• Alzheimer;
  • apolipoprotein E;
  • attention;
  • learning;
  • memory


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

The ε4 allele of apolipoprotein E (apoE4) is the predominant genetic risk factor for late-onset Alzheimer's disease (AD) and is also implicated in cognitive deficits associated with normal aging. The biological mechanisms by which APOE genotype affects cognitive processes or AD pathogenesis remain unclear, but interactions of apoE with amyloid β peptide (Aβ) are thought to play an important role in mediating apoE's isoform-specific effects on brain function. Here, we investigated the potential isoform-dependent effects of apoE on behavioral and cognitive performance in human apoE3 and apoE4 targeted-replacement (TR) mice that also overexpress the human amyloid precursor protein (APP). Beginning at 6–7 months of age, female APP-Yac/apoE3-TR (‘poE3’) and APP-Yac/apoE4-TR (‘poE4’) mice were tested on a battery of tests to evaluate basic sensorimotor functioning, spatial working memory, spatial recognition, episodic-like memory and attentional processing. Compared with apoE3 mice, a generalized reduction in locomotor activity was observed in apoE4 mice. Moderate, but significant, cognitive impairments were also detected in apoE4 mice in the novel object-location preference task, the contextual fear conditioning test, and a two-choice visual discrimination/detection test, however spontaneous alternation performance in the Y-maze was spared. These results offer additional support for the negative impact of apoE4 on both memory and attention and further suggest that APP-Yac/apoE-TR mice provide a novel and useful model for investigating the role of apoE in mediating susceptibility to cognitive decline.

Apolipoprotein E (apoE) plays a fundamental role in cholesterol transport and lipid homeostasis (Mahley 1988; Weisgraber and Mahley 1996). In the brain, apoE is synthesized primarily within astrocytes (Boyles et al. 1985) and is involved in neuronal development, repair and remodeling (Poirier 1994). In humans, three isoforms of apoE are expressed: apoE2, E3 and E4, encoded by the corresponding allelic variants ε2, ε3 and ε4. It is well established that the ε4 allele is a major genetic risk factor for Alzheimer's disease (AD) (Corder et al. 1993; Saunders et al. 1993). Evidence also indicates that apoE4 is linked to poorer outcome after brain injuries (Mahley et al. 2006). Moreover, APOE status has been reported to be predictive of neuropsychological test performance in both demented and non-demented populations (Reed et al. 1994; Roses et al. 1994; Craft et al. 1998), and apoE4 is associated with more rapid cognitive decline in healthy elderly individuals (Caselli et al. 1999; Flory et al. 2000).

Initial investigations of transgenic mouse models in which expression of human apoE isoforms were under control of either the neuron-specific-enolase (NSE) or glial fibrillary acidic protein (GFAP) promoters have emphasized the differential effects of human apoE isoforms on brain function. Both NSE-apoE4 and GFAP-apoE4 mice were impaired in learning a water maze task, with deficits being more pronounced in female animals and exacerbated with increasing age (Hartman et al. 2001; Raber et al. 1998; van Meer et al. 2007). These results suggest that human apoE4 has negative effects on spatial memory processing in mice independent of its cellular source in the brain. More recent studies with human apoE targeted-replacement (TR) mice, which express apoE in a temporal and spatial pattern similar to wild-type mice (Sullivan et al. 2004), have supported these earlier findings as age- and gender-dependent impairments have also been reported in apoE4-TR mice on a range of spatial learning and memory tasks (Bour et al. 2008; Grootendorst et al. 2005).

Mounting evidence points to interactions of apoE and amyloid β peptide (Aβ) as important modulators of the pathological events associated with the amyloid cascade (Brendza et al. 2002; Manelli et al. 2004; Sadowski et al. 2006). As such, earlier groups have examined the effects of human apoE in transgenic mice expressing a mutant form of the human amyloid precursor protein (APP) (e.g. Fagan et al. 2002; Holtzman et al. 1999), but to date there has been no study of the effects of apoE status on cognitive functions in mouse models expressing the nonmutated form of h-APP characteristic of most cases of late-onset AD. While mice that slightly overexpress wild-type human APP exhibit normal neuromotor and cognitive function up to 15 months of age (Murai et al. 1998), we investigated whether the isoform-specific effects of h-apoE on behavioral performance would be exacerbated in these genomic h-APP mice. Thus, neurological functions of adult h-APP mice with TR of apoE3 or apoE4 were assessed in the grip strength, rotarod, open-field and home-cage activity tests. In the same mice, learning and memory abilities were evaluated in the Y-maze spontaneous alternation, contextual fear conditioning, and novel object-location preference tasks; sustained attention was assessed in a two-choice visual discrimination task.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References


Because a parallel objective of this study was to establish a robust translational model of apoE4-associated behavioral impairment that could be used for drug discovery purposes, we chose to focus this initial behavioral characterization solely to female mice and to compare human-apoE4-expressing vs. human-apoE3-expressing animals. It was hypothesized that this preclinical approach should yield improved translational value as several clinical reports suggest that the association between the apoE4 allele and the risk of developing age-related cognitive impairments or dementia is particularly increased in women (Farlow 1997; Farrer et al. 1997; Ghebremedhin et al. 2001; Mortensen and Høgh 2001; Bartrés-Faz et al. 2002), a finding supported by preclinical data indicating more pronounced effects of apoE4 in female mice (Grootendorst et al. 2005; Hartman et al. 2001; Raber et al. 1998). Moreover, the fact that cognitive deficits observed in apoE4 carriers have been established in comparison with non-apoE4 carriers provided a reasonable rationale for limiting the current comparison with apoE-TR mice, as opposed to those of mice expressing wild-type murine apoE.

APP-Yac/apoE3-TR (‘poE3’) and APP-Yac/apoE4-TR (‘poE4’) bigenic mice were generated at Taconic Farms, Inc (Germantown, NY, USA) by crossing transgenic mice overexpressing the human APP protein (APP-Yac mice) with either human apolipoprotein E3 or apolipoprotein E4 TR mice (apoE3-TR and apoE4-TR mice). The generation of both APP-Yac (Lamb et al. 1993) and apoE-TR mice (Sullivan et al. 2004) has been described in detail previously. Parent lines (APP-Yac, apoE3-TR and apoE4-TR) were all backcrossed to C57BL/6NTac for at least eight generations and brought to homozygosity at Taconic. The following steps were then taken to create the bigenic mice used in the current study: (1) mating of h-APP+/+/TR-apoE−/−× h-APP−/−/TR-apoE+/+ to generate h-APP+/−/TR-apoE+/− mice, (2) mating of h-APP+/−/TR-apoE+/−× h-APP+/−/TR-apoE+/− to obtain h-APP+/+/TR-apoE+/+ mice, which were identified/confirmed by test mating and (3) mating of confirmed h-APP+/+/TR-apoE+/+ together to obtain the cohorts used for the current study. All breeding was performed at Taconic. Upon arrival to our facility, mice (apoE3, n = 24; apoE4, n = 16) were singly housed and allowed to acclimate for 4 weeks prior to start of any experimental procedures. The rather unequal sample sizes used in the current study owe simply to the complex breeding scheme and resultant animal availability, and were not intended to counteract a possible heterogeneity in performance. Mice were 6–7 months of age at the beginning of behavioral testing. The housing room was maintained on a 12/12 h light/dark cycle, with lights on at 7:00 a.m. Animals were provided with small, plastic nesting ‘igloos’ and given ad libitum access to food and water, with the exception of a 3-week period during two-choice visual discrimination testing when animals were kept on a food restriction regimen (see below).

