Testosterone Down-Regulates Ornithine Aminotransferase Gene and Up-Regulates Arginase II and Ornithine Decarboxylase Genes for Polyamines Sy
http://www.100md.com
内分泌学杂志 2005年第2期
Université Claude Bernard, Faculté de Médecine Lyon R.T.H. Laennec (O.L.), Laboratoire de Physiopathologie Métabolique et Rénale, Institut National de la Santé et de la Recherche Médicale Unité 499, 69372 Lyon Cedex 08; Université Claude Bernard, Centre de Génétique Moléculaire et Cellulaire (J.-J.D.), Centre National de la Recherche Scientifique, Unité Mixte de Recherche 5534, 69622 Villeurbanne; and Hospices Civils de Lyon (O.B., H.D.), Service de Radioanalyse, Centre de Médecine Nucléaire, H?pital Neuro-Cardiologique and Institut National de la Santé et de la Recherche Médicale, Equipe Mixte de Recherche 0322, 69394 Lyon Cedex 03, France
Address all correspondence and requests for reprints to: Dr. Olivier Levillain, Université Claude Bernard, Faculté de Médecine Lyon RTH Laennec, U 499 Institut National de la Santé et de la Recherche Médicale, 7, rue G. Paradin, 69372 Lyon Cedex 08, France. E-mail: Olivier.Levillain@laennec.univ-lyon1.fr.
Abstract
The enzymes ornithine aminotransferase (OAT) and ornithine decarboxylase (ODC) share L-ornithine as a common substrate and arginase II produces this amino acid. In the murine kidney, testosterone induced ODC gene expression and polyamine production, but it is unknown how OAT gene is expressed under androgen treatment. These experiments were designed to study the influence of testosterone on the renal expression of OAT gene. Pharmacological and physiological doses of testosterone were injected into female and castrated male mice. Total RNA and soluble proteins extracted from whole kidneys were analyzed by Northern and Western blots, respectively. The results clearly indicate that pharmacological doses of testosterone simultaneously down-regulated the level of OAT protein and up-regulated the expression of arginase II and ODC genes. Variations of the levels of OAT protein and arginase II mRNA and protein were strongly correlated with testosteronemia. Orchidectomy increased the renal level of OAT protein and decreased that of ODC and arginase II. These effects were reversed by injecting a physiological dose of testosterone into castrated male mice. In conclusion, OAT and ODC genes are inversely regulated by testosterone in the mouse kidney. Consequently, in kidneys of testosterone-treated mice, L-arginine-derived ornithine produced by arginase II might be preferentially used by ODC for putrescine production rather than by OAT. This metabolic fate of L-ornithine was facilitated by decreasing OAT gene expression. In contrast, in female and castrated male mice devoided of testosterone, OAT gene is highly expressed and L-ornithine is converted into L-glutamate
Introduction
IN THE MURINE kidney, the nucleus-encoded mitochondrial arginase II (EC 3.5.3.1) catalyzes the irreversible hydrolysis of L-arginine into urea and L-ornithine in the cortical and medullary proximal straight tubules (CPST and OSPST, respectively) (1). L-Arginine-derived ornithine is either metabolized into polyamines or transaminated. In the first pathway, polyamine synthesis is controlled by the key enzyme ornithine decarboxylase (ODC; EC 4.1.1.17) that converts L-ornithine into putrescine. In the male mouse kidney, ODC gene is highly expressed in the proximal convoluted tubule (PCT) and, to a lesser extent, in CPST and OSPST (2, 3, 4). In contrast, in the female mouse kidney, none of the nephron segments expressed ODC gene (5). In the second pathway, transamination of L-ornithine requires the key enzyme ornithine aminotransferase (OAT; EC 2.6.1.13). OAT is a nucleus-encoded mitochondrial protein that catalyzes a reversible reaction leading to the production of L-glutamate and glutamate--semialdehyde in the presence of L-ornithine and -ketoglutarate. A high OAT activity has been reported in the murine kidney (6, 7). Moreover, we observed that OAT gene was more expressed in female than in male mouse kidneys (O. Levillain, personal unpublished data). In a detailed study, we showed that gabaculine, a suicide inhibitor of OAT, deeply reduced L-ornithine decarboxylation, suggesting that OAT might be expressed in PCT, CPST, and OSPST of the female mouse kidney (5). In the male mouse nephron, although OAT protein has been detected in the different renale zones (Levillain, O., personal unpublished data), its precise distribution within the nephron remains to be identified.
It is now well-known that ODC is a highly androgen-inducible protein in the murine kidney (8). Administration of pharmacological doses of testosterone to female mice induced a rapid and prolonged increase in renal ODC mRNA, protein, and enzyme activity over a few days; enhanced the renal production of putrescine; and sharply increased putrescine excretion in urine (9, 10, 11). Testosterone induced the expression of ODC gene in the three subsegments of the proximal tubule: PCT, CPST, and OSPST (11). In addition to ODC, several genes are overexpressed by testosterone in the mouse kidneys (12). Renal arginase II activity is increased by injecting testosterone into female and castrated male mice (7, 13). However, at present, it remains unknown whether testosterone regulates arginase II gene at the transcriptional and/or at the translational levels. Moreover, to our knowledge, the influence of androgens on the expression of OAT gene in the murine kidney has never been analyzed.
Studying the regulation of OAT gene by testosterone in the mouse kidney is particularly relevant because OAT and ODC share L-ornithine as a common substrate, and the metabolic fate of L-arginine-derived ornithine under such a physiological condition is not known. The mitochondrial colocalization of arginase II and OAT supports the view that L-ornithine generated by arginase II might be immediately transaminated by OAT rather than decarboxylated by the cytosolic ODC. Surprisingly, the production of putrescine is dramatically enhanced in kidneys of testosterone female mice. To explain this result, we hypothesized that the expression of OAT gene could be down-regulated by testosterone to metabolize L-arginine-derived ornithine preferentially into polyamines rather than into L-glutamate. Alternatively or simultaneously, the up-take of L-ornithine from the arterial blood across the basolateral membranes of proximal tubular cells might be enhanced (14). The first hypothesis is supported by a recent report that indicates that OAT activity is decreased by 2-fold in testosterone-treated female mice but, surprisingly, remained unaffected in testosterone-treated male mice (7). Moreover, at present, it is not known whether OAT gene is regulated by physiological levels of testosterone in male mouse kidney.
The present study was designed: 1) to analyze the expression of OAT gene in kidneys of testosterone-treated female mice; 2) to test whether testosterone simultaneously induced the expression of arginase II and ODC gene and repressed that of OAT to shift L-arginine-derived ornithine into the polyamine pathway; and 3) to determine whether testosterone physiologically regulates the expression of OAT gene in male mouse kidney. For this last experiment, control, orchidectomized, and testosterone-treated orchidectomized mice were used.
Materials and Methods
Animals and treatment
Eight- to 9-wk-old adult female [30–32 g body weight (BW)] and male (35–40 g BW) OF-1 Swiss (IOPS Caw) mice from Iffa Credo (L’Arbresle sur Orge, France) had free access to tap water and standard laboratory diet (Souffirat, 20% protein, Genthon, France). Animals were housed in a room maintained at 20 C with a 12-h light, 12-h dark cycle. Mice were anesthetized (ip) using 0.1 ml/100 g BW pentobarbital sodium diluted 1:2 in 0.9% NaCl solution (Nembutal 6%, Clin Midy, Paris, France). The experiments were approved by the local Committee for Animal Experiments.
Twenty-four female mice were subdivided into five groups: one untreated (control) and four androgen-treated groups. Mice subjected to testosterone treatment were injected sc with 150 μl testosterone propionate (31 mg/ml in sesame oil, i.e. 155 μg/g BW). Injections were performed at 0800 h, and mice were treated for a period of 1, 2, 3, or 5 consecutive days. Male mice were untreated.
Young adult male mice of 30-d-old, rather than older adult male mice, were used to prevent the accumulation of testosterone in their tissues and plasma (15). Twenty-four 30-d-old male mice were subdivided into four groups of six mice: nonoperated (group I, control), sham-operated (group II), and two groups of orchidectomized mice. Eleven days later (i.e. 41 d after birth), mice of groups I, II, and III were killed, whereas mice of group IV were killed 7 d later (i.e. 18 d after orchidectomy).
Twelve 30-d-old male mice were subdivided into four groups of three mice: sham-operated (group V, control), 11-d orchidectomized (group VI), 11-d orchidectomized treated with sesame oil (group VII), and 11-d orchidectomized treated with testosterone + sesame oil (group VIII). Mice subjected to oil or testosterone treatment were injected sc with 150 μl vehicle or testosterone propionate (3.1 mg/ml in sesame oil, i.e. 15 μg/g BW or 0.55 mg/mouse). All mice were killed 48 h after the treatment.
Sampling of blood and plasma for testosterone analysis
Blood was collected in the vena cava of all male and female mice with a 25-gauge needle (Neolus, VWR, Limonest, France) mounted on a 1-ml syringe (Terumo, VWR) prealably heparinized (Heparin, Roche, Meylan, France). Blood was immediately transferred in a cold BD Vacutainer tube, centrifuged at 4000 x g for 20 min at 4 C. Plasma was frozen and stored in liquid nitrogen until testosterone measurement. Testosterone was measured by RIA after extraction by organic solvent and partition chromatography of the plasma samples as previously described (16).
Sampling of the kidneys for Western blot analyses
In different series of experiments performed to analyze the influence of sexual hormones on the renal level of arginase II protein, kidneys of control (untreated and nonoperated), sham-operated, 11- and 18-d orchidectomized male, oil- and testosterone-treated orchidectomized male mice, and control (untreated and nonoperated) and testosterone-treated female mice were rapidly removed, decapsulated; the blood contained in each kidney was removed with blotting paper (free-blood). The kidney was placed in a sterilized Eppendorf tube, frozen, and conserved in liquid nitrogen. The kidneys of testosterone-treated female mice used in this study are the same as those used to analyze the time course expression of ODC protein and published in Ref.11 .
Protein extraction and Western blot analyses of arginase II protein
Each frozen kidney was mixed at 4 C with a Turrax in the proportion of 100 mg frozen tissue/2 ml of lysing buffer (17) containing 1 mM protease inhibitor cocktail, 1 mM phenylmethylsulfonylfluoride, and 1 mM benzamidine, and then centrifuged at 10,000 x g for 30 min at 4 C. Protein concentrations were determined in the supernatant using the Bradford protein assay (18). For each kidney, 100-μg samples of soluble protein were subjected to 10% PAGE containing 0.1% sodium dodecyl sulfate (SDS) using 6 watts/gel. Ten microliters of a protein ladder (precision plus protein standards, Bio-Rad, Marnes la Coquette, France) were deposed on each gel to further verify the size of the protein of interest. Proteins were transferred to a polyvinylidene difluoride membrane (0.45 μm, Immobilon-P, Millipore, St Quentin en Yvelines, France) at 150 mA for 90 min. Proteins were fixed on the membrane with Ponceau S solution for 15 min. Immunoblots were washed twice in 1x Tris-buffered salt + 0.1% Tween 20 (TBST) and immersed twice in a blocking solution consisting of 5% fat-free milk powder in 1x TBST for 30 min.
Immunoblots were incubated with the following primary antibodies in 5% milk-1x TBST: a polyclonal rabbit antimouse-arginase II [CovalAb, dilution 1:1000 (19)], a polyclonal rabbit antimouse-OAT [CovalAb, dilution 1:1000 (19)], a polyclonal rabbit antihuman-ODC (dilution 1:500, Eurodiagnostica, Paris, France), a polyclonal rabbit antisubstractive clone A [SA; dilution 1:350 (20)], a monoclonal mouse antiglyceraldehyde-3-phosphate dehydrogenase (G3PDH; dilution 1:170, Chemicon International, Temecula, CA), a monoclonal mouse anti-?-actin (dilution 1:2000) or a monoclonal mouse anti-?-tubulin (dilution 1:1000). ?-Actin, ?-tubulin, and G3PDH were used as controls of equal loading and transfer of total proteins. SA has been identified as a medium-chain acyl-CoA synthetase, also called acetate-CoA ligase, and is a truly androgen-dependent gene (20). For this reason, SA was used as a marker of orchidectomy. The blots were washed three times for 10 min in 1x TBST and incubated for 60 min with either peroxidase-conjugated antirabbit IgG or antimouse IgG secondary antibodies (dilution 1:10,000) in 5% milk-1x TBST. Blots were washed three times for 10 min in 1x TBST, and antibody binding was revealed using an enhanced chemiluminescence (ECL) Western Blotting Kit. ECL detection was performed using X-MAT film (Kodak, Rochester, NY). Low-exposure film was scanned, and the intensity optical densitometry (IOD) of the bands was estimated using the ImagerMaster Total Lab version 2.01 program (Pharmacia, Orsay, France).
