DP

HCa

sb

vc

P

Aiirmdodersm(aptdeonriilfceewcpE

Km

M8

*EcA3lsr

Neuroscience 154 (2008) 1598–1606

0d

OPAMINERGIC MESENCEPHALIC SYSTEMS AND BEHAVIORAL

ERFORMANCE IN VERY OLD RATS

mvaa(srtmwclca(tptwcpabgt(roec

cnMdrks(2cmscrslc1

. L. SANCHEZ,a* L. B. SILVA,a E. L. PORTIANSKY,b

. B. HERENU,c R. G. GOYAc AND G. O. ZUCCOLILLIa

Institute of Anatomy, Faculty of Veterinary Sciences, National Univer-ity of La Plata, CC296, Calle 60 y 118, 1900 La Plata, Argentina

Institute of Pathology, School of Veterinary Sciences, National Uni-ersity of La Plata, La Plata, Argentina

INIBIOLP-Histology “B,” School of Medicine, National University of Lalata, La Plata, Argentina

bstract—Morphologic and functional studies describing thempact of aging on mesencephalic dopaminergic (DA) neuronsn laboratory animals are rather scanty and inconclusive. Inats, stereological studies characterizing age changes in theesencephalic DA neurons have not been documented. In or-er to fill this information gap and to determine whether the veryld rat may serve as a suitable animal model of Parkinson’sisease, we performed a stereological assessment o f t h e mes-ncephalic tyrosine hydroxylase immunoreactive (TH-ir) neu-ons in young-adult (4 – 6 months), old (22–24 months) andenile (30 –32 months) Sprague–Dawley female rats. Morpho-etric analysis of the TH-ir neurons of the substantia nigra

SN) and ventral tegmental area (VTA) was performed usingn appropriate image analysis system. Age changes in motorerformance were assessed measuring the endurance of ratso hang from a wire mesh pole or to remain on a ramp set atifferent angles to the floor. Age changes in locomotion andxploratory activity were evaluated by the open field test. Webserved a significant age-related reduction in TH-ir neuronumbers in the SN (17 and 33% reduction in old and senileats, respectively compared with young counterparts) but notn the VTA. The size of the TH-ir cells increased significantlyn both the SN and VTA of the senescent animals but THabeling intensity fell. Motor, locomotor and exploratory per-ormance deteriorated markedly in the old and senile rats asompared with young animals. These findings reveal thexistence of a moderate but significant vulnerability of mes-ncephalic DA neurons to aging in rats. This phenomenon,hich is particularly marked in the SN of very old rats, mayontribute to the age-related decline in motor and exploratoryerformance recorded in this species. © 2008 Published bylsevier Ltd on behalf of IBRO.

ey words: aging, mesencephalic, tyrosine hydroxylase,orphometry, motor tests, open field.

esencephalic dopaminergic (DA) neurons account for0% of brain DA neurons. The perikarya of these cells are

Corresponding author. Tel: �54-221-425-6735; fax: �54-221-425-3276.-mail address: hlsanchez@amc.com.ar (please cc rgoya@netverk.om.ar) (H. L. Sanchez).bbreviations: ANOVA, analysis of variance; DA, dopaminergic; DAB,,3-diaminobenzidine tetrahydrochloride; IA, interaural; ml, medial

emniscus; mp, mammillary peduncle; SN, substantia nigra; SNpc,

eubstantia nigra pars compacta; TH-ir, tyrosine hydroxylase immuno-eactive; VTA, ventral tegmental area.

306-4522/08$32.00�0.00 © 2008 Published by Elsevier Ltd on behalf of IBRO.oi:10.1016/j.neuroscience.2008.04.016

1598

ainly located within the substantia nigra (SN) and theentral tegmental area (VTA) which correspond to the A9nd A10 catecholaminergic areas described by Dahlströmnd Fuxe (1964) (Tillet and Kitahama, 1998), respectivelyFig. 1). They form two long-length DA systems, the nigro-triatal and meso-cortico-limbic systems. The former isepresented by the nigral A9 neurons projecting mainly tohe neostriatum (caudate nucleus and putamen). Theeso-cortico-limbic system is formed by the A10 neurons,hich spread out axons from the VTA to striatal, limbic andortical areas (Bentivoglio and Morelli, 2005). This has

