UBC Faculty Research and Publications

Cognition, learning behaviour and hippocampal synaptic plasticity are not disrupted in mice over-expressing… Parkinson, Pamela F; Kannangara, Timal S; Eadie, Brennan D; Burgess, Braydon L; Wellington, Cheryl L; Christie, Brian R Feb 24, 2009

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12944_2008_Article_193.pdf [ 894.85kB ]
JSON: 52383-1.0228475.json
JSON-LD: 52383-1.0228475-ld.json
RDF/XML (Pretty): 52383-1.0228475-rdf.xml
RDF/JSON: 52383-1.0228475-rdf.json
Turtle: 52383-1.0228475-turtle.txt
N-Triples: 52383-1.0228475-rdf-ntriples.txt
Original Record: 52383-1.0228475-source.json
Full Text

Full Text

ralssBioMed CentLipids in Health and DiseaseOpen AcceResearchCognition, learning behaviour and hippocampal synaptic plasticity are not disrupted in mice over-expressing the cholesterol transporter ABCG1Pamela F Parkinson†1, Timal S Kannangara†2,3,4,5, Brennan D Eadie2,3,4,5,6, Braydon L Burgess1, Cheryl L Wellington1 and Brian R Christie*2,3,4,5,7Address: 1The Department of Pathology and Laboratory Medicine, Child and Family Research Institute, University of British Columbia, Vancouver, BC, Canada, 2Division of Medical Sciences, University of Victoria, Victoria, BC, Canada, 3The Brain Research Centre, University of British Columbia, Vancouver, BC, Canada, 4The Graduate Program in Neuroscience, University of British Columbia, Vancouver, BC, Canada, 5The Department of Cellular and Physiological Sciences, University of British Columbia, Vancouver, BC, Canada, 6MD/PhD program, University of British Columbia, Vancouver, BC, Canada and 7The Department of Biology, University of Victoria, Victoria, BC, CanadaEmail: Pamela F Parkinson - parkinsonpam@gmail.com; Timal S Kannangara - timal.k@gmail.com; Brennan D Eadie - brennan.eadie@gmail.com; Braydon L Burgess - bburgess@chori.org; Cheryl L Wellington - cheryl@cmmt.ubc.ca; Brian R Christie* - brain64@uvic.ca* Corresponding author    †Equal contributorsAbstractBackground: Cognitive deficits are a hallmark feature of both Down Syndrome (DS) andAlzheimer's Disease (AD). Extra copies of the genes on chromosome 21 may also play an importantrole in the accelerated onset of AD in DS individuals. Growing evidence suggests an importantfunction for cholesterol in the pathogenesis of AD, particularly in APP metabolism and productionof Aβ peptides. The ATP-Binding Cassette-G1 (ABCG1) transporter is located on chromosome21, and participates in the maintenance of tissue cholesterol homeostasis.Results: To assess the role of ABCG1 in DS-related cognition, we evaluated the cognitiveperformance of mice selectively over-expressing the ABCG1 gene from its endogenous regulatorysignals. Both wild-type and ABCG1 transgenic mice performed equivalently on several behavioraltests, including measures of anxiety, as well as on reference and working memory tasks. No deficitsin hippocampal CA1 synaptic plasticity as determined with electrophysiological studies wereapparent in mice over-expressing ABCG1.Conclusion: These findings indicate that although ABCG1 may play a role in maintaining cellularor tissue cholesterol homeostasis, it is unlikely that excess ABCG1 expression contributes to thecognitive deficits in DS individuals.BackgroundThe central nervous system is the most sterol-rich organ inthe body, containing over 25% of total body cholesterolsis in the brain is largely regulated independently fromthat in peripheral tissues. Aberrant brain cholesterolmetabolism has been associated with compromised AβPublished: 24 February 2009Lipids in Health and Disease 2009, 8:5 doi:10.1186/1476-511X-8-5Received: 23 December 2008Accepted: 24 February 2009This article is available from: http://www.lipidworld.com/content/8/1/5© 2009 Parkinson et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 8(page number not for citation purposes)in only 2% of total body weight [1]. Cholesterol cannotcross the blood brain barrier; therefore, sterol homeosta-metabolism, cognitive function, and synaptogenesis [2].In addition, a link between brain cholesterol metabolismLipids in Health and Disease 2009, 8:5 http://www.lipidworld.com/content/8/1/5and learning and memory was recently observed with thedemonstration that mice deficient in 24-hydroxylase donot show long-term potentiation (LTP) and are pro-foundly impaired in spatial learning tasks [3]. Under-standing how genes that regulate sterol homeostasis in thebrain