Kidney preparation for Northern blot analyses
In a series of experiments designed to test the influence of testosterone on renal arginase II mRNA levels, kidneys of four untreated male, four untreated female, and nine testosterone-treated female mice were rapidly removed and decapsulated. Each free-blood kidney was placed in a sterilized Eppendorf tube, frozen, and conserved in liquid nitrogen.
RNA extraction and Northern blot analyses of arginase II and ?-actin mRNA
Each frozen kidney was homogenized in RNAxel solution (Eurobio, Cortaboeuf, France), and total RNA was extracted according to the manufacturer’s recommendations and maintained at 4 C. RNA were rinsed twice with 70% ethanol and dried in a Speed Vac. RNA were resuspended in cold 10 mM Tris-HCl and 1 mM EDTA, pH 8.0, and their concentrations were determined by absorbance at 260 nm. Fifteen micrograms of RNA samples were submitted to 1.2% agarose gel electrophoresis. The gel was treated for 20 min in 50 mM NaOH, then for 20 min in a solution containing 0.5 M Tris and 1.5 M NaCl; RNA was transferred overnight to a nylon membrane (Appligène, Illkirch, France) and immobilized using an UV cross-linker (Appligène).
Membranes were hybridized with murine 32P-labeled cDNAs corresponding to arginase II [pBSK– arginase II EcoRI-EcoRI, NCBI accession NM 009705 (21)] and ?-actin [pAL41-cytoplasmic ?-actin, PstI-PstI, (22)]. cDNA was labeled using the RTS RadPrime DNA labeling system (GIBCO BRL, Life Technologies) and -[32P]-deoxycytidine triphosphate. Hybridization was performed overnight at 65 C. Membranes were washed three times in 2x saline sodium citrate (0.3 M NaCl and 30 mM sodium citrate), 5 mM phosphate buffer, and 0.1% SDS and washed three times in 0.5x saline sodium citrate, 3 mM phosphate buffer, and 0.1% SDS. The amount of radioactivity hybridized to specific mRNA was estimated after scanning densitometry of the membranes using a PhosphoImager SI (Molecular Dynamics, Amersham, Orsay, France). Quantification of ?-actin mRNA was used as a control of equal loading and RNA transfer.
Chemicals
Salts and most chemicals, Ponceau S solution, monoclonal mouse anti-?-actin, secondary anti-IgG antibodies, and Kodak X-MAT film were purchased from Sigma (St. Quentin Fallavier, France). Protease inhibitor cocktail was purchased from Boehringer Mannheim (Strasbourg, France). Agarose Seakem GTG was from TEBU (Le Perray-en-Yvelines, France). ECL Western Blotting Kits, -[32P]-deoxycytidine triphosphate (9.25 MBq/25 μl), ImagerMaster Total Lab version 2.01 program, liquid scintillation counting mixture (Aqueous Counting Scintillant ACS II), and monoclonal mouse anti-?-tubulin antibody were purchased from Amersham (Buckinghamshire, UK; Orsay, France).
Results and statistical analyses
Values are means ± SE except when n = 2. The calculations were as follows: for each group of six or three mice and for each protein, the mean IOD of the bands was calculated (see Figs. 5 and 6). The mean IOD of the untreated (group I) (see Fig. 5) and sham-operated (group V) (see Fig. 6) mice were used as a reference (control). For each mouse, the IOD value of a given protein was divided by the mean IOD value of the control group. Consequently, this ratio value is 1 in each control group. Then, these ratios were related to those of ?-actin and/or G3PDH.
FIG. 5. Immunoblot assessing the abundance of OAT, arginase II (AII), and ODC protein in kidney of control (Cont), sham-operated (Sham), and 11- and 18-d orchidectomized male mice. Immunoblots were loaded with samples of 100 μg soluble proteins from six mice (see Table 1). Blots were probed with antibodies to OAT, arginase II, ODC, SA as a control of the efficiency of orchidectomy, and ?-actin and G3PDH as controls of protein loading and transfer. A, To simplify the figure, a representative immunoblot that corresponds to one of the six mice from each group was shown. B–D, Relative quantitation of ODC (B), OAT (C), and arginase II (D) protein levels after scanning densitometry of the immunoblots. Calculations were performed from data obtained with six mice (see Materials and Methods). Values are means ± SE. Differences between groups were statistically analyzed by Kruskal-Wallis test (ODC, P < 0.0011; OAT, P < 0.0003; arginase II, P < 0.017) and followed by Mann-Whitney test. *, P < 0.05, controls vs. 11- or 18-d; #, P < 0.05, Sham vs. 11- or 18-d. Three replicate experiments were performed.
FIG. 6. Immunoblot assessing the abundance of OAT, arginase II (AII), and ODC in the kidney of sham-operated, orchidectomized, oil-treated (Oil), and testosterone-treated (Oil+T) male mice. Immunoblots were loaded with samples of 100 μg soluble proteins from three mice (see Table 1). Blots were probed with antibodies to OAT, arginase II, ODC, SA as a control of the efficiency of orchidectomy, and ?-actin and G3PDH as controls of protein loading and transfer. A, To simplify the figure, a representative immunoblot that corresponds to one of the three mice from each group was shown. B–D, Relative quantitation of ODC (B), OAT (C), and arginase II (D) to G3PDH protein after scanning densitometry of the immunoblots. Calculations were performed from data obtained with three mice (see Materials and Methods). Values are means ± SE. Differences between groups were statistically analyzed by Kruskal-Wallis test (ODC, P < 0.014; OAT, P < 0.032; arginase II, P < 0.026) and followed by Mann-Whitney test. *, P < 0.05, Sham vs. orchidectomy; #, P = 0.05, Oil vs. T + Oil. Three replicate experiments were performed.
Where appropriate, statistical differences were assessed using the Kruskal-Wallis and/or the Mann-Whitney tests (StatView SE+Gr and StatView 5) at the 95% level of significance. For correlation analyses, the correlation coefficient r2 was calculated with Microsoft Excel, and P was determined from tables at the 95% level of significance.
Results
Regulation of OAT, arginase II, and ODC genes by testosterone in female mouse kidneys
The goal of these series of experiments was to test whether testosterone simultaneously induced the expression of arginase II and ODC genes and repressed that of OAT. In addition, it has been verified whether a correlation might be established between testosteronemia and the levels of OAT and arginase II mRNA and proteins.
Time-course effect of testosterone on the expression of OAT protein in female mouse kidneys.
In a previous report (11), we demonstrated that pharmacological doses of testosterone induced a sharp increase in ODC mRNA and protein levels in the female mouse kidneys. For this reason, ODC was used as a control to prove the efficiency of the hormonal treatment. As expected, in these experiments, testosterone treatment induced a progressive increase in the level of the 51- and 53-kDa ODC proteins (Fig. 1A, left). In male and female mouse kidneys, anti-OAT antibody revealed a 48-kDa protein that corresponds to the expected size of the OAT polypeptide (Swiss-Prot P29758 mouse: 48,354 Da; Fig. 1A, left). The renal content of ?-actin, a 44-kDa protein, was not affected by testosterone treatment (Fig. 1A, left). The relative level of OAT to ?-actin protein was about 2-times higher in female than in male mouse kidneys (Fig. 1A, right, Mann-Whitney, P < 0.05). Testosterone treatment induced a progressive and sharp decrease in the amount of OAT protein (Fig. 1A, right, Kruskal-Wallis, P < 0.0013). The relative abundance of OAT to ?-actin protein was about 6-times lower in the kidneys of female mice treated for 5-d with testosterone compared with that of the controls (Fig. 1A, right; Mann-Whitney, P < 0.009).
FIG. 1. Immunoblot assessing the abundance of OAT, arginase II, ODC, ?-actin, and ?-tubulin protein in kidney of male (M), female (F), and testosterone-treated female mice. Female mice were injected sc with 150 μl testosterone propionate (31 mg/ml in sesame oil). Injections were performed every day at 0800 h for a period of 1, 2, 3, and 5 consecutive days. Each lane corresponds to one different mouse, 100 μg (A), and 50 μg (B) of soluble protein. Three immunoblots were performed for (A) and (B). Left, Representative immunoblots probed with antibodies to OAT, arginase II (AII), ODC, ?-tubulin, and ?-actin and revealed by ECL after exposure to an x-ray film. Right, Relative abundance of OAT to ?-actin protein (A) and arginase II to ?-tubulin (B) protein after scanning densitometry of the immunoblots. Values are means ± SE; n = 5 films analyzed in one of the three experiments (A), whereas n = 3 replicate experiments (B). Differences were first statistically analyzed by Kruskal-Wallis followed by Mann-Whitney test: A: *, P < 0.05, M vs. F; **, P < 0.009, F vs. 5-d testosterone-treated females. B: *, P < 0.05, M vs. F; #, P < 0.01, F vs. 5-d testosterone-treated females.
Time-course effect of testosterone on the expression of arginase II protein in female mouse kidneys.
In soluble proteins extracted from male, female, and testosterone-treated female mice, and analyzed by Western blotting, antiarginase II antibody revealed a 38-kDa protein that corresponds to the predicted size of the arginase II polypeptide (Swiss-Prot O08691 mouse: 38,878 Da, Fig. 1B, left). The renal content of ?-tubulin, a 55-kDa protein, was not affected by testosterone treatment (means ± SE, females: 158 ± 77, males: 173.7 ± 51.7; 1-d: 168.7, 2-d: 108.7, 3-d: 110.9, and 5-d: 119.3 x 103 pixels; Kruskal-Wallis P = 0.0734, Fig. 1B, left). The relative level of arginase II to ?-tubulin protein was 3.7-times higher in female than in male mouse kidneys (Fig. 1A, right, Mann-Whitney, P < 0.05). Testosterone treatment induced a progressive increase in arginase II protein level in female mouse kidney (Fig. 1B, right, Kruskal-Wallis, P < 0.0002). The amount of arginase II protein reached its highest level between d 3 and d 5. On d 5, the abundance of arginase II relative to ?-tubulin proteins was 2.8 times higher than in untreated female mouse kidney (Fig. 1B, right, Mann-Whitney, P < 0.01).
Time-course effect of testosterone on the synthesis of arginase II mRNA in female mouse kidneys.
In total RNA extracted from mouse kidneys and submitted to Northern blotting, the murine arginase II cDNA probe allowed detection of the 1.8-kb arginase II mRNA (Fig. 2A). The size of the arginase II mRNA is in good agreement with a previous study (23). There was no statistical difference in the content of the 2.1-kb ?-actin mRNA between the different groups of mice (Fig. 2B, right and Ref.11). The level of arginase II mRNA was higher in female than in male kidneys (Fig. 2B, left, Mann-Whitney, P < 0.021). Testosterone treatment induced a progressive and important increase in the level of arginase II mRNA between d 1 and d 5 (Fig. 2B, left, Kruskal-Wallis, P < 0.01). The highest level of arginase II mRNA was reached on d 5 and was about 3-times higher than in untreated female mice (Fig. 2B, left, Mann-Whitney, P < 0.034).
FIG. 2. Northern blot analysis of arginase II and ?-actin mRNA levels in male (M), female (F), and testosterone-treated female mouse kidneys. Female mice were treated as described in Fig. 1 legend. A, Each line corresponds to 15 μg total RNA extracted from one mouse kidney. B, Quantitation of arginase (AII) and ?-actin mRNA levels. The amount of radioactivity hybridized was estimated after scanning densitometry of the membranes using a PhosphoImager. Left, 1.8-kb arginase II mRNA; right, 2.1-kb ?-actin mRNA used as a control. Values are means ± SE, except when n = 2. Differences between groups were statistically analyzed by Kruskal-Wallis (P < 0.01) and followed by Mann-Whitney test. *, P < 0.021, M vs. F; #, P < 0.034, F vs. 5-d testosterone-treated females.
