Pflugers Arch - Eur J Physiol (2004) 448: 85–92 DOI 10.1007/s00424-003-1220-y

RENAL FUNCTION , BO DY FLUID S

Andrea Carranza . Carlos F. Mendez . Marta Barontini . Susana Nowicki

Insulin enhances L-dopa renal proximal tubule uptake: a regulatory mechanism impaired in insulin resistance Received: 4 April 2003 / Revised: 24 October 2003 / Accepted: 26 November 2003 / Published online: 12 February 2004 # Springer-Verlag 2004

Abstract A stimulatory role for insulin in the uptake of neutral amino acids has been reported for a variety of tissues. Here we examine the effect of insulin on L-dopa uptake by proximal tubule cells (PT cells) isolated from control and fructose-fed rats (FR-rats, 10% w/v fructose solution in tap water), a model of insulin resistance. Insulin (200 μU/ml) increased L-dopa uptake into PT cells by about 50% (705±186 vs.1117±140 pmol L-dopa/mg protein per minute) (p<0.05). The higher uptake correlated with a 40% increase in the number of high-affinity L-dopa transport sites (L-dopa 0.2 μM) (0.59±0.05 vs. 0.82 ±0.09 pmol L-dopa/mg protein per minute), without changing their affinity. The effect of insulin was not modified by ouabain (1 mM), nocodazole (1–10 μM) or colchicine (50–100 μM), whereas it was abolished by cytochalasin D or latrunculin B (both 1 μM). This suggests that the process is independent of Na+,K+ATPase activity or the microtubule network but that it requires the integrity of the actin cytoskeleton. L-dopa transport by the low-affinity transport sites (L-dopa 5 μM) was not affected by insulin, neither was the effect of insulin observed in PT cells isolated from FR-rats. In line with this, FR-rats showed lower renal L-dopa reabsorption as compared to control animals (81±4 vs. 51±9%). Taken together, our results support the involvement of insulin in the multifactorial regulation of renal L-dopa reabsorption. Keywords Actin cytoskeleton . Amino acid transport . Insulin . Insulin resistance . L-dopa . Microtubule network A. Carranza . M. Barontini . S. Nowicki (*) Centro de Investigaciones Endocrinológicas (CONICET), Hospital de Niños R. Gutierrez, Gallo 1360 (C1425EFD) , Buenos Aires, Argentina e-mail: [email protected] Fax: +54-11-49635930 C. F. Mendez Departamento de Bioquímica Humana, Facultad de Medicina, Universidad de Buenos Aires, Paraguay 2155 (C1121ABG), Buenos Aires, Argentina

Introduction Recent studies on catecholaminergic systems have highlighted the importance of non-neuronal biosynthesis of dopamine. Although non-neuronal dopamine synthesis and release have been proved to take place in pancreatic exocrine cells [26], in the gastrointestinal tract [25] and in human amniotic epithelial cells [9], the best studied nonneuronal catecholaminergic system to date is the renal Ldopa-dopamine system [1, 23]. L-dopa (L-3,4-dihydroxyphenylalanine) is an aromatic amino acid and the immediate precursor of all the endogenous catecholamines. The renal proximal tubules reabsorb L-amino acids almost completely from the glomerular filtrate. Early studies demonstrated that the uptake of L-dopa into the proximal tubules is an active process, with high structural specificity [7]. In previous studies from our laboratory the transport of L-dopa into proximal tubules has been shown to be impaired in aged rats [2], inhibited by β2-adrenergic receptor signalling [6] and by high glucose [5]. The reabsorbed L-dopa is converted into dopamine by the cytoplasmic aromatic Lamino acid decarboxylase [23], the intracellular availability of L-dopa being the limiting step in renal dopamine synthesis [2, 23]. Renal dopamine acts, in turn, as an autocrine-paracrine agent that plays an important role in the regulation of sodium excretion [1]. Several different major amino acid transport systems have been reported as responsible for the uptake of neutral amino acids in the kidney [12]. A stimulatory role of insulin in the uptake of neutral amino acids has been reported for a variety of tissues including skeletal and cardiac muscle, smooth muscle cells, hepatocytes and other animal cells [13, 14, 24]. Early studies on the distribution of insulin receptors along the rat nephron showed that the greatest density of high-affinity binding sites occurs in the proximal convoluted tubule [4] where insulin regulates both metabolism and transport of different substrates. Stimulation of the Na+-H+ exchanger [11] and Na+,K+-ATPase [10] have been attributed to insulin in the proximal tubules.

86

If L-dopa behaves as a neutral amino acid for proximal tubule uptake, it is likely that renal L-dopa reabsorption may be regulated by insulin, and this regulatory mechanism could be altered in conditions where the function of insulin is impaired. Indeed, previous results from our laboratory have described an increased renal excretion of L-dopa in streptozotocin-induced diabetic rats [5]. The aim of the present study was to assess whether insulin modulates the uptake of L-dopa into renal proximal tubule cells and to explore the possible mechanisms involved in this regulatory process.

