B R A I N R E S E A R C H 1 4 0 8 ( 2 0 1 1 ) 8 1 – 8 7

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te /b ra i n res

Research Report

Functional significance of aldehyde dehydrogenase ALDH1A1to the nigrostriatal dopamine system

David W. Andersona, Rebecca C. Schraya, Gregg Duesterb, Jay S. Schneidera,⁎aDept. of Pathology, Anatomy and Cell Biology, Thomas Jefferson University, Philadelphia, PA 19107, USAbDevelopment and Aging Program, Sanford-Burnham Medical Research Institute, La Jolla, CA 9203, USA

A R T I C L E I N F O

⁎ Corresponding author at: Department of PatPhiladelphia, PA 19107, USA. Fax: +1 215 923

E-mail address: jay.schneider@jefferson.e

0006-8993/$ – see front matter © 2011 Elsevidoi:10.1016/j.brainres.2011.06.051

A B S T R A C T

Article history:Accepted 21 June 2011Available online 26 June 2011

Aldehyde dehydrogenase 1A1 (ALDH1A1) is a member of a superfamily of detoxificationenzymes found in various tissues that participate in the oxidation of both aliphatic andaromatic aldehydes. In the brain, ALDH1A1 participates in the metabolism ofcatecholamines including dopamine (DA) and norepinephrine, but is uniquely expressedin a subset of dopaminergic (DAergic) neurons in the ventral mesencephalon where itconverts 3,4-dihydroxyphenylacetaldehyde, a potentially toxic aldehyde, to 3,4-dihydroxyphenylacetic acid, a non toxic metabolite. Therefore, loss of ALDH1A1expression could be predicted to alter DA metabolism and potentially increaseneurotoxicity in ventral mesencephalic DA neurons. Recent reports of reduced levels ofexpression of both Aldh1a1 mRNA and protein in the substantia nigra (SN) of Parkinson'sdisease patients suggest possible involvement of ALDH1A1 in this progressiveneurodegenerative disease. The present study used an Aldh1a1 null mouse to assess theinfluence of ALDH1A1 on the function and maintenance of the DAergic system. Resultsindicate that the absence ofAldh1a1 did not negatively affect growth and development of SNDA neurons nor alter protein expression levels of tyrosine hydroxylase, the DA transporteror vesicular monoamine transporter 2. However, absence of Aldh1a1 significantly increasedbasal extracellular DA levels, decreased KCl and amphetamine stimulated DA release anddecreased DA re-uptake and resulted in more tyrosine hydroxylase expressing neurons inthe SN than in wildtype animals. These data suggest that in young adult animals withdeletion of the Aldh1a1 gene there is altered DA metabolism and dysfunction of the DAtransporter and DA release mechanisms.

© 2011 Elsevier B.V. All rights reserved.

Keywords:Aldehyde dehydrogenaseMicrodialysisDopamineMouse

1. Introduction

Aldehyde dehydrogenase 1 (ALDH1A1) and aldehyde dehy-drogenase 2 (ALDH2) are members of a diverse family ofenzymes that participates in the oxidation of a variety of

hology, Anatomy and Cell3808.du (J.S. Schneider).

er B.V. All rights reserved

aldehydes. In the brain, ALDH1A1 is significantly involved inthe metabolism of biogenic aldehydes, norepinephrine, dopa-mine (DA) and gamma-aminobutyric acid (Maring et al., 1985).3,4-Dihydroxyphenylacetaldehyde (DOPAL), a potentially toxicmetabolite of DA is formed by the oxidative deamination of DA

Biology, Thomas Jefferson University, 1020 Locust Street, 521 JAH,

.

82 B R A I N R E S E A R C H 1 4 0 8 ( 2 0 1 1 ) 8 1 – 8 7

catalyzed by monoamine oxidases (Burke et al., 2003). DOPAL,an intermediate in the degradation of DA, is then oxidized to3,4-dihydroxyphenylacetic acid (DOPAC) by ALDH1A1 orreduced to 3,4-dihydroxyphenylethanol (DOPET) by aldose/aldehyde reductase. Potentially damaging oxygen radicals areformed by this process (Adams et al., 2001). Although impairedDA transmission may influence ALDH activity and/or changesin ALDH-mediated metabolism may affect DA levels in nervecell bodies and terminal fields in the basal ganglia and limbicsystem (Galter et al., 2003), data directly supporting this arelacking.