Testosteronemia in untreated and androgen-treated mice.
In untreated female mice, testosteronemia was very low (0.19 + 0.07 ng/ml, n = 7); whereas in male mice, testosteronemia was about 12-times higher than in females (2.43 ± 1.76 ng/ml, n = 6). As shown in Table 1, the individual level of plasma testosterone in six other male and nine sham-operated male mice varied in a broad range. This variation has been attributed earlier to the pulsatile release of testosterone (24). As expected, in response to injections of pharmacological doses of testosterone, testosteronemia sharply increased in a time-dependent manner (Fig. 3, Kruskal-Wallis, P < 0.0001), and its level was 52- (d 1) and 114-times (d 5) higher compared with that of the males.
TABLE 1. Biological parameters in control, sham-operated, orchidectomized, oil- and testosterone-treated male mice
FIG. 3. Testosteronemia in untreated male (M), female (F), and androgen-treated female mice. Female mice were treated as described in Fig. 1 legend. Testosteronemia was measured by RIA as previously described (16 ). Values are means ± SE, with the number of animals tested given in parentheses.
Relationship between testosteronemia and the expression of OAT and arginase II genes.
To test the link between testosteronemia and the expression of OAT and arginase II genes in the murine kidney, the relative levels of OAT protein, arginase II mRNA, and protein of untreated male and testosterone-treated female mice were plotted against the level of plasma testosterone. The ratio of OAT to ?-actin protein was negatively correlated with testosteronemia (Fig. 4A; r2 = 0.918, P < 0.01). In contrast, the ratios of arginase II to ?-actin mRNA and arginase II to ?-tubulin protein were positively and linearly correlated with testosteronemia (Fig. 4B, r2 = 0.913; and Fig. 4C, r2 = 0.971, respectively, P < 0.01). Therefore, altogether, these results strongly support that pharmacological levels of testosterone simultaneously down-regulate the expression of OAT gene and up-regulate that of arginase II and ODC.
FIG. 4. Correlation between testosteronemia and the relative level of OAT protein (A), arginase II (AII) mRNA (B), and arginase II protein (C) in male and 1- to 5-d testosterone-treated female mice. To test the link between testosteronemia and the expression of arginase II and OAT genes in the murine kidney, the ratios of OAT to ?-actin protein, arginase II to ?-actin mRNA, and arginase II to ?-tubulin protein were calculated and plotted against the level of plasma testosterone. Values are means ± SE. A, r2 = 0.918; B, r2 = 0.913; C, r2 = 0.971; P < 0.01 in all cases. Closed lozenge, 5-d testosterone; open lozenge, 3-d testosterone; open circle, 2-d testosterone; closed circle, 1-d testosterone; open square, untreated female; closed square, untreated male.
Physiological regulation of OAT, arginase II, and ODC genes by testosterone in male mouse kidney
These experiments were conducted to determine whether physiological levels of testosterone regulate the expression of OAT and arginase II genes. To achieve this goal, male mice were castrated to abolish the endogenous production of androgens by testes, and then testosterone propionate was injected into these mice to reverse the effects of orchidectomy. In these experiments, the level of OAT and arginase II proteins was analyzed by Western blotting.
Influence of orchidectomy on biological parameters in male mice.
Mice of groups I, II, and III had similar body weight, whereas mice of group IV were heavier than the others because they were killed 1 wk later (Table 1, Mann-Whitney, group III vs. IV, P < 0.007). The absolute mass of the left and right kidneys was diminished by 30–34% in orchidectomized mice compared with control and sham-operated mice (Table 1, Mann-Whitney, group III and IV vs. I and II, P < 0.004 or less). Orchidectomy diminished the relative kidney mass (RKM, mass of two kidneys/BW) by about 30%. Testosteronemia did not differ between male mice of groups I and II but was undetectable in orchidectomized mice (Table 1, Mann-Whitney, group III and IV vs. I and II, P < 0.004).
Influence of orchidectomy on the level of OAT, arginase II, and ODC proteins in male mouse kidneys.
The 62-kDa androgen-dependent SA polypeptide is abundantly expressed in kidneys of control male mice and was used as a marker of orchidectomy (20). Similar levels of SA protein were detected in kidneys of control and sham-operated mice (Fig. 5A), whereas orchidectomy induced a considerable decrease (90–95%) in SA protein levels (Fig. 5A). The highly androgen-inducible ODC protein was expressed at the same level in kidneys of control and sham-operated male mice (Fig. 5, A and B). Orchidectomy induced a 90–95% decrease in the level of ODC protein compared with that of control and sham-operated mice (Fig. 5B, Mann-Whitney, P < 0.004). The renal content of ?-actin was similar in the four groups of mice (Fig. 5A, quantitation not shown). The level of OAT protein was not affected by the sham operation (Fig. 5C). In contrast, the level of OAT protein was increased by 2.3- and 3-times in mice castrated during 11- and 18-d periods, respectively, compared with the controls (Fig. 5C, Mann-Whitney, P < 0.004). The level of arginase II protein was similar in kidneys of control and sham-operated mice, but it decreased by about 75% and 54%, respectively, in kidneys of mice castrated during 11- and 18-d periods (Fig. 5D, Mann-Whitney, P < 0.011).
Effect of a physiological dose of testosterone on biological parameters in orchidectomized mice.
The reduction of the absolute and relative renal mass, as well as the very low level of plasma testosterone, proved the efficiency of orchidectomy (Table 1, Mann-Whitney, P < 0.05). A single dose of testosterone enhanced the RKM (Table 1, Mann-Whitney, P < 0.05), whereas oil had no influence on the kidney weight and RKM. Androgen treatment restored testosteronemia to the range of physiological level (Table 1, groups I and II) but was higher than testosteronemia measured in group V (Table 1).
Effect of a physiological dose of testosterone and oil on the level of OAT, arginase II, and ODC proteins in orchidectomized mice.
As observed in the previous experiment (Table 1 and Fig. 5), orchidectomy significantly decreased the level of SA, ODC, and arginase II proteins and enhanced that of OAT (Fig. 6, A–D, Kruskal-Wallis, P < 0.032 or less; followed by Mann-Whitney, P < 0.05 for each protein). The injection of a single dose of oil did not affect the renal level of SA, ODC, OAT, and arginase II proteins compared with that of the sham-operated mice (Fig. 6, A–D). In contrast to oil, a single injection of a physiological dose of testosterone induced a sharp increase in the level of ODC protein (Fig. 6, A and B, Mann-Whitney, P < 0.05), tended to restore the level of arginase II protein to that of the control values (Fig. 6, A and D, Mann-Whitney, P < 0.05), and decreased by 2-times the level of OAT protein (Fig. 6, A and C, Mann-Whitney, P < 0.05). In contrast to ODC, the level of SA protein was partially restored by testosterone injection (Fig. 6A). The same results were found when using ?-actin as a control of protein loading and transfer (data not shown).
In conclusion, in the murine kidney, our results strongly suggest that testosterone physiologically up-regulated not only ODC gene, but also arginase II gene, and simultaneously down-regulated OAT gene. In orchidectomized mice, the lack of testosterone led to a reduction of the renal level of ODC and arginase II proteins and enhanced that of OAT. These effects were reversed by injecting a physiological dose of testosterone into castrated male mice.
Discussion
The anabolic hormone testosterone is involved in a multitude of physiological processes and produces a variety of biochemical changes in several organs. The effects of testosterone have been extensively analyzed in the murine kidney because androgens lead to renal hypertrophy. In addition, administration of testosterone to male and female mice induces the renal expression of several genes coding for ?-glucuronidase, ODC, alcohol dehydrogenase, arginase II, kidney androgen-regulated protein, RP2 (mouse androgen kidney or MAK), SA, and transferase II (12, 13, 25, 26). ODC gene is highly sensitive to testosterone, which induces, within hours, a sharp increase in ODC mRNA, protein, and enzyme activity (10, 11). As a physiological consequence, polyamines are synthesized; the renal content of putrescine, spermidine, and spermine is enhanced; and the excess of putrescine is excreted in urine (9, 10). In addition to ODC, OAT gene is highly expressed by the murine kidney. Both enzymes use and compete for the substrate L-ornithine, which is provided, in part, by the mitochondrial arginase II. The present study was designed to determine how OAT and arginase II genes are regulated in kidneys of testosterone-treated mice for the following reasons: 1) the regulation of OAT gene by androgens has not been investigated in the murine kidney; 2) it is not known whether arginase II gene is regulated by androgens at the transcriptional and/or at the translational levels; 3) in unstimulated renal cells, the expression of ODC gene is low, whereas that of OAT is very high; and 4) in kidneys of testosterone-treated mice, the overproduction of putrescine raises the metabolic fate of L-ornithine supplied by arginase II and the balance between OAT and ODC gene expression.
Our results clearly reveal that, in kidneys of female mice, testosterone inversely regulated the expression of OAT and ODC genes. Indeed, pharmacological doses of testosterone deeply diminished the level of OAT protein in a dose-dependent manner, whereas the expression of arginase II and ODC genes was strongly enhanced. The excellent correlations depicted between testosteronemia and the levels of OAT and arginase II mRNA and proteins strongly support these findings. The inverse hormonal regulation of OAT and ODC genes in kidneys of testosterone-treated female mice might favor putrescine synthesis to the detriment of L-glutamate production. The biochemical events might appear as follows: 1) in kidneys of androgen-treated mice, the expression of ODC gene is rapidly increased by about 1000 times and becomes more expressed than OAT gene (12); 2) the sharp down-regulation of OAT gene by testosterone deeply diminished L-ornithine transamination into L-glutamate; 3) induction of arginase II gene expression by testosterone enhances arginase II activity and leads to hydrolyze larger amounts of L-arginine into L-ornithine; 4) L-ornithine produced by arginase II in mitochondria can be transported into the cytosol; and 5) L-ornithine is converted in putrescine by the cytosolic ODC (Fig. 7A). Because testosterone exerts anabolic effects, the new production of polyamines by the kidneys may contribute to general physiological needs such as cell growth, hypertrophy, mRNA and protein synthesis, membrane stability, and a variety of cell functions in the different organs (27).
FIG. 7. Renal fate of L-arginine-derived ornithine in testosterone-treated female mouse and untreated male mouse (A) and in untreated female mouse and castrated male mice (B). A, In the presence of testosterone, ODC and arginase II (AII) genes are highly expressed, and L-arginine is a precursor for polyamine synthesis (thick line), whereas OAT gene expression is decreased (hatched line). B, In contrast, in untreated female mice and orchidectomized mice, ODC gene is not expressed (hatched line), and OAT gene is highly expressed, leading to L-glutamate production. P5CDH, Pyrroline-5-carboxylate dehydrogenase; GS, glutamine synthase; NAD+, nicotinamide adenine dinucleotide; NADH,H+, reduced nicotinamide adenine dinucleotide; NADP+, nicotinamide adenine dinucleotide phosphate; NADPH,H+, reduced nicotinamide adenine dinucleotide phosphate; GLDH, glutamate dehydrogenase.
We felt it more relevant to determine whether OAT gene was physiologically regulated by a physiological level of plasma testosterone and to examine simultaneously the expression of ODC and arginase II genes in mouse kidneys. We first analyzed the influence of gender on the basal expression of these three genes. OAT, ODC, and arginase II genes were expressed in untreated male mouse kidneys (see also Ref.11). In contrast, in kidneys of untreated female mice, ODC gene was not expressed (see also Refs.5 and 11), whereas OAT and arginase II genes were more expressed than in those of the males. The very low level of plasma testosterone in females, compared with that of the males, might be responsible for the lack of ODC gene expression and the very high level of OAT protein in the female mouse kidney. OAT metabolizes L-ornithine into L-glutamate, which can be converted either into L-glutamine and/or energy by entering into the citric acid cycle (Fig. 7B). This result confirms the constitutive and inverse regulation of OAT and ODC gene expression by a physiological level of testosterone. In contrast, at present, it is difficult to explain why kidneys of female mice exhibited higher levels of arginase II mRNA, protein, and enzyme activity than those of the males (7, 13). Although arginase II gene is more expressed in female than in male mouse kidney, arginase II activity has been known for a long time to be increased by pharmacological doses of testosterone. In addition, our data reveal that arginase II gene was regulated at the transcriptional and translational levels by high doses of testosterone. Altogether, these results strongly indicate that arginase II gene is controlled by testosterone.