Merck's Institutional Animal Care and Use Committee (IACUC) approved the research protocols, in accordance with the Guide for the Care and Use of Laboratory Animals as adopted by the NIH (pub. no. 85-23, revised 1996), and every effort was made to minimize any potential pain or discomfort during all phases of experimentation.

Behavioral experiments

All animals were tested on each individual assay in the order listed below; with the exception that home-cage diurnal activity monitoring was performed last.

Grip strength test

Testing was performed on a Stoelting grip strength meter (Wood Dale, IL, USA) that consisted of a force transducer with a 4-cm-wide triangular wire grip bar oriented parallel to, and 3 cm above, a 25 cm × 25 cm black base platform. Mice were gently removed from their cages by the base of the tail and lowered toward the bar until both forepaws gripped onto the triangular handle. Subjects were then pulled away by the tail horizontally in one smooth motion until they released the handle. The peak force (g) of each animal's grip was measured automatically by the apparatus and recorded. Five scores were recorded per animal in consecutive sequence separated by approximately 30-second intervals.


The rotarod apparatus (Ugo Basile; Comerio VA, Italy) consisted of an elevated rod (3 cm in diameter) capable of rotating at different speeds and separated into five equal segments by plastic dividers. The rotation speed of the rod can be either fixed or variable (i.e. accelerating). Mice were first habituated to the apparatus by placing them onto the rod as it rotated at a fixed speed of 4 r.p.m. for 300 seconds. If a mouse fell from the rod during this habituation session it was replaced on the apparatus until the complete session time elapsed. Following habituation, the rod was accelerated from 2 to 40 r.p.m. over 180 seconds and latency to fall off the rod was recorded. This acceleration procedure was run three times for each animal with a 45-min inter-experiment interval.

Open field

The apparatus consisted of a 27.9 cm × 27.9 cm, clear Plexiglas arena equipped with three 16-beam infrared arrays (Med Associates; St Albans, VT, USA). Mice were acclimated to the experimental test room for at least 30 min prior to testing. To start a session, a mouse was placed into the center of the arena and allowed to freely explore for a total of 60 min. The mouse's movements were tracked and recorded automatically via Med Associates software to provide measurements of total distance traveled as well as vertical beam breaks that assessed incidences of rearing behavior. The open field was cleaned of any feces or urine between animals and wiped down with a dilute solution of ethanol.

Home-cage diurnal activity

The apparatus used was a set of 16 OxyMax® Metabolic Activity Monitoring chambers (Columbus Instruments; Columbus, OH, USA). Each chamber consisted of a self-contained unit capable of providing continuous measurements of an individual mouse's total activity and feeding behavior. Monitoring occurred over a 3-day period. Each subject was placed into an individual chamber on Day 1, with free access to food and water during the course of the experiment. Subjects were maintained under a normal 12:12 h light:dark cycle. All measurements were sampled periodically (at approximately 12-min intervals) and automatically recorded via the oxymax Windows V3.22 software. Activity measures over the final 24-h period were parceled into 2-h bins and these were used to express diurnal activity levels.

Y-maze spontaneous alternation

The apparatus is a White Plexiglas Y-shaped maze (i.e. three equidistant arms). Each arm is 30 cm long, 10 cm wide and has 20-cm high walls. Two Y-mazes were used in parallel. A computer-assisted video-tracking system was used to record the distance traveled and the sequence of arm entries during testing (CleverSys Inc., VA, USA). Mice were acclimated to the testing room for at least 30 min prior to testing. At the beginning of the session, the mouse was placed in the center of the Y-maze and allowed to freely explore the maze for a period of 6 min. The total number of arm entries and the sequence of arm entries were automatically recorded by the video-tracking system. An alternation was defined as three successive visits to the three arms of the maze (i.e. ABC, ACB, BAC, BCA, CAB or CBA). The percent alternation corresponded to the number of alternations × 100 divided by the number of possible alternations (i.e. total arm entries–2). One animal was excluded from the analysis on the basis that it failed to make more than two-arm entries.

Novel object-location preference task

The apparatus is a black Plexiglas, square maze with walls 50 cm wide and 40 cm high. Stickers and paper cutouts of varying shapes and colors were affixed to the interior surfaces of the maze walls in order to increase the spatial distinctiveness of the test environment. White Plexiglas floor inserts were used to increase subject contrast and aid in automated behavioral analysis. Four arenas were employed in parallel in the same testing room. Subjects were acclimated to the testing room for at least 15 min prior to testing. Testing occurred over a 4-day period. During the first 2 days, subjects were acclimated to the empty arena for 15 min. On the 3rd day (i.e. the training session), subjects were placed back into the test arena for another 15-min session, however, the arena now contained two identical objects (20-ml glass vials with white plastic caps), spaced evenly apart, and equidistant from the back and side walls. Following a 24-h delay, subjects were placed back into the arena with the same two objects and allowed to freely explore for a total of 3 min. On this last session (i.e. the test session), one of the objects was displaced to the opposite side of the arena, whereas the second object occupied the same location as did the previous day. During each session, a computer-assisted video-tracking system (CleverSys Inc., Reston, VA, USA) was used to record both activity (i.e. distance traveled) and object-directed exploratory behavior (i.e. number of object contacts and total object exploration time). Object exploration was defined as a behavioral bout during which an animal's snout entered into a preset zone extending 1 cm around the outside edge of each stimulus object. A recognition index was calculated by dividing the total time spent exploring the displaced object ×100 by the total time spent exploring both objects during the fourth and final test session. A recognition index of 50% would, therefore, correspond to equal exploration of both objects. Subjects were excluded from the analysis if they failed to explore both stimulus objects for a total of at least 10 seconds during either training or test sessions. Four animals were excluded from this study based on this criterion.

