Toxicology 223 (2006) 209–218

Urinary concentrating mechanism and Aquaporin-2 abundancein rats chronically treated with aluminum lactate

Stella Mahieu a, Nestor Millen a, Marıa del Carmen Contini a, Marcela Gonzalez a,Sara M. Molinas b, Marıa Monica Elıas b,∗

a Fisiologıa Humana, Facultad de Bioquımica y Ciencias Biologicas, Universidad Nacional del Litoral, Paraje El Pozo,3000 Santa Fe, Argentina

b Farmacologıa, Departamento de Ciencias Fisiologicas, Facultad de Ciencias Bioquımicas y Farmaceuticas,Universidad Nacional de Rosario, Consejo Nacional de Investigaciones Cientıficas y Tecnologicas (CONICET),

Suipacha 531, 2000 Rosario, Argentina

Received 3 January 2006; received in revised form 3 March 2006; accepted 26 March 2006Available online 4 April 2006

Abstract

The aim of this work was to study the effects of chronic administration of aluminum (Al) on the urinary concentrating and dilutingmechanisms in the distal tubules and collecting ducts. Male Wistar rats were chronically treated with aluminum lactate for 12 weeks(0.575 mg Al/100 g of body weight, i.p., three times per week). After 12 weeks, renal function of control and Al-treated rats wasevaluated by clearance techniques. To study urinary concentrating mechanisms, renal function was also measured in control andAl-treated rats deprived of water, after the administration of desmopressin (vasopressin agonist) and after the infusion of hypertonicssSAAht©

K

1

a

f

m

0

aline at increasing infusion rates. Sodium and water balance were impaired. We found decreased urinary concentrating ability inituations in which endogenous (thirst or infusion of hypertonic saline) or exogenous plasma antidiuretic hormone was increased.olute-free water formation, measured during the infusion of hypotonic saline showed normal transport in the thick ascending limb.quaporin-2 (AQP2) expression was measured by Western blot to evaluate water permeability in collecting ducts. We found thatl produced downregulation of AQP2 in plasma membranes and intracellular vesicles, that could account for the impaired waterandling. Administration of desmopressin increased AQP2 in plasma membranes, suggesting that Al did not impair trafficking ofhis protein, but could interfere with AQP2 synthesis.

2006 Elsevier Ireland Ltd. All rights reserved.

eywords: Aluminum; Renal failure; Urine concentration and dilution; Aquaporin-2

. Introduction

The effect of renal failure on aluminum (Al)ccumulation in different organs and the subsequent

∗ Corresponding author. Tel.: +54 341 4393400;ax: +54 341 4804598.

E-mail addresses: smahieu@fbcb.unl.edu.ar (S. Mahieu),elias@fbioyf.unr.edu.ar (M.M. Elıas).

systemic toxicity, has been well studied, but thereare few reports about Al nephrotoxicity. Al is mainlyexcreted in urine and different mechanisms of renalAl elimination such as glomerular filtration (Yokeland McNamara, 1985), tubular reabsorption of filteredAl (Burnatowska-Hledin et al., 1985) and excretionin the distal tubules (Monteagudo et al., 1988) havebeen suggested. Moreover, it was also reported thatAl accumulated in kidney promotes degeneration inrenal tubular cells and thus could induce nephrotoxicity

300-483X/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.tox.2006.03.018

210 S. Mahieu et al. / Toxicology 223 (2006) 209–218

(Ebina et al., 1984; Cacini and Yokel, 1988; Bertholf etal., 1989; Stein et al., 1987; Roy et al., 1991).

In previous works we have reported that chronicadministration of Al by parenteral exposure increased therenal tubular phosphate reabsorption, with an increase inthe apparent maximal velocity without modifications inthe apparent affinity (Km) (Mahieu and Calvo, 1998).We also described that Al accumulated in renal tis-sue due to chronic administration modified other tubu-lar transport mechanisms, such as the organic aniontransport system (Mahieu et al., 2003). This transportsystem plays a critical role in protecting against thetoxic effects of anionic substances and it is a targetfor the actions of many drugs and environmental tox-icants (Sweet et al., 2001). Moreover, data obtainedin cortical homogenates from Al-treated rats indicatedthat both glutathione and glutathione-S-tranferase activ-ity were significantly decreased, while lipoperoxidationwas significantly increased (Mahieu and Calvo, 1998).These results reinforce the pro-oxidant role of Al (Exley,2004).

Other results from our laboratory indicated thatchronic administration of Al affected simultaneously thephosphorous and calcium balance. The administrationof calcitriol to Al-treated rats normalized phosphorousbalance. These findings suggest that Al could modifythe phosphorous metabolism acting directly on intestine,kidney and bone, or indirectly through possible changesin Vitamin D3 levels (Mahieu et al., 2004).

We also observed that several changes in rat renal

tered Al dose (tap water: 1 �mol Al/l; food: 36.3 �mol Al/gwet weight).

The Al dose, the frequency and the route of administra-tion used in this work were described previously (Ellis et al.,1979; Ballanti et al., 1989; Mahieu et al., 2003). Al-treatedrats (Al) received an intraperitoneal injection of sterile solu-tion of aluminum lactate (Merck) containing 2.7 g/l of Al and26.7 g/l of anion lactate, pH 7.0, during 12 weeks (0.575 mgAl/100 g of body weight, three times per week). Aluminumlactate was dissolved in deionized water and the solution wasfiltered through a 0.2 �m microfilter (Nalgene, USA). Al con-centration was measured in the solution. At physiological pH,aluminum lactate is relatively stable at concentrations up to10−3 M (Tapparo et al., 1995). Control rats (C) were injectedwith a sterile solution of sodium lactate (Merck) containing26.7 g/l of anion lactate, pH 7.0 (5.68 mg of lactate/100 g ofbody weight, i.p.) during the same period and frequency. Thesodium lactate control treated rats did not present differencesas compared with control rats without any treatment in allthe experiments described in this manuscript. No differenceswere observed in food intake and body weight during thetreatment between control and Al-treated rats. Experimentswere performed in accordance with the Guiding Principles inthe Use of Animals in Toxicology, adopted by the Society ofToxicology in 1989 that were approved by each institution’scommittees.