Materials and methods In vivo studies Male Sprague-Dawley rats weighing 180–230 g were used. Animals were maintained at 22°C on a 12:12-h dark-light cycle, were fed a commercial standard laboratory chow with the following composition (wt/wt): 20% proteins, 3% fat, 2% fibre, 6% minerals, and 69% starch and vitamin supplements, and had free access to tap water. Animals were randomly divided into two groups: control and experimental. The experimental group was composed of rats made resistant to insulin-mediated glucose disposal by a fructose-enriched diet (FR-rats) [18]. These animals had free access to a 10% w/v fructose solution in tap water for 4 weeks before the day of the experiment and continued on the standard diet [27]. On the day of the study rats were placed in metabolic cages for 24-h urine collection. The animals’ weight, as well as food and water intake were measured at the end of the urine collection period. Urine was collected into 100-ml polyethylene tubes containing 500 μl of 6 N HCl. Blood samples from 4-h-fasted rats were taken by tail bleeding. Plasma was immediately separated by centrifugation (6000×g, 10 min, 4°C). Samples were stored at −20°C until assayed. Glucose, insulin, creatinine and catecholamine concentrations were measured in plasma, and creatinine and catecholamines were also measured in urine. Rats were acclimated to the procedure of blood pressure measurement throughout the experimental period. Indirect systolic blood pressure was measured daily by the tail-cuff method. The average systolic blood pressure values of the last 2 days of the study were used for statistical comparison.

In vitro studies After urine collection rats were killed by decapitation, kidneys were removed and renal cortical tissue immediately isolated and processed. Proximal tubule cell suspensions were prepared for measurement of the uptake of L-dopa or immunoblotting studies.

Preparation of proximal tubule cells Proximal tubule cells were prepared as previously described [6]. Briefly, kidney cortex was isolated, the tissue was minced on ice to a paste-like consistency which was digested with 0.7 mg/ml collagenase (Type V, Sigma Chemical, St. Louis, Mo., USA) in 10 ml Dulbecco’s Modified Eagle’s Medium supplemented with 20 mM HEPES and 24 mM NaHCO3; (pH 7.4). Incubation was carried out in a shaking water bath at 37°C for 30 min in an atmosphere of 95% O2/5% CO2. It was cooled on ice and poured through graded sieves (180–75–53 and 38 μm in pore size) to obtain a cell suspension. Proximal tubule cells were washed to separate the remaining blood cells and traces of collagenase and the final pellet was resuspended

in Krebs buffer (in mM: NaCl 120, KCl 4.7, MgSO4 1.2, CaCl2 2.4, NaHCO3 24, KH2PO4 1.2, EDTA 0.5, ascorbic acid 0.57 and glucose 11, pH 7.4). Each determination was performed using 50 μl of the cell suspension containing 150–250 μg protein. This preparation is known to consist mainly of proximal tubule cells [32]. The quality of each preparation was monitored by microscopy and cell viability was assessed by Trypan blue exclusion.

Transport of L-dopa into tubule cells The transport of L-dopa was determined as previously described [6]. Briefly, cells were preincubated for 20 min in Krebs buffer in the presence of an inhibitor of aromatic L-amino acid decarboxylase activity, 3-hydroxybenzylhydrazine (HBH; 250 μM) [6]. The concentration of HBH was selected as the lowest concentration that inhibited dopamine release to the incubation medium. In experiments aimed to study sodium dependence of L-dopa uptake, sodium chloride was replaced by choline chloride, and sodium bicarbonate by Tris-(hydroxymethyl)-aminomethan (TRIS base). In experiments aimed to evaluate the role of the cytoskeleton (nocodazole 1–10 μM, colchicine 50–100 μM, cytochalasin D 1 μM, latrunculin B 1 μM), or the effect of terbutaline (1 μM) or ouabain (1 mM), drugs were added to the medium before the 20-min preincubation. Unless specified, insulin was added to the medium 5 min before L-dopa. L-dopa uptake was started by the addition of Ldopa to the incubation medium and incubation was carried out for 20 min at room temperature. After centrifuging (4°C, 60×g, 3 min) and rapid removal of uptake solution, cells were immediately rinsed twice with ice-cold Krebs solution, and centrifuged (4°C, 60×g, 3 min). The pellet was resuspended in 200 μl of 0.3 N HClO4, disrupted (Sonifier Cell Disruptor, Heat Systems, Ultrasonics) and stored at −20°C until assayed. L-dopa concentrations were measured by HPLC-ED as described bellow.

Immunoblotting studies For immunoblotting studies, proximal tubule cells were resuspended in 1 ml of ice-cold homogenization buffer (20 mM Tris-HCl, pH 7.4) containing a protease inhibitor cocktail (Protease and phosphatase inhibitor cocktail for use with mammalian cell and tissue extracts, Sigma, St. Louis, Mo., USA). Cells were disrupted by three cycles of freezing and thawing and briefly sonicated [28]. To obtain cell membrane fractions, the lysates were centrifuged at 100,000×g (4°C, 60 min). The pellets were resuspended in homogenization buffer plus 1% (vol/vol) Triton X-100, incubated at room temperature for 30 min and centrifuged at 100,000×g (4°C, 60 min). The supernatants (considered as membrane fractions) containing equal amounts of protein (40 μg) were diluted in Laemmli buffer and subjected to SDS-PAGE and electrotransferred onto polyvinylidene difluoride (PVDF) membranes. Western blot analysis with antibodies against the insulin receptor β subunit (rabbit polyclonal IgG, Santa Cruz Biotechnology, Santa Cruz, Calif., USA) was performed. For sequential staining of the same blot, the antibodies were first stripped off (20-min incubation in 10% acetic acid and 40% methanol), then extensively washed and finally incubated with antibodies against α-tubulin (rabbit polyclonal IgG, Santa Cruz Biotechnology). The signal from peroxidase-conjugated anti-rabbit polyclonal antibodies (Peroxidase-linked anti rabbit IgG, Amersham Biosciences, Buckinghamshire, UK) was examined with an enhanced chemiluminescence method (ECL Plus Western blotting analysis system, Amersham Biosciences). The X-ray films were scanned using a HP ScanJet 5100C and HP Precision Scan software (Hewlett-Packard, Palo Alto, Calif., USA). The images obtained were analysed using NIH Image 1.57 software. For each experiment, data are expressed as the percentage of the signal from control animals.