ong been recognized as an oversimplification because SNontains not only neurons projecting to the striatum, butlso neurons that innervate cortical and limbic areasBjörklund and Dunnett, 2007). Thus, dense projections ofhe ventral and intermediate sheets of the substantia nigraars compacta (SNpc) and ventro-lateral VTA innervatehe caudate-putamen (Gerfen, 1992). DA neurons locatedithin medial SNpc and middorsal VTA project to the nu-leus accumbens and olfactory tubercle, whereas dopaminerojections to the amygdaloid nuclei arise from lateral SNpcnd SN lateral (Fallon and Loughlin, 1995). VTA DA cellodies distributed in a loosely topographical fashion and ni-ral DA neurons of the dorsal-most sheet of SNpc innervate

he prefrontal, cingulated, perirhinal and entorhinal corticesFallon, 1988). In addition, sparse DA projections have beeneported arising from SN and VTA to connect with neurons ofther neocortical fields (Berger et al., 1991), cerebellum (Ikait al., 1992), hypothalamus, hippocampus, pallidum and lo-us coeruleus (Fallon and Moore, 1978a,b).

In the adult rat, the total number of mesencephalic DAells bilaterally is 40,000 – 45,000, with about half of theeurons located in the SN (German and Manaye, 1993).orphologic and functional studies in laboratory animalsescribing the impact of aging on mesencephalic DA neu-ons are rather scanty and inconclusive. In squirrel mon-eys, it has been reported that aging causes either aignificant loss of tyrosine hydroxylase immunoreactiveTH-ir; i.e. DA neurons) cells in the SN (McCormack et al.,004) or no change (Irwin et al., 1994). Despite this dis-repancy both studies reported a marked reduction ofotor activity in the aged monkeys. In rhesus monkeys, a

ignificant (50.3%) age-related loss of TH-ir neurons whichorrelated with a marked fall in motor performance waseported (Emborg et al., 1998). The only stereologicaltudy in aging mice we are aware of reported a significant

oss of TH-ir neurons in the SN of 104-week-old animals asompared with 8-week-old counterparts (Tatton et al.,991). We are unaware of stereological studies on the mes-

ncephalic DA neurons of aging rats and have only detected

acr6dm

loTbtmraoD

A

Ymcfiw1swsmlsd

NNiit

K

AibrcatooW4

I

SiaswaU(3Mm

I

TS

Fopt 10, cereb(

H. L. Sanchez et al. / Neuroscience 154 (2008) 1598–1606 1599

study reporting neurochemical but not morphologichanges (histologically assessed) in nigrostriatal TH-ir neu-ons of 24 to 25-month-old male rats as compared with 5 to-month-old counterparts (Emerich et al., 1993). The authorsocumented a significant reduction in locomotor activity andotor performance in the aged animals.

In the present study we employed an unbiased stereo-ogical approach to quantitate the impact of old and veryld age on the number and morphology of mesencephalicH-ir neurons. Since direct evaluation of nigral DA functiony the apomorphine- or amphetamine-induced rotation

ests can only be performed in unilaterally lesioned ani-als (not in intact animals with symmetric nigral DA neu-

on loss), we assessed age-changes in motor performancend exploratory activity in an attempt to gain a general idean the correlation between morphometric changes in nigralA neurons and those in behavioral parameters.

EXPERIMENTAL PROCEDURES

nimals and specimen collection

oung (5–6 months), old (22–24 months), and senile (30–32onths) Sprague–Dawley female rats, raised in our aging rat

olony, were used for morphological studies. For motor and openeld tests, slightly different age groups were employed. Animalsere housed in a temperature-controlled room (21�2 °C) on a2-h light/dark cycle. Rats were housed in groups of four intainless steel cages (32�32�18 cm, L�W�H). Food and waterere available ad libitum. In our rat colony, the average 50%urvival time for females, studied in groups of 50–60 animals, is 31onths (range, 30–32 months). Animal experiments were done fol-

owing the Guidelines on the Use of Animals in Neuroscience Re-earch (the Society of Neuroscience) using IACUC approved proce-

ig. 1. Schematic representation of coronal sections of the midbrain sf the mammillary body (MB) showing the distribution of TH-ir neeriaqueductal gray (PAG), third ventricle (3V), posterior commissure (he distribution of TH-ir neuron bodies in the VTA correspond to the AR), aqueduct (Aq), commissure superior colliculus (CSC).

ures (approval date 06/01/04) and Animal Welfare Guidelines of the w

ational Institutes of Health (INIBIOLP’s Animal Welfare Assuranceo A5647-01). The minimum number of rats compatible with detect-

ng reasonable (�25%) differences in the different parameters stud-ed was used and every effort was made to avoid or at least minimizehe suffering of animals during the experiments performed.

