. Blood sampling and corticosterone measurement For determination of basal activity of the HPA axis, morning and evening corticosterone concentrations were measured in blood samples that were collected within 2 h after lights-on (morning) and 1 h before lights-off (evening). In a pilot study, we investigated the circadian rhythm of corticosterone in mice by obtaining 12 blood samples evenly distributed over 24 h. Immediately after light onset, corticosterone concentrations were low and remained low until about 14.00 h, then started to rise gradually over the day, reaching maximum levels 1–2 h before lights-off. Thereafter, concentrations started to decline (data not shown). Reactivity of the HPA axis to an acute challenge was determined by measuring corticosterone secretion in blood samples that were collected 5, 15, 30, and 60 min after placing the mouse in a novel clean cage without sawdust. During the period of blood sampling, the animals remained in that cage. Blood samples were collected by tail incision, a method that allows estimation of basal (resting) concentrations of corticosterone and multiple samples from the same mouse without anesthesia [6,12]. Briefly, a small incision at the base of the tail with a razor blade allows a 25–50 ␮l blood sample, within 90 s after opening of the animal’s cage. Each animal provided maximally three blood samples. Corticosterone was measured in plasma according to the manufacturer’s instructions with a 125 I-corticosterone radioimmunoassay for rats and mice (ICN Pharmaceuticals, Inc., New York, USA). 2.5. Estrous cycle determination The stages of the estrous cycle are reflected in the vaginal cytology and correlate closely with circulating levels of estrogen and progesterone in the bloodstream [8,9,29]. Thus, daily examination of vaginal cell types can be used to determine the regularity of hormone release, or estrous cycling. Estrous cycles in mice typically last 4 or 5 days and are divided into four distinct stages: proestrus, estrus, metestrus, and diestrus. Proestrus is associated with relatively high levels of circulating estrogen, estrus with intermediate levels, and diestrus with low levels of estrogen

170

J. Grootendorst et al. / Behavioural Brain Research 148 (2004) 167–177

(A)

60

young apoE0/0 50

aged apoE0/0

Latency (s)

40

30

20

10

0 1 2 3

day 1

4 5 6 7

8 9 10 11

day 2

12 13 14 15

16 17

day 3

18 19 20

21 22 23 trials

day 4

(B) 60

young wild-type aged wild-type

50

Latency (s)

40

30

20

10

0 1 2 3

day 1

4 5 6 7

8 9 10 11

day 2

12 13 14 15

16 17

day 3

18 19 20

21 22 23 trials

day 4

Fig. 1. In female apoE0/0 and wild-type mice, water maze performance was dependent on genotype and age. (A) Aged female apoE0/0 mice showed impaired water maze performance compared to their younger counterparts. (B) In wild-type mice, water maze performance was similar for young and aged mice. Data are presented per trial as mean latency in seconds ± S.E.M. to reach the platform.

[8,9,29]. In aging females, these cycles became prolonged and eventually ceased [30]. To determine whether female mice in this study were cycling regularly, stages of the estrous cycle were recorded by obtaining vaginal smears each morning between 09.00 and 10.30 h for 21 consecutive days. Vaginal epithelial cells were rolled onto a slide, air-dried, and subsequently slides were incubated with 1:20 diluted Giemsa solution for 7 min at room temperature, rinsed with tap water, and air-dried. Proestrus is marked by pink staining of epithelial cells, estrus by masses of dark staining cornified cells, metestrus by some cornified cells and lots of leukocytes, whereas diestrus is characterized by dark-stained leukocytes with scattered

epithelial cells. Cycling activity was divided into three categories (adapted from Frick et al. [9]): (i) regular cycling, which denoted a 4- to 5-day estrous cycle in which the estrus phase was observed at least three times during the sampling period, (ii) irregular cycling, which denoted a prolonged (>5 days) cycle in which the estrus phase was observed only once, and (iii) no cycling, which denoted the absence of estrus throughout the sampling period. 2.6. Statistics Data were subjected to analysis of variance, when appropriate with repeated measurements followed by post-hoc

J. Grootendorst et al. / Behavioural Brain Research 148 (2004) 167–177

Least Significant Difference test. Since latency and distance measures of the water maze training trials provided similar results, we present latency data only. The distribution of the swim patterns (persistent, concentric, and random) and results of the vaginal smears were compared by Chi-square test. Correlations of swim patterns with performance were tested by Spearman’s rho. Statistical difference was accepted at P < 0.05. Data are presented as mean ± S.E.M.