Two-choice operant visual discrimination

Food restriction procedure. Five days prior to the beginning of training, mice were placed in clean cages and were given free access to water only. Daily food rations (regular chow) were calculated as a function of each animal's weight loss/gain from the previous day in order to maintain a stable weight of approximately 85% of free feeding weight. Additionally, about ten 20-mg casein pellets were added to the daily food rations prior to the training period so that mice could habituate to the rewards. The food restriction procedure was maintained for the entire training and testing period.

Habituation and shaping. The testing apparatus is a standard sized Med Associates (St Albans, VT, USA) operant chamber with Plexiglas side walls, stainless steel end-walls and solid steel floor. A pellet receptacle is placed in the center of one end-wall with a small, yellow stimulus light located directly above the pellet receptacle. Two retractable lever-press devices are located on either side on the opposite wall of the pellet receptacle. There are large stimulus lights located directly above each lever to operate as direct cues during the two-choice visual attention task. On the first day of testing, food-restricted mice were exposed to the operant chambers for 45 min without access to the operant levers. During this habituation session, mice received 24 exposures to a 1-second receptacle-light stimulus with an intertrial interval of 120 seconds. Each light stimulus was paired with the delivery of a food pellet. Twenty-four hours after the habituation session, all mice began the shaping phase, which consisted of daily 30-min sessions for 2 days with a maximum cutoff of 80 trials per session. During each trial, one lever was presented at a time concurrently with the random illumination of either the left or right stimulus light. A press to the extended lever resulted in the delivery of a food reward and the presentation of a light stimulus over the food receptacle. Once an animal retrieved the food pellet, a 10-second intertrial interval elapsed before the start of the next trial.

Visual discrimination acquisition. Visual discrimination training began 24 h after the second shaping session. During discrimination training, each trial began with the presentation of both levers; however, only one lever was cued by an illuminated stimulus light that remained lit for the duration of the entire trial. Pairings of the light-lever were random and counterbalanced across trials. A correct response was scored after a subject responded to the lever paired with the stimulus light. After a correct response, a food reward was delivered and the receptacle light remained on until the pellet was retrieved. Upon retrieval of the pellet, a 10-second inter-trial interval (ITI) was initiated during which the house light remained on. After an incorrect choice, no reward was given and the house light was switched off to initiate a 30-second ITI. After a trial began, if an animal failed to respond within 30 seconds, an omission was scored, both levers were retracted, and the house light extinguished to initiate a 30-second ITI. Each daily session ended when animals completed a total of 80 trials or once 45 min had elapsed. Visual discrimination training continued until a group average of 80% correct or greater was achieved on two consecutive sessions. Animals that did not complete at least 40 trials within a single session after the fourth day of discrimination training were excluded from the analysis.

Sustained attention testing. Two-choice visual attention trials were conducted 24 h following the completion of discrimination learning. The attention sessions followed an identical protocol to the discrimination procedure, with the exception that four separate trial types were employed in which the duration of the light stimulus paired with each lever presentation was varied, such that the stimulus cue now remained illuminated for only 10, 2, 1 or 0.5 seconds. These four trial types were presented 20 times each in a given session, with trial type and lever position being counterbalanced across trials. Each attention test session was completed within the allotted 45-min time frame, with a mean session length of 35 min. Total lever presses, correct and incorrect choices, omissions and reaction times were collected automatically via med-pc IV software.

Contextual fear conditioning

The testing apparatus was a standard sized Med Associates operant chamber (St Albans, VT, USA) with Plexiglas side walls and stainless steel end-walls. The floor consisted of steel bars 4.8 mm in diameter and spaced 1.6 cm apart. A small yellow stimulus light was located on top of the center panel of the right wall, and a speaker (Mallory-Sonalert; Indianapolis, IN, USA) was located directly to the right of the light. A scrambled electric shock could be delivered through the grid floor. A fan was also attached to the sound attenuation chamber and was used to provide masking noise. Eight fear conditioning chambers were used in parallel and the FreezeScan® computer-assisted video-tracking system was used to assess freezing behavior (CleverSys Inc., Reston, VA, USA). Prior to testing, all subjects were acclimated to the testing room for at least 15 min. Testing occurred over two sessions separated by a 24-h delay. To begin Session 1, a subject was placed in an illuminated chamber and allowed to explore for 2 min. At the end of this 2-min period, a 2-second, 0.7-mA shock was delivered via the grid floor. A second, identical shock followed 2 min later, and the subject was subsequently given one additional minute before being removed from the chamber. During the entire session, freezing behavior (i.e. lack of displacement, body movement, head turning and grooming) was recorded via the FreezeScan® system. These experimental parameters were chosen based on the data obtained from preliminary studies investigating freezing responses to various combinations of shock number and intensity. Increasing immobility in response to the electric shocks can be taken as an index of acquisition of fear conditioning. Session 2 occurred 24 h later and consisted of a 3-min period during which the subject was allowed to freely explore the chamber in the absence of any foot shock. Behavior was once again automatically recorded and the total percentage of time that mice spent freezing was used as an index of retention of the aversive experience.

Statistical analysis

All data are presented as mean ± SEM. Differences between mean values were determined using two-tailed t-test or appropriate analysis of variance (anova) procedures as indicated in the text. Post hoc comparisons were performed using Student t-test with appropriate Bonferoni correction factors. Differences of P < 0.05 were considered significant.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

Neurological assessment: grip strength and rotarod tests

At the beginning of behavioral testing, ‘poE3’ mice were slightly, but significantly, heavier than ‘poE4’ mice, with mean weights of 28.6 ± 0.56 and 25.8 ± 0.57 g, respectively (t = 3.56,df = 38, P < 0.01). Despite this weight difference, it can be seen in Fig. 1a that both groups recorded comparable grip strength scores [apoE 3 = 454.7 ± 22.3 vs .apoE 4 = 421.3 ± 21.2, t-test (P > 0.1)]. In addition, when assessed for their latency to fall from the rotarod under a continually accelerating procedure (Fig. 1b), repeated measures anova yielded a significant effect of trial number (F2,37 = 13.53, P < 0.001), but no effect of genotype (F1,38 = 1.42, P > 0.05) nor any genotype x trial interaction (F2,37 = 2.97, P > 0.05).


Figure 1. Grip strength and rotarod performance in ‘poE3’ and ‘poE4’ mice. (a) Mean grip strength score averaged over five successive trials. apoE3, n = 24; apoE4, n = 16. (b) Mean latency to fall from rotarod across each of the three trials. apoE3, n = 24; apoE4, n = 16.