affect cognitive and neuronal function may there-fore lead to insights into several disorders of the centralnervous system.ABCG1 is the founding member of the ABCG subclass ofthe ATP-Binding-Cassette (ABC) transporter superfamily[4]. Like several other members of the ABCG family,ABCG1 functions in the regulation of sterol homeostasisand high density lipoprotein (HDL) metabolism. ABCG1is widely expressed in a variety of tissues, with the highestlevels of expression observed in macrophage-rich tissuessuch as the spleen, lungs, and thymus; correspondingly,the majority of studies concerning ABCG1 functionfocused on the peripheral tissues. There are, however, sev-eral lines of evidence supporting the idea that ABCG1 mayhave an important role in the CNS, particularly concern-ing neurodegenerative disorders. ABCG1 is abundantlyexpressed in the brain, including the hippocampal forma-tion [5,6]. Additionally, ABCG1 mRNA and protein levelsare significantly increased in DS cortex, and over-expres-sion of ABCG1 affects Aβ production in transfected cells[7]. Interestingly, selective over-expression of ABCG1 invivo did not affect Aβ or amyloid levels in the PDAPPmouse model of AD (Burgess et al., 2008), suggesting thatover-expression of ABCG1 is unlikely to contribute to theearly onset of AD in the DS population. As ABCG1 doesalter cholesterol homeostasis in the brain (Burgess et al.,2008), it therefore remains possible that excess ABCG1may influence neuronal physiology and contribute to thecognitive deficits in DS individuals. The purpose of thepresent study was to investigate the role of ABCG1 on cog-nition, learning behaviour and synaptic plasticity in thehippocampus.MethodsMiceAll procedures were approved by the University of BritishColumbia Animal Care Committee and are in accordancewith the Canada Council on Animal Care. ABCG1 bacte-rial artificial chromosome (BAC) transgenic mice weregenerated as previously described [6]. All animals weregroup housed in standard cages in a colony maintained at21°C. Animals were maintained on a 12-hour light/darkcycle with access to food and water ad libitum.ABCG1 Protein Expression AssayTo investigate protein expression of ABCG1 in adult brain,mice were anesthetized by intraperitoneal administrationmg/kg ketamine (Bimeda-MTC) mixture, then perfusedfor 7 min with phosphate buffered saline (PBS) contain-ing 2500 U/L Heparin. Brains were collected and dissectedinto cortex, hippocampus, and cerebellum, and snap fro-zen at -80°C until analysis. ABCG1 protein expressionwas detected in a crude total membranes preparation asdescribed [8]. Briefly, tissues were homogenized in 5 vol-umes of lysis buffer (50 mM mannitol, 2 mM EDTA, 50mM Tris HCl pH 7.6, and Complete protease inhibitor(Roche)), and centrifuged at 500 × g to pellet nuclei anddebris. Between 400–450 μl of supernatant was layeredonto 600 μl of fractionation buffer (300 mM mannitol, 2mM EDTA, 50 mM Tris HCl pH 7.6) and centrifuged at100,000 × g for 45 min to pellet total membranes. Mem-branes were resuspended in 150–200 μl of lysis buffer.SDS was added to a final concentration of 1% prior toTris-glycine SDS-PAGE and immunoblotted with antibod-ies against ABCG1 (Novus) and Na/K-ATPase (Novus) asan internal loading control.Behavioural AnalysisBehavioural analysis cohorts included a total of 22 femalemice, with 11 ABCG1 BAC Tg mice and 11 wild-type lit-termate controls. Phenotypic assessment of ABCG1 BACTg mice was carried out using the SHIRPA protocol [9] toscreen for any differences in baseline phenotype, as thesecould markedly affect performance on cognitive tests.Locomotor activity, exploratory behaviour and generalanxiety were analyzed using the open field test. A circulararena with a diameter of 90 cm was set up directly under-neath an overhead digital camera (Logitech QuickCamPro 5000) in a brightly lit room in an undisturbed, quietlocation. A room divider shielded the experimenter andcomputer from the animal's view for the duration of thetask. Animals were brought into the testing room, andgiven 5–10 minutes to adjust to the room's environmentin their home cages. Each mouse was placed in the arenaindividually, and allowed to freely explore for 5 minutes,while its activities were tracked and recorded using ANY-maze™ (Stoelting Co.). Upon completing the task, themouse was removed from the arena by the experimenterand returned to the home cage. The number of defeca-tions