illing and specimen collection

nimals were anesthetized with ketamine hydrochloride (40 mg/kg;.p.) plus xylazine (8 mg/kg; i.m.) and perfused transcardially withuffered saline–4% formaldehyde solution. The brain was carefullyemoved from the cranium, equilibrated in a cryoprotectant solutionontaining 30% sucrose, 0.1 M PB (0.1 M Na2HPO4 buffer) in H2Ond stored at �20 °C until processing. Each brain was trimmed downo a block containing the whole mesencephalon, from the rostral levelf mammillary body (interaural (IA) line 4.20 mm) to the caudal levelf the pontomesencephalic grove (IA line 2.28 mm) (Paxinos andatson, 1998). The block was then serially cut into coronal sections

0 �m thick on a freezing microtome.

mmunohistochemistry

tereological assessment of brain structures was carried out us-ng a simplified variant of the optical fractionator method (West etl., 1991). In each animal, one of every four mesencephalic serialections was sampled throughout this structure. The sectionsere submitted to free floating immunohistochemistry using annti-TH monoclonal antibody (Calbiochem, Inc., La Jolla, CA,SA). As a detection system, the Vectastain Universal ABC kit

Vector Laboratories, Inc., Burlingame, CA, USA) was used, with,3-diaminobenzidine tetrahydrochloride (DAB; Sigma, St. Louis,O, USA) used as chromogen. Sections were dehydrated andounted for light microscopy assessment.

mage analysis

he images of the hemimesencephalon (objective 1.25�), the wholeN and VTA (objective 4�) and the TH-ir neurons (objective 20�)

he distribution of TH-ir neuron bodies. (a) Coronal section at the levelies in the SN correspond to the A9, cerebral peduncle (cp), mp,oronal section at the level of the interpeduncular nucleus (ip) showingral peduncle (cp), ml, central linear nucleus raphé (CL), red nucleus

howing turon bodpc). (b) C

ere captured with a digital color video camera (EvolutionVF, QIm-

aBmbafia

ssdtToadug

wmilau

wsppsTsocsp1

an

M

Fc(

dbatt

tasbppwai

O

R(ng

S

ThmKuaPa

Mn

TdbAo

FrSnmdm

H. L. Sanchez et al. / Neuroscience 154 (2008) 1598–16061600

ging, Surrey, BC, Canada) attached to a microscope (OlympusX50, Olympus Ltd., Tokyo, Japan) and connected to a high perfor-ance computer loaded with the Image Pro Plus v6.2 (Media Cy-ernetics, Silver Spring, MD, USA) image analysis software. Im-ges were captured with a pixel depth of 24 bits, RGB and TIFF

ormat. The hemimesencephalon and the SN and VTA were dig-tally circumscribed and their respective TH-ir and mesencephalicreas determined.

The sections from different animals corresponding to theame level were aligned using neuroanatomical markers. Mea-urements were thus recorded for the overall structure and forifferent rostrocaudal levels. The recognition of forebrain struc-ures in co-stained sections was accomplished by inspection ofH-ir adjacent sections and the Paxinos and Watson (1998) atlasf the rat brain. The VTA and SN were outlined for each sectionnd analyzed according to TH-ir adjacent sections. In order toelimit the SN from the VTA, the following tracts of fibers weresed from rostral to caudal: mammillary peduncle (mp), emer-ence of the III cranial pair (3n) and medial lemniscus (ml) (Fig. 1).

Immunoreactive perikarya showing clearly shaped nucleiere characterized using the following parameters: cellular area,ajor and minor cell axis, perimeter and cell roundness. TH-

mmunoreactive neuronal bodies were counted. The intensity ofabeling, as determined by DAB staining (optical density), waslso recorded. The total number of TH-ir perikarya was estimatedsing the following equation:

N � ssf1 �

t�1

n

x

here, N�total estimated number of TH-ir perikarya per region;sf�section sampling fraction (1/4); n�number of sections sam-led per region; x�number of TH-ir perikarya counted per sam-led section x (from sampled section # 1 to # n). Each sampledection was divided into frames, images of all frames captured andH-ir perikarya counted. Therefore, the number of objects in eachampled section was not estimated (from an area sampling fractionr asf) but determined by full counting. Similarly, objects wereounted throughout the height (thickness) of sampling sections in-tead of being estimated from a height sampling fraction or hsf, asroposed by the original optical fractionator method (West et al.,991).

The entire SN and VTA areas were used for counting cellsnd neuronal density was expressed as the number of countedeuronal bodies per mm2.

otor coordination tests

emale rats of different ages were submitted to a battery of motoroordination and strength tests adapted in part from Wallace et al.1980).