Download figure to PowerPoint

Locomotor activity: open-field and home-cage diurnal activity measurements

To examine locomotor activity in ‘poE3’ and ‘poE4’ mice in response to a novel environment, animals were placed into an open-field arena for 1 h and their movements were automatically recorded via an infrared (IR) tracking system (Fig. 2a,b). The two groups showed clear habituation in terms of both distance traveled and rearing behavior over the course of the experimental session, with peak activity occurring in the first 5-min period and declining steadily thereafter [repeated measures anova (F11,28 = 37.67, P < 0.001) and (F11,28 = 7.03, P < 0.001), respectively. Neither dependent variable was associated with a significant group × time interaction [(F11,28 = 1.36, P > 0.05) and (F11,28 = 1.90, P > 0.05)], but it can be seen from Fig. 2a that ‘poE4’ mice displayed decreased ambulatory activity compared with ‘poE3’ mice across the entire 1-h session. Indeed a significant group effect (F1,38 = 19.53, P < 0.001) was observed with respect to distance traveled, but not with respect to the number of rearings (F1,38 = 0.02, P > 0.05).


Figure 2. Open-field and home-cage diurnal activity in ‘poE3’ and ‘poE4’ mice. (a) Spontaneous exploratory activity in a novel open-field environment (mice introduced into the open field at t = 0 min). Inset bar graphs represent total distance traveled over the 60-min exploration phase. apoE3, n = 24; apoE4, n = 16. (b) Number of rearing occurrences during each 5-min open-field activity bin. Inset bar graphs represent total rearing instances over the course of the 60-min session. apoE3, n = 24; apoE4, n = 16. (c) Home-cage diurnal activity in singly housed mice. apoE3, n = 16; apoE4, n = 15. (d) Total activity over 24-h home-cage assessment period. apoE3, n = 16; apoE4, n = 15. *P < 0.05, apoE3 vs. apoE4.

Download figure to PowerPoint

To assess diurnal activity levels, ‘poE3’ and ‘poE4’ mice were housed in home-cage-like metabolic chambers for 3 days and activity was recorded over the final 24-h period. In both mouse lines, we observed a distinct diurnal pattern with increased activity counts during the dark phase of the cycle (Fig. 2c). Repeated measures anova performed on the bi-hourly binned data showed a significant effect of both group (F1,26 = 12.16, P < 0.01) and time (F11,16 = 22.16, P < 0.001), but a nonsignificant interaction (F11,16 = 1.34, P > 0.05). The finding that ‘poE4’ mice showed a significant reduction in home-cage activity over the 24-h test period is also shown in Fig. 2d, in which total activity counts have been pooled across all 12 time bins. In addition to recording endogenous activity levels, our experimental setup permitted the monitoring of feeding behavior, with both groups of mice showing similar numbers of eating bouts and total amounts of food consumed (data not shown; P > 0.05).

Working memory: Y-maze spontaneous alternation task

Spatial working memory was assessed in ‘poE3’ and ‘poE4’ mice by quantifying spontaneous alternation behavior in a standard Y-maze. The results, depicted in Fig. 3, show that although ‘poE4’ mice made significantly fewer arm entries compared with ‘poE3’ mice (t = 5.08, df = 37, P < 0.01), there was no difference in their rate of alternation (P > 0.1). Moreover, there was no significant correlation between the number of arms mice entered in the maze and spontaneous alternation performance (r[39] = 0.13, P > 0.05).


Figure 3. Y-maze spontaneous alternation behavior in ‘poE3’ and ‘poE4’ mice. (a) Total arm entries during the 6-min Y-maze session. apoE3, n = 23; apoE4, n = 16. (b) Alternation rate expressed as a percentage of total possible alternations. apoE3, n = 23; apoE4, n = 16. *P < 0.05, apoE4 vs. apoE3.

Download figure to PowerPoint

Spatial recognition memory: novel object-location preference task

The novel object-location preference task was used to assess spatial recognition memory in ‘poE3’ and ‘poE4’ mice. This task is based on the spontaneous tendency of mice to preferentially explore a displaced vs. a nondisplaced object within a familiar test environment (Thinus-Blanc et al. 1996). As shown in Fig. 4a, there were no statistically significant differences between the two groups during the training session in either the total number of investigatory episodes or the overall time spent exploring the two stimulus objects (P > 0.05 in both cases). When exploratory behavior was assessed 24 h later (Fig. 4b), ‘apoE3’ mice showed a clear preference for investigating the newly displaced object during the test session, regardless of whether total object contacts or total exploration time was used in calculating a recognition index [chance level = 50, (t = 4.66,df = 20, P < 0.01) and (t = 5.31,df = 20, P < 0.01) for total contacts and total time, respectively]. In contrast, although ‘apoE4’ mice displayed a tendency to explore the displaced object at a slightly higher rate compared with the nondisplaced object, they did not show statistically significant evidence of detecting a spatial change (P > 0.05). Further analysis also confirmed that ‘apoE4’ mice have significantly worse object-location memory compared with ‘apoE3’ animals when recognition index is derived from the number of object approaches and/or contacts (t = 2.39,df = 34, P < 0.05). However, when recognition performance was based on the total time that mice spent exploring each object during the test session, the difference between groups just failed to reach statistical significance (P = 0.058).


Figure 4. Spatial recognition memory performance of ‘poE3’ and ‘poE4’ mice in the novel object-location preference task. (a) Exploratory behavior during the training session expressed as both the total number of object approaches (left axis) and the total time spent sniffing (right axis) the two stimulus objects. apoE3, n = 21; apoE4, n = 15. (b) Memory performance during the Test session expressed in terms of recognition index: D × 100/(D + ND), where D and ND are the mean number of approaches or exploration time directed at displaced and nondisplaced objects, respectively. Dashed line represents chance-level performance (i.e. 50%) when mice explore similarly both object categories. apoE3, n = 21; apoE4, n = 15. *P < 0.05, apoE4 vs. apoE3; P = 0.058, apoE4 vs. apoE3.

Download figure to PowerPoint

Sustained attention: two-choice operant visual discrimination

To examine attentional abilities in ‘poE3’ and ‘poE4’ mice, subjects were first trained to acquire a simple two-choice visual discrimination rule, upon which the contribution of attentional processes to task performance could be altered by varying the duration of the relevant discriminatory stimulus. Following 8 days of training, during which the to-be-attended-to stimulus cue remained illuminated for the entire length of each trial, baseline visual discrimination performance in ‘apoE3’ and ‘apoE4’ mice, as calculated by averaging each group's mean percent correct score during the final training session, was 89.0% ± 2.31% and 88.9% ± 2.96%, respectively. In addition to measuring choice accuracy, reaction time was assessed by recording the latency to make a lever-press response following the initiation of each trial. Upon reaching criterion on the visual discrimination task, average response latencies during the final training session were 5.86% ± 0.66% and 9.07% ± 0.77% in ‘apoE3’ and ‘apoE4’ groups, respectively (P < 0.05). Two sessions comprised of mixed stimulus-duration trials were then administered, and the results are depicted in Fig. 5 as an average of both days of testing. As expected, choice accuracy (Fig. 5a) was found to decline in both groups of mice in conjunction with a decrease in stimulus duration (F3,22 = 46.0, P < 0.01). A significant group effect was observed, as ‘apoE4’ animals made fewer correct responses compared with ‘apoE3’ animals (F1,24 = 4.33, P < 0.05), however, the interaction between stimulus duration and genotype was itself nonsignificant (P > 0.05). Figure 5b shows average reaction times for both groups of mice as a function of stimulus trial type (duration). For all mice there was a clear effect of trial type, with response latencies increasing as the duration of the stimulus cue was shortened (F3,22 = 26.4, P < 0.01). Latency scores also showed a significant group effect (F1,24 = 4.66, P < 0.05), with ‘apoE4’ animals being slower to respond compared with ‘apoE3’ animals, but there was no significant interaction between group and stimulus duration (P > 0.05).