Thereafter, we studied the physiological regulation of OAT gene by testosterone in male mouse kidney. Orchidectomy, rather than drug inhibitors, was used to stop the production of androgen hormones by testes. Proteins were analyzed at 11 and 18 d after the operation to guarantee a complete depletion of testosterone and to reduce the effects due to the surgery. As expected, the lack of testosterone led to a sharp decrease in the level of ODC and arginase II proteins. It appeared that arginase II gene was less sensitive to androgens than that of ODC because the level of arginase II protein was not totally abolished by orchidectomy. This result suggests that the basal level of arginase II gene expression is not under the control of androgens. In contrast to ODC and arginase II, the level of OAT protein was highly enhanced by orchidectomy, supporting the idea that testosterone physiologically down-regulates OAT gene. The effects of orchidectomy on gene expression were reversed by injecting a single dose of testosterone (0.55 mg/mouse) that shifted testosteronemia to physiological levels. This dose was calculated from the data obtained from testosterone-treated female mice (Fig. 3). As expected, a low level of testosteronemia restored the expression of ODC protein. A physiological level of testosteronemia clearly diminished the level of OAT protein and reached the level measured in the sham-operated group. Our results demonstrated, for the first time, that OAT gene is physiologically down-regulated by testosterone. Finally, testosterone increased, and almost restored, the renal level of arginase II protein. In the past, Kochakian and Stettner (13) reported that a pharmacological dose of 14–15 mg testosterone given per castrated-male Swiss mouse enhanced the renal arginase II activity and concluded that arginase II gene was controlled by testosterone. However, here, we demonstrated, for the first time, that testosterone physiologically up-regulated arginase II gene in the mouse kidney. Surprisingly, Manteuffel-Cymborowska et al. (7), who gave a dose of 4.5 mg testosterone per male Swiss mouse, did not observe changes in the renal arginase II activity.
It is generally assumed that testosterone and steroid hormones act in different biological programs through the binding of intracellular receptors, which act as ligand-inducible transcription factors on specific DNA elements of target genes. The receptors for steroid hormones belong to a single receptor superfamily, which includes receptors for androgens, estrogens, progesterone, glucocorticoids, mineralocorticoids, retinoic acid, thyroid hormone, and vitamin D. All steroid receptors share similar structural and functional characteristics with respect to regulating the transcriptional activity of specific genes (28). The androgen receptor has been cloned and characterized, and specific antibodies-antiandrogen receptors allowed to identify its nuclear localization in a variety of animal species and tissues, including cells of the proximal and distal tubules of the rat kidney (29). In mouse kidney, androgens induce the expression of kidney androgen-regulated protein and ODC genes that possess an androgen response element (ARE) in the 5' flanking region of their genes (30). Recently, the promoter region of the murine, rat, and human OAT genes has been sequenced. Numerous 5'-AGGTCA-like motifs corresponding to the consensus binding site for several members of the nuclear receptor superfamily were located from nucleotides –222 to –205 and 366–396 of the rat OAT gene and are relatively well conserved in mouse OAT gene (31, 32). At present, we do not know whether the AGGTCA-like motif mediates the androgen response. Whatever, the physiological consequence of the presence of testosterone was a down-regulation of OAT gene expression. The promoter region of the murine arginase II gene does not contain an ARE but possesses numerous potential binding sites for enhancer and promoter elements, including AP1, NF-KB, SP1, and CRE-BP2 (33). Consequently, it is difficult to understand the molecular mechanism through which testosterone induces the biological effect on arginase II gene in the absence of ARE. Therefore, several possibilities might account for this observation: 1) the increase in arginase II gene expression was due to a direct or indirect effect of testosterone; 2) the promoter of arginase II gene possesses an ARE distal to the sequenced region; 3) testosterone binds to a cellular membrane receptor as recently reported (34); 4) testosterone binds to a glucocorticoid receptor or another receptor belonging to this superfamily; and 5) testosterone induces a cascade of events that are secondarily responsible for the increase of arginase II gene expression. Recently, it has been reported that SHBG, which complexes testosterone, binds to a SHBG-receptor on cell membranes of several tissues and therefore activates a G-coupled protein that modulates the production of cAMP (34). In addition, in RAW 264.7 cells, the level of arginase II mRNA was enhanced by dexamethasone and dibutyryl cAMP (31, 35). Further experiments will be performed to clarify the mechanisms involved in regulating arginase II gene by testosterone.
In conclusion, we demonstrated, for the first time, that OAT gene was pharmacologically and physiologically down-regulated by testosterone in male and female mouse kidneys. ODC and OAT genes were inversely regulated by testosterone. In the presence of high levels of the androgen hormone, the synthesis of polyamines was favored, whereas that of L-glutamate is suspected to be decreased. In contrast, in control female mouse kidney, the lack of ODC suggests that L-arginine-derived ornithine might be metabolized by OAT to produce L-glutamate.
Acknowledgments
The authors are indebted to Dr. Sidney M. Morris (University of Pittsburgh, Pittsburgh, PA) and Dr. Anna Meseguer (Centre d’Investigacions en Bioquimica i Biologia Molecular, Hospital Universitari Vall d’Hebron, Barcelona, Spain), who kindly provided, respectively, pBSK– arginase II plasmid containing mouse arginase II cDNA [full-length cDNA encoding for the murine arginase II type, GenBank accession no. AF032466 (21 )], and the rabbit anti-SA antibody (20 ). O.L. is indebted to Marie Thérèse Ducluseau and Prof. Jean-Fran?ois Nicolas, who kindly gave access to materials for RNA extraction; Prof. Dr. Bruno Claustrat for discussion; Jocelyne Vial for assistance; and Bernard Marchand for his contribution in the animal house.
References
Levillain O, Hus-Citharel A, Morel F, Bankir L 1992 Localization of urea and ornithine production along mouse and rabbit nephrons: functional significance. Am J Physiol (Renal Fluid Electrolyte Physiol) 263:F878–F885
Pegg AE, Seely J, Zagon IS 1982 Autoradiographic identification of ornithine decarboxylase in mouse kidney by means of -[5-14C] difluoromethylornithine. Science 217:68–70
Koibuchi N, Matsuzaki S, Kitani T, Fujisawa H, Suzuki M 1990 Localization by immunohistochemistry of renal ornithine decarboxylase in the mouse with and without testosterone treatment. Endocrinol Jpn 37:555–561
Levillain O, Hus-Citharel A 1998 Ornithine decarboxylase along the mouse and rat nephron. Am J Physiol Renal Physiol 43:F1020–F1028
Levillain O, Diaz J-J, Reymond I, Soulet D 2000 Ornithine metabolism along the female mouse nephron: localization of ornithine decarboxylase and ornithine amino-transferase. Pflügers Arch Eur J Physiol 440:761–769
Alonso E, Rubio V 1989 Participation of ornithine aminotransferase in the synthesis and catabolism of ornithine in mice. Biochem J 259:131–138
Manteuffel-Cymborowska M, Chmurzynska W, Peska M, Grzelakowska-Sztabert B 1995 Arginine and ornithine metabolizing enzymes in testosterone-induced hypertrophic mouse kidney. Int J Biochem Cell Biol 27:287–295
Berger FG, Szymanski P, Read E, Watson G 1984 Androgen-regulated ornithine decarboxylase mRNAs of mouse kidney. J Biol Chem 259:7941–7946
Persson L 1981 Putrescine turnover in kidneys of testosterone-treated mice. Biochim Biophys Acta 674:204–211
Pajunen AEI, Isomaa VV, J?nne OA, Bardin CW 1982 Androgenic regulation of ornithine decarboxylase activity in mouse kidney and its relationship to changes in cytosol and nuclear androgen receptor concentrations. J Biol Chem 257:8190–9198
Levillain O, Greco A, Diaz J-J, Augier R, Didier A, Kindbeiter K, Catez F, Cayre M 2003 Influence of testosterone on the regulation of ornithine decaboxylase, antizyme and spermidine/spermine-N1-acetyl-transferase gene expression in female mouse kidney. Am J Physiol Renal Physiol 285:F498–F506
Catterall JF, Kontula KK, Watson CS, Sepp?nen PJ, Funkenstein B, Melanitou E, Hickok NJ, Bardin CW, J?nne OA 1986 Regulation of gene expression by androgens in murine kidney. Recent Prog Horm Res 42:71–109
Kochakian CD, Stettner CE 1948 Effect of testosterone propionate and growth hormone on the arginase and phosphatases of the organs of the mouse. Am J Physiol 155:262–264
Silbernagl S 1992 Amino acids and oligopeptides. In: Seldin DW, Giebish G, eds. The kidney: physiology and pathophysiology. 2nd ed. New York: Raven Press, Ltd; 2889–2920
Sanchez-Capelo A, Penafiel R, Tovar A, Galindo JD, Cremades A 1994 Postnatal development of ornithine decarboxylase and polyamines in the mouse kidney: influence of testosterone. Biol Neonate 66:119–127
Déchaud H, Lejeune H, Garoscio-Cholet M, Mallein R, Pugeat M 1989 Radioimmunoassay of testosterone not bound to sex-steroid-binding protein in plasma. Clin Chem 35:1609–1614
Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685
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
Levillain O, Hus-Citharel A, Garvi S, Peyrol S, Reymond I, Mutin M, Morel F 2004 Ornithine metabolism in male and female rat kidney mitochondrial expression of ornithine aminotransferase and arginase II. Am J Physiol Renal Physiol 286:F727–F738
Aresté C, Meliá MJ, Isern J, Tovar JL, Meseguer A 2004 Sex steroid regulation and identification of different transcription units of the SA gene in mouse kidney. J Endocrinol 183:101–114
Morris Jr SM, Kepka-Lenhart D, Chen LC 1998 Differential regulation of arginases and inducible nitric oxide synthase in murine macrophage cells. Am J Physiol 38:E740–E747
Hanahan D 1983 Studies on transformation of E. coli with plasmids. J Mol Biol 166:557–580
Ozaki M, Gotoh T, Nagasaki A, Miyanaka K, Takeya M, Fujiyama S, Tomita K, Mori M 1999 Expression of arginase II and related enzymes in the rat small intestine and kidney. J Biochem (Tokyo) 125:586–593
Bartke A, Steele RE, Musto N, Caldwell BV 1973 Fluctuations in plasma testosterone levels in adult male rats and mice. Endocrinology 92:1223–1228
King D, Sun YH, Lingrel JB 1986 Amino acid sequence of the testosterone-regulated mouse kidney RP2 protein deduced from its complementary DNA sequence. Nucleic Acids Res 14:5159–5170
Melia MJ, Bofill N, Hubank M, Meseguer A 1998 Identification of androgen-regulated genes in mouse kidney by representational difference analysis and random arbitrarily primed polymerase chain reaction. Endocrinology 139:688–695
Pegg AE, McCann PP 1982 Polyamine metabolism and function. Am J Physiol 243:C212–C221
Schuchard M, Landers JP, Sandhu NP, Spelsberg TC 1993 Steroid hormone regulation of nuclear proto-oncogenes. Endocr Rev 14:659–669
Takeda H, Chodak G, Mutchnik S, Nakamoto T, Chang C 1990 Immunohistochemical localization of androgen receptors with mono- and polyclonal antibodies to androgen receptor. J Endocrinol 126:17–25
Crozat A, Palvimo JJ, Julkunen M, J?nne OA 1992 Comparison of androgen regulation of ornithine decarboxylase and S-adenosylmethionine decarboxylase gene expression in rodent kidney and accessory sex organs. Endocrinology 130:1131–1144
Shull JD, Pennington KL, Pitot HC, Boryca VS, Schulte BL 1992 Isolation and characterization of the rat gene encoding ornithine aminotransferase. Biochim Biophys Acta 1132:214–218
Shull JD, Esumi N, Colwell AS, Pennington KL, Jendoubi M 1995 Sequence of the promoter region of the mouse gene encoding ornithine aminotransferase. Gene 162:275–277
Shi O, Kepka Lenhart D, Morris Jr SM, O’Brien WE 1998 Structure of the murine arginase II gene. Mamm Genome 9:822–824
Heinlein CA, Chang C 2002 The roles of androgen receptors and androgen-binding proteins in nongenomic androgen actions. Mol Endocrinol 16:2181–2187
Gotoh T, Sonoki T, Nagasaki A, Terada K, Takiguchi M, Mori M 1996 Molecular cloning of cDNA for nonhepatic mitochondrial arginase (arginase II) and comparison of its induction with nitric oxide synthase in a murine macrophage-like cell line. FEBS Lett 395:119–122(Olivier Levillain, Jean-J)
Address all correspondence and requests for reprints to: Dr. Olivier Levillain, Université Claude Bernard, Faculté de Médecine Lyon RTH Laennec, U 499 Institut National de la Santé et de la Recherche Médicale, 7, rue G. Paradin, 69372 Lyon Cedex 08, France. E-mail: Olivier.Levillain@laennec.univ-lyon1.fr.