in the arena was recorded for each mouse, and thearena was cleaned with a mild soap following each test toavoid transfer of olfactory cues between animals. Distancetraveled, average speed and path length serve as measuresof locomotor activity and exploratory behaviour. Anxietywas assessed through measurement of time spent aroundthe edges of the arena relative to the time spent in the cen-tre, and also by freezing and defecation.Emotionality was specifically measured using the elevatedPage 2 of 8(page number not for citation purposes)of 600 mg/kg Avertin (Sigma-Aldrich) or intramuscularadministration of a 20 mg/kg xylazine (Bayer) and 150plus maze: an elevated maze with 2 open arms and 2enclosed arms, set up in a room in an undisturbed, quietLipids in Health and Disease 2009, 8:5 http://www.lipidworld.com/content/8/1/5location. The experimenter and computer were behind aroom divider, hidden from the animal's view for the dura-tion of the task. Animals were given 5–10 minutes in theirhome cages to adjust to the testing room. Each mouse wasinitially placed in the centre of the maze, and allowed 5minutes of free exploration. The animal's activities weretracked and recorded by an overhead camera (LogitechQuickCam Pro 5000) and ANY-maze™. Upon completionof the task, the mouse was returned to its home cage andnumber of defecations was recorded before the maze wascleaned with a mild soap prior to the next trial.Assessment of Learning and MemoryHippocampal-dependent learning and memory wasassessed using several variations of the Morris Water Mazetask. The water maze consisted of a circular white pool(100 cm diameter, 54 cm height) filled with water to aheight of 36 cm. The water was rendered opaque usingwhite, non-toxic tempera paint. A circular platform (14cm diameter, 35 cm height) was submerged in the pool(about 1 cm below the surface), providing a ledge uponwhich the mice could step to escape from the water. Foreach trial, a mouse was released into the pool at one ofthree release positions and allowed to search for the plat-form. Each trial lasted a maximum of 60 seconds, afterwhich the mouse was manually guided to the platformand allowed to remain on the platform for another 10 sec-onds. All trials were tracked using an overhead digitalcamera (Logitech QuickCam Pro 5000) and ANY-maze™computer software.The visible platform task was used to screen for any visualdeficits, as well as motor problems (such as inability toswim) that may hinder an animal's ability to completewater maze training trials. The platform was located about2 cm above the surface of the water, and a visually con-spicuous marker, such as a small coloured flag, wasattached to the top of the platform. The platform locationwas changed for each of the 4 trials. Mice were releasedinto the pool and allowed 60 seconds to locate the plat-form, while being tracked and recorded by ANY-maze™.The experimenter noted any apparent deficits in vision ormotor behaviors.The reference memory task was used to assess learningand long-term memory for a fixed platform location. Dis-tinct geometric shapes were attached to the walls of thepool on three sides, and the experimenter and computersystem were hidden behind a dividing wall throughoutthe experiment to avoid influence of extraneous cues.Mice were trained for 4 trials per day, for a total of 5 days.Twenty-four hours after the last training session, a probetrial was administered, in which the platform wasquadrant of the pool that had previously contained theplatform was measured as an indication of memory reten-tion.ElectrophysiologyFor in vitro electrophysiology, aged animals were anesthe-tized with isofluroane and decapitated. The brain wasremoved and immersed in ice-cold sucrose artificial cere-brospinal fluid (sACSF; pH7.2) containing (in mM):110.00 sucrose, 60.00 NaCl, 3.00 KCl, 1.25 NaH2PO4,28.00 NaHCO3, 0.50 CaCl2, 7.00 MgCl2, 5.00 dextroseand 0.60 ascorbate, oxygenated with 95%O2/5%CO2.Transverse hippocampal slices (350 μm) were generatedusing a Vibratome Sectioning System 1500 (Ted Pella,Redding, CA). Slices were gently transferred to an incuba-tion chamber filled with oxygenated normal artificial cer-ebrospinal fluid (ACSF; pH 7.2) containing (in mM):125.0 NaCl, 2.5 KCl, 1.25 NaH2PO4, 25.0 NaHCO3, 2CaCl2, 1.3 MgCl2, and 10 dextrose, and maintained at30°C for a minimum of 1 hour post-dissection. Sliceswere then transferred to a recording chamber