While ALDH1A1 is strongly and specifically expressed inventral mesencephalic DA neurons in rodents (Westerlundet al., 2005) and man (Galter et al., 2003), its functionalsignificance to the DA system and its possible relevance toParkinson's disease (PD) is incompletely understood. mRNAencoding cytosolic ALDH1A1 is highly expressed in humansubstantia nigra (SN) and ventral tegmental area (VTA)dopaminergic (DAergic) neurons, with the virtual absence ofthis gene in neighboring non-DAergic cells (Galter et al., 2003).In brains from PD patients, there was a markedly lowerexpression of Aldh1a1 mRNA in tyrosine hydroxylase (TH)-positive and DA transporter (DAT)-positive neurons in the SN,compared to non-PD controls. In contrast, Aldh1a1 expressionwas not significantly decreased in TH- or DAT-positive neuronsin the VTA in most PD patients (Galter et al., 2003). Based onthese data, it was suggested that a down-regulation of Aldh1a1in SN DA neurons in PD might reflect a type of compensatoryresponse of these cells to decrease the rate of DA degradation.However, reduced ALDH1A1 expression in the SN could alsocontribute to the development of PD rather than be aconsequence of PD. With reduced ALDH1A1 expression, it ispossible that the degradation ofDAwould be slowed at the levelof DOPAL. An accumulation of DOPAL could be potentially toxictoDAneuronsandcould render theseneuronsmore susceptibleto aldehyde toxicity and degeneration.

Although current evidence suggests that ALDH1A1 isimportant for DA neuron function, there is little detailavailable on this, primarily due to absence of an appropriateanimal model for study. The present investigation character-izes the influence of ALDH1A1 on DAergic function using anAldh1a1 (Raldh1) null mousemodel (Fan et al., 2003) utilizing invivo microdialysis and post-mortem evaluations of pre-andpost-synaptic DAergic markers.

A B

ALDH1A1

GAPDH

W

250bp 600bp

Aldh1a1-/- Aldh1a1-/-Wildtype

Fig. 1 – Expression of Aldh1a1 mRNA and protein in Aldh1a1−/− aAldh1a1−/− animals did not contain a functional copy of the Aldh1gene expression (NEO cassette, 600 bp;Aldh1a1, 250 bp). (B)Westesignificant loss of ALDH1A1 expression in the Aldh1a1−/− mice cocontrols (using GAPDH) for all lanes are shown. (C) Immunohistonigra pars compacta showed significant loss ALDH1A1-immuno

2. Results

All mice used were genotyped by PCR to confirm the presenceof the Neo cassette which deleted the expression of theALDH1A1 protein in the Aldh1a1−/− animals (Fig. 1A). Westernblotting was used to visualize the effective deletion ofALDH1A1 protein in the striatum (Fig. 1B) and immunohisto-chemistry verified the lack of ALDH1A1 expression in neuronsin the SN in Aldh1a1−/− animals (Fig. 1C).

2.1. Microdialysis

Basal extracellular fluid (ECF) DA levels in the striatum in awakeAldh1a1−/− mice (23.39±4.43 nM) were significantly higher thanthose observed in wildtype animals (7.59±0.86 nM; t(12)=3.023,p=0.0106) (Fig. 2A). Perfusion of the probe with 120 mM KClresulted in significant release of DA in both Aldh1a1−/− animals(402.0±71.49 nM) and wildtype control animals (383.4±100.5 nM). However, Aldh1a1−/− animals had significantly lessKCl-stimulated DA release (67%) than did wildtype animals(t(12)=2.814, p=0.0156), relative to basal ECF DA levels (Fig. 2B).Perfusion of the probewith 50 μM D-amphetamine also resultedin significant release of DA in both Aldh1a1−/− animals (232.7±75.85 nM) and wildtype control animals (218.4±48.05 nM).There was significantly less (68%) amphetamine-stimulatedDA release inAldh1a1−/− animals compared to wildtype animals(t(12)=2.857, p=0.0144), relative to basal ECF DA levels (Fig. 2C).

2.2. Tissue DA levels

There were no statistically significant differences in the levelsof striatal DAmeasured in post-mortem tissue fromAldh1a1−/−

or wildtype animals (Fig. 3A). Aldh1a1−/− animals had asignificantly lower level of striatal DOPAC when compared towildtype animals (t(13)=2.257, p=0.0419; Fig. 3B). However, anindex of DA turnover (DOPAC/DA ratio) showed no statisticallysignificant differences between Aldh1a1−/− animals and wild-type controls (Fig. 3C).