2.2. Experiments in metabolic cages

After the 12 weeks of Al treatment, the rats (n = 10) werehoused in metabolic cages that allowed accurate determina-tion of urine volume and food intake. The animals received

function after Al chronic exposure, may occur but with-out a marked alteration in haemodynamic parameters orglomerular filtration rate (Mahieu et al., 2003). We havedemonstrated that renal water and sodium handling wasimpaired in balance studies. These changes were asso-ciated with increased levels of Al accumulated in renaltissue. The purpose of the present work was to intro-duce further insight on the effects of Al nephrotoxicity.We evaluated the alterations in the urinary concentrat-ing and diluting mechanisms in the distal tubules andcollecting ducts.

2. Materials and methods

2.1. Animals and treatments

Male adult Wistar rats weighing between 300 and 350 gwere used. The animals were housed in a temperature(22–24 ◦C) controlled room with 12-light:12-h dark cycle. Ani-mals were allowed free access to tap water and a standard dietcontaining approximately 65 mmol/kg sodium, 140 mmol/kgpotassium and 23% protein. Aluminum exposure from foodand tap water resulted negligible with respect to the adminis-

demineralized water and standard diet that contained 66 mmolNa/kg. Daily sodium balance was calculated as sodium intakeminus sodium excretion. After 2 days of adaptation, daily waterand sodium balance were measured during the last 2 days ofthe 12 week period. Urine was collected during the last 24 hand creatinine excretion and urine osmolality were analyzed.

Another experiment was performed to assess the urinaryconcentration ability after 12 weeks of Al treatment. Controland Al-treated rats were divided into two subgroups (n = 10per subgroup): normal hydrated rats and animals that weredeprived of food and water for 24 h. Urine was collected inmetabolic cages during the last 4 h of this period and bloodwas obtained from the tail vein. The urine to plasma osmo-lality ratio (Uosm/Posm), and the urine volume per minute (Vm)were calculated.

2.3. Renal function measurements

The effects of Al on renal function were examined on inde-pendent conditions as follows.

2.3.1. Protocol 1At the end of the experiment (12 weeks) control (n = 10)

and Al-treated rats (n = 10) were anesthetized by injection

S. Mahieu et al. / Toxicology 223 (2006) 209–218 211

of pentobarbital (50 mg/kg b.w., i.p.) and were prepared forrenal clearance studies as previously reported (Mahieu andCalvo, 1998; Torres et al., 1986). The femoral vein and thefemoral artery were cannulated. The bladder was exposedthrough a small abdominal incision and cannulated with acatheter (internal diameter: 3 mm) for urine collection. A solu-tion containing inulin (1 g/100 ml), sodium p-aminohippurate(PAH, 0.3 g/100 ml) and isotonic saline was infused through thefemoral vein catheter at a rate of 4.1 ml/h using a constant infu-sion pump (Sage Instruments, model 341-B-syringe pump).After equilibration for 60 min, urine samples were obtainedduring two periods of 30 min. Blood samples were obtainedfrom the femoral artery at the midpoint of each collectionperiod.

Urinary volume was measured. Glomerular filtration rate(GFR) was determined by inulin clearance and renal plasmaflow was estimated by PAH clearance. Creatinine, inulin,PAH, sodium and osmolality were measured in serum andurine samples. Fractional excretion of water (FEH2O) andsodium (FENa) were calculated by conventional formulae.Al, urea and aldosterone were determined in serum samples.Alkaline phosphatase (AP) activity was measured in urinesamples.

2.3.2. Protocol 2: renal function after a single dose ofdesmopressin

Control (n = 7) and Al-treated rats (n = 7) were prepared forclearance studies as reported above. All rats received a singledose (0.4 �g/rat, i.v.) of desmopressin acetate, Ferring Phar-maceuticals (1 desamino-8-d-arginine vasopressin, DDAVP, anspecific vasopressin V2 receptor agonist), followed by the infu-sion of physiologic saline containing inulin (4.1 ml/h), throughtf3aFte

2d

aiuir0Nv2pcuo

The reabsorption of water in excess of solutes (hypertonicurine) was calculated as T c

H2O (ml/min) = Cosm − Vm (Grosmanet al., 1984).

To quantify the capacity of solute-free water formation(CH2O), another group of control (n = 6) and Al-treated rats(n = 6) were deprived of food for 24 h, but were allowed freeaccess to water. Before the experiment, all the rats receivedthree doses of a solution containing glucose (1.5%, w/v),ethanol (1.3%, v/v) and sodium chloride (0.13%, w/v) throughan orogastric catheter in amounts equivalent to 3% of ratbody weight at 20 min intervals; this procedure initiated awater diuresis. Then, the rats were prepared for clearanceexperiments as previously described. The water diuresis wasmaintained by the infusion of a solution containing inulin(1 g/100 ml) and NaCl (0.45 g/100 ml). The different infusionrates and the protocol were as described above. The renalfunction parameters measured were the same as describedbefore. CH2O (ml/min) was also calculated (CH2O = Vm − Cosm)(Grosman et al., 1984).