Figure 5. Attentional performance of ‘poE3’ and ‘poE4’ mice in the two-choice visual discrimination task. (a) Choice accuracy for each stimulus-duration trial type expressed as a percentage of correct responses to the cued lever. Dashed line represents chance-level performance (i.e. 50%). apoE3, n = 15; apoE4, n = 11. (b) Reaction time expressed as a latency to make a lever-press response following illumination of the stimulus cue light. apoE3, n = 15; apoE4, n = 11. *P < 0.05, apoE4 vs. apoE3.

Download figure to PowerPoint

Episodic-like memory: contextual fear conditioning

The contextual fear conditioning test was used to assess hippocampal-dependent, episodic-like memory (Anagnostaras et al. 2001; Phillips and LeDoux, 1992) in ‘poE3’ and ‘poE4’ mice. It can be seen in Fig. 6a that both groups of mice showed increases in freezing behavior in response to the footshocks delivered during the first exposure to the training context. Repeated measures anova confirmed a statistically significant effect of training time (F4,35 = 98.8, P < 0.01), but no group effect (F1,38 = 1.96, P > 0.05). In addition, the group x time interaction during training just failed to reach statistical significance (F4,35 = 2.49, P = 0.06). However, at the end of the conditioning session, it can be seen that the two groups are exhibiting very comparable amounts of immobility (‘apoE 3′ mice = 57.9% ± 3.94%;‘apoE 4′ mice = 58.5% ± 5.12%). Figure 6b depicts the behavioral response of mice to the conditioning context alone (in the absence of any footshock) when animals were tested in a retention trial the following day. With respect to their baseline levels of freezing (averaged over the first 2-min period prior to footshock during the conditioning session), both ‘apoE3’ and ‘apoE4’ mice spent a much greater percentage of time immobile throughout the 3-min retest session, indicative of retention of the conditioning procedure. However, the strength of this fear memory was reduced in the ‘apoE4’ group, as these mice were found to freeze significantly less than those in the ‘apoE3’ group in response to the contextual conditioning cues (t = 2.26,df = 40, P < 0.05).


Figure 6. Performance of ‘poE3’ and ‘poE4’ mice in the contextual fear conditioning task. (a) Acquisition of the conditioned response during training in response to aversive shock stimuli. Behavior is expressed as a percentage of total time spent freezing (i.e. immobile, except for normal respiration) over each 1-min time bin. Mice were introduced to the novel conditioning chamber at t = 0 and 2-second duration electric footshocks were delivered at t = 2 min and t = 4 min. apoE3, n = 24; apoE4, n = 16. (b) Retention of the contextual fear memory. The bars on the right represent the time spent freezing to the conditioned context 24 h following acquisition, in the absence of footshock. For comparison sake, baseline-freezing levels (recorded during the first two minutes exposure to the novel context during training) are shown on the left. apoE3, n = 24; apoE4, n = 16. *P < 0.05, apoE4 vs. apoE3.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

This is the first report of isoform-specific effects of human apoE on both cognitive and noncognitive behavioral measures in an animal model that also expresses nonmutated, genomic human APP. We report that ‘poE4’ female mice show impairments in contextual fear memory and spatial recognition memory, reduced attentional capacity and slower reaction times, as well as a generalized decrease in activity levels as compared with ‘poE3’ mice.

Prior to characterization on the cognitive tasks, ‘poE3’ and ‘poE4’ mice were first assessed for their gross neurological functions, as sensorimotor anomalies can confound behavioral analysis (Crawley 1999). Whereas ‘apoE4’ mice were slightly lighter than ‘apoE3’ mice, there were no differences between the two lines in terms of grip strength or latency to fall from the rotarod, nor were any ataxic or otherwise abnormal neurologic phenotypes observed throughout the entire test period. As mentioned above, though, a general pattern emerged in which ‘apoE4’ mice were found to show significant reductions in overall activity levels. This difference manifested itself not only as less distance traveled in activity tests, but also as fewer arm entries committed during Y-maze spontaneous alternation testing. Nonetheless, ‘apoE3’ and ‘apoE4’ mice showed comparable vertical rearing events, rates of habituation to the open field, as well as comparable patterns of diurnal activity. Interestingly, reduced locomotor activity has also been reported for female apoE4 TR mice (Bour et al. 2008; Grootendorst et al. 2005). In contrast, Raber and colleagues (1998) failed to observe a difference in total distance traveled in the open-field test between mice expressing human apoE3 or apoE4 in neurons (NSE-apoE mice), while instead detecting a reduction in rearing events in NSE-apoE4 transgenics. This suggests that the physiological distribution of human apoE expression in genetically modified mouse models may have significant effects on the manner in which APOE genotype can influence locomotor and exploratory activity. Most importantly, however, the results of our sensorimotor testing make it highly unlikely that impairments in balance, strength, coordination or other noncognitive factors can account for the current differences observed between ‘poE3’ and ‘poE4’ mice on tests of learning, memory and attention.

The ability to remember and process spatial information is impaired in humans with AD (Lawrence and Sahakian 1995). In the present study, we found evidence for impaired spatial memory in female ‘poE4’ mice compared with ‘apoE3’ mice when performance was assessed in the novel object-location preference task, which is based on an animal's tendency to spend more time exploring an object encountered in a novel location vs. an identical object encountered in a familiar location (Ennaceur et al. 1997; Thinus-Blanc et al. 1996). Whereas ‘poE3’ mice displayed a significant preference for exploring the displaced object more so than the nondisplaced object during the test session, ‘apoE4’ mice did not differentiate between the two objects. Moreover, the greater recognition index in the apoE3 group compared with the apoE4 group is reflective of more robust long-term spatial recognition memory in ‘apoE3’ mice. Although ‘apoE4’ mice displayed generalized decreases in locomotor activity across several tasks, both groups showed comparable numbers of contacts and total time spent exploring the objects during the acquisition phase. While these data argue against potential disparities in object-directed exploratory behaviors, and are instead indicative of a spatial recognition memory impairment in female ‘poE4’ mice, the possible contribution of differences in novelty preference or neophobia to this observed effect cannot be entirely ruled out. Nonetheless, the observation that both groups also showed equal exploration of the stimulus objects upon first presentation during the training session, together with the fact that these results are in agreement with previous findings in female h-apoE TR mouse lines showing intact spatial recognition performance in mice expressing h-apoE3 but not in mice expressing h-apoE4 (Grootendorst et al. 2005), suggest a true difference in spatial memory abilities.