2.3. Tyrosine hydroxylase positive neuron number

Stereological estimates of the number of tyrosine hydroxylase-immunoreactive (TH-IR) neurons in the substantia nigra pars

C Wildtype

ildtype

Aldh1a1-/-

nd wildtype (WT) mice. (A) Representative genotyping gel.a1 gene, but were positive for the Neo insert used to disruptrn blot analysis of ALDH1A1 protein in the striatum showed ampared to normal expression levels in WT mice. Loadingchemical analysis of ALDH1A1 expression in the substantiareactivity when compared to WT animals.

Wildtype Aldh1a1-/-

Wildtype Aldh1a1-/-

Wildtype Aldh1a1-/-

0

10

20

30 *E

xtra

cellu

lar

Do

pam

ine

(nM

)

0

1000

2000

3000

4000

5000

6000

% C

han

ge

fro

mB

asel

ine

Do

pam

ine

% C

han

ge

fro

mB

asel

ine

Do

pam

ine

*

0

500

1000

1500

2000

2500

3000

3500

*

A

B

C

Fig. 2 – Analysis of dopamine (DA) levels in the striatum offreely moving awake mice by in vivo microdialysis. (A) Basalextracellular levels of DA were significantly higher in theAldh1a1−/− mice than wildtype (WT) mice. (B) KCl-stimulatedDA release was significantly reduced in Aldh1a1−/− micecompared to WT mice. (C) Amphetamine stimulated DArelease was significantly reduced in Aldh1a1−/− micecompared to WT mice. *p<0.05 vs. wildtype. (n=8 and 6respectively for Aldh1a1−/− and WT).

Wildtype Aldh1a1-/-

Wildtype Aldh1a1-/-

Wildtype Aldh1a1-/-

0

2

4

6

8

10

12

μg/g

wet

tis

sue

μg/g

wet

tis

sue

0.00

0.05

0.10

0.15

DO

PA

C:D

A R

atio

0.00

0.25

0.50

0.75

1.00

1.25

1.50

*

A

B

C

Fig. 3 – Comparison of tissue dopamine (DA) and metabolitelevels in the striatum of Aldh1a1−/− and wildtype mice.(A) Striatal DA levels in Aldh1a1−/− and wildtype (WT) micewere not significantly different. (B) The level of3,4-dihydroxyphenylacetic acid (DOPAC) in the Aldh1a1−/−

mouse was significantly lower than in WT mice (*p<0.05).(C) Dopamine turnover was reduced in Aldh1a1−/− mice whencompared to WT, but the difference was not statisticallysignificant. (n=10 and 6 respectively for Aldh1a1−/− and WT).

83B R A I N R E S E A R C H 1 4 0 8 ( 2 0 1 1 ) 8 1 – 8 7

compacta indicated more TH-IR neurons in Aldh1a1−/− animals(5377±52.5) compared to wildtype controls (4590±145.9; t(6)=5.075, p=0.0023; Fig. 4). Stereological estimates of Nissl stainedneurons in the substantia nigra pars compacta were notsignificantly different between Aldh1a1−/− animals (5929±501.3) compared to wildtype controls (5886±286.0; Fig. 4— allvalues are mean±SEM).

2.4. Striatal protein analysis

Striatal tissue was analyzed by Western blot to assess thelevels of TH, DAT, and vesicular monoamine transporter 2(VMAT2) protein. Expression levels of TH inAldh1a1−/− animals(0.80±0.04 o.d.) were not significantly different from wildtypeanimals (0.87±0.04 o.d.). Expression levels of DAT also showedno statistically significant change in Aldh1a1−/− animals

compared to wildtype controls (1.96±0.08 o.d. vs. 1.91±0.06o.d. respectively). Similarly, VMAT2 expression levels were notstatistically significantly different in Aldh1a1−/− animalscompared to wildtype controls (0.94±0.06 o.d. vs. 0.75±0.02o.d. respectively) although this trended towards significance(p=0.052) (Fig. 5, all values are stated as mean±SEM.).

2.5. Dopamine uptake

Dopamine uptake in striatal homogenates from Aldh1a1−/−

animals, measured during the linear phase, was significantlyslowed compared to uptakemeasured in striatal homogenatesfrom wildtype controls. Striatal DA uptake in Aldh1a1−/−

striatal homogenateswas 30.29%±5.72 less than that observedin wildtype controls (t(12)=4.672, p=0.0005) (Fig. 6).