superfusedat a rate of 2 ml/min with 30°C, oxygenated ACSF. Arecording electrode (0.7–1.5 MΩ) filled with ACSF wasplaced under visual guidance into the striatum radiatumof the hippocampal CA1 region using an OlympusBX51W1 microscope. Field excitatory postsynaptic poten-tials (fEPSPs) were evoked using monophasic negativecurrent pulses (120 μs, 10–80 μA) supplied to the CA1Schaeffer Collaterals via a concentric bipolar stimulatingelectrode (FHC, Bowdoin, ME) connected to a digitalstimulus isolation unit (Getting Instruments, San Diego,CA). Responses were acquired at 100 Hz using a Multi-Clamp 700B amplifier (Molecular Devices, Sunnyvale,CA) and collected on a computer for offline analysis. Foreach slice, stimulus intensity was adjusted to yield 50–55% of the maximal response. Prior to baseline measure-ments, a paired-pulse protocol (50 ms inter-stimulusinterval) was employed. Baseline measurements were col-lected using individual fEPSPs evoked every 15 seconds. Asteady baseline of 15 minutes was required for allresponses. Following baseline acquisition, a conditioningprotocol was applied: high frequency stimulation (HFS:four bursts of 50 pulses at 100 Hz, 30 s between bursts) toinduce long-term potentiation. Immediately after the con-ditioning protocol, baseline measurements were acquiredfor a minimum of 1 hour. Following the hour of postacquisition, a second paired-pulse protocol (50 ms inter-stimulus interval) was conducted. Computed results wereprocessed for statistical analysis using Clampfit (Molecu-lar Devices, Sunnyvale, CA), Excel 2007 (Microsoft) andStatistica 7.0 (Statsoft). For all studies, data was presentedas means ± standard error of the mean (S.E.M.) and com-pared using unpaired t-tests. Differences were consideredPage 3 of 8(page number not for citation purposes)removed from the pool and the mouse was allowed toswim freely for the full 60 seconds. Time spent in thesignificant when P < 0.05.Lipids in Health and Disease 2009, 8:5 http://www.lipidworld.com/content/8/1/5ResultsABCG1 Expression in Brains of Transgenic MiceAs previously reported [6], over-expression of ABCG1 wasobserved in the brains of ABCG1 transgenic mice thatwere generated using a BAC insertion of the full humanABCG1 gene (ABCG1 BAC Tg mice), with protein levelsbetween 3–8 fold higher than that observed in wild-typelittermate control animals (Figure 1A). The level of over-expression varied with brain region, with greater over-expression in the cortex and hippocampus as compared tothe cerebellum. This verified that our transgene does infact significantly increase ABCG1 protein levels in thebrain over endogenous levels.Baseline Behavioural Analysis using SHIRPAA cohort of 22 female mice (n = 11 ABCG1 BAC Tg, n = 11wild-type; approximately 10 months of age) were assessedusing the categories described by the SHIRPA protocol [9]to test whether over-expression of ABCG1 resulted in anygross phenotypic abnormalities. Observations madeusing the primary screen of the SHIRPA protocol showedthat animals expressing the ABCG1 BAC transgene had asimilar baseline phenotype to wild-type littermates (Fig-ure 1B). ABCG1 BAC Tg mice did not display any abnor-malities in physiological measures such as heart rate,body tone and grip strength or in behavioural measuressuch as irritability, aggression and exploratory activity.Anxiety and General Locomotor ActivityABCG1 BAC Tg mice (n = 11) and wild-type littermates (n= 11; approximately 12 months of age) were next evalu-ated for anxiety level and general locomotor activity, asboth factors can interfere with learning tasks and thereforecan drastically affect performance during cognitive assess-ment. The single-trial Open Field Test revealed no differ-ence between groups in time spent in different locationsin the arena. Both ABCG1 BAC Tg and wild-type micespent the majority of their time exploring the edges of theopen arena, and generally avoided the brightly lit centre ofthe arena. No differences were observed when mice werealso analyzed for average speed in the arena, distance trav-elled and time spent immobile or frozen in the arena (Fig-ure 2), indicating that ABCG1 BAC Tg mice have normalexploratory and locomotor behaviour and do not showincreased anxiety levels compared to wild-type animals.The Elevated Plus Maze, a behavioural task specificallydesigned to evaluate anxiety, was administered to verifythat anxiety levels are not altered in