The contextual fear conditioning test has been used extensively in mice and rats to assess the integrity of hippocampal-dependent memory systems (Anagnostaras et al. 2001), dysfunction of which may underlie the episodic memory deficits that characterize AD and related dementias (Gold and Budson 2008). In this task, the strength of the pairing between the conditioned stimulus (i.e. chamber context) and unconditioned stimulus (i.e. electric footshock) is inferred by the animal's conditioned fear response (i.e. freezing behavior). Thus, the greater the freezing to the conditioning context, in the absence of aversive shock stimuli, the greater the strength of the contextual fear memory. We have shown here that ‘poE4’ mice exhibit less freezing to the conditioning context following a 24-h retention interval than do ‘poE3’ mice. The comparable levels of freezing observed at the end of the acquisition session imply that the difference observed on the test session does not reflect a generalized loss of fear, or the inability to acquire a conditioned fear response. The fact that, at the time of retention testing, ‘apoE4’ mice freeze at substantially higher levels than those seen at pre-shock baseline also indicates that significant memory of the conditioning procedure has been consolidated and recalled. Nonetheless, these results do suggest that the strength of the contextual fear association is subtly, but significantly, reduced in ‘apoE4’ mice. Whether the extent of this impairment might be exacerbated by decreasing the salience of the conditioning context, decreasing the number of pairings between conditioned and unconditioned stimuli or increasing the length of the retention delay, all of which have been shown to affect the nature or appearance of contextual fear memory deficits in transgenic mouse models of AD (Corcoran et al. 2002; Dineley et al. 2002; Kimura and Ohno 2009), remains a topic for future investigation.

Despite the impairments observed in ‘poE4’ mice on both the contextual fear conditioning and novel object-location preference tasks, no difference was detected between ‘apoE4’ and ‘apoE3’ mice in terms of spontaneous alternation behavior in the Y-maze. Although all three tasks are known to be affected by hippocampal dysfunction (Anagnostaras et al. 2001; Lalonde 2002; Thinus-Blanc et al. 1996), spontaneous alternation is thought to reflect, in part, the operation of intact spatial working or short-term memory, whereas the contextual fear and novel object-location paradigms require the successful consolidation and recall of more long-lasting memory traces. These results raise the possibility that the brain circuitry mediating spontaneous alternation performance in mice is somehow less vulnerable to the negative consequences of human apoE4 expression than are the respective neural systems underlying longer-term spatial recognition and episodic-like memory. Another possibility that must be considered, is that the difference in the total number of arm entries committed by ‘apoE3’ and ‘apoE4’ mice during the test session could impact the final rates of alternation observed, perhaps by affecting the overall level of proactive interference resulting from previous arm visits.

It has become increasingly clear in recent years that the deficits which characterize AD are often accompanied by dysfunction in certain aspects of attentional processing (Lawrence and Sahakian 1995), yet in most animal models of the disease this cognitive domain has gone unexamined. This may be explained, in part, by the fact that assessment of attentional functioning in rodents has typically involved the use of behavioral tests requiring several weeks to months of training (De Bruin et al. 2006; Patel et al. 2006). We report here that, compared with ‘apoE3’ mice, ‘poE4’ mice display impairments on a two-choice discrimination task that measures aspects of sustained visual attention. Training and testing on this task can be completed within only 2–3 weeks, and performance has been shown to be sensitive to both prefrontal cortex damage and cholinergic blockade (Dillon et al. 2009). Both ‘apoE4’ and ‘apoE3’ mice readily acquired the visual discrimination rule, but ‘poE4’ mice showed a significant impairment as the attentional demands of the task increased. Although deficits in both choice accuracy and reaction time were apparent in ‘apoE4’ mice only at the 2- and 10-second stimulus-duration trials, ‘loor’ effects are likely to have prevented potential differences at the shorter stimulus durations, at which both groups are performing at chance levels. Nonetheless, these data provide the first evidence of an isoform-specific effect of apoE on attentional functioning in mice. In this respect, it is worth noting that the observed effects in ‘poE4’ mice recapitulate some of the changes in speed-of-processing (O’Hara et al. 2008) and visuospatial attention (Greenwood et al. 2005) that have been reported in non-demented human subjects carrying the ε4 allele.

Although previous reports have shown deleterious effects of human apoE4 expression on learning and memory in different genetically modified mouse models, the current study is the first to examine of the effects of apoE status on cognitive performance in a mouse line expressing the nonmutated form of human APP characteristic of most cases of late-onset AD. In vitro studies have shown that apoE binds to Aβ in an isoform-specific manner (Strittmater et al. 1993) and that it influences Aβ fibrillization (Wisniewski et al. 1994), while in vivo studies have shown that apoE colocalizes with Aβ deposits in the brains of demented patients (Namba et al. 1991) and that an increased gene dose of the ε4 allele correlates with increased number of Aβ-containing senile plaques (Rebeck et al. 1993). Additionally, numerous animal studies have shown that apoE is an important cofactor in amyloid-related neuropathology and support the hypothesis that apoE's ability to affect the structure and clearance of Aβ underlies its role as the main genetic risk factor for late-onset AD (Brendza et al. 2002). Given the many complex interactions between apoE and Aβ, the identification of a mouse model in which isoform-specific effects of h-apoE on cognitive function occur against a backdrop of genomic h-APP expression should prove valuable toward efforts in clarifying how apoE status mediates susceptibility to cognitive decline.