87 Table 1 In vivo parameters from control and fructose-fed rats (FRrats). Values are means ±SEM; n=7 animals/group In vivo parameter

Control

Body weight (g) 457±9 Daily food intake (g) 10.5±4.9 Daily water intake (ml) 25±2 Plasma glucose (mg/%) 152±10 Plasma insulin (μU/ml) 40.3±7.2 Diuresis (ml/24 h) 27.3±3.7 Creatinine clearance (ml/min per 100 g) 0.79±0.06 Urine osmolarity (mosmol/l) 278±54 Systolic blood pressure (mmHg) 123±1

FR-rats 439±23 9.5±2.4 100±10* 155±8 93.0±16.0* 77.8±5.8* 0.78±0.12 209±14 134±2*

*p<0.05 vs. control Chemical determinations Plasma samples were assayed for glucose content and for creatinine levels, using commercial kits (Wiener, Argentina). Insulin was determined in plasma samples by radioimmunoassay as previously reported [17]. Protein concentrations were determined by the method of Bradford, using bovine serum albumin as a standard [3]. Urine osmolarity was measured on a Fiske Osmometer (Fisk Associates, Massachusetts, USA).

Fig. 1 Effect of phenylalanine on L-dopa uptake by proximal tubule cells. Freshly isolated rat proximal tubule cells were preincubated for 20 min in Krebs buffer in the absence (open bars) or presence of 40 μM L-phenylalanine (L-Phen, filled bars). Increasing concentrations of L-dopa were added to the medium and incubation was prolonged for 20 min. Results are expressed as the percentage of the intracellular L-dopa content. Each bar represents the mean ±SEM of three experiments performed in triplicate. (*p<0.05 vs. the same L-dopa concentration added in the absence of L-Phen) post hoc test. The criteria for significant differences was set at p<0.05. The lines of the Scatchard plots were determined by linear regression and the coefficients of linear regression (r) were 0.80 or better in all cases.

Results Catechol assays

In vivo experiments

The catechols in 10-μl aliquots of urine, 350 μl of plasma or 200 μl of cell homogenates were determined as reported previously [8]. Briefly, catechols in the samples were partially purified by batch alumina extraction, separated by reverse-phase high-pressure liquid chromatography using a 4.6×250 mm Zorbax RxC18 column (DuPont Company, USA) and quantified amperometrically by the current produced upon exposure of the column effluent to oxidizing and then reducing potentials in series using a triple-electrode system (ESA, Bedford, Mass., USA). Recovery through the alumina extraction step averaged 70–80% for dopamine, 45–55% for Ldopa and 40% for 3,4-dihydroxyphenylacetic acid (DOPAC). Catechol concentrations in each sample were corrected for recovery of an internal standard, dihydroxybenzylamine. Levels of L-dopa, dopamine and DOPAC were further corrected for differences in recovery of the internal standard and of these catechols in a mixture of external standards. The limit of detection was about 15 pg/volume assayed for each catechol.

Data analysis All data are presented as the mean ±SEM. Comparisons between groups were performed using the unpaired Student’s t test or by oneway analysis of variance (ANOVA) followed by Newman-Keuls Table 2 Plasma concentration and daily urinary output of Ldopa, dopamine and DOPAC in control and FR-rats. Values are means ±SEM; n=7 animals/ group. (DA Dopamine, DOPAC 3,4-dihydroxyphenylacetic acid)

*p<0.05 vs. control

Plasma (pg/ml) concentration of:

Urinary output of:

Body weight, food consumption, plasma glucose, glomerular filtration rate (measured as creatinine clearance) and urine osmolarity did not differ between control and FRrats. FR-rats had higher daily water consumption, urine output, plasma insulin concentrations and systolic blood pressure (p<0.05 in every case, Table 1). Table 2 shows plasma concentrations and daily urinary output of L-dopa, dopamine and the dopamine metabolite DOPAC, and the ratio between the urinary excretion of Ldopa and dopamine plus DOPAC, in control and FR-rats. Plasma concentrations of L-dopa were significantly lower in FR-rats, but no changes were found in plasma levels of dopamine or DOPAC. The excretion of L-dopa, dopamine and DOPAC was similar in both groups. However, the ratio between the excretion of L-dopa and dopamine plus DOPAC was higher in FR-rats (p<0.05). The delivery of L-dopa at the proximal tubules (calculated as plasma L-dopa × creatinine clearance) was decreased in FR-rats (in pg L-dopa per minute per 100 g: control 687±95; FR-rats 347±125) (p<0.05). Tubular LCompound