Suspension from a horizontal wire mesh pole. The timeuring which the rats could sustain their own weight was determinedy placing the animals on a horizontal wire mesh pole and immedi-tely rotating the pole so that the animals were left suspended fromhe wire mesh 70 cm over a water tank. The latency for the animalso fall was recorded as the average of three consecutive tests.

Performance on a wire mesh ramp set at different angles tohe floor. A 90 cm long by 42 cm wide metal ramp set at differentngles to the floor was used. The ramp was covered by a centraltrip of wire mesh (65�20 cm) to offer the animals a grip and thease of the ramp was submerged into water up to 15 cm torevent the animals from descending to the floor. Animals werelaced on the central strip and the latency for them to fall to theater was recorded as the average of three consecutive tests. Ifn animal lasted up to 120 s, it was removed from the ramp and

ts performance was recorded as maximal.nb

pen field test

ats were individually placed in a Plexiglas square arena45�45�35 cm) divided into nine equal sectors on the floor. Theumber of sector crossings (with all four paws), episodes of self-rooming and rearing was recorded for 5 min.

tatistical analysis

he analysis of variance (ANOVA) followed by the Bonferroni postoc test was used to evaluate intra- and inter-group differences inorphometric variables (Zar, 1984), while the non-parametricruskal-Wallis one-way ANOVA followed by the Dunn’s test wassed to analyze behavioral data. Significant differences betweenge groups for each parameter were defined as those with�0.05. Highly significant differences were defined as those withP�0.01.

RESULTS

orphologic age changes in TH-ir mesencephaliceurons

here was a macroscopically detectable age-related re-uction in the TH-ir staining in coronal sections of the SNut not of the VTA (Figs. 2 and 3, respectively, left panels).significant age-related reduction in neuronal density was

bserved in both regions (Tables 1 and 2).

ig. 2. TH-ir neurons corresponding to the A9 nigral area in femaleats of different ages. Left panels show a low magnification view of theN in coronal midbrain sections. An age-related reduction in TH-irigral cross-sectional area can be observed. Right panels show higheragnification views of the same sections. A reduction in TH-ir neuronensity and labeling intensity is evident in senile animals. Y: young (5onths). O: old (24 months). S: senile (32 months). For further tech-

ical details see Materials and Methods. Left and right scalears�0.3 mm and 20 �m, respectively.

aVaT

(2p3tcs

tnacw

A

Inep

FdcrcnO2

Fgyfaov

Tb

P

CAAAPRN

ilg*

Tb

P

CAAAPRN

ilg*

H. L. Sanchez et al. / Neuroscience 154 (2008) 1598–1606 1601

Histologically, the more conspicuous age change wasn increase in the size of TH-ir neurons in both the SN andTA (Figs. 2 and 3, respectively, right panels and Tables 1nd 2, respectively). In the senile rats, nigral but not VTAH-ir neurons showed a significant increase in roundness

ig. 3. TH-ir neurons corresponding to the A10 VTA in female rats ofifferent ages. Left panels show a low magnification view of the VTA inoronal midbrain sections. TH-ir VTA cross-sectional area is compa-able in the three sections shown. Right panels show higher magnifi-ation views of the same sections. As in the SN, a reduction in TH-ireuron density and labeling intensity is evident in senile animals.ther details are as in Fig. 2. Left and right scale bars�0.3 mm and0 �m, respectively.

able 1. Morphometric parameters corresponding to TH-ir neuronalodies in the SN of rats of different age

arameter Young (n�7) Old (n�6) Senile (n�7)

ell area (�m2) 160.98�13.48 162.11�12.03 230.27�17.01*spect 2.17�0.02 2.08�0.20 1.77�0.06xis (major) (�m) 20.75�0.63 21.48�1.38 22.74�0.65xis (minor) (�m) 10.34�0.60 10.97�0.72 13.17�0.75*erimeter (�m) 55.65�2.79 54.93�3.07 63.75�2.44*oundness 1.60�0.03 1.55�0.09 1.43�0.01*euronal density/(mm2)

716.58�39.63 549.50�15.09* 517.90�14.26*

Data are expressed as mean�S.E.M.; numbers in parenthesesndicate n values for the corresponding group. Asterisks represent theevel of significance of the differences between the indicated ageroup and the corresponding young group.

c

yP�0.05.

Figs. 2 and 3, respectively, right panels and Tables 1 and, respectively). The number of TH-ir neurons declinedrogressively with age in the SN, showing a 17% (NS) and3% (P�0.05) reduction in the old and senile rats, respec-ively as compared with the young animals (Fig. 4). Inontrast, the number of TH-ir neurons in the VTA was notignificantly affected by age (Fig. 4).