In summary, we report here on the phenotypic characterization of novel transgenic mouse lines and provide evidence that the apoE4 genotype is associated with a generalized reduction in locomotor activity, and subtle impairments in spatial recognition, contextual fear memory, and attentional processing when compared with animals expressing human apoE3. Interestingly, these effects appear comparable in many ways to the moderate deficits previously reported in human apoE4 carriers on neuropsychological test measures. Whereas previous studies using other transgenic mice have pointed to the importance of both age and gender in influencing the apoE4-related cognitive decline (Bour et al. 2008; Grootendorst et al. 2005; Raber et al. 1998; van Meer et al. 2007), it is important to note that our present study was not designed to investigate age- and gender-dependent effects of apoE on behavior. Therefore, further study is necessary to clarify the contributions of these factors to the behavioral phenotype observed in our current mouse model. Preliminary findings from our laboratory do, however, suggest that cognitive impairments in ‘poE4’ vs. ‘apoE3’ mice are more pronounced in females than in males (Kornecook et al. 2008). While future experiments should also address the neuropathological mechanisms underlying the behavioral changes, our data support the notion that APP-Yac/apoE4-TR mice represent an excellent model for facilitating our understanding of apoE's effects on both normal and diseased brain functioning.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  • Anagnostaras, S.G., Gale, G.D. & Fanselow, M.S. (2001). Hippocampus and contextual fear conditioning: recent controversies and advances. Hippocampus 11, 817.
  • Bartrés-Faz, D., Junqué, C., Moral, P., López-Alomar, A., Sánchez-Aldeguer, J. & Clemente, I.C. (2002). Apolipoprotein E gender effects on cognitive performance in age-associated memory impairment. J Neuropsychiatry Clin Neurosci 14, 8083.
  • Bour, A., Grootendorst, J., Vogel, E., Kelche, C., Dodart, J.-C., Bales, K., Moreau, P.-H., Sullivan, P. M. & Mathis, C. (2008). Middle-aged human apoE4 targeted-replacement mice show retention deficits on a wide range of spatial memory tasks. Behav Brain Res 193, 174182.
  • Boyles, J.K., Pitas, R.E., Wilson, E., Mahley, R.W. & Taylor, J.M. (1985). Apolipoprotein E associated with astrocytic glia of the central nervous system and with nonmyelinating glia of the peripheral nervous system. J Clin Invest 76, 15011513.
  • Brendza, R.P., Bales, K.R., Paul, S.M. & Holtzman, D.M. (2002). Role of apoE/Aβ interactions in Alzheimer's disease: insights from transgenic mouse models. Mol Psychiatry 7, 132135.
  • Caselli, R.J., Graff-Radford, N.R., Reiman, E.M. et al. (1999). Preclinical memory decline in cognitively normal apolipoprotein E-epsilon4 homozygotes. Neurology 53, 201207.
  • Corcoran, K.A., Lu, Y., Turner, R.S. & Maren, S. (2002). Overexpression of hAPPswe impairs rewarded alternation and contextual fear conditioning in a transgenic mouse model of Alzheimer's disease. Learning and Memory 9, 243252.
  • Corder, E.H., Saunders, A.M., Strittmatter, W.J. et al. (1993).Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science 261, 921923.
  • Craft, S., Teri, L., Edland, S.D., Kukull, W.A., Schellenberg, G., McCormick, W.C., Bowen, J.D. & Larson, E.B. (1998). Accelerated decline in apolipoprotein E-epsilon4 homozygotes with Alzheimer's disease. Neurology 51, 149153.
  • Crawley, J.N. (1999). Behavioral phenotyping of transgenic and knockout mice: experimental design and evaluation of general health, sensory functions, motor abilities, and specific behavioral tests. Brain Res 835, 1826.
  • De Bruin, N.M., Fransen, F., Duytschaever, H., Grantham, C. & Megens, A.A. (2006). Attentional performance of (C57BL/6J× 129Sv)F2 mice in the five-choice serial reaction time task. Physiol Behav 89, 692703.
  • Dillon, G., Shelton, D., McKinney, A.P., Caniga, M., Marcus, J., Kornecook, T. J. & Dodart, J. C. (2009). Prefrontal cortex lesions and scopolamine impair attention performance of C57BL/6 mice in a novel 2-choice visual discrimination task. Behav Brain Res, 204, 6776.
  • Dineley, K.T., Xia, X., Bui, D., Sweatt, J.D. & Zheng, H. (2002). Accelerated plaque accumulation, associative learning deficits, and upregulation of alpha 7 nicotinic receptor protein in transgenic mice co-expressing mutant human presenilin 1 and amyloid precursor proteins. J Biol Chem 277, 2276822780.
  • Fagan, A.M., Watson, M., Parsadanian, M., Bales, K.R., Paul, S.M., Holtzman, D.M. (2002). Human and murine ApoE markedly alters Aβ metabolism before and after plaque formation in a mouse model of Alzheimer's disease. Neurobiol Dis 9, 305318.
  • Farlow, M.R. (1997). Alzheimer's disease: clinical implications of the apolipoprotein genotype. Neurology 48, S30S34.
  • Farrer, L.A., Cupples, L.A., Haines, J.L., Hyman, B., Kukull, W.A., Mayeux, R., Myers, R.H., Pericak-Vance, M.A., Risch, N. & Van Duijn, C.M. (1997). Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. J Am Med Assoc 278, 13491356.
  • Flory, J.D., Manuck, S.B., Ferrell, R.E., Ryan, C.M., Muldoon, M.F. (2000). Memory performance and the apolipoprotein E polymorphism in a community sample of middle-aged adults. Am J Med Genet 96, 707711.
  • Ghebremedhin, E., Schultz, C., Thal, D.R., Rüb, U., Ohm, T.G., Braak, E. & Braak, H. (2001). Gender and age modify the association between APOE and AD-related neuropathology. Neurology 56, 16961701.
  • Gold, C.A. & Budson, A.E. (2008). Memory loss in Alzheimer's disease: implications for development of therapeutics. Expert Rev Neurother 8, 18791891.
  • Greenwood, P.M., Lambert, C., Sunderland, T. & Parasuraman, R. (2005). Effects of apolipoprotein E genotype on spatial attention, working memory, and their interaction in healthy, middle-aged adults: results from the National Institute of Mental Health's BIOCARD study. Neuropsychology 19, 199211.
  • Grootendorst, J., Bour, A., Vogel, E., Kelche, C., Sullivan, P.M., Dodart, J.-C., Bales, K. & Mathis, C. (2005). Human apoE targeted replacement mouse lines: h-apoE4 and h-apoE3 mice differ on spatial memory performance and avoidance behavior. Behav Brain Res 159, 114.