3. Results 3.1. Water maze performance 3.1.1. Training trials Genotype and age differentially affected the spatial water maze performance of female mice (Fig. 1; genotype × age interaction F(1, 47) = 4.13, P < 0.05). Only in female apoE0/0 mice, spatial learning was affected by age (F(1, 26) = 5.07, P < 0.05). Aged apoE0/0 mice showed increased latencies to reach the platform compared to young apoE0/0 mice on all days of training, except on day 3. In contrast, in female wild-type mice, water maze performance was unaffected by age (F(1, 21) = 1.00, P = 0.33). Also within the aged groups, apoE0/0 mice showed impaired water maze performance (F(1, 24) = 4.52, P < 0.05). In young mice, water maze performance was similar for both genotypes. All groups improved their performance (F(22, 1034) = 13.07, P < 0.0001). Water maze performance during visible platform training was not affected by genotype or age (mean latency in seconds ± S.E.M.: young apoE0/0 21.3 ± 3.1; young wild-type 18.0 ± 3.0; aged apoE0/0 22.6 ± 3.1; aged wild-type 22.7 ± 3.4). 3.1.2. General activity and search strategies during free swim trials The results of the free swim trials support and extend the findings gained by the animal’s performance during training trials (Tables 1–3). Before training, all mice showed a ran-

Table 1 General activity in the water maze during free swim trials before and during training Swim ability

ApoE0/0 Before training

Wild-type During training

Before training

During training

Travelled distance (m) Young 5.4 ± 0.7∗,# Aged 8.2 ± 0.4

9.6 ± 0.7 8.3 ± 0.5

7.9 ± 0.3∗ 10.1 ± 0.6

10.0 ± 0.4 9.7 ± 0.4

Swim velocity (cm/s) Young 9.0 ± 1.2∗,# Aged 13.7 ± 0.7

16.0 ± 1.1 14.2 ± 1.0

13.1 ± 0.5∗ 16.6 ± 1.0

16.8 ± 0.6 16.3 ± 0.7

∗ #

P < 0.05 young vs. aged within genotype. P < 0.05 young apoE0/0 vs. young wild-type; see text for details.

171

domly distributed exploration pattern of the pool (data not shown). Computerized image analysis revealed that young female mice of both genotypes had a slower swim speed and thus, shorter swim distances than their aged counterparts (Table 1; speed: F(1, 47) = 23.27, P < 0.001; distance: F(1, 47) = 23.26, P < 0.001). Young apoE0/0 mice were even slower than young wild-type mice (P < 0.05). Inspection of the videotapes revealed that young mice frequently moved along the side wall displaying an interesting “escape” behavior: they tried to climb the wall of the pool, clung to it with the forepaws, and moved in this way further along the side wall. Some mice dragged half of their body above the water level. Aged female apoE0/0 mice did not show this extraordinary “climbing” behavior (which we neither observed in young male mice in previous studies [12,16,31]; young male apoE0/0 mice often showed the so-called wall-bumping behavior [31]). Floating was never observed. Spatial training obliterated this wall-climbing behavior and no differences in general activity were observed in the second free swim trial. Then, mice of all groups travelled comparable distances (Table 3). The swim pattern displayed before the training by more than 90% of the mice (independent of age and genotype) was identified as random (data not shown). After a series of 15 training trials, young mice of both genotypes showed a similar and preferentially, persistent swim pattern (Table 2) at the free swim trial during training. This swim pattern reflects a search strategy for the location of the platform. Interestingly, the search strategies of wild-type mice changed 2 = 30.59, P < 0.0001). Wild-type mice bewith age (χ(2) 2 = 6.23, came significantly more persistent with age (χ(2) P < 0.05); while aged apoE0/0 mice showed a comparable distribution of search strategies to young apoE0/0 mice. In parallel, a significant difference in swim patterns was cal2 = 61.08, P < 0.0001). culated for the aged groups (χ(2) More aged wild-type than aged apoE0/0 mice expressed the persistent swim pattern (wild-type 90%; apoE0/0 50%; 2 = 8.45, P < 0.05). χ(2) Calculation of the standard parameters for the free swim trial (Table 3) revealed a main effect for age on platform crossings (F(1, 47) = 5.57, P < 0.05) with the least number of crossings for aged apoE0/0 mice (apoE0/0 young vs. aged: P < 0.05). All groups showed preference for the platform quadrant compared to the other three quadrants expressed as percentage of time, but genotype or age differences in percentage time spent in platform quadrant did not reach statistical significance (P = 0.09). Interestingly, the swim strategy used during the second free swim trial had no immediate effect on water maze performance, i.e. we found no correlation between swim pattern and latency to the platform in the next training trial (trial 16; Spearman’s rho = 0.186, P = 0.192). Mice showed similar latencies independent of their swim pattern. However, we detected a positive correlation of the search strategy during training with the performance in the last training trial (Spearman’s rho = 0.38, P < 0.01; mean