Control

FR-rats

L-dopa

874±106 217±36 1934±397 374±72 5730±302 12776±1661 0.018±0.002

533±56* 210±45 1186±218 469±58 5333±310 10904±604 0.031±0.004*

DA DOPAC L-dopa (ng/24 h) DA (ng/24 h) DOPAC (ng/24 h) L-dopa/(DA+DOPAC)

88

Fig. 2 Concentration-dependent effect of human insulin on L-dopa uptake by proximal tubule cells. Freshly isolated rat proximal tubule cells were preincubated for 10 min in Krebs buffer in the absence (control, open bar) or presence of increasing concentrations of human insulin (filled bars). L-dopa (final concentration 0. 2 μM) was added to the medium and incubation was prolonged for 20 min. Results are expressed as the percentage of the intracellular L-dopa content in control cells (705±186 pg L-dopa/mg protein). Each bar represents the mean ±SEM of four experiments performed in triplicate. (*p<0.05 vs. control)

dopa reabsorption, estimated as [(delivery of L-dopa at the proximal tubules − renal excretion of L-dopa)×100/delivery of L-dopa at the proximal tubules] was also lower in FR-rats (in % L-dopa reabsorption: control 81±4; FR-rats 51±9, p<0.05). In vitro experiments Proximal tubule cells freshly isolated from control rats were incubated in the presence of different concentrations of L-dopa. The uptake of L-dopa increased with increasing concentrations of the amino acid in the incubation medium (in pg L-dopa per mg protein per 20 min: 264, 396 and 498 for 0.2, 0.5 and 1 μM L-dopa respectively). The addition of L-phenylalanine, an amino acid structurally related to Ldopa, decreased the uptake of L-dopa (p<0.05) (Fig. 1). Sodium and pH dependence of L-dopa uptake by proximal tubule cells is depicted in Table 3. The uptake of L-dopa was not altered within the pH range analysed, but was impaired in the absence of sodium, since its intracellular levels were similar to those found when no Ldopa was present in the incubation medium. The effect of increasing concentrations of rat insulin (5– 200 μU/ml) on the uptake of L-dopa was evaluated by cell incubation in a medium containing 0.2 μM L-dopa. Even Table 3 Na+ and pH dependence of L-dopa uptake by proximal tubule cells. Cells were preincubated in Krebs buffer under the specified conditions. L-dopa (final concentration 0.2 μM) was added to the medium and incubation was prolonged for 20 min. Basal L-dopa

Fig. 3 Time course of the effect of human insulin on L-dopa uptake. Freshly isolated rat proximal tubule cells were preincubated with Krebs buffer in the presence of human insulin (200 μU/ml) for the indicated periods. L-dopa (final concentration 0. 2 μM) was added to the medium and incubation was prolonged for 20 min. Results are expressed as the percentage of the intracellular L-dopa content in control cells (658±157 pg L-dopa/mg protein). Each point represents the mean ±SEM of four experiments performed in triplicate

at the lowest concentration tested, rat insulin enhanced Ldopa uptake (37% over the control). The maximal effect was observed with 50 μU/ml insulin (in pg L-dopa per mg protein per 20 min, LD 332±79; LD+I 668±127, 101% over the control, p<0.05), and was insignificant with the highest concentration tested (200 μU/ml) (2% over the control). A similar pattern of insulin response was observed when human insulin was used (Fig. 2). Nevertheless, the effective concentrations of human insulin were approximately 5 times higher, with a maximal response achieved when cells were exposed to 200–300 μU/ml insulin (about 60% over the control, p<0.05). For operative reasons, human insulin (200 μU/ml) was used in all subsequent experiments The stimulatory effect of insulin on L-dopa uptake was fast, as it became evident as soon as after 1 min of preincubation and was no longer observed after 40 min of preincubation (Fig. 3). To analyse whether insulin regulates the transport of Ldopa through high- or low-affinity uptake sites, isolated proximal tubule cells were incubated in a medium intracellular L-dopa concentration (absence of L-dopa in the incubation medium) was 150±20 pg L-dopa/mg protein. Values are means ±SEM; n=4

uptake by proximal tubule cells (pg L-dopa/mg protein per 20 min) at: [Na+] 140

pH 7.4 [Na+] 140

[Na+] 80

[Na+] 0

pH 7.8

pH 7.4

pH 7.1

pH 6.8

432±32

373±17

163±37*

455±48

421±38

397±32

430±62

*p<0.05 vs. L-dopa uptake at [Na+] 140 mM

89

Fig. 4A, B Effect of insulin on the uptake of L-dopa by the highand low-affinity transport sites. Proximal tubule cells were preincubated for 5 min in Krebs buffer in the absence (control, open bar) or presence of 200 μU/ml insulin (I, filled bar). AL-dopa (LD) 0. 2 μM or B 5 μM was added to the medium and incubation was prolonged for 20 min. Results are expressed as the percentage of the intracellular L-dopa content in control cells (645±116 and 6589 ±1482 pg L-dopa/mg protein for A and B respectively). Each bar represents the mean ±SEM of four experiments performed in triplicate. (*p<0.05)