Abstract
The enzymes ornithine aminotransferase (OAT) and ornithine decarboxylase (ODC) share L-ornithine as a common substrate and arginase II produces this amino acid. In the murine kidney, testosterone induced ODC gene expression and polyamine production, but it is unknown how OAT gene is expressed under androgen treatment. These experiments were designed to study the influence of testosterone on the renal expression of OAT gene. Pharmacological and physiological doses of testosterone were injected into female and castrated male mice. Total RNA and soluble proteins extracted from whole kidneys were analyzed by Northern and Western blots, respectively. The results clearly indicate that pharmacological doses of testosterone simultaneously down-regulated the level of OAT protein and up-regulated the expression of arginase II and ODC genes. Variations of the levels of OAT protein and arginase II mRNA and protein were strongly correlated with testosteronemia. Orchidectomy increased the renal level of OAT protein and decreased that of ODC and arginase II. These effects were reversed by injecting a physiological dose of testosterone into castrated male mice. In conclusion, OAT and ODC genes are inversely regulated by testosterone in the mouse kidney. Consequently, in kidneys of testosterone-treated mice, L-arginine-derived ornithine produced by arginase II might be preferentially used by ODC for putrescine production rather than by OAT. This metabolic fate of L-ornithine was facilitated by decreasing OAT gene expression. In contrast, in female and castrated male mice devoided of testosterone, OAT gene is highly expressed and L-ornithine is converted into L-glutamate
Introduction
IN THE MURINE kidney, the nucleus-encoded mitochondrial arginase II (EC 3.5.3.1) catalyzes the irreversible hydrolysis of L-arginine into urea and L-ornithine in the cortical and medullary proximal straight tubules (CPST and OSPST, respectively) (1). L-Arginine-derived ornithine is either metabolized into polyamines or transaminated. In the first pathway, polyamine synthesis is controlled by the key enzyme ornithine decarboxylase (ODC; EC 4.1.1.17) that converts L-ornithine into putrescine. In the male mouse kidney, ODC gene is highly expressed in the proximal convoluted tubule (PCT) and, to a lesser extent, in CPST and OSPST (2, 3, 4). In contrast, in the female mouse kidney, none of the nephron segments expressed ODC gene (5). In the second pathway, transamination of L-ornithine requires the key enzyme ornithine aminotransferase (OAT; EC 2.6.1.13). OAT is a nucleus-encoded mitochondrial protein that catalyzes a reversible reaction leading to the production of L-glutamate and glutamate--semialdehyde in the presence of L-ornithine and -ketoglutarate. A high OAT activity has been reported in the murine kidney (6, 7). Moreover, we observed that OAT gene was more expressed in female than in male mouse kidneys (O. Levillain, personal unpublished data). In a detailed study, we showed that gabaculine, a suicide inhibitor of OAT, deeply reduced L-ornithine decarboxylation, suggesting that OAT might be expressed in PCT, CPST, and OSPST of the female mouse kidney (5). In the male mouse nephron, although OAT protein has been detected in the different renale zones (Levillain, O., personal unpublished data), its precise distribution within the nephron remains to be identified.
It is now well-known that ODC is a highly androgen-inducible protein in the murine kidney (8). Administration of pharmacological doses of testosterone to female mice induced a rapid and prolonged increase in renal ODC mRNA, protein, and enzyme activity over a few days; enhanced the renal production of putrescine; and sharply increased putrescine excretion in urine (9, 10, 11). Testosterone induced the expression of ODC gene in the three subsegments of the proximal tubule: PCT, CPST, and OSPST (11). In addition to ODC, several genes are overexpressed by testosterone in the mouse kidneys (12). Renal arginase II activity is increased by injecting testosterone into female and castrated male mice (7, 13). However, at present, it remains unknown whether testosterone regulates arginase II gene at the transcriptional and/or at the translational levels. Moreover, to our knowledge, the influence of androgens on the expression of OAT gene in the murine kidney has never been analyzed.
Studying the regulation of OAT gene by testosterone in the mouse kidney is particularly relevant because OAT and ODC share L-ornithine as a common substrate, and the metabolic fate of L-arginine-derived ornithine under such a physiological condition is not known. The mitochondrial colocalization of arginase II and OAT supports the view that L-ornithine generated by arginase II might be immediately transaminated by OAT rather than decarboxylated by the cytosolic ODC. Surprisingly, the production of putrescine is dramatically enhanced in kidneys of testosterone female mice. To explain this result, we hypothesized that the expression of OAT gene could be down-regulated by testosterone to metabolize L-arginine-derived ornithine preferentially into polyamines rather than into L-glutamate. Alternatively or simultaneously, the up-take of L-ornithine from the arterial blood across the basolateral membranes of proximal tubular cells might be enhanced (14). The first hypothesis is supported by a recent report that indicates that OAT activity is decreased by 2-fold in testosterone-treated female mice but, surprisingly, remained unaffected in testosterone-treated male mice (7). Moreover, at present, it is not known whether OAT gene is regulated by physiological levels of testosterone in male mouse kidney.
The present study was designed: 1) to analyze the expression of OAT gene in kidneys of testosterone-treated female mice; 2) to test whether testosterone simultaneously induced the expression of arginase II and ODC gene and repressed that of OAT to shift L-arginine-derived ornithine into the polyamine pathway; and 3) to determine whether testosterone physiologically regulates the expression of OAT gene in male mouse kidney. For this last experiment, control, orchidectomized, and testosterone-treated orchidectomized mice were used.
Materials and Methods
Animals and treatment
Eight- to 9-wk-old adult female [30–32 g body weight (BW)] and male (35–40 g BW) OF-1 Swiss (IOPS Caw) mice from Iffa Credo (L’Arbresle sur Orge, France) had free access to tap water and standard laboratory diet (Souffirat, 20% protein, Genthon, France). Animals were housed in a room maintained at 20 C with a 12-h light, 12-h dark cycle. Mice were anesthetized (ip) using 0.1 ml/100 g BW pentobarbital sodium diluted 1:2 in 0.9% NaCl solution (Nembutal 6%, Clin Midy, Paris, France). The experiments were approved by the local Committee for Animal Experiments.
Twenty-four female mice were subdivided into five groups: one untreated (control) and four androgen-treated groups. Mice subjected to testosterone treatment were injected sc with 150 μl testosterone propionate (31 mg/ml in sesame oil, i.e. 155 μg/g BW). Injections were performed at 0800 h, and mice were treated for a period of 1, 2, 3, or 5 consecutive days. Male mice were untreated.
Young adult male mice of 30-d-old, rather than older adult male mice, were used to prevent the accumulation of testosterone in their tissues and plasma (15). Twenty-four 30-d-old male mice were subdivided into four groups of six mice: nonoperated (group I, control), sham-operated (group II), and two groups of orchidectomized mice. Eleven days later (i.e. 41 d after birth), mice of groups I, II, and III were killed, whereas mice of group IV were killed 7 d later (i.e. 18 d after orchidectomy).
Twelve 30-d-old male mice were subdivided into four groups of three mice: sham-operated (group V, control), 11-d orchidectomized (group VI), 11-d orchidectomized treated with sesame oil (group VII), and 11-d orchidectomized treated with testosterone + sesame oil (group VIII). Mice subjected to oil or testosterone treatment were injected sc with 150 μl vehicle or testosterone propionate (3.1 mg/ml in sesame oil, i.e. 15 μg/g BW or 0.55 mg/mouse). All mice were killed 48 h after the treatment.
Sampling of blood and plasma for testosterone analysis
Blood was collected in the vena cava of all male and female mice with a 25-gauge needle (Neolus, VWR, Limonest, France) mounted on a 1-ml syringe (Terumo, VWR) prealably heparinized (Heparin, Roche, Meylan, France). Blood was immediately transferred in a cold BD Vacutainer tube, centrifuged at 4000 x g for 20 min at 4 C. Plasma was frozen and stored in liquid nitrogen until testosterone measurement. Testosterone was measured by RIA after extraction by organic solvent and partition chromatography of the plasma samples as previously described (16).
Sampling of the kidneys for Western blot analyses
In different series of experiments performed to analyze the influence of sexual hormones on the renal level of arginase II protein, kidneys of control (untreated and nonoperated), sham-operated, 11- and 18-d orchidectomized male, oil- and testosterone-treated orchidectomized male mice, and control (untreated and nonoperated) and testosterone-treated female mice were rapidly removed, decapsulated; the blood contained in each kidney was removed with blotting paper (free-blood). The kidney was placed in a sterilized Eppendorf tube, frozen, and conserved in liquid nitrogen. The kidneys of testosterone-treated female mice used in this study are the same as those used to analyze the time course expression of ODC protein and published in Ref.11 .
Protein extraction and Western blot analyses of arginase II protein
Each frozen kidney was mixed at 4 C with a Turrax in the proportion of 100 mg frozen tissue/2 ml of lysing buffer (17) containing 1 mM protease inhibitor cocktail, 1 mM phenylmethylsulfonylfluoride, and 1 mM benzamidine, and then centrifuged at 10,000 x g for 30 min at 4 C. Protein concentrations were determined in the supernatant using the Bradford protein assay (18). For each kidney, 100-μg samples of soluble protein were subjected to 10% PAGE containing 0.1% sodium dodecyl sulfate (SDS) using 6 watts/gel. Ten microliters of a protein ladder (precision plus protein standards, Bio-Rad, Marnes la Coquette, France) were deposed on each gel to further verify the size of the protein of interest. Proteins were transferred to a polyvinylidene difluoride membrane (0.45 μm, Immobilon-P, Millipore, St Quentin en Yvelines, France) at 150 mA for 90 min. Proteins were fixed on the membrane with Ponceau S solution for 15 min. Immunoblots were washed twice in 1x Tris-buffered salt + 0.1% Tween 20 (TBST) and immersed twice in a blocking solution consisting of 5% fat-free milk powder in 1x TBST for 30 min.
Immunoblots were incubated with the following primary antibodies in 5% milk-1x TBST: a polyclonal rabbit antimouse-arginase II [CovalAb, dilution 1:1000 (19)], a polyclonal rabbit antimouse-OAT [CovalAb, dilution 1:1000 (19)], a polyclonal rabbit antihuman-ODC (dilution 1:500, Eurodiagnostica, Paris, France), a polyclonal rabbit antisubstractive clone A [SA; dilution 1:350 (20)], a monoclonal mouse antiglyceraldehyde-3-phosphate dehydrogenase (G3PDH; dilution 1:170, Chemicon International, Temecula, CA), a monoclonal mouse anti-?-actin (dilution 1:2000) or a monoclonal mouse anti-?-tubulin (dilution 1:1000). ?-Actin, ?-tubulin, and G3PDH were used as controls of equal loading and transfer of total proteins. SA has been identified as a medium-chain acyl-CoA synthetase, also called acetate-CoA ligase, and is a truly androgen-dependent gene (20). For this reason, SA was used as a marker of orchidectomy. The blots were washed three times for 10 min in 1x TBST and incubated for 60 min with either peroxidase-conjugated antirabbit IgG or antimouse IgG secondary antibodies (dilution 1:10,000) in 5% milk-1x TBST. Blots were washed three times for 10 min in 1x TBST, and antibody binding was revealed using an enhanced chemiluminescence (ECL) Western Blotting Kit. ECL detection was performed using X-MAT film (Kodak, Rochester, NY). Low-exposure film was scanned, and the intensity optical densitometry (IOD) of the bands was estimated using the ImagerMaster Total Lab version 2.01 program (Pharmacia, Orsay, France).