ALDH1A1

TH

Wildtype ALDH1A1-/-

Fig. 4 – Photomicrographs of ALDH1A1 and tyrosine hydroxylase immunoreactivity in the substantia nigra of Aldh1a1−/− andwildtype (WT) mice. All images were taken at 4× magnification, with higher 40× magnification inset for both tyrosinehydroxylase photomicrographs. (n=6 and 6 respectively for Aldh1a1−/− and WT).

84 B R A I N R E S E A R C H 1 4 0 8 ( 2 0 1 1 ) 8 1 – 8 7

3. Discussion

The results of the present study show that absence ofALDH1A1 results in higher than normal extracellular fluidlevels of DA in the striatum and alterations in DA release inresponse to KCl or amphetamine stimulation. Tissue levels ofDA, as well as TH, DAT and VMAT2 protein expression in thestriatum were not affected by Aldh1a1 gene deletion. Dopa-mine uptake rate was significantly reduced in Aldh1a1−/−

animals compared to wildtype controls. These findingssuggest that loss of Aldh1a1 in young animals affects at least

Wildtype Aldh1a1-/-

TH

GAPDH

DAT

β-ACTIN

β-ACTIN

VMAT2

Fig. 5 – Striatal protein expression in Aldh1a1−/− andwildtype(WT)mice. Therewere no significant differences in any of theanalyzed proteins between the Aldh1a1−/− and WT mice.TH=tyrosine hydroxylase; DAT=dopamine transporter;VMAT2=vesicular monoamine transporter 2. GAPDH orβ-actin was used as loading controls. (n=6 and 6 respectivelyfor Aldh1a1−/− and WT).

some functional dynamics of DA neurons. However, theabsence of Aldh1a1 does not appear to negatively impact theinitial growth and development of SN DAergic neurons normaterially impact their survival in young animals. However,absence of Aldh1a1 in SN DA neurons does influence theability to release DA in response to a physiological signal (KCl)and to clear released DA. Surprisingly, the number of TH-IRneurons in the SN was higher in Aldh1a1−/− mice than inwildtype animals, while overall neuronal numbers remainedunchanged from wildtype controls. Even though there was asignificant increase TH-IR cell numbers in the SN, there wereno increases in levels of the striatal protein markers TH, DATor VMAT. Although the reasons for these findings arepresently unclear, it is possible that despite a larger number

Striatal DA Uptake

Wildtype Aldh1a1-/-0

10000

20000

30000

40000

50000

Sp

ecif

ic U

pta

ke (

DP

M)

*

Fig. 6 – Analysis of striatal dopamine uptake in Aldh1a1−/−

and wildtype (WT) mice. There was a significant decrease inDA uptake in striatal homogenates from Aldh1a1−/− animalscompared to wildtype control animals. *p<0.001. (n=7 and 7respectively for Aldh1a1−/− and WT).

85B R A I N R E S E A R C H 1 4 0 8 ( 2 0 1 1 ) 8 1 – 8 7

of cells expressing detectable levels of TH in the SN, thenumber of DAergic terminals in the striatum has not changedand/or a defective intracellular transport of proteinsmay existbetween the SN and the striatum in the Aldh1a1−/− mice.Taken together these data suggest that the expression ofALDH1A1 is not critical to the development or survival of DAneurons in the SN of young animals but does affect thefunctioning of the DA system.

It is not entirely clear at this time how the findings fromthese animals relate to findings in Parkinson's disease brain.TheseanimalshadnoALDH1A1 for their entire lives and it isnotclear when Parkinson's patients may develop reduced levels ofALDH1A1. However, ALDH1A1 function is highly dependentupon the availability of its cofactor NAD+ (Meyer et al., 2004),which is formed in the mitochondria, and perturbations ofmitochondrial function, specifically complex I inhibition havebeen suggested to play a role in the processes underlying theloss of the SN DAergic neurons (Olanow and Tatton, 1999;Schapira et al., 1998). Profound decreases in NAD+ levels,leading to decreased processing of aldehydes may furthercompromise the function of these neurons and increase theirsusceptibility to degeneration. Thus, some of what we haveobserved in these young adult Aldh1a1−/− animals may repre-sent at least in part a response of the nigrostriatal system to thelifetime absence ofALDH1A1.What happens to thenigrostriatalsystem in these animals as they age is not known but is thesubject of on-going follow-up studies.