Fig. 5 Scatchard analysis of the uptake of L-dopa into tubule cells. Proximal tubule cells were preincubated for 5 min in Krebs buffer in the absence (◯) or presence (●) of 200 μU/ml insulin. L-dopa (LD, 0.05–0.7 μM) was added to the medium and incubation was prolonged for 20 min. Results shown are the mean of four experiments performed in triplicate

containing 0.2 μM or 5 μM L-dopa in the absence or presence of insulin. Insulin increased the transport of Ldopa by the high-affinity uptake sites (L-dopa 0.2 μM) by 65% (p<0.05) (Fig. 4A) whereas it did not modify the transport of the amino acid by the low-affinity uptake sites (L-dopa 5 μM) (Fig. 4B). Scatchard analysis of the uptake curves obtained in the presence of different concentrations of L-dopa (0.05– 0.70 μM) showed an increase of about 40% in the number of high-affinity uptake sites in the presence of insulin (0.59±0.05 vs. 0.82±0.09 pmol per mg protein per min;

Fig. 6A–C Effect of the disruption of actin or microtubule network on the insulin-enhanced L-dopa uptake. Freshly isolated rat proximal tubule cells were preincubated for 20 min in Krebs buffer in the absence (control, open bars) or presence of cytochalasin D (Cyt), latrunculin B (Lat) (A) nocodazole (Noc) (B), colchicine (Col) (C) and/or insulin (I, 5 min). L-dopa was added to the medium and incubation was prolonged for 20 min. Results are expressed as the percentage of the intracellular L-dopa content in control cells (454 ±77 pg L-dopa/mg protein). Each bar represents the mean ±SEM of three experiments performed in triplicate. (#p<0.05 vs. L-dopa, *p<0.05 vs. L-dopa+I)

n=4, p<0.05), without changes in the Km (666±75 vs. 769 ±83 nM in the absence and presence of insulin) (Fig. 5). The stimulatory effect of insulin on L-dopa uptake was abolished by depolymerization of actin cytoskeleton by cytochalasin D or latrunculin B (both 1 μM) (Fig. 6A). In contrast, disruption of the microtubule network by nocodazole (1–10 μM) (Fig. 6B) or colchicine (50–100 μM) (Fig. 6C) did not prevent this effect. Neither of these drugs affected the transport of L-dopa in the absence of insulin. To assess if the effect of insulin on L-dopa uptake is a consequence of the stimulation of Na+,K+-ATPase activity, experiments were performed in the presence of 10−3 M

90

Fig. 7 Effect of insulin on the inhibitory effect of terbutaline on Ldopa uptake. Cells were preincubated in Krebs buffer alone (control, open bar), or with the addition of terbutaline (Ter, 1 μM, 20 min) or/ and Insulin (I, 200 μU/ml, 5 min) (filled bars). L-dopa (LD, final concentration 0. 2 μM) was added to the medium and incubation was prolonged for 20 min. Results are expressed as the percentage of the intracellular L-dopa content in control cells (645±99 pg Ldopa/mg protein). Each bar represents the mean ±SEM of three experiments performed in triplicate. (*p<0.05 vs. control and vs. LD +Ter+I)

ouabain. The insulin-elicited increase of L-dopa uptake was not affected by ouabain (100±12; 141±14 and 159 ±18% of control for L-dopa, L-dopa + insulin and L-dopa + insulin + ouabain respectively, p<0.05 vs. control). Incubation of tubule cells with 1 μM terbutaline (a β2 adrenergic agonist) reduced the uptake of L-dopa by 50%. This effect was abolished by insulin (Fig. 7). We then explored whether the effect of insulin on the uptake of L-dopa was also evident in proximal tubule cells isolated from FR-rats. Absolute values for L-dopa uptake in the absence of insulin were similar between groups (in pg L-dopa per mg protein, 377±69 and 461±97 for control and FR-rats). However, the stimulatory effect of insulin was no longer observed in FR-rats (Fig. 8). Next, we analysed if the lack of response to insulin might be due to an altered expression of insulin receptors in cells from FR-rats. The expression of the β subunit of insulin receptors was lower in membrane fractions from proximal tubule cells isolated from FR-rats. Densitometric

Fig. 8 Effect of insulin on L-dopa uptake into proximal tubule cells isolated from control rats (C), or fructose-treated rats (FR). Cells were incubated for 20 min with L-dopa 0. 2 μM in the absence (open bar) or presence of insulin (200 μU/ml, 5 min). Results are expressed as the percentage of the intracellular L-dopa content in the absence of insulin. Each bar represents the mean ±SEM of four experiments performed in triplicate. *p<0.05

Fig. 9A–C Abundance of insulin receptors in membranes from tubule cells isolated from control or fructose-fed animals. A Representative immunoblot of the β subunit of insulin receptors in membrane fractions purified from proximal tubule cells isolated from control rats (C), or fructose-treated rats (FR). B Densitometric analysis of the expression of the β subunit of insulin receptors is shown as the percentage of the immunoreactive signal of the control. Each bar is the mean ±SEM (n=6). *p<0.05 vs. control. C Anti α tubulin was used for control of plasma membrane protein loading

analysis of the 90-kDa band indicates that the expression of the β subunit of insulin receptors in membranes from FR-rats was markedly decreased to 33% of the immunoreactive signal observed in cell membranes from control animals (p<0.05) (Fig. 9).

