Metabolic Profiling of Rat Brain and Cognitive Behavioral Tasks

Metabolic Profiling of Rat Brain and Cognitive Behavioral Tasks: Potential Complementary Strategies in Preclinical Cognition Enhancement Research Dilys P. Q. Goh,† Aveline H. Neo,‡ Catherine W. Goh,‡ Chiu Cheong Aw,‡ Lee Sun New,† Woei Shin Chen,‡ Zeenat Atcha,‡ Edward R. Browne,‡ and Eric C. Y. Chan*,† Department of Pharmacy, Faculty of Science, National University of Singapore, 18 Science Drive 4, Singapore 117543, and GlaxoSmithKline R&D China, Centre for Cognition and Neurodegeneration Research, Biopolis at One-North, 11 Biopolis Way, The Helios Building #03-01/02, Singapore 138667 Received September 7, 2009

Abstract: In this study, the correlation between the metabolic profiles of rats undergoing cognition enhancement drug therapy and their associated cognitive behavioral outcomes were investigated. Male Lister Hooded rats were administered either with donepezil, galantamine, or vehicle and subjected to Atlantis watermaze training and novel object recognition tests. An UPLC/MS/MS method was developed to profile 21 neurologically related metabolites in the rat brains. Pharmacologically induced behavioral changes were compared subsequently with the metabolic fluctuations of neurologically related metabolites from multiple neurotransmitter pathways using multivariate and univariate statistical analyses. Significant improvements in cognitive behavioral outcomes were demonstrated in the rats administered with donepezil and galantamine using both AWM training (P < 0.05) and NOR (P < 0.05) tests as compared to those dosed with the vehicle. This corroborated with the significant elevation of eight prominent biomarkers after the cognitive enhancement therapy. An orthogonal partial least-squares discriminant analysis model generated using only the 8 metabolites identified as discriminating the drug-dosed rats from the vehicle-dosed rats gave a Q2 ) 0.566, receiver operator characteristic (ROC) AUC ) 1.000, using 7-fold cross validation. Our study suggests that metabolic profiling of rat brain is a potential complementary strategy to the cognitive behavioral tasks for characterizing neurobiological responses to cognition enhancement drug testing. Keywords: Alzheimer’s disease • cognitive behavioral tests • donepezil • galantamine • metabonomics • metabolic profiling • UPLC/MS/MS • chemometric

Introduction In the current experimental phase of Alzheimer’s disease (AD) therapies, many novel drug targets have surfaced. Immu* To whom correspondence should be addressed. Asst. Prof. Eric C.Y. Chan. Tel, +65 65166137; Fax, +65 67791554; E-mail, [email protected]. † National University of Singapore. ‡ GlaxoSmithKline R&D China. 10.1021/pr900795g CCC: $40.75

 2009 American Chemical Society

notherapies using humoral and cell-based approaches,1,2 vaccinations, anti-inflammatory drugs,2 granulocyte colonystimulating factor (G-CSF),3 and therapies targeting the oxidative neurotoxicity via metal chelation4 are currently being investigated for the treatment and prevention of AD. Though these novel drug targets hold new promise, molecular mechanisms underlying the neurobiology of cognitive deficits and pharmacological cognition enhancement in age-related cognitive disorders remain undefined. Inconsistent conceptualization and the “translational gap” between preclinical and clinical research5-8 are key obstacles to our understanding of the disease states. These bottlenecks implicate high rate of false positive drug candidate production and also form one crucial circumscription to discovering new therapeutics. Several domains of testing approaches have been adopted in the current preclinical research platform to assess novel drug therapies that ameliorate cognitive deficits. These approaches include the experimental animal disease models,9,10 cognitive behavioral tests,11 sensitivity analysis to pharmacological intervention,12 cell cultures from animal models,13 proteomics studies,14 neurotransmitters and neuropeptides analysis,15 quantitative magnetic resonance imaging (MRI) of transverse (T2) relaxation times16 and measurement of extracellular space size and geometry alterations.17 Yet up until today, fundamental questions regarding the validity of these preclinical measurements in predicting clinically relevant cognitive outcomes still persist. Empirical evidence obtained from earlier studies demonstrated that the pathological alterations of numerous transmitter systems are implicated in dementias and AD, resulting in theircomplexanddiversecognitivebehavioralmanifestations.18-21 Meta analysis of literatures (1966-2000) indicated that several cognitive behavioral tests, namely, Morris watermaze, radial maze, passive avoidance, and spontaneous alternation, corroborated with the roles of these classical transmitter systems.11 Nonetheless, none of these studies profiled the metabolites from multineurotransmitter systems simultaneously and their interpathway relationships might have been masked. On the basis of these findings, we hypothesized that the profiles of neurologically related metabolites from the classical transmitter systems in rat brain would correlate with and distinguish the preclinical outcomes of cognitive behavioral tests. Journal of Proteome Research 2009, 8, 5679–5690 5679 Published on Web 10/21/2009

technical notes In this study, we developed and validated an ultraperformance liquid chromatography tandem mass spectrometry (UPLC/MS/MS) method to profile the neurologically related metabolites of multiple principal transmitter pathways in the rat brain (Figure 1). The method was subsequently applied to elucidate these metabolites in the brain of rats dosed separately with donepezil (DON) and galantamine (GAL), which are drugs commonly prescribed for AD in the clinical settings. Finally, the neurologically related metabolic profiles were compared with the outcome measures of the Atlantis watermaze (AWM) training and novel object recognition (NOR) tests which are established precognitive tests adopted in pharmaceutical research.22,23 We demonstrated that pharmacologically induced changes in the cognitive behavioral patterns established through the animal tasks were associated with elevations of neurologically related metabolites from multiple neurotransmitter pathways.

Materials and Methods Materials and Reagents. GAL and DON were purchased from J.INC, India. Internal standards (IS) [acetominophen (APAP) and salicylic acid (SCA)] and the neurologically related metabolites (Figure 1) were purchased from Sigma-Aldrich (U.S.A.). Amicon Ultra-4 centrifugal filter units (3000 Da) were purchased from Millipore (U.S.A.). All other chemicals and reagents used for the experiments were of analytical grades. Animals. Male LH rats (3-4 months old, weighing 300-350 g, Harlan, UK) were housed five per cage at 20 ( 1 °C and 40 ( 2% humidity-controlled environment on a 12 h light/dark cycle, with ad libitum access to food and water. All experiments were compliant with the Singapore National Advisory Committee for Laboratory Animal Research guidelines for the use and care of animals for scientific purposes and GlaxoSmithKline animal research ethical standards. AWM Training Test. Three groups of 10 rats each were dosed intraperitoneally at 1 mL/kg with 0.5 mg/kg of DON, 1.0 mg/ kg of DON, and the vehicle (0.9% w/v sodium chloride solution), respectively. The dosing was 1 h before the first trial of each spatial cue (SC) training. SC trainings were performed over four days and the visual cue (VC) training was performed a day before the SC training. The dosing was repeated for another three separate groups of rats except that DON was replaced with GAL. The watermaze apparatus (diameter: 1.7 m, height: 0.65 m) was surrounded with permanent spatial cues to aid the formation of a spatial map necessary for escape learning. The water (26 ( 1 °C) was made opaque by adding 1 L of opacifier (1:1000, Syntran 5905, Interpolymer, USA). The pool was divided into four imaginary quadrants. The AWM platform (diameter: 20 cm) was placed in the center of one of the four quadrants. When the platform was raised, it was covered by approximately 2 cm of water and therefore invisible to the rat while swimming. A video camera was positioned directly above the pool to record each animal’s movements. Latency, path length, swim speed, and time spent in each quadrant were acquired using the Watermaze software (Actimetrics Inc.). At the start of each trial, the animal was placed into the tank facing the wall of the maze at randomly designated spatial locations. In the VC training, a curtain was drawn around the tank shielding the extra-maze spatial cues. The Atlantis platform was raised and a black cylindrical object (visual cue) was placed 40 cm above the platform. Each animal was allowed to locate the platform using the visual cue (four trials). When the 5680

