Short Stories

13 pages

Environmental enrichment promotes improved spatial abilities and enhanced dendritic growth in the rat

of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Environmental enrichment promotes improved spatial abilities and enhanced dendritic growth in the rat
  Behavioural Brain Research 163 (2005) 78–90 Research report Environmental enrichment promotes improved spatial abilities andenhanced dendritic growth in the rat Maria Giuseppa Leggio a , b , ∗ , Laura Mandolesi b , c , Francesca Federico a , d , Francesca Spirito a , b ,Benedetta Ricci b , Francesca Gelfo a , Laura Petrosini a , b a  Department of Psychology, University of Rome “La Sapienza”, Via dei Marsi 78, 00185 Rome, Italy b  IRCCS Santa Lucia Foundation, Rome c University of Naples “Parthenope”, Italy d University of Siena, Italy Received 14 February 2005; received in revised form 12 April 2005; accepted 14 April 2005Available online 23 May 2005 Abstract An enriched environment consists of a combination of enhanced social relations, physical exercise and interactions with non-social stimulithat leads to behavioral and neuronal modifications. In the present study, we analyzed the behavioral effects of environmental complexity ondifferent facets of spatial function, and we assessed dendritic arborisation and spine density in a cortical area mainly involved in the spatiallearning, as the parietal cortex. Wistar rat pups (21 days old) were housed in enriched conditions (10 animals in a large cage with toys and arunning wheel), or standard condition (two animals in a standard cage, without objects). At the age of 3 months, both groups were tested in theradial maze task and Morris water maze (MWM). Morphological analyses on layer-III pyramidal neurons of parietal cortex were performedin selected animals belonging to both experimental groups. In the radial maze task, enriched animals exhibited high performance levels, byexploitingproceduralcompetenciesandworkingmemoryabilities.Furthermore,whentherequirementsofthecontextchanged,theypromptlyreorganized their strategies by shifting from prevalently using spatial procedures to applying mnesic competencies. In the Morris water maze,enriched animals more quickly acquired tuned navigational strategies. Environmental enrichment provoked increased dendritic arborisationas well as increased density of dendritic spines in layer-III parietal pyramidal neurons.© 2005 Elsevier B.V. All rights reserved. Keywords:  Development; Spatial procedural learning; Spatial working memory; Neuronal changes; Dendritic spines; Parietal cortex 1. Introduction An enriched environment consists of a combination of enhanced social relations, physical exercise and interactionswith non-social stimuli. Enriched conditions are known tohave a significant impact on memory and learning abili-ties [8,30,46] as well as on neurogenesis and synaptoge-nesis [41,74,84]. In the 1960s, Rosenzweig, Diamond and co-workers pioneered studies on the mechanisms by whichcomplex environment and experience influence brain func-tion, demonstrating significant changes in brain anatomy ∗ Corresponding author. Tel.: +39 0649917522; fax: +39 0649917522.  E-mail address: (M.G. Leggio). [19,20,45,46,65]. More recent studies have established thatthese structural and behavioral changes are associated withchanges in brain neurochemistry and physiology. Several as-pectsofhippocampalfunctioning,suchasLTP,neurogenesis,dendritic spine growth and neurotrophin mRNA expressionare reported to be enhanced by environmental enrichment[22,25,27,40,41,74,80]. Also, in the neocortical regions, en-richment induces clear neural alterations such as increase inthickness, dendritic branching, presynaptic vesicle numberand synaptophysin levels [18,28,33,51,78].However, in spite of the copious literature on environ-mental complexity effects, whether the beneficial effects of enrichment influence all aspects of learning, or are linkedto particular components of learning abilities has not been 0166-4328/$ – see front matter © 2005 Elsevier B.V. All rights reserved.doi:10.1016/j.bbr.2005.04.009   M.G. Leggio et al. / Behavioural Brain Research 163 (2005) 78–90  79 addressed. In adult rodents, environmental enrichment ini-tiated at almost any point in the lifespan seems to im-prove performances on all kinds of tasks, i.e. spatial ornon-spatial, associative or discriminative, mnesic or proce-dural [9,52,64,72,76,78]. This non-specific pattern of be- havioral improvement probably emerges because analysesfail to split complex functions into single components.Thus, it seemed interesting to analyze the effect of envi-ronmental enrichment on specific facets of cognitive