  • Hartman, R.E., Wozniak, D.F., Nardi, A., Olney, J.W., Sartorius, L. & Holtzman, D.M. (2001). Behavioral phenotyping of GFAP-apoE3 and–apoE4 transgenic mice: apoE4 mice show profound working memory impairments in the absence of Alzheimer's-like neuropathology. Exp Neurol 170, 326344.
  • Holtzman, D.M., Bales, K.R., Wu, S., Bhat, P., Parsadanian, M., Fagan, A.M., Chang, L.K., Sun, Y., & Paul, S.M. (1999). Expression of human apolipoprotein E reduces amyloid-beta deposition in a mouse model of Alzheimer's disease. J Clin Invest 103, R15R21.
  • Kimura, R. & Ohno, M. (2009). Impairments in remote memory stabilization precede hippocampal synaptic and cognitive failures in 5XFAD Alzheimer mouse model. Neurobiol Dis 33, 229235.
  • Kornecook, T.J., McKinney, A.P., Ferguson, M.T. & Dodart, J.-C. (2008). Gender- and isoform-specific effects of apolipoprotein E on cognitive performance in mice overexpressing human APP. Program No. 638.3. 2008 Abstract Viewer/Itinerary Planner. Society for Neuroscience 2008, Washington DC: Online .
  • Lalonde, R. (2002). The neurobiological basis of spontaneous alternation behavior. Neurosci Biobehav Rev 26, 91104.
  • Lamb, B.T., Sisodia, S.S., Lawler, A.M., Slunt, H.H., Kitt, C.A., Kearns, W.G., Pearson, P.L., Price, D.L. & Gearhart, J.D. (1993). Introduction and expression of the 400 kilobase amyloid precursor protein gene in transgenic mice. Nat Genet 5, 2230.
  • Lawrence, A.D. & Sahakian, B.J. (1995). Alzheimer's disease, attention, and the cholinergic system. Alzheimer Dis Assoc Disorders 9, 4349.
  • Mahley, R.W. (1988). Apolipoprotein E: Cholesterol transport protein with expanding role in cell biology. Science 240, 622630.
  • Mahley, R.W., Weisgraber, K.H. & Huang, Y. (2006). Apolipoprotein E4: A causative factor and therapeutic target in neuropathology, including Alzheimer's disease. Proc Natl Acad Sci U S A 103, 56445651.
  • Manelli, A.M., Stine, W.B., Van Eldik, L.J. & LaDu, M.J. (2004). ApoE and Aβ1–42 interactions: effects of isoform and conformation on structure and function. J Mol Neurosci 23, 235246.
  • Mortensen, E.L. & Hogh, P. (2001) A gender difference in the association between APOE genotype and age-related cognitive decline. Neurology 57, 8995.
  • Murai, H., Pierce, J.E.S., Raghupathi, R. et al. (1998). Twofold overexpression of human β-amyloid precursor proteins in transgenic mice does not affect the neuromotor, cognitive, or neurodegenerative sequelae following experimental brain injury. J Comp Neurol 392, 428438.
  • Namba, Y., Tomonaga, M., Kawasaki, H., Otomo, E. & Ikeda, K. (1991). Apolipoprotein E immunoreactivity in cerebral amyloid deposits and neurofibrillary tangles in Alzheimer's disease kuru plaque amyloid in Creutzfeldt-Jacob disease. Brain Res 541, 163166.
  • Patel, S., Stolerman, I.P., Asherson, P. & Sluyter, F. (2006). Attentional performance of C57BL/6 and DBA/2 mice in the 5-choice serial reaction time task. Behav Brain Res 170, 197203.
  • Raber, J., Wong, D., Buttini, M., Orth, M., Bellosta, S., Pitas, R. E., Mahley, R.W. & Mucke, L. (1998). Isoform-specific effects of human apolipoprotein E on brain function revealed in ApoE knockout mice: increased susceptibility of females. Proc Natl Acad Sci U S A 95, 1091410919.
  • Rebeck, G.W., Reiter, G.S., Strickland, D.K. & Hyman, B.T. (1993). Apolipoprotein E in sporadic Alzheimer's disease: allelic variation and receptor interactions. Neuron 11, 575580.
  • Reed, T., Carmelli, D., Swan, G.E. et al . (1994). Lower cognitive performance in normal older adult male twins carrying the apolipoprotein E ε4 allele. Arch Neurol 51, 11891192.
  • Roses, A.D., Strittmatter, W.J., Pericak-Vance, M.A. et al . (1994). Clinical application of apolipoprotein E genotyping to Alzheimer's disease. Lancet 343, 15641565.
  • Phillips, R.G. & LeDoux, J.E. (1992). Differential contribution of amygdala and hippocampus to cued and contextual fear conditioning. Behav Neurosci 106, 274285.
  • Poirier, J. (1994). Apolipoprotein E in animal models of CNS injury and in Alzheimer's disease. Trends Neurosci 17, 525530.
  • Raber, J., Wong, D., Buttini, M., Orth, M., Bellosta, S., Pitas, R.E., Mahley, R.W. & Mucke, L. (1998). Isoform-specific effects of human apolipoprotein E on brain function revealed in ApoE knockout mice: Increased susceptibility of females. Proc Natl Acad Sci U S A 95, 1091410919.
  • Sadowski, M.J., Pankiewicz, J., Scholtzova, H., Mehta, P.D., Prelli, F., Quartermain, D. & Wisniewski, T. (2006). Blocking the apolipoprotein E/amyloid-β interaction as a potential therapeutic approach for Alzheimer's disease. Proc Natl Acad Sci U S A 103, 1878718792.
  • Saunders, A.M., Strittmatter, W.J., Schmechel, D. et al . (1993). Association of apolipoprotein E allele epsilon 4 with late-onset familial and sporadic Alzheimer's disease. Neurology 43, 14671472.
  • Strittmatter, W.J., Saunders, A.M., Schmechel, D., Pericak-Vance, M., Enghild, J., Salvesen, G.S. & Roses, A.D. (1993). Apolipoprotein E: high-avidity binding to β-amyloid and increased frequency of type 4 allele in late-onset Alzheimer's disease. Proc Natl Acad Sci U S A 90, 19771981.
  • Sullivan, P.M., Mace, B.E., Maeda, N. & Schmechel, D.E. (2004). Marked regional differences of brain human apolipoprotein E expression in targeted replacement mice. Neuroscience 124, 725733.
  • Thinus-Blanc, C., Save, E., Rossi-Arnaud, C., Tozzi, A. & Ammassari-Teule, M. (1996). The differences shown by C57Bl/6 and DBA/2 inbred mice in detecting spatial novelty are subserved by a different hippocampal and parietal cortex interplay. Behav Brain Res 80, 3340.
  • Van Meer, P., Acevedo, S & Raber, J. (2007). Impairments in spatial memory retention of GFAP-apoE4 female mice. Behav Brain Res 176, 372375.
  • Weisgraber, K.H. & Mahley, R.W. (1996). Human apolipoprotein E: the Alzheimer's disease connection. Faseb J 10, 14851494.
  • Wisniewski, T., Castano, E.M., Golabek, A., Vogel, T. & Frangione, B. (1994). Acceleration of Alzheimer's fibril formation by apolipoprotein E in vitro. Am J Pathol 145, 10301035.