Kidney preparation for Northern blot analyses
In a series of experiments designed to test the influence of testosterone on renal arginase II mRNA levels, kidneys of four untreated male, four untreated female, and nine testosterone-treated female mice were rapidly removed and decapsulated. Each free-blood kidney was placed in a sterilized Eppendorf tube, frozen, and conserved in liquid nitrogen.
RNA extraction and Northern blot analyses of arginase II and ?-actin mRNA
Each frozen kidney was homogenized in RNAxel solution (Eurobio, Cortaboeuf, France), and total RNA was extracted according to the manufacturer’s recommendations and maintained at 4 C. RNA were rinsed twice with 70% ethanol and dried in a Speed Vac. RNA were resuspended in cold 10 mM Tris-HCl and 1 mM EDTA, pH 8.0, and their concentrations were determined by absorbance at 260 nm. Fifteen micrograms of RNA samples were submitted to 1.2% agarose gel electrophoresis. The gel was treated for 20 min in 50 mM NaOH, then for 20 min in a solution containing 0.5 M Tris and 1.5 M NaCl; RNA was transferred overnight to a nylon membrane (Appligène, Illkirch, France) and immobilized using an UV cross-linker (Appligène).
Membranes were hybridized with murine 32P-labeled cDNAs corresponding to arginase II [pBSK– arginase II EcoRI-EcoRI, NCBI accession NM 009705 (21)] and ?-actin [pAL41-cytoplasmic ?-actin, PstI-PstI, (22)]. cDNA was labeled using the RTS RadPrime DNA labeling system (GIBCO BRL, Life Technologies) and -[32P]-deoxycytidine triphosphate. Hybridization was performed overnight at 65 C. Membranes were washed three times in 2x saline sodium citrate (0.3 M NaCl and 30 mM sodium citrate), 5 mM phosphate buffer, and 0.1% SDS and washed three times in 0.5x saline sodium citrate, 3 mM phosphate buffer, and 0.1% SDS. The amount of radioactivity hybridized to specific mRNA was estimated after scanning densitometry of the membranes using a PhosphoImager SI (Molecular Dynamics, Amersham, Orsay, France). Quantification of ?-actin mRNA was used as a control of equal loading and RNA transfer.
Chemicals
Salts and most chemicals, Ponceau S solution, monoclonal mouse anti-?-actin, secondary anti-IgG antibodies, and Kodak X-MAT film were purchased from Sigma (St. Quentin Fallavier, France). Protease inhibitor cocktail was purchased from Boehringer Mannheim (Strasbourg, France). Agarose Seakem GTG was from TEBU (Le Perray-en-Yvelines, France). ECL Western Blotting Kits, -[32P]-deoxycytidine triphosphate (9.25 MBq/25 μl), ImagerMaster Total Lab version 2.01 program, liquid scintillation counting mixture (Aqueous Counting Scintillant ACS II), and monoclonal mouse anti-?-tubulin antibody were purchased from Amersham (Buckinghamshire, UK; Orsay, France).
Results and statistical analyses
Values are means ± SE except when n = 2. The calculations were as follows: for each group of six or three mice and for each protein, the mean IOD of the bands was calculated (see Figs. 5 and 6). The mean IOD of the untreated (group I) (see Fig. 5) and sham-operated (group V) (see Fig. 6) mice were used as a reference (control). For each mouse, the IOD value of a given protein was divided by the mean IOD value of the control group. Consequently, this ratio value is 1 in each control group. Then, these ratios were related to those of ?-actin and/or G3PDH.
FIG. 5. Immunoblot assessing the abundance of OAT, arginase II (AII), and ODC protein in kidney of control (Cont), sham-operated (Sham), and 11- and 18-d orchidectomized male mice. Immunoblots were loaded with samples of 100 μg soluble proteins from six mice (see Table 1). Blots were probed with antibodies to OAT, arginase II, ODC, SA as a control of the efficiency of orchidectomy, and ?-actin and G3PDH as controls of protein loading and transfer. A, To simplify the figure, a representative immunoblot that corresponds to one of the six mice from each group was shown. B–D, Relative quantitation of ODC (B), OAT (C), and arginase II (D) protein levels after scanning densitometry of the immunoblots. Calculations were performed from data obtained with six mice (see Materials and Methods). Values are means ± SE. Differences between groups were statistically analyzed by Kruskal-Wallis test (ODC, P < 0.0011; OAT, P < 0.0003; arginase II, P < 0.017) and followed by Mann-Whitney test. *, P < 0.05, controls vs. 11- or 18-d; #, P < 0.05, Sham vs. 11- or 18-d. Three replicate experiments were performed.
FIG. 6. Immunoblot assessing the abundance of OAT, arginase II (AII), and ODC in the kidney of sham-operated, orchidectomized, oil-treated (Oil), and testosterone-treated (Oil+T) male mice. Immunoblots were loaded with samples of 100 μg soluble proteins from three mice (see Table 1). Blots were probed with antibodies to OAT, arginase II, ODC, SA as a control of the efficiency of orchidectomy, and ?-actin and G3PDH as controls of protein loading and transfer. A, To simplify the figure, a representative immunoblot that corresponds to one of the three mice from each group was shown. B–D, Relative quantitation of ODC (B), OAT (C), and arginase II (D) to G3PDH protein after scanning densitometry of the immunoblots. Calculations were performed from data obtained with three mice (see Materials and Methods). Values are means ± SE. Differences between groups were statistically analyzed by Kruskal-Wallis test (ODC, P < 0.014; OAT, P < 0.032; arginase II, P < 0.026) and followed by Mann-Whitney test. *, P < 0.05, Sham vs. orchidectomy; #, P = 0.05, Oil vs. T + Oil. Three replicate experiments were performed.
Where appropriate, statistical differences were assessed using the Kruskal-Wallis and/or the Mann-Whitney tests (StatView SE+Gr and StatView 5) at the 95% level of significance. For correlation analyses, the correlation coefficient r2 was calculated with Microsoft Excel, and P was determined from tables at the 95% level of significance.
Results
Regulation of OAT, arginase II, and ODC genes by testosterone in female mouse kidneys
The goal of these series of experiments was to test whether testosterone simultaneously induced the expression of arginase II and ODC genes and repressed that of OAT. In addition, it has been verified whether a correlation might be established between testosteronemia and the levels of OAT and arginase II mRNA and proteins.
Time-course effect of testosterone on the expression of OAT protein in female mouse kidneys.
In a previous report (11), we demonstrated that pharmacological doses of testosterone induced a sharp increase in ODC mRNA and protein levels in the female mouse kidneys. For this reason, ODC was used as a control to prove the efficiency of the hormonal treatment. As expected, in these experiments, testosterone treatment induced a progressive increase in the level of the 51- and 53-kDa ODC proteins (Fig. 1A, left). In male and female mouse kidneys, anti-OAT antibody revealed a 48-kDa protein that corresponds to the expected size of the OAT polypeptide (Swiss-Prot P29758 mouse: 48,354 Da; Fig. 1A, left). The renal content of ?-actin, a 44-kDa protein, was not affected by testosterone treatment (Fig. 1A, left). The relative level of OAT to ?-actin protein was about 2-times higher in female than in male mouse kidneys (Fig. 1A, right, Mann-Whitney, P < 0.05). Testosterone treatment induced a progressive and sharp decrease in the amount of OAT protein (Fig. 1A, right, Kruskal-Wallis, P < 0.0013). The relative abundance of OAT to ?-actin protein was about 6-times lower in the kidneys of female mice treated for 5-d with testosterone compared with that of the controls (Fig. 1A, right; Mann-Whitney, P < 0.009).
FIG. 1. Immunoblot assessing the abundance of OAT, arginase II, ODC, ?-actin, and ?-tubulin protein in kidney of male (M), female (F), and testosterone-treated female mice. Female mice were injected sc with 150 μl testosterone propionate (31 mg/ml in sesame oil). Injections were performed every day at 0800 h for a period of 1, 2, 3, and 5 consecutive days. Each lane corresponds to one different mouse, 100 μg (A), and 50 μg (B) of soluble protein. Three immunoblots were performed for (A) and (B). Left, Representative immunoblots probed with antibodies to OAT, arginase II (AII), ODC, ?-tubulin, and ?-actin and revealed by ECL after exposure to an x-ray film. Right, Relative abundance of OAT to ?-actin protein (A) and arginase II to ?-tubulin (B) protein after scanning densitometry of the immunoblots. Values are means ± SE; n = 5 films analyzed in one of the three experiments (A), whereas n = 3 replicate experiments (B). Differences were first statistically analyzed by Kruskal-Wallis followed by Mann-Whitney test: A: *, P < 0.05, M vs. F; **, P < 0.009, F vs. 5-d testosterone-treated females. B: *, P < 0.05, M vs. F; #, P < 0.01, F vs. 5-d testosterone-treated females.
Time-course effect of testosterone on the expression of arginase II protein in female mouse kidneys.
In soluble proteins extracted from male, female, and testosterone-treated female mice, and analyzed by Western blotting, antiarginase II antibody revealed a 38-kDa protein that corresponds to the predicted size of the arginase II polypeptide (Swiss-Prot O08691 mouse: 38,878 Da, Fig. 1B, left). The renal content of ?-tubulin, a 55-kDa protein, was not affected by testosterone treatment (means ± SE, females: 158 ± 77, males: 173.7 ± 51.7; 1-d: 168.7, 2-d: 108.7, 3-d: 110.9, and 5-d: 119.3 x 103 pixels; Kruskal-Wallis P = 0.0734, Fig. 1B, left). The relative level of arginase II to ?-tubulin protein was 3.7-times higher in female than in male mouse kidneys (Fig. 1A, right, Mann-Whitney, P < 0.05). Testosterone treatment induced a progressive increase in arginase II protein level in female mouse kidney (Fig. 1B, right, Kruskal-Wallis, P < 0.0002). The amount of arginase II protein reached its highest level between d 3 and d 5. On d 5, the abundance of arginase II relative to ?-tubulin proteins was 2.8 times higher than in untreated female mouse kidney (Fig. 1B, right, Mann-Whitney, P < 0.01).
Time-course effect of testosterone on the synthesis of arginase II mRNA in female mouse kidneys.
In total RNA extracted from mouse kidneys and submitted to Northern blotting, the murine arginase II cDNA probe allowed detection of the 1.8-kb arginase II mRNA (Fig. 2A). The size of the arginase II mRNA is in good agreement with a previous study (23). There was no statistical difference in the content of the 2.1-kb ?-actin mRNA between the different groups of mice (Fig. 2B, right and Ref.11). The level of arginase II mRNA was higher in female than in male kidneys (Fig. 2B, left, Mann-Whitney, P < 0.021). Testosterone treatment induced a progressive and important increase in the level of arginase II mRNA between d 1 and d 5 (Fig. 2B, left, Kruskal-Wallis, P < 0.01). The highest level of arginase II mRNA was reached on d 5 and was about 3-times higher than in untreated female mice (Fig. 2B, left, Mann-Whitney, P < 0.034).
FIG. 2. Northern blot analysis of arginase II and ?-actin mRNA levels in male (M), female (F), and testosterone-treated female mouse kidneys. Female mice were treated as described in Fig. 1 legend. A, Each line corresponds to 15 μg total RNA extracted from one mouse kidney. B, Quantitation of arginase (AII) and ?-actin mRNA levels. The amount of radioactivity hybridized was estimated after scanning densitometry of the membranes using a PhosphoImager. Left, 1.8-kb arginase II mRNA; right, 2.1-kb ?-actin mRNA used as a control. Values are means ± SE, except when n = 2. Differences between groups were statistically analyzed by Kruskal-Wallis (P < 0.01) and followed by Mann-Whitney test. *, P < 0.021, M vs. F; #, P < 0.034, F vs. 5-d testosterone-treated females.
Testosteronemia in untreated and androgen-treated mice.
In untreated female mice, testosteronemia was very low (0.19 + 0.07 ng/ml, n = 7); whereas in male mice, testosteronemia was about 12-times higher than in females (2.43 ± 1.76 ng/ml, n = 6). As shown in Table 1, the individual level of plasma testosterone in six other male and nine sham-operated male mice varied in a broad range. This variation has been attributed earlier to the pulsatile release of testosterone (24). As expected, in response to injections of pharmacological doses of testosterone, testosteronemia sharply increased in a time-dependent manner (Fig. 3, Kruskal-Wallis, P < 0.0001), and its level was 52- (d 1) and 114-times (d 5) higher compared with that of the males.