172

J. Grootendorst et al. / Behavioural Brain Research 148 (2004) 167–177

Table 2 Percentage of mice that showed persistent, concentric, or random swim strategies during the free swim trial on day 3, after 15 training trials Persistent

Young Aged

Concentric

Random

ApoE0/0 (%)

Wild-type (%)

ApoE0/0 (%)

Wild-type (%)

ApoE0/0 (%)

Wild-type (%)

75 50

62 90∗,#

17 19

23 0

8 31

15 10

Note that more aged wild-type mice showed persistent swim strategies than aged apoE0/0 mice. ∗ P < 0.05 young vs. aged wild-type. # P < 0.05 aged apoE0/0 vs. aged wild-type; see text for details.

latencies (seconds)±S.E.M. in trial 23, independent of genotype and age: persistent 13.1 ± 2.3; concentric 14.6 ± 2.8; random 34.8 ± 7.8). This indicates that mice with persistent search strategies will further decrease their latencies, while mice still showing a random search pattern after trial 15 will not improve their water maze performance over time. 3.2. Activity of the HPA axis 3.2.1. Circadian rhythm of corticosterone secretion Corticosterone secretion patterns were dependent on age and genotype (Fig. 2; age × genotype interaction (F(1, 84) = 4.30, P < 0.05). In female apoE0/0 mice, aging affected the circadian profile for corticosterone: in the evening, corticosterone concentrations increased with age (P < 0.05) while basal morning levels remained constant. In wild-type mice, age did not affect circadian secretion of corticosterone, although we observed a trend in aged mice for lower evening corticosterone concentrations compared to their younger counterparts (P = 0.10). When corticosterone concentrations were compared between the genotypes, two effects were observed. (i) Young and aged female apoE0/0 mice showed significantly elevated morning corticosterone concentrations compared to young and aged wild-type mice (P < 0.05). (ii) In the evening, aged apoE0/0 mice did reach significantly higher corticosterone concentrations compared to aged wild-type mice (P < 0.05). 3.2.2. Corticosterone response to novelty Corticosterone responses to a novel environment were determined by genotype (Fig. 3; F(1, 43) = 4.41, P < 0.05). That is, within apoE0/0 mice corticosterone responses were

affected by age (F(1, 23) = 4.52, P < 0.05), but young and aged wild-type mice showed similar corticosterone responses (P = 0.59). Aged apoE0/0 mice showed significantly elevated corticosterone concentrations compared to young apoE0/0 as well as aged wild-type mice at 30 and 60 min after introduction into the novel environment (P < 0.05). In contrast, young apoE0/0 and wild-type mice showed comparable concentrations of corticosterone at all time points measured. 3.3. Regularity of estrous cycling In young females, regularity of estrous cycles depended 2 = 9.75, P < 0.005). Of the wild-type on genotype (χ(1) mice, 92% showed a regular cycle whereas only 69% of the apoE0/0 mice had a weekly cycle and 31% showed an irregular cycle. Note that in young mice, water maze performance was similar for both genotypes. With increasing age, the estrous cycle became irregular in both genotypes (apoE0/0 50%; wild-type 60%) and in a small percentage of the mice cycling ceased (apoE0/0 29%; 2 = 17.64, P < 0.001; wild-type 10%; young vs. aged χ(1) 2 young vs. aged χ(1) = 24.03, P < 0.001). At 14–16 months of age, the genotype difference in regularity of the estrous cycle was abolished, which coincided with declined water maze performance in apoE0/0 mice, but not in wild-type mice. 3.4. Bodyweight The bodyweight of the mice was dependent on both age and genotype (F(1, 84) = 4.11, P < 0.05). Young apoE0/0

Table 3 Performance of mice in the free swim trial during training on day 3 (after training trial 15) given as latency to the former location of the submerged platform, number of crossings of this location, and the percentage of time spent in the four quadrants Free swim trial during training

ApoE0/0 Young

Latency to platform (s) Crossings of platform Time in platform quadrant (%) Time in left quadrant (%) Time in opposite quadrant (%) Time in right quadrant (%) ∗

P < 0.05 young apoE0/0 vs. aged apoE0/0.