Journal of Proteome Research • Vol. 8, No. 12, 2009

Goh et al. platform was located, the trial was stopped and the rat was left on the platform for 30 s. In the SC training, in the presence of the extra-maze spatial cues, the training performance was assessed over four days (six trials per day) with increasing difficulty (increasing platform dwell time as follows: 0.8, 1.5, 2.3, and 3.0 s on days 1, 2, 3 and 4, respectively). The Atlantis platform would rise if the rat swam within a designated trigger zone (40 cm diameter) directly above the platform for longer than the platform dwell time, or after 90 s. The starting position for each trial was randomized and each rat was allowed to search the platform for a maximum of 2 min. After which, the rat would be guided to the platform. When the platform was found, the trial stopped, and the rat was left on the platform for 30 s. The probe trial was conducted three days later. Utilizing a similar setup to the SC training but without the AWM platform, the rats were observed in the tank for 60 s. The percentage time spent by the rats in the target (where the AWM platform was placed previously in the SC training), adjacent left, adjacent right and opposite quadrants of the target, were recorded respectively. Data was analyzed by repeated-measures two way ANOVA followed by planned comparison to compare treatment versus vehicle groups using StatSoft Statistica (version 6.0) (Statsoft Inc., USA). NOR Test: 24 h Temporal Deficit Model. All animals were prehandled (8-10 min each), tailed marked, weighed and sham-dosed (intraperitoneally, twice daily) before habituation to the test caging (Tecniplast, UK) for a total of 2 days before the T1 trial. Three groups of twelve rats were each dosed intraperitoneally at 1 mL/kg with 0.5 mg/kg, 1.0 mg/kg of DON and vehicle (0.9% w/v sodium chloride solution), respectively. The NOR tests were performed over 2 days and DON and the vehicle were administered to the rats 1 h before each trial. Similarly, the experiment was repeated with GAL in place of DON using different rats. During the T1 trial, all animals were first prehabituated to the test cage without objects for 3 min. Two identical and stationary objects were subsequently positioned at the front of the cage and at equal distances from both sides. Within a 3-min interaction period, the total time spent exploring both objects was scored by a trained observer and the video data was recorded. For the T2 trial (24 h post-T1 trial), animals were briefly habituated in the test cage for 3 min, and presented subsequently with one familiar and one novel object for 3 min. The types and positions of the novel objects were randomly assigned. Total exploration time for each object was scored as described in T1 trial. The discrimination index (d2) was obtained using eq 1. Object exploration was only scored when the animal’s nose or mouth was in direct contact with either of the objects. Climbing or resting on the objects was not scored. Discrimination index (d2) )

Novel Time - Familiar Time Total Exploration Time (1)

Statistical analysis was performed using StatSoft Statistica (version 6.0) and all data were checked for normality prior to analysis. A one-way ANOVA followed by Fisher LSD posthoc comparison was used to compare treatment groups versus vehicle groups for the T1 trial and d2 index. For the T2 trial, a

Strategies in Cognition Enhancement Research

technical notes

Figure 1. Chemical structures of (A) DON, (B) GAL, (C) APAP, (D) SCA, (E) histamine (HIST), (F) glycine (GLY), (G) myoinositol (MYO), metabolites related to the (H) cholinergic, (I) dopaminergic/noradrenergic (catecholamines), (J) glutamatergic, and (K) serotonergic pathways. Journal of Proteome Research • Vol. 8, No. 12, 2009 5681

technical notes repeated-measures ANOVA followed by planned comparison was used to compare familiar versus novel object per treatment group. Animal Dosing for the Metabolic Profiling Study. On the basis of the doses that demonstrated the most optimum AWM training test responses, 2 groups of 10 rats each were dosed intraperitoneally at 1 mL/kg with 0.5 mg/kg of DON and 1.0 mg/kg of GAL, respectively for 4 days. For each drug regimen, 5 control rats were dosed at 1 mL/kg with the vehicle (0.9% w/v sodium chloride solution). Four days postdosing, both treated and control rats were decapitated at least 45 min after the last dose of GAL and DON, respectively. Similarly, on the basis of the optimum doses obtained for the NOR test, 2 groups of 10 rats each were dosed intraperitoneally at 1 mL/kg with 0.5 mg/kg of DON and 0.5 mg/kg of GAL, respectively for 2 days. For each drug regimen, five control rats were dosed at 1 mL/kg with the vehicle (0.9% w/v sodium chloride solution). Two days postdosing, both treated and control rats were decapitated as described above. After decapitation, the skull for each rat was opened and the brain was transferred into a plastic tube and frozen at -80 °C until analysis. Sample Preparation and UPLC/MS/MS Analysis. Each brain specimen was thawed and homogenized with an equal weight of 1% formic acid (FA) in cold water with 10 µM each of the IS using the mill mixer at 25.0 Hz for 25 min. The brain homogenate was centrifuged at 7000g at 4 °C for 20 min. 1200 µL of supernatant were aliquoted and filter-centrifuged at 7000g at 4 °C for 1.5 h. The filtrate was lyophilized for 8 h using the FreeZone benchtop freeze-dry system (Labconco Corporation, MO, US). The residue was reconstituted with 120 µL of cold water with 1% FA and vortex-mixed for 2 min. 60 µL of samples was transferred into an injection vial and 2 µL was injected for UPLC-MS/MS analysis. The extracts were analyzed using a UPLC-MS/MS method developed, optimized and validated in our lab using Waters ACQUITY UPLC system (Waters) interfaced with a hybrid triple quadrupole linear ion trap mass spectrometer (QTRAPMS) equipped with TurboIonSpray ESI source (3200 QTRAP, Applied Biosystems). MS experiments were performed using both ESI positive and negative modes. All chromatographic separations were performed using a Waters ACQUITY UPLC HSS T3, 1.8 µm, 100 × 2.1 mm i.d. column. The column and autosampler temperature were maintained at 45 and 4 °C, respectively. The mobile phase flow rates were 0.2 mL/min and 0.3 mL/min for the positive and negative multiple reaction monitoring (MRM) modes, respectively. Postcolumn dilution using IPA was infused using a Shimadzu LC-10AT VP HPLC pump (Shimadzu) at a flow rate of 0.2 and 0.3 mL/min to the ESI source in the positive and negative MRM modes, respectively. Optimized mobile phase consisted of 0.1% FA in water (solvent A) and methanol (solvent B). The optimized compound-dependent MS parameters and elution conditions for all subsequent MRM experiments were illustrated in Tables 1 and 2, respectively. MS data acquisition and processing were performed using the Analyst Software 1.4.2 (Applied Biosystems). The level of each metabolite was expressed as the peak area ratio between the integrated peak areas of the analyte and the IS. UPLC/MS/MS Method Validation. To evaluate the intra- and interday precision, standard solutions comprising of 10 µM each of the 21 metabolites in 1% FA in water kept in UPLC injection vials at 4 °C were injected for LC-MS/MS analysis at different time points: 0, 8, 16, 24, 48, and 72 h. 5682