2.4. Effects of Al on the abundance and trafficking ofAquaporin-2 (AQP2)

Control and Al-treated rats, after 12 weeks of treatment,were divided into two subgroups (n = 6 per subgroup) and theywere anesthetized with sodium pentobarbital (50 mg/kg b.w.,i.p.). One subgroup received a dose of DDAVP (0.4 �g/rat in1 ml of physiologic saline) via the femoral vein and the othersubgroup received the same volume of physiologic saline. After20 min, the kidneys were excised, decapsulated, and dissectedto separate cortex from medulla (Marples et al., 1995). Mean-while, they were placed on a buffered sucrose medium (0.25 M

he femoral vein, as previously described. After equilibrationor 60 min, urine samples were obtained during two periods of0 min. Blood samples were obtained from the femoral arteryt the midpoint of each collection period. The GFR, Uosm/Posm,EH2O and FENa were determined. The urinary excretion and

he plasmatic concentration of cAMP were also measured tovaluate cAMP of renal origin.

.3.3. Protocol 3: effects of Al on renal concentrating andiluting mechanisms

To measure the urinary concentrating ability, control (n = 6)nd Al-treated rats (n = 6) were deprived of water and food dur-ng 24 h. After that, the rats were prepared for clearance studiesnder anesthesia with sodium pentobarbital (50 mg/kg b.w.,.p.), as previously described, to study the solute-free watereabsorption (T c

H2O). For this, all rats received a dose of DDAVP,.4 �g/rat, i.v., followed by the infusion of a solution containingaCl (2 g/100 ml) and inulin (1 g/100 ml) through the femoralein. The infusion rate was increased progressively (6, 9, 12,1 ml/h) every 1 h. Each different infusion rate consisted in aeriod of stabilization (30 min) followed by a period of urineollection (30 min). Blood was collected in the middle of eachrine collection period. GFR, Uosm/Posm, FEH2O, FENa, and thesmolar clearance (Cosm = Uosm/Posm × Vm) were determined.

sucrose, 10 mM Tris–HCl, pH 7.2). All subsequent steps wereperformed at 4 ◦C. The medulla was minced and homoge-nized in 0.25 M sucrose, 0.1 mM phenyl-methyl sulphonyl-fluoride, 10 mM Tris–HCl, pH 7.2 (5 ml/g of tissue) in aDounce homogenizer (five strokes) with a motor-driven Teflon-glass Potter homogenizer (10 strokes at 800 rpm). Aliquotsof the homogenates were stored at −70 ◦C until used withindays. Medullary homogenates were centrifuged at 4000 × gduring 15 min. The supernatant was centrifuged at 17,000 × gduring 30 min. The pellet corresponded to a plasma mem-brane enriched fraction, according to Boumendil-Povedin andPovedin (1983) and Coux et al. (2002). The supernatant wassubmitted to another centrifugation at 100,000 × g to obtainan intracellular vesicles enriched fraction. The pellets wereresuspended in 100 �l of buffer and protein concentration wasmeasured using the Lowry method (Lowry et al., 1951). Eachfraction of medullary plasma membranes and intracellular vesi-cles (10 �g of protein for each sample) were subjected toSDS-PAGE on polyacrylamide gels (8%). For each gel, anidentical gel was run in parallel and subjected to Coomassiestaining to ensure identical loading. The amount of proteinwas chosen after the linearity of detection had been verified.Samples were boiled for 3 min in presence of 5% 2-mercaptoethanol and 1% SDS; 5 �l of prestained molecular weight stan-

212 S. Mahieu et al. / Toxicology 223 (2006) 209–218

dards was also loaded onto the gel. The separated proteins weretransferred to nitrocellulose membranes (BioRad, Hercules,CA) in Tris–glycine transfer buffer with 20% methanol in amini-blotter (Sigma–Aldrich) (Towbin et al., 1979). Uniformblotting across the gel was verified by Coomassie brilliant bluestaining of the post blot gel.

For AQP2 detection, nitrocellulose membranes were incu-bated for 1 h at room temperature with 5% (w/v) non-fat drymilk in phosphate buffer saline (PBS), followed by 1 h incuba-tion with a rabbit polyclonal antibody to the rat kidney AQP2

(1:1000 dilution, Alpha Diagnostic International, San Anto-nio, TX, USA). After washing, membranes were incubated for1 h with alkaline phosphatase-conjugated anti-rabbit antibody(1:3000 dilution, Calbiochem-Novabiochem). Alkaline phos-phatase was detected with 5-bromo-4-chloro-3-indolyl phos-phate/nitro blue tetrazolium (Promega, Madison, WI, USA).The primary antibody revealed a 29 kDa and a 35–50 kDabands, representing non-glycosylated and glycosylated formsof AQP2, respectively. The samples were run on the same geland comparisons of band intensities were performed within onemembrane. The most intense band of each membrane was cho-sen and its density was designated as 100. Blots were digitizedusing a desktop computer Scanner (HP II Scan Jet 3400C).The signal produced was quantified using the Adobe Photo-Shop (6.0) computer program. Both bands were scanned andthe densitometry values reported represent the sum of the twobands.

2.5. Analytical methods

Determination of creatinine and urea in serum, and creati-nine and AP in urine were measured by using an autoanalyzer

3. Results

As stated in previous works, no abnormal clinicalsigns, nor changes in behavior were observed after theAl treatment. At the end of the 12 weeks Al treat-ment, no differences were observed in the mean bodyweight values ([C]: 428 ± 11 g; [Al]: 395 ± 11 g). TheAl-treated rats presented a diminished kidney weightas compared with controls ([C]: 2.77 ± 0.09 g; [Al]:2.41 ± 0.06 g, p < 0.05). Al-treated rats presented higherserum Al concentration ([C, n = 8]: 10 ± 1.5 �g/l; [Al,n = 9]: 660 ± 35 �g/l; p < 0.05) and higher amount ofthis metal in renal tissue ([C, n = 8]: 2.5 ± 0.5 �g/gwet tissue; [Al, n = 9]: 35.3 ± 1.2 �g/g wet tissue;p < 0.05).