perfor-mance.In memory and learning abilities, spatial functions are of fundamental importance to mobile organisms whose spatialmemories form and are organized in distinct but coordinatedframes, thus allowing well structured adaptive capacities.Sinceenriched-rearedanimalsdisplaysuperiorperformanceson some tasks tapping spatial functioning [28,73,75,80], itseems particularly important to identify the components of spatial behavior that are enhanced most by enriched rearing.To measure spatial working memory abilities in rats, themost suitable task appears to be the radial maze. In fact, theuse of different radial maze protocols allows distinguishingspatial working memory abilities from procedural strategies.Furthermore, Morris water maze (MWM) allows separatelyassessing (localizatory) declarative memory from (explo-rative) procedural strategies [7,49]. These different facets of spatial functioning are related to the activity of variousneuronal structures, namely, neocortical regions, such asprefrontal [16,26,62,81] and parietal [4,12,21,43,61,68] cor- tices, hippocampal circuits [10,48,49,54,55,58,67] and sub- cortical areas, such as basal ganglia [56,83] and cerebellum [15,29,59,66]. Despite clear instances where these regionscan operate independently of each other, other data indicatethat they can interact during new learning and retention of spatial information [42]. Thus, more than one structure may beinvolvedinprocessingsingleattributesofspatialfunction,suggesting the possibility for both parallel and sequentialprocessing. Taking into account the marked influence of environmental enrichment on the neocortical structures[33,35,74] and the heavy involvement of associative parietalareas in different spatial aspects, it seemed particularly in-teresting to assess dendritic arborisation and spine density inthe parietal cortex. Earlier studies have proven that dendriticalterations are a sensitive index of experience-dependentmodifications of brain structures [44,74]. Because thedendritic surface receives more than 95% of the synapseson a neuron, changes in synapse activity can be inferredfrom analysis of dendritic arborisation and spine density[35].In conclusion, our specific aim was to compare the effectsof enriched versus standard rearing conditions on variouscomponents of spatial functioning and on the morphologyof the parietal neurons. Differences between groups wouldindicate which components of spatial learning are primarilyaffected by developmental manipulations, and provide a use-fulbehavioralandanatomicalmodeltostudyplasticchangeswithin normal or injured brain. 2. Materials and methods 2.1. Animals Initially, 40 Wistar rats were used in the experiment. Useful datawerecollectedon36rats;theremainingfouranimalswerediscardedbecause their behavior froze in the very first sessions of the radialmaze task. On the 21st postnatal day, an even number of littermatesof the same dam were randomly assigned to one of the two ex-perimental groups. The first group (  N  =17) was reared in enrichedconditions (EC), and the other (  N  =19) in standard conditions (SC).Animals were kept in the respective conditions until behavioral ex-periments were carried out 2.5/3 months later and also throughoutthe whole behavioral testing. Both groups of animals received thesame type of food. Food and water were provided ad libitum untilbehavioral testing. 2.2. Rearing conditions The EC group rats were housed in groups of 10 in a large cage(100cm × 50cm × 70cm), with one extra level constructed of gal-vanized wire mesh and connected by ramps of the same materialto create two interconnected levels. The cage contained wood shav-ings, a running wheel, a shelter (a house-shaped toy with a concaveopening), plastic coloured toys and small constructions. Through-out the enrichment period, the shelter and running wheel were keptin the cage, while the toys and constructions were changed once aweek. Also once a week, the feeding boxes and water bottles weremoved to different cage points to encourage foraging and explo-rative behaviors. Furthermore, each enriched animal was handleddaily for at least 10min and allowed to forage in a large space, i.e.the whole lab room.The SC group rats were housed by twos in a standard cage(40cm × 26cm × 18cm) containing wood shavings but no objects.Feeding boxes and water bottles were kept in the same positionthroughout the animals’ life. These animals received the usual careby the stabulary staff, even if without particular and prolonged ma-nipulation. This procedure did not result in an impoverished rearingandalsostandardanimalswereaccustomedtothe“human”contact. 