Available online at www.sciencedirect.com

www.elsevier.com/locate/brainres

b r a i n r e s e a r c h 1 4 9 1 ( 2 0 1 3 ) 2 3 – 3 3

0006-8993/$ - see frohttp://dx.doi.org/10

nCorrespondence toUniversity of Para.

E-mail addresse

Research Report

Hippocampal neuronal loss, decreased GFAPimmunoreactivity and cognitive impairment followingexperimental intoxication of rats with aluminum citrate

Ademir F. Silva Juniora, Maria Socorro S. Aguiarb, Odemir S. Carvalho Juniorb,Luana de Nazare S. Santanaa, Edna C.S. Francoa, Rafael Rodrigues Limaa,Natalino Valente M. de Siqueirac, Romulo Augusto Feiod, Lilian Rosana F. Faroe,Walace Gomes-Leala,n

aLaboratory of Experimental Neuroprotection and Neuroregeneration, Institute of Biological Sciences, Federal University of Para, BrazilbLaboratory of Psychobiology, Institute of Biological Sciences, Federal University of Para, BrazilcLaboratory of Chemical Analyses, Institute of Geosciences, Federal University of Para, BrazildLaboratory of Pharmacology, Institute of Biological Sciences, Federal University of Para, BrazileDepartment of Functional Biology and Health Sciences, Faculty of Biology, University of Vigo, Vigo, Spain

a r t i c l e i n f o

Article history:

Accepted 30 October 2012

Aluminum (Al) is a neurotoxic agent with deleterious actions on cognitive processes. Never-

theless, few studies have investigated the neuropathological effects underling the Al-induced

Available online 3 November 2012

Keywords:

Aluminum citrate

Hippocampus

Memory

Learning

Neurons

Astrocytes

nt matter & 2012 Elsevie.1016/j.brainres.2012.10.0

: Laboratory of ExperimRua Augusto Correa S/N.s: leal@ufpa.br, wgomes

a b s t r a c t

cognitive impairment. We have explored the effects of acute Al citrate intoxication on both

hippocampal morphology and mnemonic processes in rodents. Adult male Wistar rats were

intoxicated with a daily dose of Al citrate (320 mg/kg) during 4 days by gavage. Animals were

perfused at 8 (G2), 17 (G3) and 31 days (G4) after intoxication. Control animals were treated with

sodium citrate (G1). Animals were submitted to behavioral tests of open field and elevated

T-maze. Immunohistochemistry was performed to label neurons (anti-NeuN) and astrocytes

(anti-GFAP) in both CA1 and CA3 hippocampal regions. There was an increase in the locomotor

activity in open field test for G2 in comparison to control group and other groups (ANOVA-

Bonferroni, Po0.05). The elevated T-maze avoidance latency (AL) was higher in all intoxicated

groups compared to control (Po0.05) in avoidance 1. These values remained elevated in

avoidance 2 (Po0.05), but abruptly decreased in G2 and G3, but not in G1 and G4 animals in

avoidance 3 (Po0.05). There were no significant differences for 1 and 2 escape latencies. There

were intense neuronal loss and a progressive decrease in GFAP immunoreactivity in the

hippocampus of intoxicated animals. The results suggest that Al citrate treatment induces

deficits on learning and memory concomitant with neuronal loss and astrocyte impairment in

the hippocampus of intoxicated rats.

& 2012 Elsevier B.V. All rights reserved.

r B.V. All rights reserved.63

ental Neuroprotection and Neuroregeneration, Institute of Biological Sciences, FederalCampus do Guama CEP: 66075-900. Belem-Para Brasil. Fax: þ55 9132 01 7741.

leal@pq.cnpq.br (W. Gomes-Leal).

b r a i n r e s e a r c h 1 4 9 1 ( 2 0 1 3 ) 2 3 – 3 324

1. Introduction

Aluminum (Al) is ubiquitously distributed over the earth

crust and is widely used in natural processes and human

activities (Nayak, 2002). This metal is present in higher

amounts in soil surrounding waste sites from coal industries,

mining and Al smelting (Jackson and Huang, 1983). Al con-

centration in the soil is frequently increased by acid rain,

which can contribute to environmental problems and food

contamination (Jackson and Huang, 1983; Nayak, 2002). Al

has access to the body through dietary, iatrogenic and

occupational exposures (Nayak, 2002; Walton, 2007). Al can

be absorbed by plant roots grown in acid soils and is also

present in several food additives (Walton, 2007). About 20% of

Al human intake occur from using Al-maiden food utensils,

including pans, pots, kettles, and trays (Lione, 1983). It has

been shown that Al intake through the diet route is about

7–10 mg/day for adult humans (Flarend et al., 2001). Al is a

contaminant of some intravenous solutions used in some

medical conditions and in medications, including antacids,

buffered aspirins, antidiarrheal products, douches and

hemorroidal (Alfrey, 1993; Klein, 2005; Lione, 1985; Poole

et al., 2008). Finally, Al contamination may occur by occupa-

tional exposure, mainly workers of Al refining, metal indus-

tries and automotive dealerships (Bjor et al., 2008; Sinczuk-

Walczak et al., 2003, 2005).

Regardless the contamination source, Al can have access to

the body via inhalation and absorption by the olfactory system

or lung epithelia and gastrointestinal tract in the case of Al

swallowing (Exley et al., 1996; Walton, 2007). From the blood

stream, Al can accumulate in different tissues, including brain,

bone, kidneys, muscle, and heart (Anthony et al., 1986; Nayak,

2002). Brain is a preferential site of Al accumulation in both gray

and white matter, mainly at some cortical regions and hippo-

campus (Kawahara, 2005; Miu et al., 2003; Walton, 2009).

Al is a highly neurotoxic compound at higher concen-

trations. Neurotoxic actions of this metal induce clinical

symptoms in humans, including memory loss, tremor, jerking

movements, impaired coordination, and sluggish motor

movement, loss of curiosity, ataxia, myoclonic jerks, and

generalized convulsions with status epilepticus (Buchta et al.,

2003; Kiesswetter et al., 2009). These symptoms are related to

several pathological conditions, such as Alzheimer’s disease,

Down syndrome with manifested Alzheimer’s disease, amyo-

trophic lateral sclerosis, Parkinsonian dementia, dialysis ence-

phalopathy and alcohol dementia (Ribes et al., 2010; Shaw and

Petrik, 2009; Tomljenovic, 2011).