Aldehyde dehydrogenase 1A1 is amember of a superfamilyof detoxification enzymes found in brain and other tissuesthat participate in the oxidation of both aliphatic and aromaticaldehydes of endogenous and exogenous cellular origin. In thebrain, aldehyde dehydrogenases were first implicated in theoxidation of catecholamine-derived aldehydes in the early1960s (Erwin and Deitrich, 1966), and are now known to besignificantly involved in the metabolism of numerous biogen-ic aldehydes including DA, norepinephrine and gamma-aminobutyric acid. Specifically, ALDH1A1 is uniquely andhighly expressed within the DAergic neurons of the SN andVTA of both rodents and man (Galter et al., 2003; Grimm et al.,2004). As such, ALDH1A1 is the primary catalyst of theirreversible oxidation of the biogenic aldehyde DOPAL toDOPAC, which is itself formed by the oxidative deaminationof DA by monoamine oxidases (MAO) (Lamensdorf et al., 2000;Tank et al., 1981), a process that also produces potentiallydamaging oxygen radicals (Adams et al., 2001). Modification ofthe processing of DOPAL by ALDH1A1 could have significantimpact on the long-term survival of DAergic neurons. Al-though absence of the Aldh1a1 gene had no negative impacton the number of DAergic neurons in the SN in young (8–12 week old) animals, it is possible that as these animals age,the long-term effects of an increased level of DOPAL could bepotentially toxic to DA neurons and could render theseneurons more vulnerable to aldehyde toxicity and degenera-tion. Since the primary role of ALDH1A1 in nigrostriatal DAneurons is the oxidation of DOPAL to DOPAC, therebyremoving a neurotoxic intermediary aldehyde from thecytoplasm before it can undergo conversion to even greatertoxic compounds such as tetrahydropapaveroline (THP), anALDH1A1 deficiency could have marked long-term effects onDA neurons. Over prolonged periods of time, ALDH1A1

deficiency could lead to behavioral/phenotypic and neurode-generative changes consistent with PD that would occur withincreasing age. Also, it is possible that other ALDHs may havea redundant function in oxidation of DOPAL to DOPAC thatprevents more severe effects in Aldh1a1 mutants.

A loss of function mutation in the Aldh1a1 gene could be arisk factor for PD either alone or in conjunction withenvironmental risk factors. Aldh1a1 mRNA levels are specif-ically down-regulated in DA neurons in PD brain (Galter et al,2003; Grunblatt et al., 2004; Mandel et al., 2005). It is notpossible to know if this decrease in Aldh1a1 gene expressionpreceded the onset of PD or was a consequence of thedegenerative process in PD. However, in the periphery,ALDH1A1 is expressed in the digestive tract and implicatedin a variety of detoxification reactions. Impaired function ofthe Aldh1a1 gene in the lining of the gastrointestinal tractcould result in toxic compounds, including aldehydes, enter-ing the circulation, entering the brain and potentially,affecting SN DAergic neurons. This, combined with thealready discussed influences of loss of the Aldh1a1 gene onthe neurochemistry of DA neurons and the resulting accumu-lation of toxic by-products of abnormal DA metabolism couldpresent a heightened risk for developing PD in individualswith loss of function mutation in the Aldh1a1 gene.

In conclusion, the present study examined the functioningof the DA system in young Aldh1a1−/− mice. Lack of Aldh1a1expression in nigrostriatal DAergic neurons significantlyincreased striatal extracellular fluid levels of DA, significantlyreduced striatal DA uptake, and significantly reduced terminalrelease of DA in response to KCl or amphetamine stimulation.Further characterization of these animals is necessary tomorefully understand how alterations in Aldh1a1 gene expressionmay be involved in the development or perpetuation of PD. Inparticular it will be important to evaluate how aging andexposure to environmental toxicants may interact withabnormalAldh1a1 gene function to influence the developmentof a Parkinsonian phenotype.

4. Experimental procedures

4.1. Animals

The method for production of Raldh1 (Aldh1a1) null mice on aC57Bl/6 background has been described previously (Fan et al.,2003). Aldh1a1−/− mice used to establish the current colonywere provided by G. Duester (Sanford-Burnham Institute, CA).Mice were genotyped by PCR using custom primers; Aldh1a1WT 5′ TAAAGACCTGGATAAGGCCATCA and 5′ ACGGTGCAC-AAAATAAACATCTG and Aldh1a1 neo 5′ AGGATCTCCTGTCAT-CTCACCTTGCTCCTG and 5′ AAGAACTCGTCAAGAAGGCGATA-GAAGGCG (BioChem, Salt Lake City, UT). Wildtype male C57Bl/6(Taconic, Germantown, NY) and age-matched Aldh1a1−/− mice,were housed in 12 h light/dark cycle with ad libitum food andwater. Mice were 8–12 weeks old when studied. All procedureswere approved by the Thomas Jefferson University InstitutionalAnimalCare andUseCommittee, andcarriedout according to theNational Institutes of Health Guide for the Care and Use ofLaboratory Animals.