2.3. Radial maze The eight-arm radial maze consisted of a round central platform(30cm in diameter) from which eight arms (12.5cm wide × 60cmlong) radiated like the spokes of a wheel. To avoid accidental falls,each arm was bordered by a clear rail (3cm high). At the end of each arm, there was a food well (5cm diameter). The entire mazewas white and stood 50cm off the floor. It was located in a well-litroom with one door and was surrounded by some extra-maze cuesheld in constant spatial relations throughout the experiment. 2.3.1. Behavioral procedure Followingtheperiodofexposuretotheirrespectivehousingcon-ditions, both groups of animals were tested in the radial maze. Priorto the habituation phase, rats were food-restricted to decrease theirweight 20–25%. However, housing conditions were not modified.In this phase, both groups of rats were gently handled for 10min aday to habituate them to the researchers and experimental setting.Followingtheinitialfoodrestrictionandhandling,a2-dayhabit-uationphasestarted.Theratwasplacedonthecentralmazeplatform  80  M.G. Leggio et al. / Behavioural Brain Research 163 (2005) 78–90 andallowedtoexplorethemazefreelyfor10minandtoeatthefeeduniformly scattered throughout the maze. According to the proce-dure chosen, the arms were totally or partially baited with a piece of purinachow.Theratwasplacedonthecentralplatformandallowedto make the runs or to explore the maze for 15min. 2.3.2. Behavioral testingFull-baited maze procedure . All maze arms were baited with apieceofpurinachowateachsession.Afterthehabituationphase,therat was placed on the central platform and allowed to make 16 runsortoexplorethemazefor15min.Theanimalmadeanerrorwhenitcompletelyenteredanarmithadalreadyvisitedinthesamesession.Theanimalsunderwenttwosessionsadayforfiveconsecutivedays.The inter-session interval was at least 3h. Forced-choice procedure.  On the first day of testing, fourarms(forexample,arms1,3,4and7)werebaitedandtheremainingarms were closed by a little door at their proximal end. The baitedarms were separated by different angles to prevent the animal fromreaching the solution by adopting a stereotyped pattern. The rat wasplaced on the central platform and allowed to explore all the openarms. Then, it spent 60s in its cage before it was put back in themaze. In the second phase, free access to all arms was allowed, butonlythefourpreviouslyclosedarmswerebaited.Thisparadigmwasrepeated for five consecutive days with a different configuration of arms closed each day. In this way, no fixed search pattern could beused. 2.3.3. Behavioral parameters Parameters used to assess the animals’ performances in the full-baitedmazeprocedurepermittedemphasizingdifferenttaskaspects,that were more closely, but not exclusively, linked to motor, spatial,or procedural components.The  motor   parameters were: number of arms visited in the timeallowed (15min), either right or wrong; total time needed to visiteight arms, either right or wrong.  Spatial  parameters were: percent-age of total errors, calculated as the percentage of wrongly visitedarms divided by the number of entries; first error, calculated as thenumber of arms visited including the first wrong visit; spatial span,calculated as the longest sequence of correctly visited arms.  Proce-dural  parameters measuring the structure of choice behavior were:45 ◦ angles, calculated as the percentage of that angle that the ratsmade in each session divided by the number of angles made.Parameters taken into account in analysing the performances inthe forced-choice procedure were the following: number of armsvisited; working memory errors, considered as entries into unbaitedarms. This parameter was further broken down into two error sub-types: across-phase errors, defined as entries into an arm that hadbeen entered during the first phase; within-phase errors, defined asre-entries into an arm visited earlier in the same session. 2.4. Morris water maze The rats in both experimental groups were placed in a circular,plastic pool (diameter 140cm) with white inside walls, located in anormally equipped laboratory room and uniformly lit by four neonlamps (40W each) suspended 3m from the ceiling. No effort wasmadetoenhance(or,viceversa,todiminish)extra-mazecues,whichwere held in constant spatial relations throughout the experiments.Thepoolwasfilledwithwater(24 ◦ C),50cmdeep,madeopaquebythe addition of 2l of milk. A white steel escape platform (10cm indiameter) was placed in the middle of one cardinal quadrant (north-west, northeast, southwest, southeast), 30cm from the side walls; itwas either submerged 2cm below or elevated 2cm above the waterlevel. Each rat was gently released into the water from pseudoran-domly varied starting points, so