There is experimental evidence in vitro (Rui and Yongjian,

2010; Zhang et al., 2008) and in vivo (Fattoretti et al., 2004;

Sreekumaran et al., 2003; Walton, 2009) suggesting that the

hippocampus is damaged following Al intoxication. Al chloride

induced oxidative damage on cells derived from hippocampus

and cortex of mice (Rui and Yongjian, 2010). In vivo, chronic Al

intoxication induces cortical and hippocampal damage with

alterations in both axons and dendritic branches (Fattoretti

et al., 2004; Sreekumaran et al., 2003; Walton, 2009). In addition

to neuronal damage, Al intoxication may affect glial cells, mainly

astrocytes (Guo-Ross et al., 1999; Nedzvetsky et al., 2006; Struys-

Ponsar et al., 2000).

Some studies have suggested chronic Al intoxication may

induce behavioral deficits, including learning and memory

impairments (Ribes et al., 2010; Xiao et al., 2011). In these

studies, mainly chronic intoxication protocols were used and

behavioral tests were not normally associated with morpho-

logical techniques to address neuronal loss and glial reactiv-

ity (Sethi et al., 2008; Xiao et al., 2011). Although chronic

intoxications are closer to the situation in which humans are

contaminated in the polluted environment, it has been

shown that acute Al intoxication causes encephalopathy in

humans (Perazella and Brown, 1993; Berend et al., 2001) and

neuropathological changes (Forrester and Yokel, 1985; Kumar,

1998) as well as damage to other organs (El-Sayed et al., 2011)

in experimental animals.

In this study, we investigated the effects of aluminum

intoxication on hippocampal neurons and astrocytes as well

as in learning and memory performances in different survival

times following acute intoxication of adult rats with Al

citrate.

2. Results

2.1. Acute aluminum-citrate intoxication increases thewalking distance in the open field test

There was no conspicuous body weight loss in most of the

animals treated with Al citrate in relation to control animals.

Animal death was rarely present in all investigated groups

(less than 1%). Nevertheless, Al citrate treatment induced

increased (Po0.05, ANOVA-Bonferroni,) locomotor activity

(increased walking distance) in Group 2 animals compared

to control (G1) and other intoxicated groups (G3 and G4)

(Fig. 1). There was no statistical difference between G3 and G4

compared to G1 (P40.05). In addition, there were no differ-

ences (P40.05) between groups for the other open field

parameters (rearing, grooming and freezing) (Fig. 1).

2.2. Aluminum-citrate intoxication induces learning andmemory deficits in the elevated T-maze test

Two behavioral parameters were evaluated using the elevated

T-maze test: inhibitory avoidance latency (IAL) and scape

latency (SL). The IAL was significantly higher in intoxicated

animals compared to control (Fig. 2A, Po0.05, Friedman test

for repeated measures). The total averages were 29.96, 103.52,

156.75 and 75.06 s in the four attempts for baseline, avoi-

dances 1, 2 and 3, respectively (w2 (3, N¼29)¼17.43, Po0.05).

As illustrated in Fig. 2, the rat permanence time in the closed

arm significantly increased from the baseline up to avoidance

2. From avoidance 3 (memory test), there was a significant

decrease (Po0.05) in this behavioral parameter in all intoxi-

cated, but not in the control group (P40.05), characterizing a

learning deficit (Fig. 2).

The IALs were compared between groups using ANOVA.

The results are illustrated in Fig. 2B. Thirty seconds after the

baseline test, animals were submitted to avoidance 1 test.

Animals from all experimental groups presented a higher

IAL, compared to baseline (Po0.05), but there were no

differences between intoxicated groups (P40.05, Fig. 2B).

Fig. 1 – Open field results following Al citrate intoxication. Average response for control animals (G1) and animals intoxicated

with a single dose of Al citrate (320 mg/Kg) and perfused at 8 days (G2), 17 days (G3) and 31 days (G4). The open field

parameter walking distance (A), rearing (B), grooming (C) and freezing (D) were evaluated in the test. There was an increase

in the walking distance in G2 animals, compared to G1 (�Po0.05, ANOVA-Bonferroni) and G3 and G4 (þPo0.05).

Fig. 2 – Elevated-T maze results following Al citrate intoxication. Average of avoidance latency response for G1, G2, G3 and G4

animals. Differences between groups for all avoidance attempts (A) and comparative analysis of avoidance attempts inside

each experimental group (B).Statistically significant comparisons with baseline (�Po0.05) or avoidance 3 (þPo0.05).

Intoxicated animals presented differences on avoidance latencies compared to baseline and control group, suggesting

Al-induced learning and memory deficits.

b r a i n r e s e a r c h 1 4 9 1 ( 2 0 1 3 ) 2 3 – 3 3 25

b r a i n r e s e a r c h 1 4 9 1 ( 2 0 1 3 ) 2 3 – 3 326

For the avoidance 2 test, there were statistical differences

between groups. There was a reduction in the IAL for animals

belonging to G2 (98.57 s), G3 (168.00 s) and G4 (146.87 s),

compared to G1 (300.00 s) (Po0.05, Fig. 2B).

The avoidance 3 test (memory test) was performed at 3 days

after avoidance 2 test. There was a statistically significant

decrease in the IAL for G2 and G3, compared to G1 (Po0.05,

Fig. 2A), which suggests a learning deficit. Animals belonging to

G4 did not present a significant IAL decrease (P40.05).

Fig. 2B illustrates a comparative analysis for all avoidances

in each experimental group. G1 animals presented a signifi-

cant IAL increase for avoidances 1, 2 and 3 (Po0.05), com-

pared to baseline. G2 animals remained a longer time in the

closed arm in avoidance 1, while a significant reduction of

IAL was observed in avoidances 2 and 3 (Po0.05), character-

izing a memory deficit. G3 animals presented increased IALs

in avoidances 1 and 2, while a decreased IAL was observed for

avoidance 3, compared to previous attempts (Po0.05). Finally,

G4 animals did not present significant statistical differences

between avoidances (Fig. 2B, P40.05).

Fig. 3 – Hippocampal Neuronal damage following Al citrate int

and CA3 hippocampal areas for animals belonging to G1 (A)–(C),

present normal morphology (A)–(C), but morphological alteratio

immunoreactivity were observed in the intoxicated animals (G2

G, J. Scale bars: A, D, G, J (300 lm); B–F, H–I, L–M (100 lm).