3.1. Effects of Al chronic administration on renalwater excretion and on the sodium balance

Data obtained after 12 weeks of Al treatment is pre-sented in Fig. 1. These results are in accordance withour previously described results (Mahieu et al., 2003).Al-treated rats showed a reduction of sodium balanceand no changes in water balance. Moreover, Al treat-ment increased the daily urine flow rate and decreasedurine osmolality. The urinary creatinine excretion after12 weeks in Al-treated rats presented a slight but signifi-cant decrease as compared with controls ([C] 19.1 ± 0.8;[Al] 15.3 ± 0.5 mg/24 h; p < 0.05, n = 10). These dataclearly indicated a disturbance in water and sodium han-

(Selectra-2, Vitalab). Serum and urine osmolality were deter-mined using Osmomat-030 osmomiter (Gonotec).

Inulin and PAH concentrations in the samples were deter-mined by Roe’s procedure (Roe et al., 1949) and Brun’s methodas modified by Waugh and Beall (1974), respectively. Sodiumand potassium were determined by flame photometry. Al con-tent in serum and in kidneys was measured by a graphitefurnace atomic absorption spectrophotometer (Perkin-Elmer5000, Fremont, CA, USA). Urinary and plasma cAMP levelswere measured by radioimmunoassay (RIA) with a commercialkit (Immunotech, France). Aldosterone in serum was measuredby RIA using a commercial kit (Coat-A-Count Aldosterone,DPC, Los Angeles, CA).

2.6. Statistical analysis

Results are expressed as mean ± S.E.M. Comparisonbetween control and treated groups was carried out usingMann–Whitney U-test or Student’s t-test as indicated. The 0.05level of probability was used as the criterion of significance inall cases. Regression lines were calculated by the method ofleast squares. Each point represents the average of triplicatedeterminations. The statistical analysis was made with Statis-tical Package for Social Sciences (SPSS) program (1995).

dling by renal tubules.The urinary concentrating ability was also measured

after 12 weeks of Al treatment under normal fluid intakeand during hydropenic condition. The results are shownin Table 1. Under normal fluid intake, control and Al-treated animals presented similar values of urine osmo-lality and Uosm/Posm. Nevertheless, Al-treated animalssubmitted to hydropenic conditions were unable to main-tain the water excretion and the Uosm/Posm as controlrats.

3.2. Effects of Al chronic administration on renalfunction, using clearance techniques

Glomerular filtration rate, estimated by inulinclearance, and cortical renal plasma flow, estimatedby PAH clearance, were measured by conventionalclearance techniques (Table 2). No differences wereobserved in these values between control and Al-treated rats. A significant increase in fractional sodiumexcretion in Al-treated rats was observed, withoutchanges in fractional excretion of water. AP val-

S. Mahieu et al. / Toxicology 223 (2006) 209–218 213

Fig. 1. Daily production of urine (A), water balance (B), urine osmolality (C), sodium urine excretion (D), and sodium balance (E) measured after12 weeks treatment with aluminum lactate (0.575 mg Al/100 g body weight, i.p.). White bars indicate the results from controls (n = 10). Black barsrepresent Al-treated rats (n = 10). Data are presented as mean ± S.E.M. *Statistically different from control values (p < 0.05). Vm: urine volume perday; Wb: water balance; [Osm]u: urine osmolality; ENa: urine sodium excretion, NaB: sodium balance.

Table 1Urinary concentrating ability in control and Al-treated rats after 12 weeks of treatment

Condition Group Vm (�l/min) [Osm]u (mOsm/l) Uosm/Posm

Normal-hydrated Control (n = 10) 5.25 ± 1.01 2061 ± 199.9 6.47 ± 0.65Al-treated (n = 10) 4.61 ± 0.64 1899 ± 134.1 6.70 ± 0.46

Hydropenic Control (n = 10) 2.62 ± 034 3051 ± 108.9 10.38 ± 0.30Al-treated (n = 10) 4.60 ± 0.42* 2214 ± 64.9* 7.52 ± 0.20*

Normal hydrated rats were maintained in metabolic cages during 24 h with free access to food and water. The hydropenic rats were kept in metaboliccages for 24 h without water and food. Urine was collected in all groups during the last 4 h. Data are expressed as mean ± S.E.M. Vm: urine volumeper minute; [Osm]u: urinary osmolality; Uosm/Posm: urine to plasma osmolality ratio.

* Statistically different from the respective control values (p < 0.05).

ues referred to inulin excreted in the urine wereincreased ([C]: 2.2 ± 0.6 UI/mg; [Al]: 5.1 ± 0.5 UI/mg,p < 0.05). Neither the levels of urea, creatinine,sodium, potassium nor osmolality in serum were dif-ferent between both groups, but Al-treated rats pre-sented higher levels of aldosterone in serum ([C,n = 10]: 330 ± 41 pg/ml; [Al, n = 9]: 606 ± 106 pg/ml;p < 0.05).

3.3. Effects of Al chronic administration on renalfunction, using clearance techniques after a singledose of DDAVP

This experiment was performed to clarify whetherthe urine concentrating impairments observed in Al-treated rats could be due to an alteration on collectingtubule response to vasopressin. We performed clearance

Table 2Renal functional parameters at the end of 12 weeks in control rats and Al-treated rats measured by clearance techniques

GFR (ml/min 100 g) ClPAH (ml/min 100 g) Vm (�l/min) [Osm]u (mOsm/l) FEH2O (%) FE Na (%)

Control (n = 10) 0.81 ± 0.02 3.15 ± 0.20 9.66 ± 1.36 1729 ± 92 0.30 ± 0.02 0.13 ± 0.03Al-treated (n = 10) 0.80 ± 0.04 2.92 ± 0.16 11.13 ± 1.53 1865 ± 115 0.37 ± 0.06 0.27 ± 0.04*

Data are expressed as mean ± S.E.M. GFR: Glomerular filtration rate; Cl PAH: p-aminohippurate clearance; Vm: urine volume per minute; FEH2O:fractional excretion of water; FENa: fractional excretion of sodium.