Goh et al. The carry-over effect was investigated by comparing the levels of the 21 metabolites obtained from consecutive analyses of the standard solution at a high concentration (100 µM) and the blank sample. The carryover effect was calculated as a percentage of the peak area ratios of the 21 metabolites in the blank sample against that of the standard solution. Stability study was conducted to investigate the potential oxidative degradation of the 21 metabolites during sample preparation and to determine the optimum concentration of antioxidant (FA). Using standard solutions containing 10 µM each of the metabolites in replacement of the brain homogenate, different concentrations of FA (1, 5, and 10%) in water were experimented as the homogenizing and reconstituting solvents. A mixture containing 10 µM each of the 21 metabolites and 2 IS in cold water prepared prior to LC/MS/MS analysis was used as control. Peak area ratios of the metabolites in the samples were compared against those in the control and analyzed statistically using the paired t test (GraphPad Prism 4, San Diego, CA). To investigate the postpreparative stability profiles of the 21 metabolites in the rat brain homogenate, 2 µL of the 100 µL of supernatant in an injection vial kept at 4 °C was injected for LC-MS/MS analysis at various time points: 0, 4, 8, 12, 36, 48, 60, and 72 h. All studies were conducted in triplicates. The peak area ratios of the metabolites in the samples at different time points were compared against those at 0 h and analyzed statistically using the paired t test (GraphPad Prism 4, San Diego, CA). Metabolic Profiling Data Analysis. Multivariate analyses were performed using SIMCA-P version 11 software (Umetrics, Umeå, Sweden). Pattern recognition methods such as principal component analysis (PCA) and orthogonal partial least-squares discriminant analysis (OPLS-DA) were employed to analyze the pareto-scaled data according to the four separate classes, namely rats that underwent DON treatment and NOR test, DON treatment and AWM training test, GAL treatment and NOR test, and GAL treatment and AWM training test, respectively. PCA was used to uncover the pharmacological effects induced by the drug therapies and to identify outliers which were eliminated from subsequent analyses. OPLS-DA was employed to visualize the optimum separation between the predefined treatment and control classes. The loadings plot and variable importance plot (VIP) of the OPLS-DA model were used to reflect the importance of metabolites contributing to the differential classifications of the rats into the treatment and control groups. Metabolites with VIP values higher than 1 were most relevant for explaining the class distinction and were extracted from the VIP of the four respective OPLS-DA models. Based on the prominent biomarkers identified from the multivariate analyses, univariate independent t tests (SPSS 16.0 for windows, SPSS Inc.) were performed to compare the means of the biomarker levels between the treatment and control groups with regards to the four respective classes. The biomarkers that were consistently elevated across the four classes as indicated by both univariate and multivariate analyses, were further selected to generate OPLS-DA model describing metabolites in rats treated with DON, GAL, and the vehicle. Receiving Operating Characteristic (ROC) plot was used to evaluate the robustness of the OPLS-DA model obtained under the stepwise method. Partial least-squares discriminant analysis (PLSDA) model was further generated and subjected to permutation validation to check the validity of the model. Univariate one-way ANOVA followed by Fisher’s LSD posthoc

technical notes


Table 1. Optimized (i) Source-Dependent Parameters and (ii) Compound-Dependent Parameters and Retention Times for the 21 Neurologically Related Metabolites, APAP, and Salicylic Acid (Internal Standards, IS) (i) Source-Dependent Parameters parameter

value

Curtain gas, psi IonSpray voltage, V Temperature, °C GS 1, psi GS 2, psi

15 5000 (Positive MRM mode); -4500 (Negative MRM mode) 600 40 40 (ii) Compound Dependent Parameters

no.

names

m/z transition

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

HIST GLY CHO GLN GABA GLU NE ALA ACh E DA TYR OMD 5-HT PHE APAP (IS) TRP 5-HIAA

112.1 > 95.0 76.0 > 76.0 103.9 > 59.5 146.9 > 84.2 104.1 > 69.0 148.3 > 84.0 170.1 > 151.9 90.1 > 44.1 146.0 > 87.0 184.2 > 166.2 154.1 > 91.1 182.0 > 136.1 168.2 > 119.2 177.0 > 160.1 166.0 > 120.0 152.0 > 110.0 205.0 > 188.1 192.2 > 146.3

19 20 21 22 23

MYO ASP SA DOPAC SCA (IS)

179.1 > 87.0 132.1 > 88.0 117.0 > 73.1 167.2 > 122.9 137.0 > 93.2

DP (V)

Negative MRM mode -43.5 -4.0 -28.0 -2.9 -20.1 -8.0 -19.0 -6.0 -29.0 -8.1

Table 2. Optimized Elution Conditions for the 21 Neurologically-Related Metabolites, APAP, and Salicylic Acid (Internal Standards, IS) in the (i) Positive and (ii) Negative MRM Modes time (min)

flow rate (mL/min)

%A

EP (V)

Positive MRM mode 27.5 7.0 20.0 6.3 36.5 7.0 28.8 2.9 15.0 3.4 27.0 4.3 12.1 4.9 32.8 3.2 27.0 3.4 17.0 4.0 17.9 6.5 25.0 6.0 15.5 5.9 15.0 6.2 20.0 2.0 31.5 7.5 31.0 4.1 37.5 4.5

%B

curvea

0.00 3.25 5.25 7.00 7.01 10.10 10.15 11.00

(i) Positive MRM mode 0.200 95.0 0.200 95.0 0.200 75.0 0.200 30.0 0.200 5.0 0.200 5.0 0.200 95.0 0.200 95.0

5.0 25.0 25.0 70.0 95.0 95.0 5.0 5.0

6 6 3 5 6 6 6 6

0.00 3.25 6.50 6.51 10.10 10.15 11.00

(ii) Negative MRM mode 0.300 95.0 0.300 95.0 0.300 10.0 0.300 5.0 0.300 5.0 0.300 95.0 0.300 95.0

5.0 5.0 90.0 95.0 95.0 5.0 5.0

6 6 6 6 6 6 6

a Different curves of the UPLC gradient were adopted to optimize the chromatographic resolution of the metabolites.

comparison analysis (SPSS software) was subsequently used to confirm the observed trend.