The analysis of SL parameter in the open arm of the

elevated T-maze did not show any statistical differences

between groups (data not shown).

2.3. Aluminum-citrate treatment induces neuronal lossin the hippocampus of adult rats

To evaluate the effect of Al citrate treatment on neuronal

density in CA1 and CA3 hippocampal regions, we performed

immunohistochemistry against NeuN� specific neuronal

marker (Mullen et al., 1992). Al citrate treatment resulted in

conspicuous neuronal loss in both CA1 and CA3 from G2, G3

and G4 animals, compared to G1 (Fig. 3). These results were

confirmed by quantitative analysis (Fig. 4). There was a

decrease in the numbers of NeuNþ cells per field in both

CA1 and CA3 of G2, G3 and G4 animals, compared to G1

(Fig. 4, Po0.05).Comparisons between G2, G3 and G4 were no

statistically significant (P40.05).

Aluminum-citrate intoxication induces a progressive

decrease in the hippocampal GFAP reactivity.

oxication. Immunohistochemistry against NeuN in the CA1

G2 (D)–(F), G3 (G)–(I) and G4 (J)–(M). In G1, CA1 and CA3 areas

ns, including cell loss, layer thinning and decreased NeuN

, G3 and G4). B–C, E–F, H–I, L–M are magnified images of A, D,

Fig. 4 – Quantitative analysis of neuronal loss following Al

citrate intoxication. Average (7SE) of NeuN immunopositive

cells/field in both control (G1) and intoxicated animals (G2,

G3 and G4). There was a decrease in the number of NeuNþ

cells/field in intoxicated animals compared to control

(���Po0.05, ANOVA-Bonferroni) in both CA1 (A) and CA3

(B) hippocampal regions. SE¼standard error.

b r a i n r e s e a r c h 1 4 9 1 ( 2 0 1 3 ) 2 3 – 3 3 27

To investigate the effect of Al citrate intoxication on

hippocampal astrocytes, we performed immunohistochemi-

stry against GFAP—a classical astrocyte marker (Gomes-Leal

et al., 2004).There was a progressive decrease on GFAP

reactivity in animals belonging to G2, G3 and G4, compared

to G1 (Fig. 5). The loss of GFAP reactivity was more intense in

G3 (Fig. 5E and F) and G4 (Fig. 5G and H) animals, but

preferentially in the last experimental group.

3. Discussion

In this study, we have investigated the behavioral and

neuropathological effects of the acute experimental intoxi-

cation of adult rats with Al citrate. Intoxicated animals

presented learning and memory deficits as well as hippo-

campal neuronal loss and astrocytic impairment, compared

to control animals.

We have used Al citrated as intoxicant. This Al compound

is suitable for experimental Al intoxication, considering that

it allows a better Al solubility and gastrointestinal absorption

compared to other Al compounds (Arnich et al., 2004;

Domingo et al., 1988; Kumar, 1998). It follows that the choice

of Al citrate in this study certainly allowed a better Al

penetration through the blood brain barrier and deposition

in the CNS (Arnich et al., 2004; Domingo et al., 1988).

One can argue against the lack of SC-intoxicated in the

same time period of G3 and G4 animals. SC is an innocuous

substance to the CNS. This has been noted in both G1 and G2

groups. In an ongoing investigation animals injected with

sodium citrate and perfused at 28 days following Al citrate

injection did not present any histological or behavioral sign

of pathological impairment (data not shown).

Al citrated intoxication induced an increase the walking

distance, but not in the other parameters of the open field

test. Other studies using different Al compounds reported

similar results, in which Al-intoxicated animals did not

present locomotor alterations for several open field test

parameters (Domingo et al., 1996; Ribes et al., 2008; Sethi

et al., 2008). Ribes et al. (2008) reported increased open field

walking distances in transgenic mice intoxicated with Al

lactate. Similar results were described by Sethi et al., 2008

using Al chloride. In this study by Seth and colleagues,

hyperexcitability was confirmed by both behavioral tests

and electrophysiological analysis. This is likely due to

increased anxiety related to the acute phase of Al intoxica-

tion. It has been reported that Al intoxication alters some

emotional aspects of animal behavior, which is reflected in

higher open field scores (Miu et al., 2003).

Al-intoxicated animals presented differences on the IAL

compared to non-intoxicated animals. This can be inter-

preted as a memory-learning deficit induced by the Al

intoxication. Similar results have been described in previous

investigations using other Al compounds (Sethi et al., 2008;

Struys-Ponsar et al., 1997). Peripheral administration of Al

chloride caused learning and memory deficits (Sethi et al.,

2008; Struys-Ponsar et al., 1997).

We have observed that G4 animals had a higher perma-

nence time in the closed arm of the elevated-T maze in the

avoidance 1, compared to baseline and control group. It has

been shown that the animal latency time is normally shorter

in the closed arm for avoidance 1 than in the subsequent

attempts, considering that in this first test animals normally

explore more the open arms, compared the closed ones

(Graeff et al., 1998; Viana et al., 1994). It is likely that the

longer permanence of the animals in the closed arms of the

elevated-T maze was a pathological consequence of the Al-

citrate intoxication.

In avoidance 2, animals belonging to all intoxicated groups

decreased the permanence time in the closed arms. This

finding suggests an effect of Al intoxication on the short term

memory. In avoidance 3, which is performed at 72 h after

avoidance 2, intoxicated animals presented a reduced per-

manence time in the closed arms compared to control group,

which suggests deficits on the long term memory. It is known

that the long term memory is consolidated in some hours

and can be recovered in days or months after the aversive

stimulus (Antonov et al., 2010; Izquierdo and Medina, 1997;

Izquierdo et al., 1999). Considering that intoxicated animals

had difficulties on the avoidance learning, it is likely that Al

caused impairment in the learning consolidation of the

Fig. 5 – Decreased GFAP immunoreactivity following Al citrate intoxication. Immunohistochemistry against GFAP in the CA1

and CA3 hippocampal areas for animals belonging to G1 (A)–(B), G2 (C)–(D), G3 (E)–(F) and G4 (G)–(H). In G1, GFAPþastrocytes

present a normal morphology and pattern of immunoreactivity (A)–(B). A considerable decrease of GFAP immunoreactivity

was observed in the intoxicated animals. B, D, F, H are magnified images of A, C, E, G. A, C, E, G (300 lm); B, D, F, H (300 lm).

b r a i n r e s e a r c h 1 4 9 1 ( 2 0 1 3 ) 2 3 – 3 328

avoidance response, which can be considered a long term

memory deficit.