17.8 3.6 46.4 21.8 17.9 13.9

± ± ± ± ± ±

Wild-type Aged

5.9 0.7∗ 4.3 3.3 6.1 2.7

26.6 2.1 39.9 24.5 19.0 16.6

± ± ± ± ± ±

Young 5.4 0.4 4.6 3.8 2.9 2.1

13.0 4.1 51.5 22.7 11.3 14.5

± ± ± ± ± ±

Aged 3.0 0.5 4.7 3.6 2.8 2.9

16.1 3.2 51.8 21.5 9.8 16.7

± ± ± ± ± ±

5.4 0.5 5.3 4.0 2.3 2.5

J. Grootendorst et al. / Behavioural Brain Research 148 (2004) 167–177

(A)

200

Corticosterone (ng/ml)

180 160

*

young apoE0/0

173

#

aged apoE0/0

140 120 100 80 60

#

#

40 20 0

morning (B)

200 180 160

Corticosterone (ng/ml)

evening

young wild-type aged wild-type

140 120 100 80 60 40 20 0

morning

evening

Fig. 2. Influence of age on basal morning and evening corticosterone concentrations (mean (ng/ml) ± S.E.M.) in female mice. (A) In apoE0/0 mice, evening corticosterone concentrations increased with aging. (B) Aging did not affect circadian secretion of corticosterone in wild-type mice. ∗ P < 0.05 young vs. aged; # P < 0.05 apoE0/0 vs. wild-type.

mice had a higher bodyweight than young wild-type mice (apoE0/0 25.0 ± 0.6; wild-type 21.0 ± 0.5). With increasing age, bodyweight increased and at 14–16 months of age, the genotype difference disappeared: all aged mice weighed approximately 30 g (apoE0/0 31.2±0.8; wild-type 30.2±0.9).

4. Discussion We showed that the effect of aging on water maze performance as well as the activity of the HPA axis depends on the apoE-genetic background of the female mice. Furthermore, we are the first to demonstrate that the regularity of the estrous cycle is age dependent in apoE0/0 mice. The impaired water maze performance of the aged female apoE0/0 mice occurred in parallel with an altered secretion pattern of corticosterone. This hormone was increased in response to a novel environment and during the circadian rise in the evening. In contrast, in young and aged female

wild-type mice, corticosterone secretion patterns were similar and so was the performance of the mice in the water maze. It seems that wild-type mice are protected by apoE from age-related decline in spatial learning behavior and neuroendocrine responses. These data support our hypothesis derived from studies in male mice that depending on the apoE-genetic background, the activity of the HPA axis is related to early-onset of cognitive decline. Learning and memory are not the only cognitive processes involved in water maze behavior. The sudden absence of the platform during the free swim trial requires perception of this novel situation, followed by a specific behavioral response (search strategy). We found some intriguing results. While in aged apoE0/0 mice spatial learning and memory for the location of the platform declined (i.e. latency to platform), search strategies remained comparable to young apoE and young wild-type mice. Spatial learning and memory of aged wild-type mice was comparable to young mice of both genotypes. Surprisingly, the aged wild-type mice

174

J. Grootendorst et al. / Behavioural Brain Research 148 (2004) 167–177 (A) 600

young apoE0/0 Corticosterone (ng/ml)

500

*#

aged apoE0/0

*#

400

300

200

100

0 5

15

30

60

time (min) (B)

600

young wild-type Corticosterone (ng/ml)

500

aged wild-type

400 300 200 100 0 5

15

30

60

time (min) Fig. 3. Corticosterone responses (mean (ng/ml) ± S.E.M.) to a novel environment in female apoE0/0 and wild-type mice. (A) Aged apoE0/0 mice showed augmented corticosterone responses 30 and 60 min after initiation of the stress response. This also resulted in a genotype-related difference at those time points. (B) In wild-type mice, corticosterone responses were not affected by age. ∗ P < 0.05 young vs. aged; # P < 0.05 apoE0/0 vs. wild-type.

showed an increase in persistent search strategies. Apparently, the more persistent behavior, i.e. perseveration, returning to the platform location, contributes to the memory performance [32]. A positive correlation between search strategy and finding the platform in subsequent training trials supports this assumption. The persistent search strategy of aged wild-type mice has positive consequences for the water maze performance and thus, might either compensate for or camouflage the early-onset of cognitive decline in learning and memory. We suggest that aged apoE0/0 mice might lack the flexibility to integrate cognitive processes. Clearly, the impairment of spatial learning and memory of aged female apoE0/0 mice as a result of declined sensory and motor capacities can be excluded. One important issue in this study is the role of corticosterone in the cognitive decline of the aging female apoE0/0 mice. Previous studies have shown that the amount of corticosterone, secreted in the context of the learning experience,