* Statistically different from control values (p < 0.05).

214 S. Mahieu et al. / Toxicology 223 (2006) 209–218

Fig. 2. Fractional excretion of water (A), urine to plasma osmolality ratio (B) and the nephrogenic cAMP levels (C) measured in controls (C; n = 7)and Al-treated rats (Al; n = 7) after treatment with aluminum lactate (0.575 mg Al/100 g body weight, i.p.) during 12 weeks. A single dose of DDAVP,[0.4 �g/rat, i.v. (desmopressin acetate)] was administered to both groups of rats. Data are presented as mean ± S.E.M. *Statistically different fromcontrol values (p < 0.05). FEH2O: fractional excretion of water; Uosm/Posm: urine to plasma osmolality ratio; cAMP-n (pg/ml): cAMP of renal origin.

studies in control and Al-treated rats after the admin-istration of a single dose of DDAVP. No changes wereobserved in GFR, urine volume per minute, and frac-tional excretion of sodium and potassium in Al-treatedanimals (data not shown). In this group urine osmolal-ity was decreased ([C, n = 7]: 1934 ± 137 mOsm/l; [Al,n = 7]: 1409 ± 56 mOsm/l; p < 0.05) and the T c

H2O/GFRwas significantly diminished ([C, n = 7]: 1.4 ± 0.18%;[Al, n = 7]: 1.01 ± 0.03%; p < 0.05). Fig. 2 shows anincreased FEH2O and decreased Uosm/Posm in Al-treated rats. Those changes were observed togetherwith a decrease in the nephrogenic cAMP levels.Sodium, potassium, and osmolality in serum remainedunchanged.

3.4. Effects of Al chronic administration on theurinary concentrating ability

The results obtained by infusing a hypertonic salineand increasing the infusion rate progressively in control

and Al-treated rats allowed us to arrive to various conclu-sions. The relationship between Tc/GFR and Cosm/GFRdid not indicate differences among treated group andcontrols (data not shown). Nevertheless, both the frac-tional excretion of water and the Uosm/Posm were dif-ferent, while the infusion rate of hypertonic saline wasincreased (Fig. 3). Table 3 shows the effect of the infu-sion of hypertonic saline at increasing infusion rates, asdescribed in Section 2, on the Uosm/Posm in control andAl-treated rats. These results, suggests that the Al-treatedrats could develop a deficiency in urinary concentratingability under different conditions in which plasma antid-iuretic hormone was increased (thirst or infusion of anhypertonic saline).

To understand more about the deficiency in the uri-nary concentrating mechanism, the ability to performsolute-free water clearance indicative of the transportfunction in the thick ascending limb of Henle’s loopwas studied. For this purpose, the relationship betweenCH2O/GFR versus Vm/GFR was measured during the

ion rate.) durinults froFEH2O:

Fig. 3. Effect of the infusion of hypertonic saline at increasing infustreatment with aluminum lactate (0.575 mg Al/100 g body weight, i.pbefore the infusion of the hypertonic saline. White bars indicate the resmean ± S.E.M. *Statistically different from control values (p < 0.05).ratio.

s in adult controls rats (C; n = 6) and Al-treated rats (Al; n = 6) afterg 12 weeks. All the rats received a dose of DDAVP (0.4 �g/rat, i.v.)m controls. Black bars represent Al-treated rats. Data are presented asfractional excretion of water; Uosm/Posm: urine to plasma osmolality

S. Mahieu et al. / Toxicology 223 (2006) 209–218 215

Table 3Effects of aluminum on the urinary concentrating ability in control rats and Al-treated rats after 12 weeks submitted to different states of hydration,measured by clearance techniques

Control rats Al-treated rats

Uosm/Posm [Osm]p (mOsm/l) Uosm/Posm [Osm]p (mOsm/l)

Normal-hydrated 6.47 ± 0.65 (n = 10) 290 ± 2 (n = 10) 6.70 ± 0.46 (n = 10) 288 ± 1 (n = 10)Hydropenic 10.38 ± 0.20 (n = 10) 300 ± 1 (n = 10) 7.52 ± 0.20* (n = 10) 295 ± 2 (n = 10)Isotonic saline solution infusion—4.1 ml/min 5.60 ± 0.22 (n = 10) 289 ± 2 (n = 10) 5.90 ± 0.26 (n = 10) 289 ± 2 (n = 10)Isotonic saline solution infusion +

DDAVP—4.1 ml/min6.69 ± 0.3 (n = 7) 288 ± 1 (n = 7) 4.91 ± 0.18* (n = 7) 287 ± 3 (n = 7)

NaCl 2% infusion + DDAVP—6 ml/min 3.59 ± 0.24 (n = 6) 302 ± 1 (n = 6) 2.76 ± 0.14* (n = 6) 302 ± 2 (n = 6)NaCl 2% infusion + DDAVP—9 ml/min 2.52 ± 0.064 (n = 6) 306 ± 2 (n = 6) 2.18 ± 0.13* (n = 6) 307 ± 2 (n = 6)NaCl 2% infusion + DDAVP—21 ml/min 1.57 ± 0.039 (n = 6) 328 ± 3 (n = 6) 1.41 ± 0.056 (n = 6) 332 ± 3 (n = 6)

Data are expressed as mean ± S.E.M. Uosm/Posm: urine to plasma osmolality ratio; [Osm]p: plasma osmolality.* Statistically different from control values (p < 0.05).