Other studies have reported learning and memory deficits

following intoxication with Al and other heavy metals, including

methylmercury (Carobrez and Bertoglio, 2005; Kaneko et al.,

2006; Maia et al., 2009; Ribes et al., 2010; Xiao et al., 2011).

In these studies, animals contaminated with ethanol and

methylmercury presented learning impairment, characterized

by reduced permanence time in the closed arms of the elevated-

T maze in the first avoidance response, characterizing short term

memory deficits (Carobrez and Bertoglio, 2005; Maia et al., 2009).

This has also been observed in avoidance 3, performed more

than 24 h after avoidance 1, characterizing long term memory

deficits (Carobrez and Bertoglio, 2005; Maia et al., 2009).

b r a i n r e s e a r c h 1 4 9 1 ( 2 0 1 3 ) 2 3 – 3 3 29

The behavioral effects above described could be related to

damage induced by Al citrate on CNS regions important for

learning and memory. We have investigated this possibility

by assessing neuronal density and GFAP immunoreactivity in

both CA1 and CA3 hippocampal regions. Decreased densities

of CA1 and CA3 neurons as well as progressive decrease on

GFAP immunoreactivity were observed in Al-intoxicated ani-

mals. These findings are supported by previous studies

in vitro and in vivo showing hippocampal damage following

Al intoxication (El-Rahman, 2003; Levesque et al., 2000;

Meshitsuka and Aremu, 2008; Sethi et al., 2008).

Chronic intoxication with aluminum sulfate induces dis-

organization of the pyramidal hippocampal layer, neuronal

loss and neurofibrillary degeneration in adult rats (El-

Rahman, 2003; Sethi et al., 2008). Other CNS areas are affected

by Al intoxication, including neocortex, olfactory bulb,

hypothalamus, striatum and cerebellum, although the hip-

pocampus is preferentially affected (Struys-Ponsar et al.,

1994; Sumathi et al., 2011).

The neurotoxic effects of Al compounds are related to a

preferential tropism of this metal for CNS structures (Oteiza

et al., 1993; Zatta et al., 1993). In the CNS, Al can interfere

with several physiological processes, inducing damage by

oxidative stress, membrane biophysics alterations, deregula-

tion of cell signaling, and impairment of neurotransmission

(Pohl et al., 2011; Verstraeten et al., 2008).

The Al citrate intoxication induced a progressive decrease

on GFAP immunoreactivity. Previous studies reported that Al

intoxication interferes with astrocyte function (Erazi et al.,

2010). Nevertheless, the Al effect on astrocyte population is

variable depending on the experimental model. Some reports

suggest that Al intoxication may decrease astrocyte function

(Guo-Ross et al., 1999; Struys-Ponsar et al., 2000; Suarez-

Fernandez et al., 1999), while other suggest the opposite

(Nedzvetsky et al., 2006; Yokel and O’Callaghan, 1998;Sethi

et al., 2008). In this study, it was clear that Al citrate

intoxication induces astrocyte damage. The impairment

in astrocyte function may contribute to neuronal damage.

In vitro studies suggest that astrocyte damage contribute

to neuronal loss (Aremu and Meshitsuka, 2005; Meshitsuka

and Aremu, 2008; Sass et al., 1993). It has been suggested that

Al treatment impair astrocytes to protect neurons from

glutamate-induced excitotoxicity (Sass et al., 1993). The

progressive accumulation of Al in astrocytes may protect

neurons in early times after intoxication. Nevertheless, the

pathological accumulation of Al may damage astrocytes in

later time points. This is in agreement with the described loss

of GFAP immunoreactivity in later survival times following Al

intoxication.

An apparent drawback of our study is the lack of quanti-

tative analysis for GFAP immunoreactivity. Nevertheless,

changes in GFAP immunoreactivity can be qualitatively

addressed, depending on the experimental paradigm. We

have performed qualitative analysis of GFAP immunoreactiv-

ity following spinal cord injury (Gomes-Leal et al., 2004),

showing important differences on GFAP immunoreactivity

between gray and white matter in this acute neural disorder.

Sometimes, increased GFAP immunoreactivity is due to a

better exposure of epitopes rather than real increase in

astrocyte number. We have shown a conspicuous decrease

in GFAP immunoreactivity (Fig. 5). This dramatic decrease in

GFAP immunoreactivity rendered very difficult to recognize

cell bodies of GFAPþ cells for counting. We believe that in this

case the qualitative analysis is enough to illustrate the

astrocyte alteration.

Further studies, using electron microscopy and double

immunofluorescence for co-labeling cell death and glial as

well as neuronal markers should be performed to confirm

this hypothesis.

The hippocampus is an important neurogenic region in the

adult brain (Aimone et al., 2011). It has been suggested that

the hippocampal adult neurogenesis may contribute to

mechanisms of learning and memory (Aimone et al., 2011;

Koehl and Abrous, 2011; Sahay et al., 2011). There is a

possibility that Al intoxication may impair hippocampal

adult neurogenesis, contributing to learning and memory

deficits. This hypothesis should be investigated in further

studies.

4. Experimental procedures

4.1. Experimental animals

Male adult Wistar rats (250–300 g) were obtained from the

Federal University of Para Central Animal Facility. All animals

were housed under standard conditions (25 1C, 12 h light–

dark cycle) with food and water available ad libitum. All

experimental procedures were carried out in accordance with

the Principles of Laboratory Animal Care (NIH publication No.

86–23, revised 1985) and European Commission Directive

86/609/EEC for animal experiments under license of the

Ethics Committee on Experimental Animals of the Federal

University of Para. All possible efforts were made to avoid

animal suffering and distress.

4.2. Preparation of aluminum citrate solution

The Al citrate solution was prepared in the Laboratory of

Chemical Analysis from the Geoscience Institute of Federal

University of Para. In short, 14.23 g of citric acid plus 17.89 g

Al chloride were mixed in deionized water. The solution pH

was adjusted up to 7.0 using ammonia hydroxide. The

solution was stirred under 60 1C for some minutes and a

final volume of 500 mL obtained by adding deionized water.