is essential for memory consolidation processes [5,44,45]. Experiences beyond this context or aberrant secretion of corticosterone both impair memory consolidation [5]. In the present study, we found in the female apoE0/0 mice an elevated circadian rise and an enhanced and prolonged stress-induced rise in corticosterone. This aberrant pattern is indicative for a disturbed regulation of the HPA axis, either at the adrenal or the brain corticosteroid receptor level. Previously, we have detected a similar relationship between impaired spatial learning and high corticosterone in young male apoE0/0 mice [12,13,16]. Actually, alterations of corticosterone secretion patterns by a previous history of stressful experiences provided the strongest evidence; whereas na¨ıve young male apoE0/0 mice expressed high corticosterone and impaired spatial learning, previous stress in apoE0/0 mice decreased their corticosterone responsiveness and concurrently, improved spatial learning in the water maze and the circular hole board [12,13,16]. These

J. Grootendorst et al. / Behavioural Brain Research 148 (2004) 167–177

studies support the notion that apoE is a factor involved in the outcome of corticosterone signaling to the brain. How this action of apoE occurs requires further study. It has been generally accepted that the stages of the estrous cycle are reflected in the vaginal cytology and correlate closely with circulating levels of estrogen and progesterone in the bloodstream [29]. We monitored the regularity of the estrous cycle for two reasons: (1) to determine if the regularity of the estrous cycle will change with age and the apoE-genetic background; and (2) if the regularity of the estrous cycle and the concomitant changes in sex hormones affect water maze performance. Indeed, we found that estrous cycles changed with age and genotype of the mice. In young mice, we observed a lower percentage of regular cycling in apoE0/0 mice and both genotypes showed a good water maze performance. In aged mice, the percentage of regular estrous cycle was similar for both genotypes while impaired water maze performance was observed in apoE0/0 mice only. Thus, the regularity of the estrous cycle alone cannot account for the observed age-related impairment in water maze performance of the apoE0/0 mice. However, we cannot exclude that a regular cycle and its concomitant secretion of sex hormones, like estrogen, at young age protects young female apoE0/0 mice from impaired cognitive performance, which is observed in young male apoE0/0 mice [12,13,16,31,39,40,57]. Previously, we described that young male apoE0/0 mice showed impaired water maze performance, which coincided with increased corticosterone responses to a novel environment [12]. In line with this reasoning is the recent finding that estrogen treatment normalized the augmented response of the HPA axis to stress [2,7,18]. In aged rats, six weeks of estrogen treatment increased brain glucocorticoid receptors and corrected the disturbances in HPA axis regulation [7]. Thus, hormones related to the estrous cycle can affect the reactivity of the HPA axis, which also influences cognitive performance. Considering the facts presented above, it can be hypothesized that due to the modulating role of estrogens on corticosteroid secretion, young female apoE0/0 mice are not cognitively impaired. If in aging apoE0/0 mice, this modulating role of estrogens disappears or diminishes, increased corticosterone secretion and impaired cognitive performance is observed. Another factor that might be affected in an apoE genotypedependent way is the lipid metabolism. ApoE is mainly known for its prominent role in lipid metabolism [23], although since the identification of the apoE ε4 allele as a risk factor for the development of Alzheimer’s disease, a whole new research field has developed to unravel the functions of apoE in the brain. Male and female apoE0/0 mice spontaneously start to develop atherosclerosis at 3 months of age. These mice have an impaired cholesterol catabolism which results in a considerable increase of plasma cholesterol (up to five times normal) [37,38,53]. Consequently, already young apoE0/0 mice show morphological changes in the brain, which reflect disturbances of their lipid metabolism, like regional differences in lipid distribution, cerebral

175

lipid depositions, and age-related congophilic inclusions [43]. Since young female apoE0/0 mice were cognitively non-impaired, but not young male apoE0/0 mice [12,31], it is unlikely that the disturbed lipid metabolism is critically connected to the cognitive deficit observed in aged female apoE0/0 mice. Moreover, the spatial learning deficit of young male apoE0/0 mice was eliminated by a short history of stress prior to testing [12,13,16]. Repeated exposure to rats over the course of two weeks is unlikely to change the degree of atherosclerosis, but did decrease corticosteroid secretion patterns in apoE0/0 mice only [12]. This further supports the key role for corticosteroids in cognitive processes depending on the apoE-genetic background. Taken together, it seems that young female apoE0/0 mice are protected from cognitive impairment, which develops during the aging process in concordance with enhanced and prolonged surges in corticosterone.