Consistent with the TH-ir neuron numbers recorded inhe SN and VTA of the three age groups, the mean coronaligral, but not VTA, TH-ir area decreased significantly withge (Fig. 5, main panel). It is of interest that the meanoronal TH-ir area of the whole mesencephalon increasedith age (Fig. 5, inset).

ge changes in motor and behavioral performance

n the horizontal mesh pole suspension test a highly sig-ificant age-related decline in performance was observedven between young and adult animals (Fig. 6, loweranel). The same age-related decline in performance was

ig. 4. Effect of age on mesencephalic TH-ir neurons numbers. Theraph shows the quantitation of TH-ir neurons in the SN and VTA ofoung, old and senile female rats. TH-ir neuron counting was per-ormed both manually and automatically using an appropriate imagenalysis software (see Materials and Methods for further details). Barsver columns represent S.E.M. values. Numbers over bars indicate nalues for the corresponding group. An asterisk over a column indi-

able 2. Morphometric parameters corresponding to THir neuronalodies in the VTA of rats of different age

arameter Young (n�7) Old (n�6) Senile (n�7)

ell area (�m2) 148.86�0.77 174.86�5.19 177.51�2.46*spect 1.76�0.07 2.04�0.07 2.08�0.02xis (major) (�m) 18.26�0.51 21.16�0.02 22.61�1.05xis (minor) (�m) 10.61�0.12 10.95�0.37 10.99�0.46erimeter (�m) 51.77�0.40 54.32�0.55 58.66�0.84*oundness 1.48�0.01 1.61�0.03 1.51�0.03euronal density/(mm2)

449.12�14.03 379.06�10.09 233.98�11.53*

Data are expressed as mean�S.E.M.; numbers in parenthesesndicate n values for the corresponding group. Asterisks represent theevel of significance of the differences between the indicated ageroup and the corresponding young group.P�0.05.

ates a significant (P�0.05) difference respective to the correspondingoung group.

rstcsrsir

smchuiwa

Fhlir

2smnti

mna(nrdIcnarTtmmD

FaM me animd

H. L. Sanchez et al. / Neuroscience 154 (2008) 1598–16061602

ecorded in the ramp test when the angle of the ramp waset at 88° (Fig. 6 upper panel); however when the angle ofhe ramp was set at 70°, the 2- and 7-month-old ratsonsistently achieved maximal performance (120 s) andhowed no difficulties for climbing and descending theamp (down to the water level). In contrast the old andenescent animals did not show a dramatic improvementn their performance as compared with that on the 88°amp (Fig. 6, inset of upper panel).

In the open field test, rearing, sector crossing andelf-grooming activity was recorded in 2-, 9-, 20- and 26-onth-old females. Rearing and sector crossing frequen-

ies showed a marked age-related decline which attainedigh significance in 20 and 26-month-old animals (Fig. 7,pper and lower panel, respectively). Self-grooming activ-

ty also showed an age-associated decline but this changeas more gradual than those recorded for sector-crossingnd rearing (Fig. 7, inset).

DISCUSSION

or several years the aging female Sprague–Dawley ratas been our model for studies related to tuberoinfundibu-

ar DA neuron aging in the hypothalamus and for themplementation of restorative interventions on these neu-

ig. 5. Effect of age on TH-ir mesencephalic cross-sectional areas innd VTA in young, old and senile female rats. Average areas were estethods for further details). Inset, overall mesencephalic area in the saetails are as in Fig. 4.

ons in aging animals (Goya et al., 1990; Sanchez et al., t

003; Herenu et al., 2007). The present results extend ourtudies to the mesencephalic DA neurons of this animalodel. The data reveal significant age changes in the TH-ireurons of the female rat mesencephalon, particularly inhe SN. As expected, the changes were more conspicuousn the senile animals.

Our morphologic data in the young rats are in agree-ent with those of the literature. Thus, the number of TH-ireurons recorded in the SN and VTA of adult animals is ingreement with those reported by German and Manaye1993). Also, the cell area of the mesencephalic TH-ireurons of young rats is in general agreement with thoseeported by Poirier et al. (1983) who recorded neuroniameters ranging from 7.5–35 �m in the SN of adult rats.