4.3. Al citrate intoxication and experimental groups

Animals were intoxicated with a daily single dose of Al citrate

(320 mg/kg in 1 and 2 mL of deionized water) by gavage

during 4 days. Each animal received 1280 g of Al at the end

of treatment. Control animals received sodium citrate in the

same volume. Experimental animals were perfused 8, 17 and

31 days after the last Al citrate or sodium citrate doses. The

experimental groups are described in Table 1. The intoxica-

tion protocol followed a previous published paper (Kumar,

1998). The chosen survival times allow us to investigate both

behavioral and histological impairment in the first month

after Al intoxication.

Table 1 – Experimental groups, aluminum dose and survival times.

Groups Treatment Dose (mg/kg) N1 of doses Behavioral test time (days) Survival times (days)

1 SC� 320 4 1(n¼7) 8(n¼7)

2 AC� 320 4 1(n¼7) 8(n¼7)

3 AC 320 4 10(n¼7) 17(n¼7)

4 AC 320 4 24(n¼7) 31(n¼8)

n SC¼Sodium citrate; AC¼Aluminum citrate.

b r a i n r e s e a r c h 1 4 9 1 ( 2 0 1 3 ) 2 3 – 3 330

4.4. Behavioral tests

Behavioral tests were performed in order to evaluate how Al

intoxication influenced both locomotor and learning and

memory performances in the different experimental groups.

They were performed at 24 h after the last dose of Al citrate

or Sodium citrate for animals belonging to G1 and G2 during 3

consecutive days. Animals belonging to G3 and G4 were

submitted to behavioral analysis at 10 and 24 days after the

last Al dose, respectively. The behavioral tests were per-

formed during 3 consecutive days, like previously described

for G1 and G2 animals.

The following behavioral testes were used:

4.5. Open field

In this behavioral test, we followed a protocol published

elsewhere (Bresnahan et al., 1987). This test was used in

order to evaluate possible toxic effects of Al citrate that could

affect locomotor performance of the animals, which could

influence their performance in specific behavioral tests to

address Al-induced learning and memory deficits.

The test was performed in an open field with 60�60�50

cm dimensions and containing 25 square subdivisions of

equal size. Experiments consisted of three trials (5 min each)

in which the animal was removed from its cage and placed at

the center of the open field. The animals’ motor performance

was recorded by a video camera (Sony, USA) and analyzed by

the software Any Maze Stoeltings. The behavioral para-

meters recorded included the number of occurrences of the

exploratory behavior of standing up on the hind legs (rear-

ing), body self-cleaning (grooming), the rat behavior in which

he becomes static for about 10 s (freezing), latency (time

taken to leave the starting point) and distance travelled in

the open field.

4.6. Elevated T-maze

In this test, we followed a previously published protocol,

which is routinely used by our group (Viana et al., 1994; Graeff

et al., 1998). This behavioral test addresses a kind of emo-

tional memory of the animal, related to the environment

imposed by the open arm of the elevated-T maze apparatus,

which is bright and high (Viana et al., 1994; Graeff et al., 1998).

Rodents are aversive to these characteristics and keep a vivid

memory of the aversive situation (time in the open arm)

during the test (Viana et al., 1994; Graeff et al., 1998).

Immediately following the open field test, animals were

maintained at the end of the closed arm with their heads

turned toward the mazes center. The time (latency) used for

the animal to get out from the closed arm was recorded over

300 s (baseline). Further, the first attempt of inhibitory avoid-

ance (IA1) was performed followed by a second IA 30 s later

(IA2). The escape test was performed following the IA 1 and 2

tests. In this test, the time used for the animal to get out from

the open arm was recorded. The tests for memory retention

of IA and escape in the elevated-T maze were performed 72 h

later. At this time, one more IA and escape tests were

performed, as previously described.

4.7. Perfusion and histological analysis

Following the described survival times, animals were deeply

anesthetized with ketamine hydrochloride (72 mg/kg, i.p.)

and xylazine hydrochloride (9 mg/kg, i.p.) and transcardially

perfused with heparinized 0.9% phosphate-buffered saline

(PBS) followed by 4% paraformaldehyde. Surgical manipula-

tion was performed only after both the corneal and the paw

withdraw reflexes were abolished. Brains were post-fixed for

24 h in the same fixative and cryoprotected in different

gradients of sucrose–glycerol solutions over 7 days. The

tissue was then frozen in Tissue Tek, and 30 mm coronal

sections were cut using a cryostat (Carl Zeiss Micron,

Germany). Sections were mounted onto gelatinized slides

and stored in a freezer at �20 1C.

4.8. Immunohistochemistry

To investigate the effect of Al citrate intoxication on hippo-

campal neurons and astrocytes in the different experimental

groups, we performed a series of immunohistochemical

procedures. Mature neuronal bodies (Mullen et al., 1992) were

recognized by the antibody anti-NeuN (1:100, Chemicon-

Millipore, USA). Astrocytosis (Gomes-Leal et al., 2004) was

evaluated by using an antibody anti-GFAP (1:1000, DAKO,

USA). The GFAP immunoreactivity was used to qualitatively

evaluate the effects of Al intoxication on Astrocytes. The

intensity of GFAP immunolabeling and alterations on astro-

cyte morphology were used as the main criteria for addres-

sing astrocyte impairment (Gomes-Leal et al., 2004).

The Immunolabeling protocol used in this study was

detailed elsewhere (Gomes-Leal et al., 2004). Briefly, slide-

mounted sections were removed from the freezer, kept in a

heating oven at 37 1C for 30 min and rinsed once in 0.1 M PBS

for 5 min. To improve labeling intensity, sections were then

pretreated in 0.2 M boric acid (pH 9.0) previously heated to

65 1C for 25 min. This temperature was maintained constant

over the pretreatment period. Sections were further allowed

b r a i n r e s e a r c h 1 4 9 1 ( 2 0 1 3 ) 2 3 – 3 3 31

to cool for about 20 min and were incubated under constant

agitation in 1% hydrogen peroxide in methanol for 20 min.

Sections were rinsed 3 times (5 min each) in 0.05% PBS/Tween

(Sigma, Saint Louis, MO, USA) and incubated with normal

serum in PBS for 1 h. Without further rinsing, sections were

then incubated with the primary antibody diluted in PBS for

24 h, rinsed in PBS/Tween solution for 5 min (3 times), and

incubated with appropriate secondary antibody for 2 h. As a

negative control, PBS, rather than the primary antibody, was

used. Sections were rinsed again for 5 min (3 times) and

incubated in an avidin – biotin – peroxidase complex (ABC

Kit, Vector Laboratories) for 2 h. Sections were then rinsed 4

times (3 min each rinse) and DAB-reacted according to a

protocol published elsewhere (Gomes-Leal et al., 2004). After

the DAB reaction, sections were rinsed 3 times (3 min each) in

0.1 M phosphate buffer, dehydrated using alcohols and

xylene, and coverslipped. Some sections were also counter-

stained with cresyl violet.