CE (V)

20.0 5.0 24.1 24.0 21.5 21.1 9.5 19.0 20.0 14.0 32.2 18.0 24.7 15.3 17.0 23.0 14.5 19.0 -24.0 -18.0 -16.8 -15.0 -22.0

CXP (V)

R.T. (min)

2.5 3.0 5.2 3.4 2.5 2.0 2.0 5.5 3.3 2.2 2.5 2.3 2.9 2.2 1.7 2.5 2.2 2.1

1.13 1.30 1.30 1.31 1.33 1.35 1.39 1.41 1.61 1.60,1.88 2.08 2.77 3.36 3.91 5.51 5.93 6.43 6.84

-1.0 -1.6 -1.0 -2.3 -1.4

0.91 0.92 1.84 5.32 6.68

Results UPLC-MS/MS Method Validation. Intra- and interday reproducibility of the method was summarized in Table 3. Precision data as judged from the 15% RSD was satisfactory. The calculated carryover effects by comparing the peak area ratios of the 21 metabolites and 2 IS in 100 µM standard solution were less than 0.01 and 0.1% for the positive and negative MRM modes, respectively. Although the carryover effect was marginal and acceptable, blank solutions were injected in between every 3 biological sample analyses to further reduce carryover effects between samples. From the oxidation stability study, 1% FA was determined to be the most optimum concentration of antioxidant as it yielded the highest peak area ratios which were not significantly different (P > 0.05) from that of the control. Hence, 1% FA in water was employed as the homogenizing and reconstituting solvent in all metabolic profiling experiments. The postpreparative stability study demonstrated that the metabolites were stable for up to 36 h when kept in dark condition at 4 °C. Differences in the peak area ratios of the 9 representative metabolites extracted at 0 and 36 h were not statistically significant (P > 0.05). Therefore, the total duration of the sample preparation and analysis of each batch of samples was kept within 36 h at 4 °C to obtain optimum results for our study. Journal of Proteome Research • Vol. 8, No. 12, 2009 5683

technical notes

Goh et al.

Table 3. Intra- and Interday Precision of the Peak Area Ratios of the Nine Representative Neurologically Related Metabolites in Standard Solutions neurologically related metabolites

CHO GABA NE ACh DA 5-HT MYO SA DOPAC

parametersa

S.D. C.V. S.D. C.V. S.D. C.V. S.D. C.V. S.D. C.V. S.D. C.V. S.D. C.V. S.D. C.V. S.D. C.V.

(%) (%) (%) (%) (%) (%) (%) (%) (%)

intraday

interday

0.063 7.188 0.011 4.600 0.026 5.447 0.379 7.236 0.062 4.459 0.151 6.938 0.007 4.497 0.003 2.708 0.007 4.497

0.092 10.234 0.028 10.253 0.049 7.356 0.406 7.704 0.127 8.567 0.176 8.358 0.014 9.166 0.005 4.299 0.015 9.707

a Standard deviation (S.D.) is the square root of the variance while coefficient of variation (C.V.) is defined as its standard deviation divided by its mean.

Animal StudysAWM Training. The effects of DON and GAL on AWM test performance in LH rats are shown in Figure 2A-H. For the groups of rats treated with 0.5 mg/kg of DON and 1.0 mg/kg of GAL respectively, the average latency and path length required to locate the hidden watermaze platform by the fourth day of the spatial cue (SC) training were significantly different from the vehicle groups (P < 0.05) (Figure 2A,B and E,F). For the rats treated with 1.0 mg/kg of DON, differences in the average latency required to find the hidden platform by the fourth day was also significantly different from the vehicle group (Figure 2B). In terms of swim speed in the SC training, all the test groups showed comparable values throughout the study (Figure 2C and G). In contrast, the percentage time spent in the target and opposite quadrants in the probe trial were significantly different in all test groups (P < 0.05) (Figure 2D and H). With regards to the latency and path length measurements in the SC trainings, AWM demonstrated that rats treated with 0.5 mg/kg of DON and 1.0 mg/kg of GAL yielded prominent behavioral differences from the control groups. Our results also demonstrated that these two doses of DON and GAL yielded comparable latency, path length, and swim speed values (Figure 2). Animal StudysNOR Test. The effects of DON and GAL on NOR test performance in LH rats were shown in Figure 3A-F, respectively. The exploration time of the T1 trial was similar across treatment and control groups (Figure 3A and D). For the groups of rats treated with 0.5 mg/kg of DON and 0.5 mg/ kg of GAL, their T2 trial novel object exploration time (P < 0.01) and T2 novel and familiar objects discrimination index (P < 0.001) were significantly different from the respective vehicle groups (Figure 3B,C and E,F). For the groups of rats treated with 1.0 mg/kg of DON and 1.0 mg/kg of GAL (Figure 3B and E), the T2 trial novel object exploration time was also found to be significantly different from the vehicle groups (P < 0.05). The NOR test demonstrated that rats treated with 0.5 mg/kg of DON and 0.5 mg/kg of GAL, respectively yielded more prominent cognitive behavioral differences from the control groups. 5684


Metabolic Profiling Data Analysis. From the PCA scores plots of the four respective classes, 6 out of 60 rats were identified as outliers as they fell out of the Hotelling’s eclipse and/or exceeded the D-critical line in the distance to model X (DModX) plot. As such, data points of 2 rats were omitted from each group of rats that underwent DON treatment and NOR test, GAL treatment and NOR test, and GAL treatment and AWM, respectively. Clear clustering of treatment and control groups were observed by the PCA scores plots for all four respective classes. The OPLS-DA models of the four respective treatment classes demonstrated evident clusters of treatment and control groups (data not shown). Generally, all 21 neurologically related metabolites (Figure 1) were up-regulated across the four classes. VIP values greater than 1 were observed for ten metabolites namely, ACh, PHE, TYR, GLU, GABA, GLN, SA, TRP, 5-HT, and 5-HIAA. Univariate independent t tests indicated that eight out of these ten biomarkers (i.e., ACh, PHE, TYR, GABA, SA, TRP, 5-HT, and 5-HIAA) were consistently and significantly elevated (P < 0.10) across the four treatment classes (Table 4). Defined clusters of the 54 rats dosed with DON, GAL, and vehicle could be elucidated from the OPLS-DA model constructed using the eight biomarkers (Figure 4A). The R2X, R2Y and cumulative Q2 values of the OPLS-DA model were 0.822, 0.618, and 0.566 respectively, suggesting that the discriminate model fitted the data well. ROC was also used to evaluate the robustness of the OPLS-DA model (Figure 4B). The percentage sensitivity and specificity of the ROC curve were 100 and 93.75%, respectively. The AUC of the ROC curve was 1.00, suggesting that the discriminate model was robust. However, as the model statistics (Q2 and AUC) might be poor estimates of the robustness of the model where metabolite variables were preselected to discriminate between the groups, A PLSDA model was further generated to check the validity and degree of over fit for our model. Based on the plot of the correlation coefficient between the original Y and the permuted Y (100 permutations) versus the cumulative R2 and Q2, the intercepts [R2 (0.0362) and Q2 (-0.148) when correlation coefficient is zero] of the regression lines indicated that there was no overfitting of our data. Univariate one-way ANOVA confirmed that levels of these eight biomarkers were significantly different between the treatment and control groups (P < 0.05).