86 B R A I N R E S E A R C H 1 4 0 8 ( 2 0 1 1 ) 8 1 – 8 7

4.2. Microdialysis

Animals were anesthetized with Ketamine/Xylazine (i.p., 8:1)and placed in a custom mouse stereotaxic apparatus. Usingflat skull coordinates, a microdialysis probe cannula (Sci-Pro,Sanborn, NY) was affixed to the skull overlying the striatum atthe following coordinates relative to Bregma: 0.4 mm rostral,1.8 mm lateral and 2.0 mm below the skull surface. Animalsrecovered from surgery overnight and the following morning,a dialysis probe (2 mm membrane) (Sci-Pro, Sanborn, NY) wasinserted into the striatum. The freely moving animal wastethered to a liquid swivel (Instech Laboratories, Inc., Plym-outh Meeting, PA) attached to a syringe pump. Prior toinsertion, probes were calibrated in vitro at 37 °C to verifyfunctionality and to calculate relative recovery of DA andDOPAC. After insertion, probes were perfused with sterileartificial cerebrospinal fluid (aCSF, sodium chloride 145 mM;potassium chloride 2.8 mM; magnesium chloride 1.2 mM; andcalcium chloride 1.2 mM) at a rate of 1.5 ml/min. Approxi-mately 22 μl of dialysate was collected and analyzed every15 min until a stable DA baseline (defined as 3 consecutivesamples within approximately 10% variation) was achieved.The perfusate was then switched to aCSF containing 120 mMKCl for 30 min and then switched back to standard aCSF andsamples were collected until DA levels re-stabilized. Followingreturn to baseline, the probe was perfused with aCSF contain-ing 50 μM D-amphetamine sulfate for 30 min. The perfusatewas then switched back to standard aCSF and samples werecollected until DA levels once again returned to a stablebaseline level.

Dopamine and DOPAC levels were measured in dialysateusing high-pressure liquid chromatography with electro-chemical detection (HPLC-EC, ESA Coulochem III). Sampleswere separated using a C18 reverse-phase column andMD TMmobile phase (ESA, Chelmsford, MA) consisting of sodiumdihydrogen phosphate 75 mM; 1-Octanesulfonic Acid 1.7 mM;.01% TEA; EDTA 25 μM; 10%Acetonitrile, pH 3.0 at a flow rate of0.5 ml/min. Peak heights were analyzed using EZChromChromatography software (Scientific Software Inc., SanRamon, CA) and data processing was accomplished usingPrism GraphPad (La Jolla, CA). Data were calculated as peakrelease.

At the conclusion of a microdialysis study, the animal waseuthanized and the brain quickly removed. The striatum onthe side opposite to the probe site was collected for post-mortem analysis of DA and metabolite levels using methodspreviously described (Anderson et al., 2001). The remainingtissue including the striatum on the side used for micro-dialysis and ventral mesencephalon was post-fixed in 4%paraformaldehyde for verification of the probe insertion siteand for cell counting studies (vide infra).

4.3. Tissue processing and cell counting

Serial coronal sections (30 μm thick) were cut frozen throughthe rostral–caudal extent of the SN and collected in 0.1 Mphosphate-buffered saline (PBS). Every third section wasprocessed for TH-IR. Adjacent sections were processed forALDH1A1-immunoreactivity (ALDH1A1-IR) and the remainingsection was stained for Nissl substance to detect total