that it faced the center of the pool.The rat was allowed to swim around to find the platform. Blocks of four trials were presented to each rat, two blocks of trials per day.On reaching the platform, each rat was allowed to remain there for30s before it was placed in the water again for the next trial. If arat failed to locate the platform within 120s, it was guided there bythe experimenter and allowed to stay there for 30s. In the first foursessions (Trials 1–16), the platform was hidden in the northwestpool quadrant (Place 1); in the next two sessions (Trials 17–24), theplatform was kept visible in the northeast quadrant (cue phase); inthe final four sessions (Trials 25–40), the platform was hidden inthe northeast quadrant (Place 2) [49]. The rats’ trajectories in thepoolweremonitoredbyavideocameramountedontheceiling.Theresulting video signal was relayed to a monitor, allowing both on-andoff-lineanalyses,andtoanimageanalyzer(Ethovision,Noldus,Wageningen, The Netherlands). The  x  and  y  coordinates of the rat’sposition were sampled and stored on disk. 2.4.1. Behavioral parameters Parameters taken into account in analysing the MWM perfor-mances were successes in finding the platform, latencies to findthe platform and swimming trajectories. Spatial and temporal dis-tribution of swimming trajectories, path length, swimming speed,percentage of time spent in inner or outer annuli, headings (devi-ation between the rat’s actual direction when leaving the edge of the tank and a straight line from the start location to the tank pointcontaining the platform) were considered to divide exploration be-havior into five main categories: circling (peripheral swimming atthe tank wall); extended searching (swimming in all pool quad-rants, visiting the same areas more than once); restricted searching(swimming in some pool quadrants but not visiting other tank areasat all); restricted circling (swimming in some peripheral quadrants,not visiting central tank areas at all); direct finding (swimming di-rectlytowardtheplatformwithoutanyexplorationaroundthepool). 2.5. Analysis of neuronal morphology Since behavioral outcomes of single specimens of both experi-mental groups were rather homogeneous and there was no animalthat performed at odds with the others, randomly selected animals(EC:  N  =3; SC:  N  =3) were chosen for the neuronal morphologi-cal analysis. To this aim, we used the in vivo Golgi-like filling of the neurons described by Jiang et al. [36]. This simple and reliablemethod allows high efficiency of labeling and a complete visualiza-tionofdendriticarborizations.Furthermore,byrequiringverysmallinjection sites and affording a good spatial resolution, it allows to-pographic analyses of the neuronal circuits related to the injectionarea.UnliketheclassicalGolgimethod,thisinvivomethodpermitsthe characterization of all components of a selected projection. Theanimals were anesthetized with a solution of ketamine (90mg/kg)and xylazine (15mg/kg) i.p. administered. They received a pari-etal cortical injection 1mm below the pial surface, according to thestereotaxic co-ordinates derived from the atlas of Paxinos and Wat-son [57] (ML  − 5, AP  − 1.5) of 10% biotinylated dextran amineBDA (Molecular Probes, Eugene, OR) and 10mM NMDA (Sigma,St. Louis, MO) in 0.01M phosphate buffer (total injected volumesranging from 0.1 to 0.2  l).   M.G. Leggio et al. / Behavioural Brain Research 163 (2005) 78–90  81Fig. 1. Golgi-like labeling of rat primary somatosensory cortex showing tracer injection. (A) Dark column marks BDA+NMDA injection site in area SI; (B)a higher magnification of the (A) insert. Scale bars: 1mm. After a survival time of 72h, the animals were anesthetized andtranscardially perfused with phosphate buffered saline followed by4%paraformaldehyde.Afterremoval,thebrainswerecryoprotectedin 30% buffered sucrose and cut on a freezing microtome into coro-nal sections of 50  m as proposed in the srcinal report by Jianget al. [36]. Most of the BDA labeled neurons were evidenced usingthe avidin-biotin complex. Some sections were counterstained withthionin to visualize the cytoarchitectonic features of the cortex. Theinjectionareacenteredinprimarysomatosensorycortexshowedev-idenceofnecrosis,probablyduetotheexcitotoxicactionofNMDA(Fig. 1). Most labeled cells were found in the cortex surrounding theexperimentallesionwherenosignsofexcitotoxicitywerenoted.The radial distribution of labelled neurons was comparable in con-trol and enriched animals. In both groups the labelled parietal neu-rons chosen for the morphological analysis were located in layer IIIof parietal cortex from AP − 2.8 to − 3.8, areas corresponding to thesecondary somatosensory