4.9. Qualitative analysis

All sections stained with the different histological methods

were surveyed by light microscopy (Olympus BX41). Illustra-

tive images from all experimental groups were obtained

using a digital camera (Olympus Evolt E-330) attached to the

microscope.

4.10. Quantitative analysis

We used 3 adjacent coronal sections (per animal) from the

brain of animals belonging to all investigated experimental

group to count the number of neuronal bodies (NeuNþ cells)

per field using a rectangular 0.075�0.25 mm grid (objective

40X) in the eyepiece of a microscope. The coronal sections

contained the anterior hippocampus and were located

at �3.60 mm posterior to Bregma (Paxinos et al., 1980).

In the 40X objective, this grid corresponds to an area of

0.01875 mm2. We grid was placed in both CA1 and CA3

hippocampal regions and 3 adjacent fields per (Section 3

sections/animal and 8 animals/survival time) were counted.

Cell bodies were counted in specific grid square, whey they

were more than 50% inside the square. Otherwise, they were

included in the adjacent square. Only clearly immunolabeled

cell bodies were counted, according to the guidelines pub-

lished in our previous investigations (Franco et al., 2012). The

countings were performed blinded to the experimenter.

4.11. Statistical analysis

Averages and standard errors were calculated for all counts.

Comparisons between different groups were assessed by

analysis of variance (ANOVA) with Bonferroni post-hoc test

for both histological and behavioral data. In addition, Fried-

man test for repeated measures and Kruskall–Wallis test

were used to confirm the behavioral results. Statistical sig-

nificance was accepted for Po0.05. All statistical analyses

were performed using the BioEstat 5.0 software (Sociedade

Civil Mamiraua/CNPQ-Brazil).

5. Conclusion

In conclusion, acute Al citrate intoxication induces conspic-

uous learning and memory deficits as well as neuronal loss

and decreased GFAP immunoreactivity in the hippocampus

of adult rats. These pathological alterations likely underlie

the behavioral impairment described. Further studies

should investigate the effects of Al intoxication on other glial

cells, for example microglia, as well as the use of anti-

inflammatory and anti-oxidants in order to decrease both

hippocampal damage and behavioral deficits. In addition, the

effect of aging must be investigated, considering that hippo-

campal impairment is present in elderly people with Alzhei-

mer disease.

Acknowledgments

This work was supported by the Brazilian National Council for

Scientific and Technological Development (CNPq) and Fundac- ~ao

de Amparo A Pesquisa do Estado do Para (FAPESPA). We are

grateful to PROPESP-UFPA for further financial support.

r e f e r e n c e s

Aimone, J.B., Deng, W., Gage, F.H., 2011. Resolving new memories:a critical look at the dentate gyrus, adult neurogenesis, andpattern separation. Neuron 70, 589–596.

Alfrey, A.C., 1993. Aluminum toxicity in patients with chronicrenal failure. Ther. Drug Monit. 15, 593–597.

Anthony, J., Fadl, S., Mason, C., Davison, A., Berry, J., 1986.Absorption, deposition and distribution of dietary aluminiumin immature rats: effects of dietary vitamin D3 and food-bornechelating agent. J. Environ. Sci. Health., Part B 21, 191–205.

Antonov, I., Kandel, E.R., Hawkins, R.D., 2010. Presynaptic andpostsynaptic mechanisms of synaptic plasticity and meta-plasticity during intermediate-term memory formation inAplysia. J. Neurosci. 30, 5781–5791.

Aremu, D.A., Meshitsuka, S., 2005. Accumulation of aluminum byprimary cultured astrocytes from aluminum amino acidcomplex and its apoptotic effect. Brain Res. 1031, 284–296.

Arnich, N., Cunat, L., Lanhers, M.C., Burnel, D., 2004. Comparativein situ study of the intestinal absorption of aluminum,manganese, nickel, and lead in rats. Biol. Trace Elem. Res.99, 157–171.

Berend, K., van der Voet, G., Boer, W.H., 2001. Acute aluminumencephalopathy in a dialysis center caused by a cementmortar water distribution pipe. Kidney Int. 59, 746–753.

Bjor, O., Damber, L., Edstrom, C., Nilsson, T., 2008. Long-termfollow-up study of mortality and the incidence of cancer in acohort of workers at a primary aluminum smelter in Sweden.Scand. J. Work Environ. Health 34, 463–470.

Bresnahan, J.C., Beattie, M.S., Todd 3rd, F.D., Noyes, D.H., 1987.A behavioral and anatomical analysis of spinal cord injuryproduced by a feedback-controlled impaction device. Exp.Neurol. 95, 548–570.

Buchta, M., Kiesswetter, E., Otto, A., Schaller, K.H., Seeber, A.,Hilla, W., Windorfer, K., Stork, J., Kuhlmann, A., Gefeller, O.,Letzel, S., 2003. Longitudinal study examining the neurotoxicity ofoccupational exposure to aluminium-containing welding fumes.Int. Arch. Occup. Environ. Health 76, 539–548.

b r a i n r e s e a r c h 1 4 9 1 ( 2 0 1 3 ) 2 3 – 3 332

Carobrez, A.P., Bertoglio, L.J., 2005. Ethological and temporalanalyses of anxiety-like behavior: the elevated plus-mazemodel 20 years on. Neurosci. Biobehav. Rev. 29, 1193–1205.

Domingo, J.L., Gomez, M., Llobet, J.M., Corbella, J., 1988. Com-parative effects of several chelating agents on the toxicity,distribution and excretion of aluminium. Hum. Toxicol. 7,259–262.

Domingo, J.L., Llorens, J., Sanchez, D.J., Gomez, M., Llobet, J.M.,Corbella, J., 1996. Age-related effects of aluminum ingestionon brain aluminum accumulation and behavior in rats. LifeSci. 58, 1387–1395.

El-Sayed, W.M, Al-Kahtani, M.A, Abdel-Moneim, A.M., 2011.Prophylactic and therapeutic effects of taurine againstaluminum-induced acute hepatotoxicity in mice 192, 880–886.