Discussion This study shows that insulin enhances the uptake of Ldopa into renal proximal tubule cells from control rats, and that this regulatory mechanism is absent in cells isolated from animals with an impaired insulin function. The effective concentrations of rat insulin to enhance the uptake of L-dopa (5–100 μU/ml) were within the range of physiological plasma levels under fasting conditions. The use of human insulin required higher concentrations (5 times higher) as compared to rat insulin. This may be explained by a lower affinity of the rat receptor for the human hormone. The fact that the highest insulin concentration tested produced no significant effect on Ldopa uptake might be attributed to down regulation of insulin receptors. Previous reports describing that the number of insulin receptors is inversely related to the ambient insulin concentration [29, 31] support this explanation. In our experimental conditions, the effect of insulin on L-dopa uptake remained unchanged in the presence of ouabain. This result indicates that the effect of insulin on L-dopa transport is independent from the stimulation of Na+,K+-ATPase activity, though insulin-induced stimulation of Na+,K+-ATPase activity in proximal tubule cells has been reported [10]. We have previously characterized the transport of Ldopa into freshly isolated rat proximal tubule cells as a

91

time-, and concentration-dependent process, reaching its maximal levels at 10 μM L-dopa. The uptake of L-dopa is driven by two different sites, a high-affinity site (Km 0.316 μM) and a low-affinity site (Km 1.53 μM) [6]. The concentrations of L-dopa used in the present study (0.2 μM and 5 μM) were selected so as to match the range of the Km of the high- and low-affinity uptake sites, respectively. Insulin had no effect on the uptake of L-dopa by the lowaffinity uptake sites. Conversely, as shown by the Scatchard plot, the increase in the uptake of L-dopa in the presence of insulin was associated with an increase in the number of high-affinity uptake sites without changes in the affinity constant. This observation supports the physiological importance of our finding since the L-dopa uptake site modulated by insulin operates within nanomolar L-dopa levels, which are close to L-dopa plasma concentrations (Table 2; [8]). Previous publications have reported that the transport of L-dopa into proximal tubules is mediated by a stereospecific aromatic amino acid transport system [7]. On the other hand, Soares-da-Silva has suggested that L-dopa is transported by the organic cation/H+ exchanger in isolated renal cortical tubules [30] whereas it is taken up through a sodium-independent amino acid transporter in LLC-PK1 cells [33]. Our results demonstrate that L-dopa uptake is: (1) blunted in the presence of L-phenylalanine, (2) not affected by changes in the pH of the incubation medium and (3) impaired in the absence of sodium. This indicates that the transport of L-dopa into rat isolated tubule cells, driven by an amino acid transporter as described by Chan [7], is a sodium-dependent process. Because our experiments were performed on isolated proximal tubule cells, we cannot identify whether L-dopa transport is driven through the apical or basolateral membrane. Changes in plasma membrane amino acid transport activity in renal cortex have been suggested to result from the translocation of carriers to and from intracellular stores [16]. Furthermore, insulin-induced enhancement of amino acid transport has been correlated with the recruitment of preformed carriers to the plasma membrane [13]. Similarly, the increase in the rate of glucose transport elicited by insulin appears to be mainly due to an increase in the number of GLUT-4 glucose transporters in the plasma membrane [35]. There is increasing evidence in favour of the involvement of actin filaments in the translocation of vesicles containing GLUT 4 transporters to the plasma membrane in response to insulin stimulation. Preincubation of myocytes or adipocytes with agents that depolymerize actin filaments inhibits insulin-dependent glucose transport [21, 36, 37] and impairs the insertion of GLUT-4 glucose transporters in the plasma membrane [35]. On the other hand, data reviewed by Hamm Alvarez suggest that the insertion of channels and transporters at the apical surface of kidney epithelia is dependent on the integrity of the microtubule system [15]. In our hands the effect of insulin was not altered by nocodazole or colchicine. Instead, in agreement with previous data regarding insulininduced glucose transport [21, 35, 36, 37], the effect of insulin on L-dopa uptake was blunted by the disruption of the actin network. Taking these data into consideration, it is tempting to speculate that stimulation of insulin receptors could trigger the translocation of amino acid