and posterior parietal cortices.Aresearcherunawareofspecimenidentitymadethemorpholog-ical analysis with the aid of the Neurolucida software (Microbright-field,Colchester,VT).Thissystemallowscomputer-assistedrecon-struction of the dendritic arbor, thus providing accurate measure-ment of dendritic arborisation. Neurons were selected when theirlabeling was uniform and extending into the most distal branchesof apical and basal dendrites, where spines were clearly marked.Furthermore, the predominant plan of the dendritic arbors of theseneurons should be parallel to the plan of the section [36].Accordingtothesecriteria,inlayerIIIofparietalcortex,10neu-rons were selected from the enriched animals and 10 neurons fromthe standard housed animals (enriched group: rat EC1, three neu-rons; rat EC6, four neurons; rat EC12, three neurons and standardgroup: rat SC4, four neurons; rat SC11, three neurons; rat SC15,three neurons). Apical and basal dendritic trees were examined sep-arately by means of Sholl Analysis of sphere intersection [71]. Thenumber of intersections of dendrites with a series of concentricspheres at 10  m intervals from the center of the cell body wascounted for each neuron. Total dendritic length (in   m) was thesum of the lengths of all processes passing through each shell. Ashell was the volume contained out to a given radius. Total numberof branches was calculated by summing the dendritic nodes startingfrom the cell body.Spine number was calculated by summing all the spines fromthe initial protrusion of apical or basal dendrites. Spine density wascountedasthenumberofspinespersegmentsof10  mlengthalongthe entire dendritic extension. All protrusions of the dendritic mem-brane were considered in our study, regardless of their shape oractual function. 2.6. Statistical analysis Metric unit results of animals belonging to the two experimentalgroups were first tested for homoscedasticity of variance and thencompared using one-way or two-way “p × q” analyses of variance(ANOVAs) with repeated measures, followed by multiple compar-isons using Duncan’s tests.For morphological analyses, to take into account the correlationof data within the same animal, a nested-design ANOVA was ap-plied. Accordingly, the main factor was Group (between-subjectswith two levels), while Animal was a nested factor within Group.Sincenot-equalnumberofneurons( n )wasexaminedineachanimaland the degrees of freedom of error for each animal are  n − 1, thetotal degrees of freedom for the mean square error (denominator forthe computation of   F  -statistic) resulted 14. 3. Results 3.1. Radial maze: full-baited procedure SC group performed worse than the EC group for totaltime and number of arms visited (Fig. 2A). The EC group performed better than SC group even on the parameters mostlinked to the computation of spatial factors, as displayed inFig. 2B. In particular, the EC group made a lower number of   82  M.G. Leggio et al. / Behavioural Brain Research 163 (2005) 78–90 Fig. 2. Motor (A), spatial (B) and procedural (C) parameters used to assess radial maze performances of the enriched (  N  =17) and standard (  N  =19) animalsin the full-baited paradigm. In this and in the following figures, vertical bars indicate mean standard error (S.E.M.).  * P <0.05;  *** P <0.0001. total errors than SC group. The first errors of EC group wereperformed at the sixth to eighth entry and of the SC rats atthe fifth to sixth entry by. We found the same picture whenwe analyzed correct visits span. The correct sequences of SCgroup were shorter (five to six correct visits) than those of the EC group (six to seven correct visits).The general distribution of the 45 ◦ angles performed byboth experimental groups revealed that EC animals made agreater number of 45 ◦ angles (Fig. 2C), even if the 45 ◦ an-gles increased in both groups as the sessions went by (one-wayANOVAs:ECgroup F  9,144  =6.97, P <0.0001;SCgroup F  9,162  =6.81,  P <0.0001). Statistical comparisons are shownin Table 1. 3.2. Radial maze: forced-choice procedure When the animals faced the first phase of the task, a slightbetween groups difference was found, i.e. the EC animalsperformed better than the SC group even if the easier task (onlyfourarmsopen)markedlyloweredthenumberoferrorsin both groups of animals.In the second phase, performed after 60s, once morethe EC group performed better than the SC group. Inparticular, although both groups decreased incorrect en-tries (Fig. 3A) as sessions went by (one-way ANOVAs:EC group  F  (1,64)  =9.69,  P =0.0000; SC group  F  (1,72)  =5.78, P =0.0005), the two groups were significantly different, as
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!