n aging rats, Emerich et al. (1993) failed to detect agehanges in number, area or length of A8, A9 or A10 TH-ireurons in 24 to 25-month-old male Sprague–Dawley ratss compared with 5 to 6-month-old counterparts. Theseesults are at variance with ours in the same age groups.he authors did not perform a stereological evaluation of

he animals but a histological analysis in three coronalesencephalic sections per animal. This discrepancyight reflect a sex-related difference in the vulnerability ofA neurons to aging. However, it seems more likely that

main panel shows the average TH-ir cross-sectional area of the SNmeans of an appropriate image analysis software (see Materials and

al groups. * P�0.05; ** P�0.01 respective to young counterpart. Other

rats. Theimated by

he disagreement between our morphometric data and

tedclaamtttetil

2isad

ctTStevT

Fhsrc

H. L. Sanchez et al. / Neuroscience 154 (2008) 1598–1606 1603

hose of these workers stems from methodological differ-nces. Despite the lack of agreement in morphologicalata, our results agree with those of Emerich et al. (1993)oncerning the marked reduction in motor performance,ocomotion and exploratory activity recorded in the oldnimals. In a study employing histofluorescence to char-cterize catecholaminergic nigral neurons in 3- and 26-onth-old Fischer 344 male rats, a significant reduction in

he number of nigral fluorescent neurons was reported inhe aged animals (Felten et al., 1992). To our knowledge,here are no morphometric data documented for DA mes-ncephalic neurons in senile (�28 months) rats nor arehere morphometric rat studies assessing sex-differencesn the impact of aging on mesencephalic DA neurons. The

ig. 6. Effect of age on hanging endurance and ramp test performaorizontal mesh pole. The upper panel shows the ability of rats (expreet at 88° to the horizontal. Inset, shows the same test using a ramp sepresent number of animals per group. Double asterisks indicate a horresponding 2-month-old control.

imited biochemical data available suggest that from 2 to c

4–26 months of age, male rats show more severe deficitsn striatal TH activity and D2 binding, while female ratshow a more excessive decrease in striatal DA along withsubstantial increase in their DOPAC/DA ratios (Fernan-

ez-Ruiz et al., 1992).The age-related reduction observed here in the TH-ir

ross-sectional area of the SN but not the VTA, is consis-ent with the age-related decrease in nigral but not VTAH-ir neuron numbers recorded in the same animals.ince we did not determine the expression of TH in the

arget sites, i.e. basal ganglia, a correlation could not bestablished between TH expression in the VTA and SNersus TH expression in the corresponding target regions.he significant increase of the total (TH-ir and non TH-ir)

er panel shows the resistance of rats of different ages to fall from aeconds to fall to the water at the base of ramp) to remain on a ramp° angle. Error bars represent S.E.M. whereas numbers over columnsificant difference (P�0.01) between the indicated age group and the

nce. Lowssed in set at a 70ighly sign

ross-sectional mesencephalic area in the senile rats sug-

gab(a

aeiaeadtaats

pfatcwipauoeovnab

Ffa

H. L. Sanchez et al. / Neuroscience 154 (2008) 1598–16061604

ests that gliosis may have developed in the mesenceph-lon of these animals. The presence of reactive gliosis haseen reported in animal models of Parkinson’s diseaseWang et al., 2006) as well as in the aging brain (Condend Streit, 2006).

Although there is controversy as to whether aging isssociated with a loss of nigral DA neurons, there is gen-ral agreement that motor performance decreases signif-

cantly during aging in laboratory animals (McCormack etl., 2004; Irwin et al., 1994; Emborg et al., 1998; Emericht al., 1993). The present results agree with the latter viewnd reveal that a fall in motor performance is alreadyetectable between 2 and 7 months of age in rats. Clearly,he deterioration of motor performance observed in theged rats is the result of age-related alterations that occurt different levels of the neuromuscular unit. Most likely,he dysfunctionality or loss of DA neurons in the SN of

ig. 7. Effect of age on open field activities. The lower panel shows trequency of rearing events in animals of different ages. Inset, shows ths in Fig. 6.

enile rats accounts for a fraction of their decline in motor t

erformance, with sarcopenia being a major contributingactor in this decline (Doherty, 2003). Although a directssessment of nigral DA function deterioration with age byhe apomorphine- or amphetamine-induced rotation testould not be performed in our animal model (intact ratsith bilateral DA neuron loss), in a previous study on the

mpact of very old age on the hypothalamic DA neuronopulation in female rats we could demonstrate a markedge-related deterioration of hypothalamic DA function (reg-lation of serum prolactin levels) without a substantial lossf hypothalamic TH� neurons (Sanchez et al., 2003; Her-nu et al., 2007). Therefore, the possibility exists that partf the mesencephalic TH-ir neurons present in the old andery old female rats studied here were dysfunctional DAeurons. On the other hand, it is also possible that duringging, part of the mesencephalic DA neurons remain via-le but undergo a downregulation in the expression of TH

ncy of sector crossings in the same animals. Upper panel shows thency of self-grooming episodes the same age groups. Other details are