Acknowledgements This study was supported by the Internationale Stichting Alzheimer Onderzoek (ISAO, 679756-0270) and the Netherlands Organization for Scientific Research (Grants NWO 970-10-007 and 015-01-076; NDRF-STIGON 01480-005). The technical assistance of Maaike Kempes and Marc Fluttert is gratefully acknowledged.

References [1] Anderson R, Higgins GA. Absence of central cholinergic deficits in ApoE knockout mice. Psychopharmacology (Berl) 1997;132(2):135– 44. [2] Carey MP, Deterd CH, de Koning J, Helmerhorst F, de Kloet ER. The influence of ovarian steroids on hypothalamic–pituitary–adrenal regulation in the female rat. J Endocrinol 1995;144(2):311–21. [3] Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, et al. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 1993;261(5123):921–3. [4] Dalm S, Grootendorst J, de Kloet ER, Oitzl MS. Quantification of swim patterns in the Morris water maze. Behav Res Methods Instrum Comput 2000;32(1):134–9. [5] de Kloet ER, Oitzl MS, Joels M. Stress and cognition: are corticosteroids good or bad guys? Trends Neurosci 1999;22(10):422–6. [6] Durschlag M, Wurbel H, Stauffacher M, Von Holst D. Repeated blood collection in the laboratory mouse by tail incision—modification of an old technique. Physiol Behav 1996;60(6):1565–8. [7] Ferrini M, Piroli G, Frontera M, Falbo A, Lima A, De Nicola AF. Estrogens normalize the hypothalamic–pituitary–adrenal axis response to stress and increase glucocorticoid receptor immuno-reactivity in hippocampus of aging male rats. Neuroendocrinology 1999; 69(2):129–37. [8] Flurkey K, Randall PK, Sinha YN, Ermini M, Finch CE. Transient shortening of estrous cycles in aging C57BL/6J mice: effects of spontaneous pseudopregnancy, progesterone, l-dihydroxyphenylalanine, and hydergine. Biol Reprod 1987;36(4):949–59. [9] Frick KM, Burlingame LA, Arters JA, Berger-Sweeney J. Reference memory, anxiety and estrous cyclicity in C57BL/6NIA mice are affected by age and sex. Neuroscience 2000;95(1):293–307.

176

J. Grootendorst et al. / Behavioural Brain Research 148 (2004) 167–177

[10] Frick KM, Fernandez SM, Bulinski SC. Estrogen replacement improves spatial reference memory and increases hippocampal synaptophysin in aged female mice. Neuroscience 2002;115(2):547–58. [11] Gallagher M, Burwell R, Burchinal M. Severity of spatial learning impairment in aging: development of a learning index for performance in the Morris water maze. Behav Neurosci 1993;107(4):618– 26. [12] Grootendorst J, de Kloet ER, Dalm S, Oitzl MS. Reversal of cognitive deficit of apolipoprotein E knockout mice after repeated exposure to a common environmental experience. Neuroscience 2001;108(2):237– 47. [13] Grootendorst J, de Kloet ER, Vossen C, Dalm S, Oitzl MS. Repeated exposure to rats has persistent genotype-dependent effects on learning and locomotor activity of apolipoprotein E knockout and C57Bl/6 mice. Behav Brain Res 2001;125(1/2):249–59. [14] Grootendorst J, Kempes MM, Lucassen PJ, Dalm S, de Kloet ER, Oitzl MS. Differential effect of corticosterone on spatial learning abilities in apolipoprotein E knockout and C57BL/6J mice. Brain Res 2002;953(1/2):281–5. [15] Grootendorst J, Mulder M, Haasdijk E, de Kloet ER, Jaarsma D. Presence of apolipoprotein E immunoreactivity in degenerating neurones of mice is dependent on the severity of kainic acid-induced lesion. Brain Res 2000;868(2):165–75. [16] Grootendorst J, Oitzl MS, Dalm S, Enthoven L, Schachner M, de Kloet ER, et al. Stress alleviates reduced expression of cell adhesion molecules (NCAM, L1), and deficits in learning and corticosterone regulation of apolipoprotein E knockout mice. Eur J Neurosci 2001;14(9):1505–14. [17] Horsburgh K, McCarron MO, White F, Nicoll JA. The role of apolipoprotein E in Alzheimer’s disease, acute brain injury and cerebrovascular disease: evidence of common mechanisms and utility of animal models. Neurobiol Aging 2000;21(2):245–55. [18] Kudielka BM, Schmidt-Reinwald AK, Hellhammer DH, Kirschbaum C. Psychological and endocrine responses to psychosocial stress and dexamethasone/corticotropin-releasing hormone in healthy postmenopausal women and young controls: the impact of age and a two-week estradiol treatment. Neuroendocrinology 1999;70(6):422– 30. [19] Lancashire C, Mileusnic R, Rose SP. Apolipoprotein E antibodies affect the retention of passive avoidance memory in the chick. Neural Plast 1998;6(3):29–40. [20] Luine VN, Richards ST, Wu VY, Beck KD. Estradiol enhances learning and memory in a spatial memory task and effects levels of monoaminergic neurotransmitters. Horm Behav 1998;34(2):149–62. [21] Lupien SJ, de Leon M, De Santi S, Convit A, Tarshish C, Nair NP, et al. Cortisol levels during human aging predict hippocampal atrophy and memory deficits. Nat Neurosci 1998;1(1):69–73. [22] Magri F, Locatelli M, Balza G, Molla G, Cuzzoni G, Fioravanti M, et al. Changes in endocrine circadian rhythms as markers of physiological and pathological brain aging. Chronobiol Int 1997;14(4):385–96. [23] Mahley RW. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science 1988;240(4852):622–30. [24] Markham JA, Pych JC, Juraska JM. Ovarian hormone replacement to aged ovariectomized female rats benefits acquisition of the Morris water maze. Horm Behav 2002;42(3):284–93. [25] Masliah E, Mallory M, Ge N, Alford M, Veinbergs I, Roses AD. Neurodegeneration in the central nervous system of apoE-deficient mice. Exp Neurol 1995;136(2):107–22. [26] Masliah E, Samuel W, Veinbergs I, Mallory M, Mante M, Saitoh T. Neurodegeneration and cognitive impairment in apoE-deficient mice is ameliorated by infusion of recombinant apoE. Brain Res 1997;751(2):307–14. [27] McEwen BS, Sapolsky RM. Stress and cognitive function. Curr Opin Neurobiol 1995;5(2):205–16. [28] Morano MI, Vazquez DM, Akil H. The role of the hippocampal mineralocorticoid and glucocorticoid receptors in the hypothalamo–