TABLE 1. Biological parameters in control, sham-operated, orchidectomized, oil- and testosterone-treated male mice
FIG. 3. Testosteronemia in untreated male (M), female (F), and androgen-treated female mice. Female mice were treated as described in Fig. 1 legend. Testosteronemia was measured by RIA as previously described (16 ). Values are means ± SE, with the number of animals tested given in parentheses.
Relationship between testosteronemia and the expression of OAT and arginase II genes.
To test the link between testosteronemia and the expression of OAT and arginase II genes in the murine kidney, the relative levels of OAT protein, arginase II mRNA, and protein of untreated male and testosterone-treated female mice were plotted against the level of plasma testosterone. The ratio of OAT to ?-actin protein was negatively correlated with testosteronemia (Fig. 4A; r2 = 0.918, P < 0.01). In contrast, the ratios of arginase II to ?-actin mRNA and arginase II to ?-tubulin protein were positively and linearly correlated with testosteronemia (Fig. 4B, r2 = 0.913; and Fig. 4C, r2 = 0.971, respectively, P < 0.01). Therefore, altogether, these results strongly support that pharmacological levels of testosterone simultaneously down-regulate the expression of OAT gene and up-regulate that of arginase II and ODC.
FIG. 4. Correlation between testosteronemia and the relative level of OAT protein (A), arginase II (AII) mRNA (B), and arginase II protein (C) in male and 1- to 5-d testosterone-treated female mice. To test the link between testosteronemia and the expression of arginase II and OAT genes in the murine kidney, the ratios of OAT to ?-actin protein, arginase II to ?-actin mRNA, and arginase II to ?-tubulin protein were calculated and plotted against the level of plasma testosterone. Values are means ± SE. A, r2 = 0.918; B, r2 = 0.913; C, r2 = 0.971; P < 0.01 in all cases. Closed lozenge, 5-d testosterone; open lozenge, 3-d testosterone; open circle, 2-d testosterone; closed circle, 1-d testosterone; open square, untreated female; closed square, untreated male.
Physiological regulation of OAT, arginase II, and ODC genes by testosterone in male mouse kidney
These experiments were conducted to determine whether physiological levels of testosterone regulate the expression of OAT and arginase II genes. To achieve this goal, male mice were castrated to abolish the endogenous production of androgens by testes, and then testosterone propionate was injected into these mice to reverse the effects of orchidectomy. In these experiments, the level of OAT and arginase II proteins was analyzed by Western blotting.
Influence of orchidectomy on biological parameters in male mice.
Mice of groups I, II, and III had similar body weight, whereas mice of group IV were heavier than the others because they were killed 1 wk later (Table 1, Mann-Whitney, group III vs. IV, P < 0.007). The absolute mass of the left and right kidneys was diminished by 30–34% in orchidectomized mice compared with control and sham-operated mice (Table 1, Mann-Whitney, group III and IV vs. I and II, P < 0.004 or less). Orchidectomy diminished the relative kidney mass (RKM, mass of two kidneys/BW) by about 30%. Testosteronemia did not differ between male mice of groups I and II but was undetectable in orchidectomized mice (Table 1, Mann-Whitney, group III and IV vs. I and II, P < 0.004).
Influence of orchidectomy on the level of OAT, arginase II, and ODC proteins in male mouse kidneys.
The 62-kDa androgen-dependent SA polypeptide is abundantly expressed in kidneys of control male mice and was used as a marker of orchidectomy (20). Similar levels of SA protein were detected in kidneys of control and sham-operated mice (Fig. 5A), whereas orchidectomy induced a considerable decrease (90–95%) in SA protein levels (Fig. 5A). The highly androgen-inducible ODC protein was expressed at the same level in kidneys of control and sham-operated male mice (Fig. 5, A and B). Orchidectomy induced a 90–95% decrease in the level of ODC protein compared with that of control and sham-operated mice (Fig. 5B, Mann-Whitney, P < 0.004). The renal content of ?-actin was similar in the four groups of mice (Fig. 5A, quantitation not shown). The level of OAT protein was not affected by the sham operation (Fig. 5C). In contrast, the level of OAT protein was increased by 2.3- and 3-times in mice castrated during 11- and 18-d periods, respectively, compared with the controls (Fig. 5C, Mann-Whitney, P < 0.004). The level of arginase II protein was similar in kidneys of control and sham-operated mice, but it decreased by about 75% and 54%, respectively, in kidneys of mice castrated during 11- and 18-d periods (Fig. 5D, Mann-Whitney, P < 0.011).
Effect of a physiological dose of testosterone on biological parameters in orchidectomized mice.
The reduction of the absolute and relative renal mass, as well as the very low level of plasma testosterone, proved the efficiency of orchidectomy (Table 1, Mann-Whitney, P < 0.05). A single dose of testosterone enhanced the RKM (Table 1, Mann-Whitney, P < 0.05), whereas oil had no influence on the kidney weight and RKM. Androgen treatment restored testosteronemia to the range of physiological level (Table 1, groups I and II) but was higher than testosteronemia measured in group V (Table 1).
Effect of a physiological dose of testosterone and oil on the level of OAT, arginase II, and ODC proteins in orchidectomized mice.
As observed in the previous experiment (Table 1 and Fig. 5), orchidectomy significantly decreased the level of SA, ODC, and arginase II proteins and enhanced that of OAT (Fig. 6, A–D, Kruskal-Wallis, P < 0.032 or less; followed by Mann-Whitney, P < 0.05 for each protein). The injection of a single dose of oil did not affect the renal level of SA, ODC, OAT, and arginase II proteins compared with that of the sham-operated mice (Fig. 6, A–D). In contrast to oil, a single injection of a physiological dose of testosterone induced a sharp increase in the level of ODC protein (Fig. 6, A and B, Mann-Whitney, P < 0.05), tended to restore the level of arginase II protein to that of the control values (Fig. 6, A and D, Mann-Whitney, P < 0.05), and decreased by 2-times the level of OAT protein (Fig. 6, A and C, Mann-Whitney, P < 0.05). In contrast to ODC, the level of SA protein was partially restored by testosterone injection (Fig. 6A). The same results were found when using ?-actin as a control of protein loading and transfer (data not shown).
In conclusion, in the murine kidney, our results strongly suggest that testosterone physiologically up-regulated not only ODC gene, but also arginase II gene, and simultaneously down-regulated OAT gene. In orchidectomized mice, the lack of testosterone led to a reduction of the renal level of ODC and arginase II proteins and enhanced that of OAT. These effects were reversed by injecting a physiological dose of testosterone into castrated male mice.
Discussion
The anabolic hormone testosterone is involved in a multitude of physiological processes and produces a variety of biochemical changes in several organs. The effects of testosterone have been extensively analyzed in the murine kidney because androgens lead to renal hypertrophy. In addition, administration of testosterone to male and female mice induces the renal expression of several genes coding for ?-glucuronidase, ODC, alcohol dehydrogenase, arginase II, kidney androgen-regulated protein, RP2 (mouse androgen kidney or MAK), SA, and transferase II (12, 13, 25, 26). ODC gene is highly sensitive to testosterone, which induces, within hours, a sharp increase in ODC mRNA, protein, and enzyme activity (10, 11). As a physiological consequence, polyamines are synthesized; the renal content of putrescine, spermidine, and spermine is enhanced; and the excess of putrescine is excreted in urine (9, 10). In addition to ODC, OAT gene is highly expressed by the murine kidney. Both enzymes use and compete for the substrate L-ornithine, which is provided, in part, by the mitochondrial arginase II. The present study was designed to determine how OAT and arginase II genes are regulated in kidneys of testosterone-treated mice for the following reasons: 1) the regulation of OAT gene by androgens has not been investigated in the murine kidney; 2) it is not known whether arginase II gene is regulated by androgens at the transcriptional and/or at the translational levels; 3) in unstimulated renal cells, the expression of ODC gene is low, whereas that of OAT is very high; and 4) in kidneys of testosterone-treated mice, the overproduction of putrescine raises the metabolic fate of L-ornithine supplied by arginase II and the balance between OAT and ODC gene expression.
Our results clearly reveal that, in kidneys of female mice, testosterone inversely regulated the expression of OAT and ODC genes. Indeed, pharmacological doses of testosterone deeply diminished the level of OAT protein in a dose-dependent manner, whereas the expression of arginase II and ODC genes was strongly enhanced. The excellent correlations depicted between testosteronemia and the levels of OAT and arginase II mRNA and proteins strongly support these findings. The inverse hormonal regulation of OAT and ODC genes in kidneys of testosterone-treated female mice might favor putrescine synthesis to the detriment of L-glutamate production. The biochemical events might appear as follows: 1) in kidneys of androgen-treated mice, the expression of ODC gene is rapidly increased by about 1000 times and becomes more expressed than OAT gene (12); 2) the sharp down-regulation of OAT gene by testosterone deeply diminished L-ornithine transamination into L-glutamate; 3) induction of arginase II gene expression by testosterone enhances arginase II activity and leads to hydrolyze larger amounts of L-arginine into L-ornithine; 4) L-ornithine produced by arginase II in mitochondria can be transported into the cytosol; and 5) L-ornithine is converted in putrescine by the cytosolic ODC (Fig. 7A). Because testosterone exerts anabolic effects, the new production of polyamines by the kidneys may contribute to general physiological needs such as cell growth, hypertrophy, mRNA and protein synthesis, membrane stability, and a variety of cell functions in the different organs (27).
FIG. 7. Renal fate of L-arginine-derived ornithine in testosterone-treated female mouse and untreated male mouse (A) and in untreated female mouse and castrated male mice (B). A, In the presence of testosterone, ODC and arginase II (AII) genes are highly expressed, and L-arginine is a precursor for polyamine synthesis (thick line), whereas OAT gene expression is decreased (hatched line). B, In contrast, in untreated female mice and orchidectomized mice, ODC gene is not expressed (hatched line), and OAT gene is highly expressed, leading to L-glutamate production. P5CDH, Pyrroline-5-carboxylate dehydrogenase; GS, glutamine synthase; NAD+, nicotinamide adenine dinucleotide; NADH,H+, reduced nicotinamide adenine dinucleotide; NADP+, nicotinamide adenine dinucleotide phosphate; NADPH,H+, reduced nicotinamide adenine dinucleotide phosphate; GLDH, glutamate dehydrogenase.
We felt it more relevant to determine whether OAT gene was physiologically regulated by a physiological level of plasma testosterone and to examine simultaneously the expression of ODC and arginase II genes in mouse kidneys. We first analyzed the influence of gender on the basal expression of these three genes. OAT, ODC, and arginase II genes were expressed in untreated male mouse kidneys (see also Ref.11). In contrast, in kidneys of untreated female mice, ODC gene was not expressed (see also Refs.5 and 11), whereas OAT and arginase II genes were more expressed than in those of the males. The very low level of plasma testosterone in females, compared with that of the males, might be responsible for the lack of ODC gene expression and the very high level of OAT protein in the female mouse kidney. OAT metabolizes L-ornithine into L-glutamate, which can be converted either into L-glutamine and/or energy by entering into the citric acid cycle (Fig. 7B). This result confirms the constitutive and inverse regulation of OAT and ODC gene expression by a physiological level of testosterone. In contrast, at present, it is difficult to explain why kidneys of female mice exhibited higher levels of arginase II mRNA, protein, and enzyme activity than those of the males (7, 13). Although arginase II gene is more expressed in female than in male mouse kidney, arginase II activity has been known for a long time to be increased by pharmacological doses of testosterone. In addition, our data reveal that arginase II gene was regulated at the transcriptional and translational levels by high doses of testosterone. Altogether, these results strongly indicate that arginase II gene is controlled by testosterone.