carriers from intracellular pools to the plasma membrane through the actin cytoskeleton, resulting in the enhanced uptake of L-dopa. The short preincubation time needed for insulin to elicit its maximal effect (2–5 min) supports this hypothesis and disregards the possibility of promotion of protein and/or RNA synthesis to explain the increase in the number of high-affinity L-dopa uptake sites. Previous results from our laboratory have demonstrated that stimulation of β2-adrenergic receptors results in a reduction in the uptake of L-dopa in isolated renal tubule cells [6]. The involvement of β adrenergic receptors in amino acid transport has also been reported to occur in the jejunum and the blood brain barrier [22, 34]. Here we confirm that the β2-adrenergic agonist terbutaline decreases the uptake of L-dopa in tubule cells. This effect was completely reversed by insulin. Opposite effects between β2-adrenergic agonists and insulin in other systems have already been described. Functional studies demonstrated that insulin counteracts the catecholaminestimulated accumulation of intracellular cAMP [19]. The underlying mechanism was postulated as an increase in tyrosine kinase activity subsequent to insulin receptor stimulation, which results in phosphorylation of Tyr350, Tyr354 and Tyr364 of the β2-adrenergic receptor. Consequently, the receptor uncouples from the stimulatory G protein controlling adenylyl cyclase [20]. Our finding of the counter regulatory effect of insulin on the decrease of L-dopa uptake mediated by β2-adrenergic receptors adds support to the involvement of insulin in the multifactorial regulation of L-dopa reabsorption. The interactions between the intracellular signalling pathways triggered by stimulation of insulin- and β2-adrenergic receptors deserves further investigation. The stimulatory effect of insulin on the uptake of L-dopa was abolished in tubule cells isolated from FR-rats. Among other possibilities, the lower expression of the β subunit of insulin receptors we found in renal cortex from FR-rats may explain the lack of response to insulin in this model. Experiments performed in vivo showed no difference in urinary excretion of L-dopa, dopamine or dopamine metabolites between control or FR-rats whereas a lower level of plasma L-dopa was found in the latter group. The lower L-dopa plasma concentration together with an unchanged glomerular filtration rate may explain the lower L-dopa delivery at the proximal tubules in FR-rats. The fact that FR-rats had a lower delivery of L-dopa at the proximal tubules together with similar urine L-dopa excretion suggests impaired reabsorption of the amino acid. In addition, since DA and its metabolite DOPAC originate in the tubular cell from the reabsorbed L-dopa, any changes in the ratio L-dopa/(DA+DOPAC), expressed as daily urinary output, will reflect either a change in Ldopa reabsorption or L-dopa decarboxylase activity. The high activity of L-dopa decarboxylase in renal cortex makes the second possibility improbable [2, 23]. Consequently, the increased ratio L-dopa/(DA+DOPAC) constitutes additional evidence for altered L-dopa reabsorption in FR-rats. Our results show that insulin, at concentrations similar to those found in plasma of fasted rats, enhances the uptake of L-dopa in renal tubule cells isolated from control

92

animals by increasing the maximal L-dopa transport capacity through the high-affinity sites. Stimulation of Ldopa uptake by insulin proved to be absent in tubule cells isolated from insulin-resistant rats; furthermore L-dopa reabsorption was also impaired in in vivo studies performed under insulin resistance conditions. Acknowledgements We thank Dr. Pablo Damiano and Dr. Liliana Karabatas for skillful assistance with blood pressure and insulin measurements. This project was supported by grants from the Ministerio de Salud de la Nación, Becas de Investigación “Ramón Carrillo–Arturo Oñativia” (2000–2001); Agencia Nacional de Promoción Científica y Técnica (PICT 05–08658); and CONICET PIP573. M.B. Barontini, C.F. Mendez and S. Nowicki are Senior Investigators from CONICET, Argentina.

References 1. Aperia A (2000) Intrarenal dopamine: a key signal in the interactive regulation of sodium metabolism. Annu Rev Physiol 62:621–647 2. Armando I, Nowicki S, Aguirre J, Barontini M (1995) A decreased tubular uptake of dopa results in defective renal dopamine production in aged rats. Am J Physiol 268:F1087– F1092 3. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254 4. Butlen DS, Vadrot S, Roseau S, Morel F (1988) Insulin receptors along the rat nephron: [125I]insulin binding in microdissected glomeruli and tubules. Pflugers Arch 412:604–612 5. Carranza A, Karabatas L, Barontini M, Armando I (2001) Decreased tubular uptake ofL-3,4-dihydroxyphenylalanine in streptozotocin-induced diabetic rats Horm Res 55:282–287 6. Carranza A, Nowicki S, Barontini M, Armando I (2000)L-dopa uptake and dopamine production in tubular cells are regulated by beta (2) adrenergic receptors. Am J Physiol 279:F77–F83 7. Chan YL (1976) Cellular mechanisms of renal tubular transport of L-dopa and its derivatives in the rat: microperfusion studies. J Pharmacol Exp Ther 199:17–24 8. Eisenhofer G, Goldstein DS, Stull R, Keiser HR, Sunderland T, Murphy DL, Kopin IJ (1986) Simultaneous liquid – chromatographic determination of 3,4-dihydroxyphenylglycol, catecholamines and 3,4-dihydroxyphenylalanine in plasma, and their response to inhibition of monoamine oxidase. Clin Chem 32:2030–2033 9. Elwan MA (1998) Synthesis of dopamine fromL-3,4-dihydroxyphenylalanine by human amniotic epithelial cells. Eur J Pharmacol 354:R1–R2 10. Féraille E, Carranza ML, Rousselot M, Favre H (1994) Insulin enhances sodium sensitivity of Na-K-ATPase in isolated rat proximal convoluted tubule. Am J Physiol 267:F55–F62 11. Gesek FA, Schoolwerth AC (1991) Insulin increases Na+-H+ exchange activity in proximal tubules from normotensive and hypertensive rats. Am J Physiol 260:F695–F703 12. Gonska T, Hirsch JR, Schlatter F (2000) Amino acid transport in the renal proximal tubule. Amino Acids 19:395–407 13. Goshima K, Masuda A, Owaribe K (1984) Insulin-induced formation of ruffling membranes of KB cells and its correlation with enhancement of amino acid transport. J Cell Biol 98:801– 809 14. Guidotti GG, Borghetti AF, Gazzola GC (1978) The regulation of amino acid transport in animal cells. Biochim Biophys Acta 515:329–366 15. Hamm-Alvarez SF, Sheetz MP (1998) Microtubule-dependent vesicle transport: modulation of channel and transporter activity in liver and kidney. Physiol Rev 78:1109–1129 16. Hensley CB, Mircheff AK (1994) Complex subcellular distribution of sodium- dependent amino acid transport systems in kidney cortex and LLC-PK1/CL4 cells. Kidney Int 45:110–122