neuronal numbers. Immunohistochemistry was performedon free floating sections. Tissues were incubated in PBScontaining 3% hydrogen peroxide and 0.3% Triton X-100 toinhibit endogenous peroxidase activity followed by 1 h blockat room temperature in 5% (w/v) non-fat powderedmilk (NFM)containing Triton X-100 (NFM-T). Sections were then incubat-ed overnight at 4 °C with rabbit anti-TH (1:1,000; Pel-FreezeInc., Rogers, AR) or rabbit anti-ALDH1A1 (1:10,000; AbcamCambridge MA) in NFM-T. After washing in NFM-T, sectionswere incubated in biotinylated goat anti-rabbit antibody(1:1000 Jackson Immunoresearch, West Grove, PA) in NFM-Tfor 2 h at room temperature, washed in 0.1 M PBS andincubated in ABC reagent (Vector Labs, Burlingame, CA) for1 h at room temperature. Tissue was washed again in 0.1 MPBS and visualized using Vector SG substrate, according to themanufacturer's protocol (Vector Labs, Burlingame, CA). Sec-tions were mounted in sequential order, dried, cleared, andcoverslipped. Slides were coded and analyzed blindly toestimate the number of TH-IR cells in the SNc using Stereo-Investigator software (MicroBrightField, Inc., Williston, VT)and an Olympus BX-60 microscope. The region of interest(ROI) was first outlined at 4× magnification and 100×magnification was used for counting. The system randomlyoverlaid grids of 195 μm×87 μm on the ROI. Counting frames(40.2 μm×40.2 μm) were used for counting neurons with clearcytoplasmic staining and a distinct nucleus contained withinthe counting frame. Cell counts were only accepted with aGundersen CE of <0.1.

4.4. Protein analyses

Membrane and cytoplasmic proteins were extracted from thestriatumofAldh1a1−/− andwildtype C57Bl/6 controls (MEM-PerExtraction Kit, Pierce, Rockford, IL). Membrane protein frac-tions (for DAT and VMAT2) and cytosolic protein fractions (forTH and ALDH1A1) were separated on 4–12% polyacrylamidegels (Invitrogen, Inc.) and transferred to 0.2 μm nitrocellulosemembranes (Biorad, Hercules, CA). A rat monoclonal antibodyraised against the N-terminus of DAT (1:10,000, Millipore,Billerica, MA) and VMAT2 (1:500, Abcam, Cambridge, MA) wereused to detect membrane proteins. Rabbit polyclonal anti-bodies were used to detect the cytoplasmic proteins; TH(1:10,000, Pel-Freeze Inc., Rogers, AR), and ALDH1A1 (1:10,000,Abcam, Cambridge, MA). Blots were blocked for 1 h in Tris-buffered saline (TBS) containing 5% non fat milk and 0.1%Tween 20 for 1 h at room temperature and then washed withTBS containing 0.1% Tween 20 (TTBS). Membranes were thenincubated in appropriate secondary antibodies in TTBS for 1 hat room temperature (1:20,000, Pierce, Rockford, IL), washed inTTBS, and bands were visualized using chemiluminescence(Pierce Pico or Dura). Optical density of bands was quantifiedusing MCID software (Cambridge, UK).

4.5. Dopamine uptake

Dopamine uptake was estimated in whole striatal homoge-nates using the protocol of Lau et al. (Lau, 2003). Total uptakewasmeasured after adding 100 μl of fresh striatal homogenateto tubes containing assay buffer (50 mM Tris–HCl, pH 7.4;120 mMNaCl; 5 mMKCl; and 0.32 M sucrose), 100 μMpargyline

87B R A I N R E S E A R C H 1 4 0 8 ( 2 0 1 1 ) 8 1 – 8 7

and 0.5 μM [3H]DA (specific activity=38.7 Ci/mmol, PerkinElmer, Shelton, CT). Nonspecific uptake was assessed by theaddition of 100 μM mazindol (incubated at 37 °C for 3 min) tothe uptake assay buffer. Tubes containing samples werevortexed and incubated at 37 °C for 3 min. The reaction wasterminated by placing the tubes in ice and then samples werevacuum filtered onto Whatman GF/B filters using a BrandelCell Harvester. Following several rinses with cold assay buffer,filters were removed, dried, and added to scintillation vialscontaining 10 ml of scintillation fluid (ScintiSafe, FisherScientific, Pittsburgh, PA), and radioactivity was quantifiedby liquid scintillation spectrometry (Packard TriCarb 2200).

4.6. Data analyses

All data are reported as mean±standard error unless other-wise stated. Statistical analyses for each component studywere performed using unpaired Student's t test, two-tailed(GraphPad, Prism, La Jolla, CA).

Acknowledgments

This research was supported by the American Parkinson'sDisease Association and the F.M. Kirby Foundation, Inc.

R E F E R E N C E S

Adams Jr., J.D., Chang, M.L., Klaidman, L., 2001. Parkinson's disease—redox mechanisms. Curr. Med. Chem. 8, 809–814.