Fig. 4. A representative Western blot analysis of renal AQP2 protein in medullary plasma membrane enriched fraction (A) and in medullaryintracellular vesicles enriched fraction (C). The samples were obtained from controls (C, n = 6) and Al-treated rats (Al, n = 6). These groups weresubdivided into two groups, one received 1 ml of physiologic saline (PS) and the other desmopressin (DDAVP), i.v. 10 �g of protein of each sampleswas resolved by SDS-PAGE together with prestained molecular weight markers and blotted onto nitrocellulose and probed with the primary antibody.AQP2 expression in medullary plasma membrane enriched fraction (B) and in medullary intracellular vesicles enriched fraction (D), was expressedin arbitrary densitometry units as percent of maximal (see Section 2). Statistical analysis was conducted on groups of samples which were runon the same gel. Data are expressed as mean ± S.E.M. (n = 6 per experimental group). *p < 0.05 compared with C + PS; #p < 0.05 compared withC + DDAVP.

216 S. Mahieu et al. / Toxicology 223 (2006) 209–218

infusion of hypotonic saline at increasing infusion ratesas described in Section 2. The relationship betweenboth parameters were not statistically different nei-ther when the values were adjusted to a model ofsaturable transport rate (Michaelis–Menten) nor usingthe Lineweaver–Burks model (data not shown). Theseresults indicated that altered urinary concentrating abil-ity might be due to impaired transport of water in thecollecting duct.

3.5. Effects of Al on the abundance and traffickingof Aquaporin-2

To address whether the alterations observed in urinaryconcentrating ability in Al-treated rats was associatedwith changes in abundance of AQP2 or with a failureof trafficking of this protein to plasma membranes, westudied the abundance of AQP2 in plasma membrane andintracellular vesicles enriched fractions by Western blot.Fig. 4A shows a representative Western blot for AQP2 inmedullary plasma membranes of control and Al-treatedrats after 12 weeks. The effect of administration ofDDAVP is also shown. There was a significant decreasein AQP2 abundance in plasma membranes of Al-treatedrats relative to controls. DDAVP induced an incrementin both the glycosylated and non-glycosylated AQP2abundance as compared with the rats that receivedphysiologic saline, either in control and in Al-treatedanimals.

Fig. 4C shows a representative Western blot for AQP2

ability after 12 weeks of Al treatment. Although a reduc-tion in the sodium balance and an increase in the excre-tion of this cation, sodium and potassium serum concen-tration were maintained as control, possibly at expense ofthe increased aldosterone levels. As the potassium serumconcentration remained as control rats in Al-treated rats,we could put aside the role of hypokalemia as a factorinvolved in the impaired urinary concentrating ability(Bank and Ainedjian, 1964). In spite of the high levelsof serum aldosterone found in Al-treated rats, the frac-tional excretion of sodium was increased. One possibleexplanation for this could be due to alterations in sodiumchannels or transporters induced by Al. Suwalsky etal. (2002) reported that Al produced alterations insodium transport in isolated toad skin. They foundinactivation of sodium channels, alterations in ATPaseand a diminished permeability of this cation. Furtherstudies would be necessary to confirm this hypothesis.

On the other hand, the decrease in reabsorption ofsolute-free water in the collecting ducts, together withdecreased urine to plasma osmolality ratio, and theincrease in the urine volume observed in Al-treated ratsin response to DDAVP administration, suggest that alu-minum could impair the response of collecting ductsto the presence of antidiuretic hormone (AVP). AVPactions are mediated in kidneys by V2 receptors thatare coupled to the production of cAMP. In a previouswork, we have reported a decrease in the levels of cAMPof renal origin in Al-treated rats (Mahieu and Calvo,1998), and this change is also observed in our present

in intracellular vesicles enriched fraction. AQP2 wasdecreased in vesicles from Al-treated rats both with theadministration of physiologic saline or DDAVP.

4. Discussion

Our previous works indicated that several changesmay occur in rat renal functions after chronic exposureto Al, but this takes place without a marked changein glomerular filtration rate and in blood renal flow,although a large amount of aluminum was accumulatedin renal tissue (Mahieu et al., 2003). We have previouslydetected renal alterations of sodium and water and in theurine concentrating capacity by using balance studies.

The present work was performed to gain insight aboutthe mechanisms involved in the impaired water andsodium handling. As the chronically treated rats did notpresent modifications in food and water intake, we sup-posed that the renal changes could be associated to the Alaccumulated in renal tissue. The increased diuresis with-out changes in GFR, the decreased urine osmolality andUosm/Posm indicated an impaired urinary concentrating

experiments. Thus, AVP actions could contribute to thisphenomenon. Moreover, the osmotic gradient producedby the thick ascending limb of Henle’s loop could alsobe involved. The Na–K–2Cl cotransporter is regulated byAVP (Molony et al., 1987; Kim et al., 1999). So a reducedcAMP generation could reduce this cotransporter, con-tributing to the impaired urinary concentrating ability.

It is well known that water reabsorption in the col-lecting ducts also depends on the water permeabil-ity of these tubules. Impaired T c

H2O can be either aconsequence of modifications in water permeability ofthe collecting ducts or a diminished medullary tonic-ity due to a decreased capacity for electrolyte trans-port in the ascending limb of Henle’s loop, or anincreased medullary blood flow resulting in “washout”of the cortex-to-papilla concentration gradient. Impairedwater permeability of the collecting duct or medullary“washout” would be expected to enhance CH2O. On theother hand, if electrolyte transport in the ascending limbwere decreased, CH2O would be decreased. These exper-iments were performed by infusing a hypotonic saline,a situation in which endogenous AVP must be absent or

S. Mahieu et al. / Toxicology 223 (2006) 209–218 217

in minimum concentration in plasma; CH2O measuredin Al-treated rats was indistinguishable from that seenin the control group at any rate of urine flow. Thesefindings indicate that the impaired urinary concentrat-ing ability in Al-treated rats would not be the result of adisturbance in the function of the thick ascending limbof Henle’s loop. The alternative could be a failure in theresponse to AVP in the collecting duct or a failure in thepermeability to water in the luminal membranes of theseducts. We found decreased urinary concentrating abil-ity in situations in which endogenous (thirst or infusionof hypertonic saline) or exogenous plasma antidiuretichormone was increased.