ABCG1 BAC Tg mice.Wild-type and ABCG1 BAC Tg mice spent comparableamounts of time in the different locations within themaze, with both groups displaying preference for theclosed arms of the maze (Figure 2). There were no differ-ences between groups in average speed or distanceLearning and MemoryMice expressing the ABCG1 BAC transgene were next eval-uated against wild-type littermates on the Morris WaterMaze spatial reference memory task. The task consisted ofa 5-day acquisition phase in which mice were given 4 tri-als per day and a maximum time of 60 seconds per trial tolocate a platform fixed in the North quadrant. Twenty-four hours later, a probe trial was administered to testrecall of the platform location. ABCG1 BAC Tg mice dis-played nearly identical average escape latency, or time tolocate the platform, over each day of acquisition as com-pared to wild-type animals, and the learning curve wasvery similar (Figure 3). These results indicate that ABCG1over-expression does not significantly affect learning dur-ing the acquisition phase of this water maze task.Repeated-measures ANOVA revealed a significant effect oftraining day, (p < 0.0001), but not group (p = 0.67). Per-formance on the single-trial probe test was also compara-ble between groups, with both wild-type and ABCG1 BACTg mice spending the greatest amount of time searching inthe probe quadrant (North), which previously containedthe platform during training days. Both groups displayedstrong preference for the North quadrant, indicating thatboth wild-type and ABCG1 BAC Tg mice were capable oflearning the task thoroughly, and were able to remembervisual cues associated with the platform location duringtraining trials. Additionally, the comparable performanceof ABCG1 BAC Tg and wild-type mice on both learningduring the acquisition phase or on memory recall in theprobe trial, also suggests that over-expression of ABCG1alone does not negatively affect hippocampal neuronfunction involved in the learning and memory requiredfor this task.Excess ABCG1 does not alter synaptic plasticityTo test the capacity for synaptic plasticity in mice over-expressing ABCG1, we monitored the induction of long-term potentiation by applying high-frequency stimulation(four bursts of 50 pulses at 100 Hz, 30 s between bursts)to the Shaeffer collaterals of the CA1. Hippocampal slicesfrom both wild-type and ABCG1 BAC Tg animals demon-strated robust long-term potentiation, suggesting intactsynaptic plasticity in transgenic animals (Figure 3C). Inaddition to long-term potentiation, a paired-pulse proto-col (50 inter-stimulus interval) was employed before andafter the high-frequency stimulation tetanus as a measureof presynaptic properties [10]. Standard paired-pulsefacilitation was observed in both wild-type and ABCG1BAC Tg animals (Figure 3D). Taken together, these resultsindicate that the selective over-expression of ABCG1 doesnot affect synaptic plasticity in the CA1.DiscussionPage 4 of 8(page number not for citation purposes)traveled in the maze. Time spent immobile, an indicatorof anxiety, was also similar for both groups.The present experiments investigated the effects of ABCG1over-expression on both baseline behavioural phenotypeLipids in Health and Disease 2009, 8:5 http://www.lipidworld.com/content/8/1/5Page 5 of 8(page number not for citation purposes)ABCG1 BAC Tg mice over-express the ABCG1 transporter in the cortex, hippocampus, and cerebellum, and show normal behaviourFigure 1ABCG1 BAC Tg mice over-express the ABCG1 transporter in the cortex, hippocampus, and cerebellum, and show normal behaviour. A) Crude membrane fractions were extracted from dissected brain regions and were subjected to Western blot. Blots were probed by polyclonal antibodies recognizing human and murine ABCG1 or Na/K-ATPase as a loading control. ABCG1 protein levels are increased 6-fold in cortex (Cor) and hippocampus (Hipp) and 3-fold in cerebellum (Cer) of ABCG1-BAC-Tg (Tg) mice relative to baseline levels in the equivalent region from wild-type (WT) controls. B) SHRPA primary screen on ABCG1 BAC Tg (ABCG1; n = 11) and wild-type (WT; n = 11) mice. Primary screen involves physiological profiling of each mouse, using a number of test categories and assigning a rating for each mouse in each category. All p values are non-significant when compared to WT animals.Lipids in Health and Disease 2009, 8:5 http://www.lipidworld.com/content/8/1/5Page 