Thereafter, we studied the physiological regulation of OAT gene by testosterone in male mouse kidney. Orchidectomy, rather than drug inhibitors, was used to stop the production of androgen hormones by testes. Proteins were analyzed at 11 and 18 d after the operation to guarantee a complete depletion of testosterone and to reduce the effects due to the surgery. As expected, the lack of testosterone led to a sharp decrease in the level of ODC and arginase II proteins. It appeared that arginase II gene was less sensitive to androgens than that of ODC because the level of arginase II protein was not totally abolished by orchidectomy. This result suggests that the basal level of arginase II gene expression is not under the control of androgens. In contrast to ODC and arginase II, the level of OAT protein was highly enhanced by orchidectomy, supporting the idea that testosterone physiologically down-regulates OAT gene. The effects of orchidectomy on gene expression were reversed by injecting a single dose of testosterone (0.55 mg/mouse) that shifted testosteronemia to physiological levels. This dose was calculated from the data obtained from testosterone-treated female mice (Fig. 3). As expected, a low level of testosteronemia restored the expression of ODC protein. A physiological level of testosteronemia clearly diminished the level of OAT protein and reached the level measured in the sham-operated group. Our results demonstrated, for the first time, that OAT gene is physiologically down-regulated by testosterone. Finally, testosterone increased, and almost restored, the renal level of arginase II protein. In the past, Kochakian and Stettner (13) reported that a pharmacological dose of 14–15 mg testosterone given per castrated-male Swiss mouse enhanced the renal arginase II activity and concluded that arginase II gene was controlled by testosterone. However, here, we demonstrated, for the first time, that testosterone physiologically up-regulated arginase II gene in the mouse kidney. Surprisingly, Manteuffel-Cymborowska et al. (7), who gave a dose of 4.5 mg testosterone per male Swiss mouse, did not observe changes in the renal arginase II activity.
It is generally assumed that testosterone and steroid hormones act in different biological programs through the binding of intracellular receptors, which act as ligand-inducible transcription factors on specific DNA elements of target genes. The receptors for steroid hormones belong to a single receptor superfamily, which includes receptors for androgens, estrogens, progesterone, glucocorticoids, mineralocorticoids, retinoic acid, thyroid hormone, and vitamin D. All steroid receptors share similar structural and functional characteristics with respect to regulating the transcriptional activity of specific genes (28). The androgen receptor has been cloned and characterized, and specific antibodies-antiandrogen receptors allowed to identify its nuclear localization in a variety of animal species and tissues, including cells of the proximal and distal tubules of the rat kidney (29). In mouse kidney, androgens induce the expression of kidney androgen-regulated protein and ODC genes that possess an androgen response element (ARE) in the 5' flanking region of their genes (30). Recently, the promoter region of the murine, rat, and human OAT genes has been sequenced. Numerous 5'-AGGTCA-like motifs corresponding to the consensus binding site for several members of the nuclear receptor superfamily were located from nucleotides –222 to –205 and 366–396 of the rat OAT gene and are relatively well conserved in mouse OAT gene (31, 32). At present, we do not know whether the AGGTCA-like motif mediates the androgen response. Whatever, the physiological consequence of the presence of testosterone was a down-regulation of OAT gene expression. The promoter region of the murine arginase II gene does not contain an ARE but possesses numerous potential binding sites for enhancer and promoter elements, including AP1, NF-KB, SP1, and CRE-BP2 (33). Consequently, it is difficult to understand the molecular mechanism through which testosterone induces the biological effect on arginase II gene in the absence of ARE. Therefore, several possibilities might account for this observation: 1) the increase in arginase II gene expression was due to a direct or indirect effect of testosterone; 2) the promoter of arginase II gene possesses an ARE distal to the sequenced region; 3) testosterone binds to a cellular membrane receptor as recently reported (34); 4) testosterone binds to a glucocorticoid receptor or another receptor belonging to this superfamily; and 5) testosterone induces a cascade of events that are secondarily responsible for the increase of arginase II gene expression. Recently, it has been reported that SHBG, which complexes testosterone, binds to a SHBG-receptor on cell membranes of several tissues and therefore activates a G-coupled protein that modulates the production of cAMP (34). In addition, in RAW 264.7 cells, the level of arginase II mRNA was enhanced by dexamethasone and dibutyryl cAMP (31, 35). Further experiments will be performed to clarify the mechanisms involved in regulating arginase II gene by testosterone.
In conclusion, we demonstrated, for the first time, that OAT gene was pharmacologically and physiologically down-regulated by testosterone in male and female mouse kidneys. ODC and OAT genes were inversely regulated by testosterone. In the presence of high levels of the androgen hormone, the synthesis of polyamines was favored, whereas that of L-glutamate is suspected to be decreased. In contrast, in control female mouse kidney, the lack of ODC suggests that L-arginine-derived ornithine might be metabolized by OAT to produce L-glutamate.
Acknowledgments
The authors are indebted to Dr. Sidney M. Morris (University of Pittsburgh, Pittsburgh, PA) and Dr. Anna Meseguer (Centre d’Investigacions en Bioquimica i Biologia Molecular, Hospital Universitari Vall d’Hebron, Barcelona, Spain), who kindly provided, respectively, pBSK– arginase II plasmid containing mouse arginase II cDNA [full-length cDNA encoding for the murine arginase II type, GenBank accession no. AF032466 (21 )], and the rabbit anti-SA antibody (20 ). O.L. is indebted to Marie Thérèse Ducluseau and Prof. Jean-Fran?ois Nicolas, who kindly gave access to materials for RNA extraction; Prof. Dr. Bruno Claustrat for discussion; Jocelyne Vial for assistance; and Bernard Marchand for his contribution in the animal house.
References
Levillain O, Hus-Citharel A, Morel F, Bankir L 1992 Localization of urea and ornithine production along mouse and rabbit nephrons: functional significance. Am J Physiol (Renal Fluid Electrolyte Physiol) 263:F878–F885
Pegg AE, Seely J, Zagon IS 1982 Autoradiographic identification of ornithine decarboxylase in mouse kidney by means of -[5-14C] difluoromethylornithine. Science 217:68–70
Koibuchi N, Matsuzaki S, Kitani T, Fujisawa H, Suzuki M 1990 Localization by immunohistochemistry of renal ornithine decarboxylase in the mouse with and without testosterone treatment. Endocrinol Jpn 37:555–561
Levillain O, Hus-Citharel A 1998 Ornithine decarboxylase along the mouse and rat nephron. Am J Physiol Renal Physiol 43:F1020–F1028
Levillain O, Diaz J-J, Reymond I, Soulet D 2000 Ornithine metabolism along the female mouse nephron: localization of ornithine decarboxylase and ornithine amino-transferase. Pflügers Arch Eur J Physiol 440:761–769
Alonso E, Rubio V 1989 Participation of ornithine aminotransferase in the synthesis and catabolism of ornithine in mice. Biochem J 259:131–138
Manteuffel-Cymborowska M, Chmurzynska W, Peska M, Grzelakowska-Sztabert B 1995 Arginine and ornithine metabolizing enzymes in testosterone-induced hypertrophic mouse kidney. Int J Biochem Cell Biol 27:287–295
Berger FG, Szymanski P, Read E, Watson G 1984 Androgen-regulated ornithine decarboxylase mRNAs of mouse kidney. J Biol Chem 259:7941–7946
Persson L 1981 Putrescine turnover in kidneys of testosterone-treated mice. Biochim Biophys Acta 674:204–211
Pajunen AEI, Isomaa VV, J?nne OA, Bardin CW 1982 Androgenic regulation of ornithine decarboxylase activity in mouse kidney and its relationship to changes in cytosol and nuclear androgen receptor concentrations. J Biol Chem 257:8190–9198
Levillain O, Greco A, Diaz J-J, Augier R, Didier A, Kindbeiter K, Catez F, Cayre M 2003 Influence of testosterone on the regulation of ornithine decaboxylase, antizyme and spermidine/spermine-N1-acetyl-transferase gene expression in female mouse kidney. Am J Physiol Renal Physiol 285:F498–F506
Catterall JF, Kontula KK, Watson CS, Sepp?nen PJ, Funkenstein B, Melanitou E, Hickok NJ, Bardin CW, J?nne OA 1986 Regulation of gene expression by androgens in murine kidney. Recent Prog Horm Res 42:71–109
Kochakian CD, Stettner CE 1948 Effect of testosterone propionate and growth hormone on the arginase and phosphatases of the organs of the mouse. Am J Physiol 155:262–264
Silbernagl S 1992 Amino acids and oligopeptides. In: Seldin DW, Giebish G, eds. The kidney: physiology and pathophysiology. 2nd ed. New York: Raven Press, Ltd; 2889–2920
Sanchez-Capelo A, Penafiel R, Tovar A, Galindo JD, Cremades A 1994 Postnatal development of ornithine decarboxylase and polyamines in the mouse kidney: influence of testosterone. Biol Neonate 66:119–127
Déchaud H, Lejeune H, Garoscio-Cholet M, Mallein R, Pugeat M 1989 Radioimmunoassay of testosterone not bound to sex-steroid-binding protein in plasma. Clin Chem 35:1609–1614
Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685
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
Levillain O, Hus-Citharel A, Garvi S, Peyrol S, Reymond I, Mutin M, Morel F 2004 Ornithine metabolism in male and female rat kidney mitochondrial expression of ornithine aminotransferase and arginase II. Am J Physiol Renal Physiol 286:F727–F738
Aresté C, Meliá MJ, Isern J, Tovar JL, Meseguer A 2004 Sex steroid regulation and identification of different transcription units of the SA gene in mouse kidney. J Endocrinol 183:101–114
Morris Jr SM, Kepka-Lenhart D, Chen LC 1998 Differential regulation of arginases and inducible nitric oxide synthase in murine macrophage cells. Am J Physiol 38:E740–E747
Hanahan D 1983 Studies on transformation of E. coli with plasmids. J Mol Biol 166:557–580
Ozaki M, Gotoh T, Nagasaki A, Miyanaka K, Takeya M, Fujiyama S, Tomita K, Mori M 1999 Expression of arginase II and related enzymes in the rat small intestine and kidney. J Biochem (Tokyo) 125:586–593
Bartke A, Steele RE, Musto N, Caldwell BV 1973 Fluctuations in plasma testosterone levels in adult male rats and mice. Endocrinology 92:1223–1228
King D, Sun YH, Lingrel JB 1986 Amino acid sequence of the testosterone-regulated mouse kidney RP2 protein deduced from its complementary DNA sequence. Nucleic Acids Res 14:5159–5170
Melia MJ, Bofill N, Hubank M, Meseguer A 1998 Identification of androgen-regulated genes in mouse kidney by representational difference analysis and random arbitrarily primed polymerase chain reaction. Endocrinology 139:688–695
Pegg AE, McCann PP 1982 Polyamine metabolism and function. Am J Physiol 243:C212–C221
Schuchard M, Landers JP, Sandhu NP, Spelsberg TC 1993 Steroid hormone regulation of nuclear proto-oncogenes. Endocr Rev 14:659–669
Takeda H, Chodak G, Mutchnik S, Nakamoto T, Chang C 1990 Immunohistochemical localization of androgen receptors with mono- and polyclonal antibodies to androgen receptor. J Endocrinol 126:17–25
Crozat A, Palvimo JJ, Julkunen M, J?nne OA 1992 Comparison of androgen regulation of ornithine decarboxylase and S-adenosylmethionine decarboxylase gene expression in rodent kidney and accessory sex organs. Endocrinology 130:1131–1144
Shull JD, Pennington KL, Pitot HC, Boryca VS, Schulte BL 1992 Isolation and characterization of the rat gene encoding ornithine aminotransferase. Biochim Biophys Acta 1132:214–218
Shull JD, Esumi N, Colwell AS, Pennington KL, Jendoubi M 1995 Sequence of the promoter region of the mouse gene encoding ornithine aminotransferase. Gene 162:275–277
Shi O, Kepka Lenhart D, Morris Jr SM, O’Brien WE 1998 Structure of the murine arginase II gene. Mamm Genome 9:822–824
Heinlein CA, Chang C 2002 The roles of androgen receptors and androgen-binding proteins in nongenomic androgen actions. Mol Endocrinol 16:2181–2187
Gotoh T, Sonoki T, Nagasaki A, Terada K, Takiguchi M, Mori M 1996 Molecular cloning of cDNA for nonhepatic mitochondrial arginase (arginase II) and comparison of its induction with nitric oxide synthase in a murine macrophage-like cell line. FEBS Lett 395:119–122(Olivier Levillain, Jean-J)