AQP2, AQP3 and AQP4 are water channels expressedin collecting duct principal cells that mediate waterosmotic transport in this renal segment. It has beendescribed that this function promotes a reabsorption of10–20% of the filtered water. This fraction of the freewater, which is regulated by AVP, determines the urineosmolality. AQP2 is expressed in the apical membranesand in subapical vesicles and its expression is regu-lated by AVP via V2 receptor (Nielsen et al., 1995;Matsumura et al., 1997). This receptor is coupled to thecAMP second messenger system. The short time regula-tion (minutes) involves the translocation of AQP2 fromintracellular vesicles to the apical plasma membrane.This phenomenon is dependent on cAMP and is reversedafter AVP is dissociated from V2 receptor. Thus, therapid increase in water permeability seen with AVP stim-ulation may be attributed to the increased trafficking ofA1iepamoaAtAirtfidlmrti

receptors (Pun et al., 1990). A similar effect of Al couldbe responsible of affecting the interaction of AVP withits V2 receptors. However, the primary impairment maylie at the level of membrane G protein that acts as a trans-ducer between AVP receptor and adenylate cyclase. It iswell known that in all the receptor–protein interactions,the magnesium is an essential cofactor for proper func-tions (Spiegel et al., 1992; Simonds, 1999). The effectiveionic radii of aluminum and magnesium are comparable,thus the substitution of magnesium with aluminum inphysiologic sites could be a common pathogenic mech-anism involved in the toxicity of this metal (Miller etal., 1989; Mac Donald and Martin, 1988). Work is inprogress in our laboratory to address this possibility.

Acknowledgement

This work was supported by the grant CAID + D(Universidad Nacional del Litoral, Argentina).

References

Ballanti, P., Mocetti, P., Della Rocca, C., Bonucci, E., Costantini, S.,Giordano, R., Ioppolo, A., Montovani, A., 1989. Experimentalaluminum intoxication and parathormone: effects on the miner-alization process. Miner. Electrolyte Metab. 15, 233–240.

Bank, N., Ainedjian, H.S., 1964. A micropuncture study of the renalconcentrating defect of potassium depletion. Am. J. Physiol. 206,1347–1353.

Bertholf, R., Herman, M., Savory, J., Carpenter, R., Sturgill, C.B.,

QP2 to apical plasma membranes (Marples and Frokiar,999; Knepper, 1997). The long time effect (hours, days)s also dependent on cAMP, and promotes AQP2 mRNAxpression resulting in increased abundance of AQP2rotein in the cells (Marples and Frokiar, 1999; Terris etl., 1996). Our results indicated that the presence of alu-inum in renal tissue was associated to a downregulation

f AQP2 channels, that could be responsible for the alter-tions observed in these rats. We observed a reduction ofQP2 abundance in the plasma membrane fraction and in

he intracellular vesicles fraction. As DDAVP increasedQP2 abundance in plasma membranes and decreased

t in intracellular vesicles in both control and Al-treatedats we could suggest that there is no impairment in AQP2rafficking. A decline in AQP2 level with a normal traf-cking to plasma membranes in Al-treated rats could beue to impaired AQP2 synthesis. The decrease in cAMPevels previously described, could be at least one of the

echanisms responsible for these observations. Otheresults have described that the effects of aluminum onhe response to parathyroid hormone (PTH) in kidneys tightly associated to changes in the affinity for PTH

Katsetos, C.D., Vandenberg, S.R., Wills, M., 1989. A long termintravenous model of aluminum maltol toxicity in rabbits: tis-sue distribution, hepatic, renal and neuronal cytoskeletal changesassociated with systemic exposure. Toxicol. Appl. Pharmacol. 98,58–74.

Boumendil-Povedin, E.F., Povedin, R.A., 1983. Isolation of basolat-eral and brush border membranes from the rabbit kidney cortex.Vesicle integrity and membrane sideness of the basolateral fraction.Biochim. Biophys. Acta 735, 86–94.

Burnatowska-Hledin, M.A., Mayor, G.H., Low, K., 1985. Renal han-dling of aluminum in the rat. Clearance and micropuncture studies.Am. J. Physiol. 249, F192–F197.

Cacini, W., Yokel, R.A., 1988. Accumulation of aluminum by rabbitrenal cortex. Res. Commun. Chem. Pathol. Pharmacol. 59, 93–105.

Coux, G., Trumper, L., Elıas, M.M., 2002. Renal function and cor-tical (Na+, K+)-ATPase activity, abundance and distribution afterischemia-reperfusion in rats. Biochim. Biophys. Acta 1586, 71–80.

Ebina, Y., Okada, S., Hamazaki, S., Midorikawa, O., 1984. Liver,kidney and central nervous system toxicity of aluminum givenintraperitoneally to rats: a multiple dose subchronic using alu-minum nitriloacetato. Toxicol. Appl. Phamacol. 75, 211–218.

Ellis, H.A., McCarthy, J.H., Herrington, J., 1979. Bone aluminum inhaemodialysed patients and in the rats injected with aluminumchloride relationship to impaired bone mineralization. J. Clin.Pathol. 32, 832–844.

Exley, C., 2004. The pro oxidant activity of aluminum. Free Radic.Biol. Med. 13, 79–81.

218 S. Mahieu et al. / Toxicology 223 (2006) 209–218

Grosman, M.E., Elıas, M.M., Comin, E.J., Rodriguez Garay, E.A.,1984. Distal nephron function of the rat during acute aflatoxicosis.Toxicol. Lett. 21, 263–270.