6 of 8(page number not for citation purposes)ABCG1 over-expression does not alter anxiety or general locomotor activityFigure 2ABCG1 over-expression does not alter anxiety or general locomotor activity. ABCG1 BAC Tg (Tg; n = 11) and wild-type mice (WT; n = 11) were assessed for differences in anxiety or locomotor activity using the Open Field Test (A, C, E, G) and the Elevated Plus Maze (B, D, F, H). Data from each test were analyzed to include measures of time spent in maze loca-tions (A, B), mean speed (C, D) and distance (E, F) and time spent immobile (G, H) in the maze. All p values are non-significant.Lipids in Health and Disease 2009, 8:5 http://www.lipidworld.com/content/8/1/5and cognition. ABCG1 BAC Tg mice were indistinguisha-ble from wild-type animals when assessed using theSHIRPA protocol, indicating that over-expression ofABCG1 does not result in any gross metabolic distur-bances that could affect observable behavioural or physi-ological measures. Cognitive measures were alsoanxiety, general locomotor activity and spatial learningand memory were all comparable to those of wild-typeanimals. Additionally, our results show that stimulationof Schaeffer axon collaterals elicits comparable long-termpotentiation in both ABCG1 BAC Tg and wild-type ani-mals. No significant difference in paired-pulse facilitationABCG1 over-expression does not impact spatial reference memory or synaptic plasticityFigure 3ABCG1 over-expression does not impact spatial reference memory or synaptic plasticity. The Morris Water Maze was employed to assess spatial reference memory in wild-type (WT; n = 11) and ABCG1 BAC Tg (Tg; n = 11) animals. (A) Latency to platform. Each point represents an average of 4 daily trials during the training period. No significant difference was observed between the groups in the time taken to find the hidden platform. (B) Probe trial results from a single-trial test. Both control and ABCG1 over-expressing groups showed similar preference for the North (N) quadrant, which contained the platform on training trials. All p values are non-significant. (C) In vitro electrophysiology was performed on 350 μm thick hip-pocampal slices derived from aged mice over-expressing ABCG1 (Tg) and wild-type (WT) littermates. High frequency stimula-tion (HFS) was applied to the Schaeffer collaterals to induce LTP in the CA1 region. No significant difference was observed between wild-type and transgenic mice. (D) Two stimuli were applied to the Schaeffer collaterals, including paired-pulse facili-tation in the CA1 region, before and after the HFS tetanus protocol. No significant differences were observed between wild-type and transgenic mice, indicating that presynaptic release was unaffected by HFS in either mouse phenotype.Page 7 of 8(page number not for citation purposes)demonstrated to be unaffected in transgenic animals. Theperformance of ABCG1 BAC Tg mice on tasks measuringwas observed between ABCG1 BAC Tg mice and wild-typelittermates both prior to, and after tetanus stimulation,Lipids in Health and Disease 2009, 8:5 http://www.lipidworld.com/content/8/1/5suggesting that the presynaptic properties of ABCG1-over-expressing mice are intact. These data strongly suggest thatover-expression of ABCG1 alone is not sufficient to causechanges in cognition, learning and memory or hippocam-pal synaptic plasticity.While ABCG1 over-expression does not elicit a change incognition, ABCG1 over-expression appears to alter thebrain at the cellular level. Although total brain cholesterollevels are unaffected, over-expression of ABCG1 leads todecreased levels of sterol precursors and 24S-hydroxycho-lesterol in the brain, suggesting that ABCG1 suppressesflux through the cholesterol biosynthetic pathway [6].There are several possible explanations for the lack oftransgenic phenotype. It is possible that the changescaused by over-expression of ABCG1 may be too subtle tohave globally obvious effects on cognition, producingchanges at the neuronal level that may be below the detec-tion threshold of the behavioural and cognitive tests usedin our experiments. Another possible explanation for thelack of transgenic phenotype is the existence of a compen-satory pathway that acts to balance out the effects ofABCG1 over-expression and maintain normal cognitivefunctions in the brain. The ABCG4 transporter couldpotentially act in this manner, as it is highly homologousto ABCG1 and is