El-Rahman, S.S., 2003. Neuropathology of aluminum toxicity inrats (glutamate and GABA impairment). Pharmacol. Res. 47,189–194.

Erazi, H., Sansar, W., Ahboucha, S., Gamrani, H., 2010. Aluminumaffects glial system and behavior of rats. C R Biol. 333, 23–27.

Exley, C., Burgess, E., Day, J.P., Jeffery, E.H., Melethil, S., Yokel, R.A.,1996. Aluminum toxicokinetics. J. Toxicol. Environ. Health 48,569–584.

Fattoretti, P., Bertoni-Freddari, C., Balietti, M., Giorgetti, B.,Solazzi, M., Zatta, P., 2004. Chronic aluminum administrationto old rats results in increased levels of brain metal ions andenlarged hippocampal mossy fibers. Ann. N.Y. Acad. Sci. 1019,44–47.

Flarend, R., Bin, T., Elmore, D., Hem, S.L., 2001. A preliminarystudy of the dermal absorption of aluminium from anti-perspirants using aluminium-26. Food Chem. Toxicol. 39,163–168.

Forrester, T.M., Yokel, R.A., 1985. Comparative toxicity of intra-cerebroventricular and subcutaneous aluminum in the rabbit.Neurotoxicology 6, 71–80.

Franco, E.C.S, Gouveia, A., Pereira, A., Gomes-Leal, W., 2012.Modulation of microglial activation enhances neuroprotectionand functional recovery derived from bone marrow mono-nuclear cell transplantation after cortical ischemia. Neurosci.Res. 73, 122–132.

Gomes-Leal, W., Corkill, D.J., Freire, M.A., Picanco-Diniz, C.W.,Perry, V.H., 2004. Astrocytosis, microglia activation, oligoden-drocyte degeneration, and pyknosis following acute spinalcord injury. Exp. Neurol. 190, 456–467.

Graeff, F.G., Netto, C.F., Zangrossi Jr., H., 1998. The elevatedT-maze as an experimental model of anxiety. Neurosci.Biobehav. Rev. 23, 237–246.

Guo-Ross, S.X., Yang, E.Y., Walsh, T.J., Bondy, S.C., 1999. Decreaseof glial fibrillary acidic protein in rat frontal cortex followingaluminum treatment. J. Neurochem. 73, 1609–1614.

Izquierdo, I., Medina, J.H., 1997. Memory formation: the sequenceof biochemical events in the hippocampus and its connectionto activity in other brain structures. Neurobiol. Learn. Mem.68, 285–316.

Izquierdo, I., Medina, J.H., Vianna, M.R., Izquierdo, L.A., Barros, D.M.,1999. Separate mechanisms for short- and long-term memory.Behav. Brain Res. 103, 1–11.

Jackson, M.L., Huang, P.M., 1983. Aluminum of acid soils in thefood chain and senility. Sci. Total Environ. 28, 269–276.

Kaneko, N., Takada, J., Yasui, H., Sakurai, H., 2006. Memory deficitin mice administered aluminum-maltolate complex. Biome-tals 19, 83–89.

Kawahara, M., 2005. Effects of aluminum on the nervous systemand its possible link with neurodegenerative diseases. J.Alzheimers Dis. 8, 171–182.

Kiesswetter, E., Schaper, M., Buchta, M., Schaller, K.H., Rossbach, B.,Kraus, T., Letzel, S., 2009. Longitudinal study on potentialneurotoxic effects of aluminium: II. Assessment of exposureand neurobehavioral performance of Al welders in the

automobile industry over 4 years. Int. Arch. Occup. Environ.Health 82, 1191–1210.

Klein, G.L., 2005. Aluminum: new recognition of an old problem.Curr. Opin. Pharmacol. 5, 637–640.

Koehl, M., Abrous, D.N., 2011. A new chapter in the field ofmemory: adult hippocampal neurogenesis. Eur. J. Neurosci.33, 1101–1114.

Kumar, S., 1998. Biphasic effect of aluminium on cholinergicenzyme of rat brain. Neurosci. Lett. 248, 121–123.

Levesque, L., Mizzen, C.A., McLachlan, D.R., Fraser, P.E., 2000.Ligand specific effects on aluminum incorporation and toxi-city in neurons and astrocytes. Brain Res. 877, 191–202.

Lione, A., 1983. The prophylactic reduction of aluminium intake.Food Chem Toxicol. 21, 103–109.

Lione, A., 1985. Aluminum toxicology and the aluminum-containing medications. Pharmacol. Ther. 29, 255–285.

Maia, C.S., Lucena, G.M., Correa, P.B., Serra, R.B., Matos, R.W.,Menezes, F.C., Santos, S.N., Sousa, J.B., Costa, E.T., Ferreira, V.M.,2009. Interference of ethanol and methylmercury in the develop-ing central nervous system. Neurotoxicology 30, 23–30.

Meshitsuka, S., Aremu, D.A., 2008. (13)C heteronuclear NMRstudies of the interaction of cultured neurons and astrocytesand aluminum blockade of the preferential release of citratefrom astrocytes. J. Biol. Inorg. Chem. 13, 241–247.

Miu, A.C., Andreescu, C.E., Vasiu, R., Olteanu, A.I., 2003. Abehavioral and histological study of the effects of long-termexposure of adult rats to aluminum. Int. J. Neurosci. 113,1197–1211.

Mullen, R.J., Buck, C.R., Smith, A.M., 1992. NeuN, a neuronal specificnuclear protein in vertebrates. Development 116, 201–211.

Nayak, P., 2002. Aluminum: impacts and disease. Environ. Res. 89,101–115.

Nedzvetsky, V.S., Tuzcu, M., Yasar, A., Tikhomirov, A.A., Baydas, G.,2006. Effects of vitamin E against aluminum neurotoxicity inrats. Biochemistry. 71, 239–244.

Oteiza, P.I., Keen, C.L., Han, B., Golub, M.S., 1993. Aluminumaccumulation and neurotoxicity in Swiss-Webster mice afterlong-term dietary exposure to aluminum and citrate. Meta-bolism 42, 1296–1300.

Paxinos, G, Watson, CR, Emson, PC., 1980. AChE-stained horizon-tal sections of the rat brain in stereotaxic coordinates.J. Neurosci. Methods 3, 129–149.