Kim, G.H., Ecelbarger, C.A., Mitchell, C., Packer, R.K., Wade, J.B.,Knepper, M.A., 1999. Vasopressin increases Na–K–2Cl cotrans-porter expression in thick ascending limb of Henle’s loop. Am. J.Physiol. 276, F96–F103.

Knepper, M.A., 1997. Molecular physiology of urinary concentrat-ing mechanism: regulation of aquaporin water channels by vaso-pressin. Am. J. Physiol. 272, F3–F12.

Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Proteinmeasurement with the Folin phenol reagent. J. Biol. Chem. 193,265–275.

Mac Donald, T.L., Martin, R.B., 1988. Aluminum ion in biologicalsystems. Trends Biochem. Sci. 8, 15–19.

Mahieu, S., Calvo, M.L., 1998. Effect of chronic poisoning with alu-minum on the renal handling of phosphate in the rat. Toxicol. Lett.94, 47–56.

Mahieu, S., Gionotti, M., Millen, N., Elıas, M.M., 2003. Effect ofchronic accumulation of aluminum on renal function, cortical renaloxidative stress and cortical renal organic anion transport in rats.Arch. Toxicol. 77, 605–612.

Mahieu, S., Navoni, J., Millen, N., Contini, M.C., Gonzalez, M., Elıas,M.M., 2004. Effects of aluminum on phosphate metabolism in rats:a possible interaction with Vitamin D3 renal production. Arch.Toxicol. 78, 609–616.

Marples, D., Knepper, M.A., Christensen, E.I., Soren, N., 1995. Redis-tribution of aquaporin-2 water channels induced by vasopressin inrat kidney inner medullary collecting duct. Am. J. Physiol. 269,C655–C664.

Marples, D., Frokiar, J., 1999. Long term regulation of aquaporin inthe kidney. Am. J. Physiol. 273, F331–F339.

Matsumura, Y., Uchida, S., Rais, T., Sasaki, S., Marumo, F., 1997. Tran-scription regulation of aquaporin-2 water channel gene by cAMP.J. Am. Soc. Nephrol. 8, 861–867.

kidney collecting duct by inducing translocation of aquaporin-CDwater channels to plasma membrane. Proc. Natl. Acad. Sci. U.S.A.92, 1013–1017.

Pun, K.K., Ho, P.W., Lau, P., 1990. Effects of aluminum on the PTHreceptors of bone and kidney. Kidney Int. 37, 72–78.

Roe, H.H., Epstein, J.H., Goldstein, N.P., 1949. A photometric methodfor determination of inulin in plasma and urine. J. Biol. Chem. 178,839–845.

Roy, A.K., Talukder, G., Sharma, A., 1991. Similar effect in vivoof two aluminum salts on the liver, kidney, bone and brain ofRattus novergicus. Bull. Environ. Contam. Toxicol. 47, 228–295.

Simonds, W.F., 1999. G protein regulation of adenylate cyclase. TrendsPharmacol. Sci. 20, 66–73.

Spiegel, A.M., Shenker, A., Weinsten, L.S., 1992. Receptor-effectorcoupling by proteins: implications for normal and abnormal signaltransduction. Endocr. Rev. 13, 536–565.

Stein, G., Laske, V., Muller, A., Braunlich, H., Lin, B.W., Fleck, G.,1987. Aluminum induced damage of lysosomes in the liver, spleen,and kidney of rats. J. Appl. Toxicol. 7, 253–258.

Suwalsky, M., Norris, B., Kiss, T., Zatta, P., 2002. Effects of Al(III)speciation on cell membranes and molecular models. Coor. Chem.Rev. 228, 285–295.

Sweet, D.S., Bush, K., Nigam, S., 2001. The organic anion transporterfamily: from physiology to ontogeny and the clinic. Am. J. Physiol.281, F115–F125.

Tapparo, A., Solda, L., Bombi, G.G., Zambenedetti, P., Zatta, P.F.,Bertani, R., Corain, B., 1995. Analytical validation of a generalprotocol for the preparation of dose-controlled solutions in alu-minum toxicology. The Analyst 120, 2425–2429.

Terris, J., Ecelbarger, C.A., Nielsen, S., Knepper, M.A., 1996. Longterm regulation of four renal aquaporins in rats. Am. J. Physiol.271, F414–F422.

Torres, A.M., Rodriguez, J.V., Ochoa, J.E., Elıas, M.M., 1986. Ratkidney function related to tissue glutathione levels. Biochem. Phar-

Miller, J.L., Hubbard, C.M., Litman, B.J., Mac Donald, T.L., 1989.Inhibition of transducin activation and guanosine triphosphataseactivity by aluminum ion. J. Biol. Chem. 264, 243–250.

Molony, D.A., Reeves, W.D., Herbert, S.C., Andreoli, T.E., 1987. ADHincreases Na–K–2Cl entry in mouse medullary thick ascendinglimbs of Henle. Am. J. Physiol. 252, F177–F187.

Monteagudo, F.S., Isaacson, L.C., Wilson, G., Hickman, R., Folb, P.I.,1988. Aluminum excretion by the distal tubule of the pig kidney.Nephron 49, 245–250.

Nielsen, S., Chou, C.L., Marples, D., Christensen, E.I., Kishore, B.K.,Knepper, M.A., 1995. Vasopressin increases water permeability of

macol. 35, 3355–3358.Towbin, H., Staehelim, T., Gordon, J., 1979. Electroforetic transfer

of protein from polyacrylamide gels to nitrocellulose sheets: pro-cedure and some applications. Proc. Natl. Acad. Sci. U.S.A. 76,4350–4353.

Waugh, W.H., Beall, P.T., 1974. Simplified measurement of para-aminohippuric acid and other arylamines in plasma and urine.Kidney Int. 5, 429–436.

Yokel, R.A., McNamara, P.J., 1985. Aluminum bioavailability and dis-position in adult and immature rabbits. Toxicol. Appl. Phamacol.77, 344–352.