co-expressed with ABCG1 in neuronsand astrocytes [11]. Moreover, ABCG4 has also been pro-posed to be important in cholesterol homeostasis; bothABCG1-deficient and ABCG4-deficient mice each exhibitrepression of SREBP-2 target genes involved in cholesterolsynthesis, suggesting that ABCG4 does function in regula-tion of cholesterol levels. It is possible that the ABCG4pathway compensates for imbalances caused by over-expression of ABCG1, preventing major changes at thelevel of cognition. Further studies will be required todetermine the exact functions of ABCG4 in brain choles-terol homeostasis, and whether it is indeed part of a path-way that could compensate for alterations in ABCG1function.ConclusionIn conclusion, no effect of ABCG1 over-expression wasseen in hippocampal synaptic plasticity and behaviouralparameters such as anxiety and general locomotor activ-ity, or on learning and memory, suggesting that despitechanges in cholesterol flux in brain, it is unlikely thatover-expression of ABCG1 contributes to cognitive defi-cits in the DS population.Competing interestsThe authors declare that they have no competing interests.Authors' contributionsof the data and assisted with manuscript preparation. TKcarried out electrophysiology procedures, performed dataanalysis and drafted the manuscript. BE performed elec-trophysiological experiments and data analysis. BB per-formed perfusions and immunoblotting, and participatedin the planning of the study. CW and BC contributed toplanning of the experiments and discussion of results, andco-supervised the study.AcknowledgementsThe authors thank E. Wiebe, J.D. Shin, R.P. Peterson, A. Chang and J.Y. Chan for advice and technical assistance. CLW is supported by grants from the Alzheimer's Society of Canada, CIHR, and the Pacific Alzheiemer's Research Foundation. BRC is supported by grants from CIHR and NSERC and is a Michael Smith Senior Scholar.References1. Dietschy JM, Turley SD: Cholesterol metabolism in the brain.Curr Opin Lipidol 2001, 12(2):105-112.2. Hirsch-Reinshagen V, Wellington CL: Cholesterol metabolism,apolipoprotein E, adenosine triphosphate-binding cassettetransporters, and Alzheimer's disease.  Curr Opin Lipidol 2007,18(3):325-332.3. Kotti TJ, Ramirez DM, Pfeiffer BE, Huber KM, Russell DW: Braincholesterol turnover required for geranylgeraniol produc-tion and learning in mice.  Proc Natl Acad Sci USA 2006,103(10):3869-3874.4. Velamakanni S, Wei SL, Janvilisri T, van Veen HW: ABCG trans-porters: structure, substrate specificities and physiologicalroles: a brief overview.  J Bioenerg Biomembr 2007, 39(5–6):465-471.5. Nakamura K, Kennedy MA, Baldan A, Bojanic DD, Lyons K, EdwardsPA: Expression and regulation of multiple murine ATP-bind-ing cassette transporter G1 mRNAs/isoforms that stimulatecellular cholesterol efflux to high density lipoprotein.  J BiolChem 2004, 279(44):45980-45989.6. Burgess BL, Parkinson PF, Racke MM, Hirsch-Reinshagen V, Fan J,Wong C, Stukas S, Theroux L, Chan JY, Donkin J, et al.: ABCG1influences the brain cholesterol biosynthetic pathway butdoes not affect amyloid precursor protein or apolipoproteinE metabolism in vivo.  J Lipid Res 2008, 49(6):1254-1267.7. Tansley GH, Burgess BL, Bryan MT, Su Y, Hirsch-Reinshagen V,Pearce J, Chan JY, Wilkinson A, Evans J, Naus KE, et al.: The choles-terol transporter ABCG1 modulates the subcellular distri-bution and proteolytic processing of beta-amyloid precursorprotein.  J Lipid Res 2007, 48(5):1022-1034.8. Albrecht C, Soumian S, Amey JS, Sardini A, Higgins CF, Davies AH,Gibbs RG: ABCA1 expression in carotid atheroscleroticplaques.  Stroke 2004, 35(12):2801-2806.9. Rogers DC, Fisher EM, Brown SD, Peters J, Hunter AJ, Martin JE:Behavioral and functional analysis of mouse phenotype:SHIRPA, a proposed protocol for comprehensive phenotypeassessment.  Mamm Genome 1997, 8(10):711-713.10. Schulz PE, Cook EP, Johnston D: Changes in paired-pulse facilita-tion suggest presynaptic involvement in long-term potentia-tion.  J Neurosci 1994, 14(9):5325-5337.11. Tarr PT, Edwards PA: ABCG1 and ABCG4 are coexpressed inneurons and astrocytes of the CNS and regulate cholesterolhomeostasis through SREBP-2.  J Lipid Res 2008, 49(1):169-182.Page 8 of 8(page number not for citation purposes)PP participated in planning the study, carried out behav-ioural and cognitive testing, performed statistical analysis


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items