Discussion Alterations in multiple neurotransmitter systems contribute to the manifestation of cognitive behavioral disturbances in AD. The 21 metabolites chosen for metabolic profiling were mainly selected from the three principal pathways that affect AD patients, namely cholinergic, tyrosine, and glutamine systems. To provide a more comprehensive profiling of classical neurotransmitter systems, serotonergic, histaminergic pathways as well as glycine and myoinositol were also included. Metabolic pathways of multiple neurotransmitter systems were illustrated in literatures.24 Approximately 1100 peaks (21 metabolites × 54 rats, defined by a pair of m/z value and RT) were resolved using multivariate analysis. Both the PCA and OPLS-DA score plots suggested distinct differences between the control and treated rats. These results were valuable and important to define the pharmacological changes and uncover potential biomarkers. Our targeted LC-MS/MS strategy revealed consistent neurobiological trends of higher levels of metabolites in the rat brains upon treatment with DON and GAL. Eight biomarkers (ACh, PHE, TYR, GABA,


technical notes

Figure 2. Effects of DON (0.5 and 1.0 mg/kg) on AWM test performances namely, average (A) latency, (B) path length taken to find the hidden watermaze platform, (C) swim speed in the VC and SC trainings, and (D) percentage time spent by the rats in the target, adjacent left, adjacent right and opposite quadrants, respectively in the probe trial; Effects of GAL (0.5 and 1.0 mg/kg) on AWM test performances namely, average (E) latency, (F) path length taken to find the hidden watermaze platform, (G) swim speed in the VC and SC trainings and (H) percentage time spent by the rats in the target, adjacent left, adjacent right and opposite quadrants, respectively in the probe trial (VC ) visual cue; SC ) spatial cue; VEH ) vehicle). Data points represent mean ( SEM. Asterisks indicate significant difference from the vehicle group (* p < 0.05, **p < 0.01, ***p < 0.001). Journal of Proteome Research • Vol. 8, No. 12, 2009 5685

technical notes

Goh et al.

Figure 3. Effects of DON (0.5 and 1.0 mg/kg) on NOR test performances, namely average (A) NOR T1 trial objects exploration time, (B) NOR T2 trial novel and familiar objects exploration time and (C) NOR T2 novel and familiar objects discrimination (d2) index; Effects of GAL (0.5 and 1.0 mg/kg) on NOR test performances, namely average (D) NOR T1 trial objects exploration time, (E) NOR T2 trial novel and familiar objects exploration time, and (F) NOR T2 novel and familiar objects discrimination (d2) index; (G) Effects of 0.5 mg/kg DON and 0.5 mg/kg GAL on NOR test performances NOR T2 novel and familiar objects discrimination (d2) index. Data points represent mean ( SEM. Asterisks indicate significantly different from the vehicle group (* p < 0.05, **p < 0.01, ***p < 0.001).

SA, TRP, 5-HT, and 5-HIAA) were significantly and consistently elevated across the 4 treatment classes as demonstrated by both the multivariate and univariate analyses. These eight biomarkers are metabolites related to the cholinergic, dopaminergic, glutamatergic and serotonergic pathways, all of which play pivotal roles in the learning, memory, motivational and emotional processes.24 Our findings correlated well with the cogni5686


tive behavioral patterns yielded from the AWM training and NOR tests. WM tests are designed to assess spatial orientation, memory and are capable of revealing deficits in sensory, motor or motivational processes. NOR tests are designed to uncover the visual acuity, spontaneous behavior, spatial and temporal memory of the rats. Multitudinous pharmacological studies using diverse receptor agonists and antagonists suggested the

technical notes


Table 4. Summary of the Peak Area Ratios of Eight Prominent Biomarkers in 54 Male Lister Hooded Rats from the 4 Respective Treatment Classesa DON treatment and NOR test metabolites

ACh PHE TYR GABA SA TRP 5-HT 5-HIAA

test groups

T C T C T C T C T C T C T C T C

DON treatment and AWM test

N

mean

S.D.b

% fold change

N

mean

S.D.b

% fold change

9 4 9 4 9 4 9 4 9 4 9 4 9 4 9 4

48.522 33.718 368.72 301.9 68.855 50.981 51.17 43.806 67.683 51.435 54.354 45.249 2.779 2 0.46 0.373

6.33 3.2615 25.85 14.865 7.007 4.2115 2.99 3.0715 9.255 9.3375 5.414 4.0715 0.376 0.2945 0.034 0.0265

***143

10 5 10 5 10 5 10 5 10 5 10 5 10

43.648 30.236 345.21 288.23 67.905 55.754 51.861 51.927 59.631 50.923 54.381 42.829 2.572 2.114 0.474 0.406

2.929 4.759 26.09 25.63 8.04 5.111 3.372 5.234 10.87 5.58 5.268 3.605 0.331 0.22 0.042 0.031

***144

***122 ***135 **117 **132 **120 **139 ***123

10 5

GAL treatment and NOR test metabolites

ACh PHE TYR GABA SA Trp 5-HT 5-HIAA a

test groups

T C T C T C T C T C T C T C T C

mean

S.D.b

% fold change

N

mean

S.D.b

10 3 10 3 10 3 10 3 10 3 10 3 10 3 10 3

37.325 27.61 364.84 321.62 66.549 56.281 57.094 49.308 66.942 65.635 53.833 45.314 2.501 1.907 0.457 0.411

1.68 4.258 9.905 7.413 2.724 1.9914 2.786 1.0654 2.933 5.4884 2.498 2.9934 0.13 0.1614 0.022 0.0184

*135

9 4 9 4 9 4 9 4 9 4 9 4 9 4 9 4

43.37 32.278 3.827 2.967 80.446 57.892 58.899 53.029 61.166 57.037 58.951 50.549 2.649 2.507 0.503 0.443

2.55 2.377 9.116 4.505 3.585 4.065 1.966 1.62 1.712 2.742 1.974 3.074 0.082 0.101 0.019 0.028

roles of cholinergic,25-31 dopaminergic,32-37 glutamatergic,38-42 and serotonergic43-48 pathways in the NOR study, as well as in the WM test.13 As expected, our targeted metabolic profiling, AWM and NOR experiments were found to be complementary and corroborated with earlier research findings. Simultaneous profiling of the 21 neurologically related metabolites in our study confirmed that multiple classical neurotransmitter systems were involved in cognitive enhancement. The behavioral patterns illustrated through the animal tasks demonstrated that DON and GAL are efficacious cognition enhancers. The pro-cognitive effects of DON and GAL were reflected in the two preclinical cognitive assays, supporting their role in learning and memory. Our findings concurred with previous preclinical studies investigating these drugs for cholinergic markers and memory performance.49-51 Though both pharmacological therapies investigated in this study are acetylcholinesterase (AChE) inhibitors, our metabolic profiling results suggested that other neurotransmitter systems were found to be affected. This was possibly due to extensive interactions among the classical transmitter pathways52,53 and therefore, AChE inhibitors might indirectly affect other neurotransmitters systems to improve cognition. Our results indicated that differences between the treatment and control

122 100 117 ***127 **122 **117

GAL treatment and AWM test

N

*113 118 116 102 119 *131 111

T, treatment; C, control; p < 0.10, underlined; *p < 0.05; **p < 0.01; ***p < 0.001.