he frequee freque

hus becoming TH negative. This is the case in monkeys,

wcaTi1itDlgi

ipusfsliieMtoap

At##

B

B

B

C

C

D

d

D

E

E

F

F

F

F

F

F

G

G

G

H

I

I

M

M

P

P

S

T

T

H. L. Sanchez et al. / Neuroscience 154 (2008) 1598–1606 1605

here the number of TH-positive neurons in the SN de-lines with age. This effect, which amounts to 40–50% inged animals, is due at least in part, to a downregulation ofH in surviving DA neurons, as detected by DA transporter

mmunostaining and neuromelanin content (Emborg et al.,998; Chu et al., 2002; McCormack et al., 2004). Interest-

ngly, insulin-like growth factor I gene therapy in the hypo-halamus of aging female rats increased tuberoinfundibularA neuron numbers and corrected the chronic hyperpro-

actinemia of the animals (Herenu et al., 2007). A similarene therapy approach in the SN of aging rats might

mprove their motor performance.The present study documents for the first time, the

mpact of very old age on the mesencephalic DA neuronopulations of rats in absence of neurological diseases bysing the criteria of modern cell-counting methods. Ithould be pointed out that most in vivo models developedor the assessment of therapeutic strategies for Parkin-on’s disease are based on the use of neurotoxins toesion nigral DA neurons in young animals. The neurolog-cal lesions studied in these models are caused by exper-mental manipulations rather than by aging, the only un-quivocal risk factor for this pathology (de Rijk et al., 1995;ayeux et al., 1995). In this context, our findings point to

he senile female rat as an interesting model of spontane-us TH-ir neuron loss and/or dysfunction as well as ofge-related deterioration of motor and exploratoryerformance.

cknowledgments—The authors thank Ms. Yolanda E. Sosa forechnical help. This work was supported in part by grantR21TW6665 from the National Institutes of Health and grant11-V142 from the National University of La Plata.

REFERENCES

entivoglio M, Morelli M (2005) The organisation and circuits of mes-encephalic dopaminergic neurons and the distribution of dopaminereceptors in the brain. In: Handbook of chemical neuroanatomy(dopamine) Vol. 21 (Dunnett SB, Bentivoglio M, Björklund A, Hök-felt T, eds). pp 1–107. Amsterdam: Elsevier Science.

erger B, Gasper P, Verney C (1991) Dopaminergic innervation of thecerebral cortex: Unexpected differences between rodents. TrendsNeurosci 14(1):21–27.

jörklund A, Dunnett SB (2007) Dopamine neuron systems in thebrain: an update. Trends Neurosci 30:194–202.

hu Y, Kompoliti K, Cochran EJ, Mufson EJ, Kordower JH (2002)Age-related decreases in Nurr1 immunoreactivity in the humansubstantia nigra. J Comp Neurol 450:203–214.

onde JR, Streit WJ (2006) Microglia in the aging brain. J NeuropatholExp Neurol 65:199–203.

ahlström A, Fuxe K (1964) Evidence for the existence of monoamine-containing neurons in the central nervous system. I. Demonstrationof monoamines in the cell bodies of the brain stem neurons. ActaPhysiol Scand 62:1–55.

e Rijk MC, Breteler MM, Graveland GA, Ott A, Grobbee DE, van derMeche FG, Hofman A (1995) Prevalence of Parkinson’s disease inthe elderly: the Rotterdam study. Neurology 45:2143–2146.

oherty TJ (2003) Aging and sarcopenia. J Appl Physiol 95:1717–1727.

mborg ME, Ma SY, Mufson EJ, Levey AI, Taylor MD, Brown WD,Holden JE, Kordower JH (1998) Age-related declines in nigralneuronal function correlate with motor impairments in rhesus mon-

keys. J Comp Neurol 401:253–265.

merich DF, McDermont P, Krueger P, Banks M, Zhao J, Marsza-lkowski J, Frydel B, Winn SR, Sanberg PR (1993) Locomotion ofaged rats: relationship to neurochemical but not morphologicalchanges in nigrostriatal dopaminergic neurons. Brain Res Bull32:477–486.

allon JH (1988) Topographic organization of ascending dopaminergic pro-jections. In: The mesocorticolimbic dopamine system (Kalivas PW, Nem-eroff CB, eds), pp 1–9. New York: New York Academy of Sciences.

allon JA, Loughlin SE (1995) Substantia nigra. In: The rat nervoussystem (Paxinos G, ed), pp 215–239. New York: Academic Press.