Anderson, D.W., Neavin, T., Smith, J.A., Schneider, J.S., 2001.Neuroprotective effects of pramipexole in young and agedMPTP-treated mice. Brain Res. 905, 44–53.

Burke, W.J., Li, S.W., Williams, E.A., Nonneman, R., Zahm, D.S.,2003. 3,4-Dihydroxyphenylacetaldehyde is the toxic dopaminemetabolite in vivo: implications for Parkinson's diseasepathogenesis. Brain Res. 989, 205–213.

Erwin, V.G., Deitrich, R.A., 1966. Brain aldehyde dehydrogenase.Localization, purification, and properties. J. Biol. Chem. 241,3533–3539.

Fan, X., Molotkov, A., Manabe, S.-I., Donmoyer, C.M., Deltour, L.,Foglio, M.H., Cuenca, A.E., Blaner, W.S., Lipton, S.A., Duester, G.,2003. Targeted disruption of Aldh1a1 (Raldh1) provides evidence

for a complex mechanism of retinoic acid synthesis in thedeveloping retina. Mol. Cell. Biol. 23, 4637–4648.

Galter, D., Buervenich, S., Carmine, A., Anvret, M., Olson, L., 2003.ALDH1 mRNA: presence in human dopamine neurons anddecreases in substantia nigra in Parkinson's disease and in theventral tegmental area in schizophrenia. Neurobiol. Dis. 14,637–647.

Grimm, J., Mueller, A., Hefti, F., Rosenthal, A., 2004. Molecular basisfor catecholaminergic neuron diversity. Proc. Natl. Acad. Sci.U. S. A. 101, 13891–13896.

Grunblatt, E., Mandel, S., Jacob-Hirsch, J., Zeligson, S., Amariglo, N.,Rechavi, G., Li, J., Ravid, R., Roggendorf, W., Riederer, P.,Youdim, M.B., 2004. Gene expression profiling of parkinsoniansubstantia nigra pars compacta; alterations inubiquitin-proteasome, heat shock protein, iron and oxidativestress regulated proteins, cell adhesion/cellular matrix andvesicle trafficking genes. J. Neural Transm. 111, 1543–1573.

Lamensdorf, I., Eisenhofer, G., Harvey-White, J., Nechustan, A.,Kirk, K., Kopin, I.J., 2000. 3,4-Dihydroxyphenylacetaldehydepotentiates the toxic effects of metabolic stress in PC12 cells.Brain Res. 868, 191–201.

Lau, Y.S., 2003. Measurement of dopamine uptake in neuronalcells and tissues. Methods Mol. Med. 79, 465–471.

Mandel, S., Grunblatt, E., Riederer, P., Amariglio, N., Jacob-Hirsch,J., Rechavi, G., Youdim, M.B., 2005. Gene expression profiling ofsporadic Parkinson's disease substantia nigra pars compactareveals impairment of ubiquitin–proteasome subunits, SKP1A,aldehyde dehydrogenase, and chaperone HSC-70. Ann. N. Y.Acad. Sci. 1053, 356–375.

Maring, J.A., Deitrich, R.A., Little, R., 1985. Partial purification andproperties of human brain aldehyde dehydrogenases.J. Neurochem. 45, 1903–1910.

Meyer, M.J., Mosely, D.E., Amarnath, V., Picklo Sr., M.J., 2004.Metabolism of 4-hydroxy-trans-2-nonenal by central nervoussystem mitochondria is dependent on age and NAD+

availability. Chem. Res. Toxicol. 17, 1272–1279.Olanow, C.W., Tatton, W.G., 1999. Etiology and pathogenesis of

Parkinson's disease. Annu. Rev. Neurosci. 22, 123–144.Schapira, A.H., Gu, M., Taanman, J.-W., Tabrizi, S.J., Seaton, T.,

Cleeter, M., Cooper, J.M., 1998. Mitochondria in the etiology andpathogenesis of Parkinson's disease. Ann. Neurol. 44, S89–S98.

Tank, A.W., Weiner, H., Thurman, J.A., 1981. Enzymology andsubcellular localization of aldehyde oxidation in rat liver.Oxidation of 3,4-dihydroxyphenylacetaldehyde derived fromdopamine to 3,4-dihydroxyphenylacetic acid. Biochem.Pharmacol. 30, 3265–3275.

Westerlund, M., Galter, D., Carmine, A., Olson, L., 2005. Tissue- andspecies-specific expression patterns of class I, III, and IVAdh andAldh 1 mRNAs in rodent embryos. Cell Tissue Res. 322, 227–236.