***120

b

% fold change

*134 ***129 **139 111 107 *117 106 114

Standard deviation (S.D.) is the square root of the variance.

groups for most of the eight biomarkers were more significant for the DON-treated rats than the GAL-treated rats. These data were consistent with in vitro biochemical tests and preclinical studies where DON was found to be 40- to 500- fold more potent than GAL in inhibiting AChE and 3-15 times higher doses of GAL were necessary to elicit similar degree of brain AChE inhibition.54 Comparative clinical trials of DON and GAL therapies also suggested that DON was relatively more efficacious55-58 than GAL, though controversies existed in other studies which demonstrated both drugs as being equally efficacious.59-61 Therefore, long-term prospective clinical studies would be required to obtain a more substantial comparison between DON and GAL with regards to their clinical efficacy. More importantly, our data demonstrated that metabolic profiling has the potential to distinguish the differential pharmacological responses resulted by different drug treatments. Such a potential of metabolic profiling could be expanded in the future to characterize differential pharmacological activities of test compounds in drug discovery. Our two-pronged approach of employing cognitive behavioral tools and metabolic profiling in this study facilitated our understanding of the pharmacological mechanisms involved and contributed to the development of more substantial Journal of Proteome Research • Vol. 8, No. 12, 2009 5687

technical notes

Figure 4. (A) Biomarker-based OPLS-DA model of 54 male Lister Hooded rats treated with DON (0.5 mg/kg for AWM and NOR tests), GAL (1.0 mg/kg for AWM test and 0.5 mg/kg NOR test), and the vehicle. (LV ) 1 + 1, R2X ) 0.822, R2Y ) 0.618 and Q2 (cum) ) 0.566) (T, treatment; C, Control). (B) ROC plot of the OPLS-DA model using 7-fold cross-validated Y predicted values.

cognition enhancement predictive models. Metabolic profiling provides both an objective and quantitative means to validate the cognitive behavioral outcomes yielded from the animal tasks. Both approaches generated consistently positive results for DON and GAL, which are drugs commonly utilized in the clinical settings for treatment of AD. This further substantiated the predictions obtained from both assays. Although only AChE inhibitors were investigated, optimistic outcomes from the current study suggested that these assays could be potentially explored to screen for pharmacological phenotypes induced by novel cognition enhancers targeting other neurotransmitter pathways. In addition to the preclinical studies, our developed UPLC-MS/MS method could also be employed in in vitro brain tissue studies to screen for fluctuations in neurotransmitter levels when brain receptors are challenged with various agonists and antagonists, contributing to lead discovery and optimization. Similar methods could also be utilized in the examination of postmortem brains from AD patients to gain a better understanding of the molecular mechanisms underlying the pathophysiology. In conclusion, our study demonstrated that pharmacologically induced changes in the cognitive behavioral patterns established through animal tasks were associated with elevations of neurologically related metabolites from multiple neurotransmitter pathways. The results of the present study suggested that both NOR and AWM tests are effective cognitive 5688


Goh et al. behavioral tests and UPLC-MS/MS metabolic profiling is a potential complementary molecular cognitive screening tool. Abbreviations: ACh, acetylcholine; AChE, acetylcholinesterase; AD, Alzheimer’s disease; ALA, L-alanine; ANOVA, analysis of variance; ASP, L-aspartic acid; APAP, acetominophen; AWM, Atlantis watermaze; CAD, collision-activated dissociation; CE, collision energy; CHO, choline; CXP, collision cell exit potential; DA, dopamine; DModX, distance to model; DON, donepezil; DOPAC, 3,4-dihydroxyphenylacetic acid; DP, declustering potential; E, epinephrine (3,4-dihydroxya-(methylaminomethyl)benzyl alcohol); EP, entrance potenetial; ESI, electrospray ionization; ESI+ve, electrospray ionization positive mode; ESIve, electrospray ionization negative mode; FA, formic acid; GABA, γ-aminobutyric acid; GAL, galantamine; GLN, Lglutamine; GLU, glutamic acid; GLY, L-glycine; GSF, granulocytestimulating factors; HIST, histamine; IS, internal standard; IPA, isopropanol; LH, Lister hooded; MeOH, methanol; MRM, multiple reaction monitoring; MRI, magnetic resonance imaging; NACLAR, National Advisory Committee for Laboratory Animal Research; NE, norepinephrine; NOR, novel object recognition; OPLS-DA, orthogonal partial least-squares discriminant analysis; PCA, principal component analysis; PHE, L-phenylalanine; ROC, receiving operating characteristic; RSD, relative standard deviation; SA, succinic acid; SCA, salicylic acid; SC, spatial cue; S/N, signal-to-noise; TRP, L-tryptophan; TYR, tyrosine; UPLC-MS/MS, ultra performance liquid chromatography tandem mass spectrometry; VC, visual cue; VIP, variable importance plot; 5-HIAA, 5-hydroxyindoleacetic acid; 5-HT, serotonin.

Acknowledgment. This work was supported by the GSK R&D centre, NUS grants (R-148-000-100-112 and R-279-000-249-646 to E.C.Y.C.), and NUS Department of Pharmacy FYP grant (R-148-000-003-001 to E.C.Y.C.). D.P.Q.G., L.S.N., and E.C.Y.C. contributed to the metabolic profiling experiments and chemometric data analysis; C.W.G., C.C.A., A.H.N., and E.R.B. contributed to the pharmacology study; whereas W.S.C. and Z.A. contributed to the preclinical cognition study. References (1) Weiner, H. L.; Frenkel, D. Immunology and immunotherapy of Alzheimer’s disease. Nat. Rev. Immunol. 2006, 6 (5), 404–416. (2) Blasko, I.; Grubeck-Loebenstein, B. Role of the immune system in the pathogenesis, prevention and treatment of Alzheimer’s disease. Drugs Aging 2003, 20 (2), 101–113. (3) Tsai, K. J.; Tsai, Y. C.; Shen, C. K. J. G-CSF rescues the memory impairment of animal models of Alzheimer’s disease. J. Exp. Med. 2007, 204 (6), 1273–1280. (4) Baum, L.; Ng, A. Curcumin interaction with copper and iron suggests one possible mechanism of action in Alzheimer’s disease animal models. J. Alzheimer’s Dis. 2004, 6 (4), 367–377. (5) Sarter, M. Preclinical research into cognition enhancers. Trends Pharmacol. Sci. 2006, 27 (11), 602–608. (6) Sarter, M. Animal cognition: Defining the issues. Neurosci. Biobehav. Rev. 2004, 28 (7), 645–650. (7) Sarter, M. Taking stock of cognition enhancers. Trends Pharmacol. Sci. 1991, 12 (12), 456–461. (8) Decker, M. W. Animal models of cognitive function. Crit. Rev. Neurobiol. 1995, 9 (4), 321–343. (9) Sarasa, M. Experimental models for Alzheimer’s disease research. Rev. Neurol. 2006, 42 (5), 297–301. (10) Pepeu, G. Overview and perspective on the therapy of Alzheimer’s disease from a preclinical viewpoint. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 2001, 25 (1), 193–209. (11) Myhrer, T. Neurotransmitter systems involved in learning and memory in the rat: a meta-analysis based on studies of four behavioral tasks. Brain Res. Rev. 2003, 41 (2-3), 268–287.