Perazella, M., Brown, E., 1993. Acute aluminum toxicity and alumbladder irrigation in patients with renal failure. Am. J. KidneyDis. 21, 44–46.

Pohl, H.R., Roney, N., Abadin, H.G., 2011. Metal ions affecting theneurological system. Met. Ions Life Sci. 8, 247–262.

Poole, R.L., Hintz, S.R., Mackenzie, N.I., Kerner Jr., J.A., 2008.Aluminum exposure from pediatric parenteral nutrition:meeting the new FDA regulation. JPEN J. Parenter. Enteral.Nutr. 32, 242–246.

Ribes, D., Colomina, M.T., Vicens, P., Domingo, J.L., 2008. Effects oforal aluminum exposure on behavior and neurogenesis in atransgenic mouse model of Alzheimer’s disease. Exp. Neurol.214, 293–300.

Ribes, D., Colomina, M.T., Vicens, P., Domingo, J.L., 2010. Impairedspatial learning and unaltered neurogenesis in a transgenicmodel of Alzheimer’s disease after oral aluminum exposure.Curr. Alzheimer Res. 7, 401–408.

Rui, D., Yongjian, Y., 2010. Aluminum chloride induced oxidativedamage on cells derived from hippocampus and cortex of ICRmice. Brain Res. 1324, 96–102.

Sahay, A., Wilson, D.A., Hen, R., 2011. Pattern separation: acommon function for new neurons in hippocampus andolfactory bulb. Neuron 70, 582–588.

Sass, J.B., Ang, L.C., Juurlink, B.H., 1993. Aluminum pretreatmentimpairs the ability of astrocytes to protect neurons fromglutamate mediated toxicity. Brain Res. 621, 207–214.

b r a i n r e s e a r c h 1 4 9 1 ( 2 0 1 3 ) 2 3 – 3 3 33

Sethi, P., Jyoti, A., Singh, R., Hussain, E., Sharma, D., 2008.Aluminium-induced electrophysiological, biochemical andcognitive modifications in the hippocampus of aging rats.Neurotoxicology 29, 1069–1079.

Shaw, C.A., Petrik, M.S., 2009. Aluminum hydroxide injectionslead to motor deficits and motor neuron degeneration. J. Inorg.Biochem. 103, 1555–1562.

Sinczuk-Walczak, H., Szymczak, M., Razniewska, G., Matczak, W.,Szymczak, W., 2003. Effects of occupational exposure to alumi-num on nervous system: clinical and electroencephalographicfindings. Int. J. Occup. Med. Environ. Health 16, 301–310.

Sinczuk-Walczak, H., Matczak, W., Razniewska, G., Szymczak, M.,2005. Neurologic and neurophysiologic examinations of work-ers occupationally exposed to aluminium. Med. Pr. 56, 9–17.

Sreekumaran, E., Ramakrishna, T., Madhav, T.R., Anandh, D.,Prabhu, B.M., Sulekha, S., Bindu, P.N., Raju, T.R., 2003. Loss ofdendritic connectivity in CA1, CA2, and CA3 neurons inhippocampus in rat under aluminum toxicity: antidotal effectof pyridoxine. Brain Res. Bull. 59, 421–427.

Struys-Ponsar, C., Florence, A., Gauthier, A., Crichton, R.R., vanden Bosch de Aguilar, P., 1994. Ultrastructural changes in brainparenchyma during normal aging and in animal models ofaging. J. Neural. Transm. Suppl. 44, 111–132.

Struys-Ponsar, C., Kerkhofs, A., Gauthier, A., Soffie, M., van denBosch de Aguilar, P., 1997. Effects of aluminum exposure onbehavioral parameters in the rat. Pharmacol. Biochem. Behav.56, 643–648.

Struys-Ponsar, C., Guillard, O., van den Bosch de Aguilar, P., 2000.Effects of aluminum exposure on glutamate metabolism: apossible explanation for its toxicity. Exp. Neurol. 163, 157–164.

Suarez-Fernandez, M.B., Soldado, A.B., Sanz-Medel, A., Vega, J.A.,Novelli, A., Fernandez-Sanchez, M.T., 1999. Aluminum-induced degeneration of astrocytes occurs via apoptosis andresults in neuronal death. Brain Res. 835, 125–136.

Sumathi, T., Shobana, C., Kumari, B.R., Nandhini, D.N., 2011.Protective role of Cynodon dactylon in ameliorating thealuminium-induced neurotoxicity in rat brain regions. Biol.Trace Elem. Res. 144, 843–853.

Tomljenovic, L., 2011. Aluminum and Alzheimer’s disease: after acentury of controversy, is there a plausible link? J. AlzheimersDis. 23, 567–598.

Verstraeten, S.V., Aimo, L., Oteiza, P.I., 2008. Aluminium and lead:molecular mechanisms of brain toxicity. Arch. Toxicol. 82,789–802.

Viana, M.B., Tomaz, C., Graeff, F.G., 1994. The elevated T-maze: anew animal model of anxiety and memory. Pharmacol. Bio-chem. Behav. 49, 549–554.

Walton, J.R., 2007. A longitudinal study of rats chronicallyexposed to aluminum at human dietary levels. Neurosci. Lett.412, 29–33.

Walton, J.R., 2009. Brain lesions comprised of aluminum-rich cellsthat lack microtubules may be associated with the cognitivedeficit of Alzheimer’s disease. Neurotoxicology 30, 1059–1069.

Xiao, F., Li, X.G., Zhang, X.Y., Hou, J.D., Lin, L.F., Gao, Q., Luo, H.M.,2011. Combined administration of D-galactose and aluminiuminduces Alzheimer-like lesions in brain. Neurosci. Bull. 27,143–155.

Yokel, R.A., O’Callaghan, J.P., 1998. An aluminum-inducedincrease in GFAP is attenuated by some chelators. Neurotox-icol. Teratol. 20, 55–60.

Zatta, P., Favarato, M., Nicolini, M., 1993. Deposition of aluminumin brain tissues of rats exposed to inhalation of aluminumacetylacetonate. Neuroreport 4, 1119–1122.

Zhang, Q.L., Boscolo, P., Niu, P.Y., Wang, F., Shi, Y.T., Zhang, L.,Wang, L.P., Wang, J., Di Gioacchino, M., Conti, P., Li, Q.Y.,Niu, Q., 2008. How do rat cortical cells cultured with aluminumdie: necrosis or apoptosis? Int. J. Immunopathol. Pharmacol. 21,107–115.