allon JH, Moore RY (1978a) Catecholamine innervation of the basalforebrain. III. Olfactory bulb, anterior olfactory nuclei, olfactorytubercle and piriform cortex. J Comp Neurol 180:533–544.

allon JH, Moore RY (1978b) Catecholamine innervation of the basalforebrain. IV. Topography of dopamine projection to the basalforebrain and neostriatum. J Comp Neurol 180:545–580.

elten DL, Felten SY, Steece-Collier K, Date I, Clemens JA (1992)Age-related decline in the dopaminergic nigrostriatal system: theoxidative hypothesis and protective strategies. Ann Neurol32:133–136.

ernandez-Ruiz J, De Miguel R, Hernandez ML, Cebeira M, RamosJA (1992) Comparisons between brain dopaminergic neurons ofjuvenile and age rats: sex-related differences. Mech Age Dev63:45–55.

erfen CR (1992) The neostriatal mosaic: Multiple levels of compart-mental organization in the basal ganglia. Annu Rev Neurosci15:285–320.

erman DC, Manaye KF (1993) Midbrain dopaminergic neurons (nu-clei A8, A9, and A10): three-dimensional reconstruction in the rat.J Comp Neurol 331:297–309.

oya RG, Lu JKH, Meites J (1990) Gonadal function in aging rats andits relation to pituitary and mammary pathology. Mech Age Dev56:77–88.

erenu CB, Cristina C, Rimoldi OJ, Becú-Villalobos D, Cambiaggi V,Portiansky EL, Goya RG (2007) Restorative effect of insulin-likegrowth factor-I gene therapy in the hypothalamus of senile rats withdopaminergic dysfunction. Gene Ther 14:237–245.

kai Y, Takada M, Ahinonaga Y, Mizuno N (1992) Dopaminergic andnon dopaminergic neurons in the ventral tegmental area of the ratproject, respectively, to the cerebellar cortex and deep cerebellarnuclei. Neuroscience 51:719–728.

rwin I, DeLanney LE, McNeill T, Chan P, Forno LS, Murphy GM, DiMonte DA, Sandy MS, Langston JW (1994) Aging and the nigro-striatal dopamine system: a non-human primate study. Neurode-generation 3:251–265.

ayeux R, Marder K, Cote LJ, Denaro J, Hemenegildo N, Mejia H,Tang MX, Lantigua R, Wilder D, Gurland B (1995) The fre-quency of idiopathic Parkinson’s disease by age, ethnic groupand sex in northern Manhattan, 1988 –1993. Am J Epidemiol142:820 – 827.

cCormack AL, Di Monte DA, Delfani K, Irwin I, Delanney LE, Lang-ston WJ, Janson AM (2004) Aging of the nigrostriatal system in thesquirrel monkey. J Comp Neurol 471:387–395.

axinos G, Watson CH (1998) The rat brain in stereotaxic coordinates.San Diego, CA: Academic Press.

oirier LJ, Giguere M, Marchand R (1983) Comparative morphology ofthe substantia nigra and ventral tegmental area in the monkey, catand rat. Brain Res Bull 11:371–397.

anchez HL, Silva LB, Portiansky EL, Goya RG, Zuccolilli GO (2003)Impact of very old age on hypothalamic dopaminergic neurons inthe female rat: A morphometric study. J Comp Neurol 458:319–325.

atton WG, Greenwood CE, Salo PT, Seniuk NA (1991) Transmittersynthesis increases in substantia nigra neurons of the agedmouse. Neurosci Lett 131:179–182.

illet Y, Kitahama K (1998) Distribution of central catecholaminergicneurons: a comparison between ungulates, humans and other

species. Histol Histopathol 13:1163–1177.

W

W

W

Z

H. L. Sanchez et al. / Neuroscience 154 (2008) 1598–16061606

allace JE, Krauter EE, Campbell BA (1980) Motor and reflexivebehavior in the aging rat. J Gerontol 35:364–370.

ang T, Zhang W, Pei Z, Block M, Wilson B, Reece JM, Miller DS,Hong JS (2006) Reactive microgliosis participates in MPP�-in-duced dopaminergic neurodegeneration: role of 67 kDa laminin

receptor. FASEB J 20:906–915.

est MJ, Slomianka L, Gundersen HJG (1991) Unbiased stereologi-cal estimation of the total number of neurons in the subdivisions ofthe rat hippocampus using the optical fractionator. Anat Rec231:482–497.

ar JL (1984) Biostatistical analysis. In: Multiple comparisons (Jerold

H, ed), pp 185–205. Englewood Cliffs, NJ: Prentice Hall.

(Accepted 9 April 2008)(Available online 16 April 2008)