[29]

[30]

[31]

[32]

[33]

[34] [35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44] [45]

[46]

[47]

[48]

pituitary–adrenal axis of the aged Fisher rat. Mol Cell Neurosci 1994;5(5):400–12. Nelson JF, Felicio LS, Osterburg HH, Finch CE. Differential contributions of ovarian and extraovarian factors to age-related reductions in plasma estradiol and progesterone during the estrous cycle of C57BL/6J mice. Endocrinology 1992;130(2):805–10. Nelson JF, Karelus K, Bergman MD, Felicio LS. Neuroendocrine involvement in aging: evidence from studies of reproductive aging and caloric restriction. Neurobiol Aging 1995;16(5):837–43. Oitzl MS, Mulder M, Lucassen PJ, Havekes LM, Grootendorst J, de Kloet ER. Severe learning deficits in apolipoprotein E-knockout mice in a water maze task. Brain Res 1997;752(1/2):189–96. Oitzl MS, Workel JO, Fluttert M, Frosch F, de Kloet ER. Maternal deprivation affects behaviour from youth to senescence: amplification of individual differences in spatial learning and memory in senescent Brown Norway rats. Eur J Neurosci 2000;12(10):3771–80. Packard MG, Teather LA. Intra-hippocampal estradiol infusion enhances memory in ovariectomized rats. Neuroreport 1997;8(14): 3009–13. Paganini-Hill A, Henderson VW. Estrogen replacement therapy and risk of Alzheimer disease. Arch Intern Med 1996;156(19):2213–7. Payami H, Zareparsi S, Montee KR, Sexton GJ, Kaye JA, Bird TD, et al. Gender difference in apolipoprotein E-associated risk for familial Alzheimer disease: a possible clue to the higher incidence of Alzheimer disease in women. Am J Hum Genet 1996;58(4):803–11. Peskind ER, Wilkinson CW, Petrie EC, Schellenberg GD, Raskind MA. Increased CSF cortisol in AD is a function of APOE genotype. Neurology 2001;56(8):1094–8. Plump AS, Smith JD, Hayek T, Aalto-Setala K, Walsh A, Verstuyft JG, et al. Severe hypercholesterolemia and atherosclerosis in apolipoprotein E-deficient mice created by homologous recombination in ES cells. Cell 1992;71(2):343–53. Quarfordt SH, Oswald B, Landis B, Xu HS, Zhang SH, Maeda N. In vivo cholesterol kinetics in apolipoprotein E-deficient and control mice. J Lipid Res 1995;36(6):1227–35. Raber J, Akana SF, Bhatnagar S, Dallman MF, Wong D, Mucke L. Hypothalamic–pituitary–adrenal dysfunction in Apoe(−/−) mice: possible role in behavioral and metabolic alterations. J Neurosci 2000;20(5):2064–71. Raber J, Wong D, Buttini M, Orth M, Bellosta S, Pitas RE, et al. Isoform-specific effects of human apolipoprotein E on brain function revealed in ApoE knockout mice: increased susceptibility of females. Proc Natl Acad Sci USA 1998;95(18):10914–9. Rao VS, Cupples A, van Duijn CM, Kurz A, Green RC, Chui H, et al. Evidence for major gene inheritance of Alzheimer disease in families of patients with and without apolipoprotein E epsilon 4. Am J Hum Genet 1996;59(3):664–75. Rissanen A, Puolivali J, van Groen T, Riekkinen PJ. In mice tonic estrogen replacement therapy improves non-spatial and spatial memory in a water maze task. Neuroreport 1999;10(6):1369–72. Robertson TA, Dutton NS, Martins RN, Roses AD, Kakulas BA, Papadimitriou JM. Age-related congophilic inclusions in the brains of apolipoprotein E-deficient mice. Neuroscience 1998;82(1):171–80. Sandi C. The role and mechanisms of action of glucocorticoid involvement in memory storage. Neural Plast 1998;6(3):41–52. Sandi C, Loscertales M, Guaza C. Experience-dependent facilitating effect of corticosterone on spatial memory formation in the water maze. Eur J Neurosci 1997;9(4):637–42. Simpkins JW, Singh M, Bishop J. The potential role for estrogen replacement therapy in the treatment of the cognitive decline and neurodegeneration associated with Alzheimer’s disease. Neurobiol Aging 1994;15(Suppl 2):S195–7. Srivastava RA, Bhasin N, Srivastava N. Apolipoprotein E gene expression in various tissues of mouse and regulation by estrogen. Biochem Mol Biol Int 1996;38(1):91–101. Srivastava RA, Srivastava N, Averna M, Lin RC, Korach KS, Lubahn DB, et al. Estrogen up-regulates apolipoprotein E (ApoE)