17. Herbert V, Lau KS, Gottlieb CW, Bleicher SJ (1965) Coated charcoal immunoassay of insulin. J Clin Endocrinol Metab 25:1375–1384 18. Hwang IS, Ho H, Hoffman BB, Reaven GM (1987) Fructoseinduced insulin resistance and hypertension in rats. Hypertension 10:512–516 19. Karoor V, Baltensperger K, Paul H, Czech MP, Malbon CC (1995) Phosphorylation of tyrosyl residues 350/354 of the beta adrenergic receptor is obligatory for counterregulatory effects of insulin. J Biol Chem 270:25305–25308 20. Karoor V, Malbon CC (1998) G-protein-linked receptors as substrates for tyrosine kinases: cross-talk in signalling. Adv Pharmacol 42:425–428 21. Khayat ZA, Tong P, Yaworsky K, Bloch RJ, Klip A (2000) Insulin-induced actin filament remodeling colocalizes actin with phosphatidylinositol 3-kinase and GLUT4 in L6 myotubes. J Cell Sci 113:279–290 22. Kreydiyyeh SI (1997) Alpha and beta adrenoceptors mediate the inhibitory effect of epinephrine on the mucosal uptake of phenylalanine in the rat jejunum. Can J Physiol Pharmacol 75:1312–1315 23. Lee MR (1993) Dopamine and the kidney. Ten years on. Clin Sci 84:357–375 24. McDowell HE, Eyers PA, Hundal HS (1998) Regulation of system A amino acid transport in L6 rat skeletal muscle cells by insulin, chemical and hyperthermic stress. FEBS Lett 441:15– 19 25. Mezey E, Eisenhofer G, Hansson S, Harta G, Hoffman BJ, Gallatz K, Palkovits M, Hunyady B (1999) Non-neuronal dopamine in the gastrointestinal system. Clin Exp Pharmacol Physiol Suppl 26:S14–S22 26. Mezey E, Eisenhofer G, Harta G, Hansson S, Gould L, Hunyady B, Hoffman BJ (1996) A novel nonneuronal catecholaminergic system: exocrine pancreas synthesizes and releases dopamine. Proc Natl Acad Sci USA 93:10377–10382 27. Miatello RM, Damiani MT, Nolly HL (1998) Cardiovascular kinin-generating capability in hypertensive fructose-fed rats. J Hypertens 16:1273–1277 28. Nowicki S, Kruse MS, Brismar H, Aperia A (2000) Dopamineinduced translocation of protein kinase C isoforms visualized in renal epithelial cells. Am J Physiol 279:C1812–C1818 29. Okabayashi Y, Maddux BA, McDonald AR, Logsdon CD, Williams JA, Goldfine ID (1989) Mechanisms of insulininduced insulin-receptor downregulation. Decrease of receptor biosynthesis and mRNA levels. Diabetes 38:182–187 30. Pinto-do-O PC, Soares-da-Silva P (1996) Studies on the pharmacology of the inward transport of L-DOPA in rat renal tubules. Br J Pharmacol 118:741–747 31. Sechi LA, Griffin CA, Zingaro L, Valentin JP, Bartoli E, Schambelan M (1997) Effects of angiotensin II on insulin receptor binding and mRNA levels in normal and diabetic rats. Diabetologia 40:770–777 32. Seri I, Kone BC, Gullans SR, Aperia A, Brenner BM, Ballermann BJ (1988) Locally formed dopamine inhibits Na+-K+-ATPase activity in rat renal cortical tubule cells. Am J Physiol 279:F666–F673 33. Soares-da-Silva P, Serrâo MP (2000) Molecular modulation of inward and outward apical transporters of L-dopa in LLC-PK1 cells. Am J Physiol 279:F736–F746 34. Takao Y, Kamisaki Y, Itoh T (1992) Beta-adrenergic regulation of amine precursor amino acid transport across the blood- brain barrier. Eur J Pharmacol 21:245–251 35. Tong P, Khayat ZA, Huang C, Patel N, Ueyama A, Klip A (2001) Insulin-induced cortical actin remodeling promotes GLUT4 insertion at muscle cell membrane ruffles. J Clin Invest 108:371–381 36. Tsakiridis T, Vranic M, Klip A (1994) Disassembly of the actin network inhibits insulin-dependent stimulation of glucose transport and prevents recruitment of glucose transporters to the plasma membrane. J Biol Chem 269:29934–29942 37. Wang Q, Bilan PJ, Tsakiridis T, Hinek A, Klip A (1998) Actin filaments participate in the relocalization of phosphatidylinositol 3-kinase to glucose transporter-containing compartments and in the stimulation of glucose uptake in 3T3-L1 adipocytes. Biochem J 331:917–928