Strategies in Cognition Enhancement Research (12) Woodruff-Pak, D. S. Preclinical experiments on cognition enhancement in Alzheimer’s disease: Drugs affecting nicotinic acetylcholine receptors. Drug Dev. Res. 2002, 56 (3), 335–346. (13) Trinchese, F.; Liu, S.; Ninan, I.; Puzzo, D.; Jacob, J. P.; Arancio, O. Cell cultures from animal models of Alzheimer’s disease as a tool for faster screening and testing of drug efficacy. J. Mol. Neurosci. 2004, 24 (1), 15–21. (14) Sowell, R. A.; Owen, J. B.; Allan Butterfield, D. Proteomics in animal models of Alzheimer’s and Parkinson’s diseases. Ageing Res. Rev. 2009, 8 (1), 1–17. (15) Tareke, E.; Bowyer, J. F.; Doerge, D. R. Quantification of rat brain neurotransmitters and metabolites using liquid chromatography/ electrospray tandem mass spectrometry and comparison with liquid chromatography/electrochemical detection. Rapid Commun. Mass Spectrom. 2007, 21 (23), 3898–3904. (16) Falangola, M. F.; Dyakin, V. V.; Lee, S. P.; Bogart, A.; Babb, J. S.; Duff, K.; Nixon, R.; Helpern, J. A. Quantitative MRI reveals agingassociated T2 changes in mouse models of Alzheimer’s disease. NMR Biomed. 2007, 20 (3), 343–351. (17) Sykova, E.; Vorisek, I.; Antonova, T.; Mazel, T.; Meyer-Luehmann, M.; Jucker, M.; Hajek, M.; Or, M.; Bures, J. Changes in extracellular space size and geometry in APP23 transgenic mice: A model of Alzheimer’s disease. Proc. Natl. Acad. Sci. U.S.A. 2005, 102 (2), 479– 484. (18) Lanari, A.; Amenta, F.; Silvestrelli, G.; Tomassoni, D.; Parnetti, L. Neurotransmitter deficits in behavioural and psychological symptoms of Alzheimer’s disease. Mech. Ageing Dev. 2006, 127 (2), 158– 165. (19) Garcia-Alloza, M.; Gil-Bea, F. J.; Diez-Ariza, M.; Chen, C. P. L. H.; Francis, P. T.; Lasheras, B.; Ramirez, M. J. Cholinergic-serotonergic imbalance contributes to cognitive and behavioral symptoms in Alzheimer’s disease. Neuropsychologia 2005, 43 (3), 442–449. (20) Schweizer, T.; Birthelmer, A.; Lazaris, A.; Cassel, J. C.; Jackisch, R. 3,4-DAP-evoked transmitter release in hippocampal slices of aged rats with impaired memory. Brain Res. Bull. 2003, 62 (2), 129– 136. (21) Kalkman, H. O.; Subramanian, N.; Hoyer, D. Extended radioligand binding profile of iloperidone: A broad spectrum dopamine/ serotonin/norepinephrine receptor antagonist for the management of psychotic disorders. Neuropsychopharmacology 2001, 25 (6), 904–914. (22) Atcha, Z.; Chen, W. S.; Ong, A. B.; Wong, F. K.; Neo, A.; Browne, E. R.; Witherington, J.; Pemberton, D. J. Cognitive enhancing effects of ghrelin receptor agonists. Psychopharmacology 2009; DOI: 10.1007/s00213-009-1620-6. (23) Fone, K. C. F. An update on the role of the 5-hydroxytryptamine receptor in cognitive function. Neuropharmacology 2008, 55 (6), 1015–1022. (24) Siegel, G. J. Basic neurochemistry: molecular, cellular, and medical aspects, 7th ed.; Elsevier: Amsterdam; Boston, 2006. (25) Rutten, K.; Prickaerts, J.; Blokland, A. Rolipram reverses scopolamine-induced and time-dependent memory deficits in object recognition by different mechanisms of action. Neurobiol. Learn. Mem. 2006, 85 (2), 132–138. (26) Abe, H.; Ishida, Y.; Iwasaki, T. Perirhinal N-methyl-D-aspartate and muscarinic systems participate in object recognition in rats. Neurosci. Lett. 2004, 356 (3), 191–194. (27) Obinu, M. C.; Reibaud, M.; Miquet, J. M.; Pasquet, M.; Rooney, T. Brain-selective stimulation of nicotinic receptors by TC-1734 enhances ACh transmission from frontoparietal cortex and memory in rodents. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 2002, 26 (5), 913–918. (28) Chopin, P.; Colpaert, F. C.; Marien, M. Effects of acute and subchronic administration of dexefaroxan, an alpha(2)-adrenoceptor antagonist, on memory performance in young adult and aged rodents. J. Pharmacol. Exp. Ther. 2002, 301 (1), 187–196. (29) Dodart, J. C.; Mathis, C.; Ungerer, A. Scopolamine-induced deficits in a two-trial object recognition task in mice. NeuroReport 1997, 8 (5), 1173–1178. (30) Bartolini, L.; Casamenti, F.; Pepeu, G. Aniracetam restores object recognition impaired by age, scopolamine, and nucleus basalis lesions. Pharmacol., Biochem. Behav. 1996, 53 (2), 277–283. (31) Ennaceur, A.; Meliani, K. A new one-trial test for neurobiological studies of memory in rats. III. Spatial vs. non-spatial working memory. Behav. Brain Res. 1992, 51 (1), 83–92. (32) Kamei, H.; Nagai, T.; Nakano, H.; Togan, Y.; Takayanagi, M.; Takahashi, K.; Kobayashi, K.; Yoshida, S.; Maeda, K.; Takuma, K.; Nabeshima, T.; Yamada, K. Repeated methamphetamine treatment impairs recognition memory through a failure of novelty-induced ERK1/2 activation in the prefrontal cortex of mice. Biol. Psychiatry 2006, 59 (1), 75–84.