J. Grootendorst et al. / Behavioural Brain Research 148 (2004) 167–177

[49]

[50]

[51]

[52]

gene expression by increasing ApoE mRNA in the translating pool via the estrogen receptor alpha-mediated pathway. J Biol Chem 1997;272(52):33360–6. Stone DJ, Rozovsky I, Morgan TE, Anderson CP, Finch CE. Increased synaptic sprouting in response to estrogen via an apolipoprotein E-dependent mechanism: implications for Alzheimer’s disease. J Neurosci 1998;18(9):3180–5. Stone DJ, Rozovsky I, Morgan TE, Anderson CP, Hajian H, Finch CE. Astrocytes and microglia respond to estrogen with increased apoE mRNA in vivo and in vitro. Exp Neurol 1997;143(2): 313–8. Tang MX, Jacobs D, Stern Y, Marder K, Schofield P, Gurland B, et al. Effect of oestrogen during menopause on risk and age at onset of Alzheimer’s disease. Lancet 1996;348(9025):429–32. Teter B, Harris-White ME, Frautschy SA, Cole GM. Role of apolipoprotein E and estrogen in mossy fiber sprouting in hippocampal slice cultures. Neuroscience 1999;91(3):1009–16.

177

[53] van Ree JH, van den Broek WJ, Dahlmans VE, Groot PH, Vidgeon-Hart M, Frants RR, et al. Diet-induced hypercholesterolemia and atherosclerosis in heterozygous apolipoprotein E-deficient mice. Atherosclerosis 1994;111(1):25–37. [54] Veinbergs I, Mallory M, Sagara Y, Masliah E. Vitamin E supplementation prevents spatial learning deficits and dendritic alterations in aged apolipoproteinE-deficient mice. Eur J Neurosci 2000;12(12):4541–6. [55] Warren SG, Juraska JM. Sex differences and estropausal phase effects on water maze performance in aged rats. Neurobiol Learn Mem 2000;74(3):229–40. [56] Woolley CS. Effects of estrogen in the CNS. Curr Opin Neurobiol 1999;9(3):349–54. [57] Zhou Y, Elkins PD, Howell LA, Ryan DH, Harris RB. Apolipoprotein-E deficiency results in an altered stress responsiveness in addition to an impaired spatial memory in young mice. Brain Res 1998;788(1/2):151–9.