technical notes (33) Hotte, M.; Thuault, S.; Lachaise, F.; Dineley, K. T.; Hemmings, H. C.; Nairn, A. C.; Jay, T. M. D1 receptor modulation of memory retrieval performance is associated with changes in pCREB and pDARPP32 in rat prefrontal cortex. Behav. Brain Res. 2006, 171 (1), 127– 133. (34) Chuhan, Y. S.; Taukulis, H. K. Impairment of single-trial memory formation by oral methylphenidate in the rat. Neurobiol. Learn. Mem. 2006, 85 (2), 125–131. (35) Hotte, M.; Naudon, L.; Jay, T. M. Modulation of recognition and temporal order memory retrieval by dopamine D1 receptor in rats. Neurobiol. Learn. Mem. 2005, 84 (2), 85–92. (36) Ventura, R.; Pascucci, T.; Catania, M. V.; Musumeci, S. A.; PuglisiAllegra, S. Object recognition impairment in Fmr1 knockout mice is reversed by amphetamine: Involvement of dopamine in the medial prefrontal cortex. Behav. Pharmacol. 2004, 15 (5-6), 433– 442. (37) Woolley, M. L.; Marsden, C. A.; Sleight, A. J.; Fone, K. C. F. Reversal of a cholinergic-induced deficit in a rodent model of recognition memory by the selective 5-HT6 receptor antagonist, Ro 04-6790. Psychopharmacology 2003, 170 (4), 358–367. (38) Winters, B. D.; Bussey, T. J. Glutamate receptors in perirhinal cortex mediate encoding, retrieval, and consolidation of object recognition memory. J. Neurosci. 2005, 25 (17), 4243–4251. (39) de Lima, M. N. M.; Laranja, D. C.; Bromberg, E.; Roesler, R.; Schroder, N. Pre- or post-training administration of the NMDA receptor blocker MK-801 impairs object recognition memory in rats. Behav. Brain Res. 2005, 156 (1), 139–143. (40) Sargolini, F.; Roullet, P.; Oliverio, A.; Mele, A. Effects of intraaccumbens focal administrations of glutamate antagonists on object recognition memory in mice. Behav. Brain Res. 2003, 138 (2), 153–163. (41) Baker, K. B.; Kim, J. J. Effects of stress and hippocampal NMDA receptor antagonism on recognition memory in rats. Learn. and Mem. 2002, 9 (2), 58–65. (42) Roullet, P.; Sargolini, F.; Oliverio, A.; Mele, A. NMDA and AMPA antagonist infusions into the ventral striatum impair different steps of spatial information processing in a nonassociative task in mice. J. Neurosci. 2001, 21 (6), 2143–2149. (43) Schiapparelli, L.; Simon, A. M.; Del Rio, J.; Frechilla, D. Opposing effects of AMPA and 5-HT1A receptor blockade on passive avoidance and object recognition performance: Correlation with AMPA receptor subunit expression in rat hippocampus. Neuropharmacology 2006, 50 (7), 897–907. (44) Pitsikas, N.; Tsitsirigou, S.; Zisopoulou, S.; Sakellaridis, N. The 5-HT1A receptor and recognition memory: Possible modulation of its behavioral effects by the nitrergic system. Behav. Brain Res. 2005, 159 (2), 287–293. (45) Pitsikas, N.; Sakellaridis, N. The 5-HT2C receptor antagonist RO 60-0491 counteracts rats’ retention deficits in a recognition memory task. Brain Res. 2005, 1054 (2), 200–202. (46) Pitsikas, N.; Rigamonti, A. E.; Cella, S. G.; Muller, E. E. The 5-HT1A receptor antagonist WAY 100635 improves rats performance in different models of amnesia evaluated by the object recognition task. Brain Res. 2003, 983 (1-2), 215–222. (47) Bonvento, G.; Scatton, B.; Claustre, Y.; Rouquier, L. Effect of local injection of 8-OH-DPAT into the dorsal or median raphe nuclei on extracellular levels of serotonin in serotonergic projection areas in the brain. Neurosci. Lett. 1992, 137 (1), 101–104. (48) Invernizzi, R.; Carli, M.; Di Clemente, A.; Samanin, R. Administration of 8-hydroxy-2-(di-n-propylamino)tetralin in raphe nuclei dorsalis and medianus reduces serotonin synthesis in the rat brain: Differences in potency and regional sensitivity. J. Neurochem. 1991, 56 (1), 243–247. (49) Hernandez, C. M.; Gearhart, D. A.; Parikh, V.; Hohnadel, E. J.; Davis, L. W.; Middlemore, M. L.; Warsi, S. P.; Waller, J. L.; Terry Jr, A. V. Comparison of galantamine and donepezil for effects on nerve growth factor, cholinergic markers, and memory performance in aged rats. J. Pharmacol. Exp. Ther. 2006, 316 (2), 679–694. (50) Dimitrova, D. S.; Getova-Spassova, D. P. Effects of galantamine and donepezil on active and passive avoidance tests in rats with induced hypoxia. J. Pharmacol. Sci. 2006, 101 (3), 199–204. (51) Barnes, C. A.; Meltzer, J.; Houston, F.; Orr, G.; McGann, K.; Wenk, G. L. Chronic treatment of old rats with donepezil or galantamine: Effects on memory, hippocampal plasticity and nicotinic receptors. Neuroscience 2000, 99 (1), 17–23. (52) Montgomery, A. M. J.; Grottick, A. J. Neurotransmitter system interactions revealed by drug-induced changes in motivated behavior. Pharmacol., Biochem. Behav. 1999, 62 (4), 643–657. (53) Nieoullon, A.; Kerkerian-Le Goff, L. Cellular interactions in the striatum involving neuronal systems using ‘classical’ neurotrans-

Journal of Proteome Research • Vol. 8, No. 12, 2009 5689

technical notes (54)

(55)

(56)

(57)

5690

mitters: Possible functional implications. Movement Disord. 1992, 7 (4), 311–325. Geerts, H.; Guillaumat, P. O.; Grantham, C.; Bode, W.; Anciaux, K.; Sachak, S. Brain levels and acetylcholinesterase inhibition with galantamine and donepezil in rats, mice, and rabbits. Brain Res. 2005, 1033 (2), 186–193. Jones, R. W.; Soininen, H.; Hager, K.; Aarsland, D.; Passmore, P.; Murthy, A.; Zhang, R.; Bahra, R. A multinational, randomised, 12week study comparing the effects of donepezil and galantamine in patients with mild to moderate Alzheimer’s disease. Int. J. Geriatr. Psychiatry 2004, 19 (1), 58–67. Aguglia, E.; Onor, M. L.; Saina, M.; Maso, E. An open-label, comparative study of rivastigmine, donepezil and galantamine in a real-world setting. Curr. Med. Res. Opin. 2004, 20 (11), 1747– 1752. Harry, R. D.; Zakzanis, K. K. A comparison of donepezil and galantamine in the treatment of cognitive symptoms of Alzheimer’s disease: A meta-analysis. Human Psychopharmacol. 2005, 20 (3), 183–187.


Goh et al. (58) Hansen, R. A.; Gartlehner, G.; Lohr, K. N.; Kaufer, D. I. Functional outcomes of drug treatment in Alzheimer’s disease - A systematic review and meta-analysis. Drugs Aging 2007, 24 (2), 155–167. (59) Hansen, R. A.; Gartlehner, G.; Webb, A. P.; Morgan, L. C.; Moore, C. G.; Jonas, D. E. Efficacy and safety of donepezil, galantamine, and rivastigmine for the treatment of Alzheimer’s disease: A systematic review and meta-analysis. Clin. Interventions Aging 2008, 3, 211–225. (60) Takeda, A.; Loveman, E.; Clegg, A.; Kirby, J.; Picot, J.; Payne, E.; Green, C. A systematic review of the clinical effectiveness of donepezil, rivastigmine and galantamine on cognition, quality of life and adverse events in Alzheimer’s disease. Int. J. Geriatr. Psychiatry 2006, 21 (1), 17–28. (61) Cumbo, E. Differential effects or rivastigmine, galantamine and donepezil on behavioral and psychological symptoms in patients with Alzheimer’s disease: 18-month, randomized, open-label trial. Primary Care Community Psychiatry 2005, 10 (3), 95–102.

PR900795G

Metabolic Profiling of Rat Brain and Cognitive Behavioral Tasks

Recommend Documents