**4. Discussion**

OAT plays a pivotal role in the intermediary metabolism because this enzyme is at the crossroad of several pathways. OAT controls the production of glutamate and ornithine, competes with the polyamine pathway, and may be a source of carbon for the renal gluconeogenesis. These pathways are essential for many physiological roles including growth and energy supply. Several reports clearly documented the hormonal regulation of OAT gene in the rat liver and kidney. In this species, the renal expression of OAT gene is upregulated by estrogens and triiodothyronine (Herzfeld & Knox, 1968; Lyons & Pitot, 1977; Mueckler & Pitot, 1983; Mueckler et al., 1984). During the post-natal development of the rat kidney, the sexual dimorphism of the expression of OAT gene increased in parallel with the endogenous synthesis of estrogens (Herzfeld & Knox, 1968). In contrast, testosterone affected OAT activity neither in liver nor in kidneys of male and female rats (Herzfeld & Knox, 1968). The authors concluded that estrogen naturally control and upregulate the expression of OAT

per group). To simplify the Figure, a representative RT-PCR that corresponds to one mouse from each group was shown. Abbreviations: sham-operated (Sham), and 11-day orchidectomized mice (11d), cyclophilin A (Cyclo A), eukaryotic initiation factor eIF4-E (eIF4-E), and oil + testosterone (Testo). Differences between groups were statistically

This experiment was performed to determine the delay requested for testosterone to downregulate the expression of OAT gene in the mouse kidney. The time-course effect of testosterone on the expression of OAT gene was analyzed at the transcriptional and posttranslational levels 8, 24, 28, and 32 hrs following hormone administration. Eight hours following testosterone treatment, the level of OAT mRNA was reduced by 8%, but did not reach statistical significance (Fig. 7A). In contrast, 24, 28, and 32 hrs after testosterone replacement, the renal levels of OAT mRNA were decreased by 45%, 40%, and 36%, respectively (Kruskal-Wallis *P* < 0.0028 followed by Mann Whitney, *P* < 0.0209 for each case). The pattern of OAT activity along this time-course study strongly resembled that of OAT mRNA except that the decrease in OAT activity was about 2-fold lower than that of OAT mRNA (Fig. 7B). Indeed, 24, 28, and 32 hrs after testosterone administration, OAT activities were decreased by 20%, 21%, and 23%, respectively (Kruskal-Wallis *P* < 0.0071 followed by Mann Whitney, *P* < 0.0209 for each case). The efficiency of the hormonal treatment was checked by measuring testosteronemia. Our results show that testosteronemia was very high 8 hrs after a single injection of the hormone and regularly decreased over the experimental period. Thirty-two hours after the treatment, testosteronemia was about 2-fold higher than the physiological concentration of testosterone reported for control male mice (Levillain et al., 2007). Corticosteronemia were quantified in the same mice to determine whether testosteronemia modified the plasma level of corticosterone. A high concentration of corticosterone was detected 8 hrs after testosterone treatment (Fig. 7C). One day later (24 and 28 hrs), the plasma level of corticosterone were decreased by 3-fold whereas at 32 hrs, it increased again by 2-fold (Fig. 7D). Unfortunately, given that it was impossible to have access to the animal room during the night, no mice

OAT plays a pivotal role in the intermediary metabolism because this enzyme is at the crossroad of several pathways. OAT controls the production of glutamate and ornithine, competes with the polyamine pathway, and may be a source of carbon for the renal gluconeogenesis. These pathways are essential for many physiological roles including growth and energy supply. Several reports clearly documented the hormonal regulation of OAT gene in the rat liver and kidney. In this species, the renal expression of OAT gene is upregulated by estrogens and triiodothyronine (Herzfeld & Knox, 1968; Lyons & Pitot, 1977; Mueckler & Pitot, 1983; Mueckler et al., 1984). During the post-natal development of the rat kidney, the sexual dimorphism of the expression of OAT gene increased in parallel with the endogenous synthesis of estrogens (Herzfeld & Knox, 1968). In contrast, testosterone affected OAT activity neither in liver nor in kidneys of male and female rats (Herzfeld & Knox, 1968). The authors concluded that estrogen naturally control and upregulate the expression of OAT

analyzed by Kruskal-Wallis test, *P* = 0.1574.

could be sampled during the dark period.

**4. Discussion** 

**3.5 Time course changes in the expression of OAT gene** 

Fig. 7. Time course effect of testosterone on the expression of ornithine aminotransferase gene in the mouse kidney. A single dose of testosterone was injected to castrated-male mice. Mice were killed 8, 24, 28, and 32 hrs after the hormonal-treatment. Untreated castrated mice were used as controls. The values are expressed as means ± SE, (n=4 mice per group). A: The level of OAT mRNA were analyzed by semi-quantitative RT-PCR. The amplified cDNAs were separated by agarose-gel electrophoresis. Band intensities were quantified and reported relative to the cyclophilin A band. To simplify the Figure, a representative RT-PCR that corresponds to one mouse from each group was shown. B: OAT activity was measured in mouse kidneys. C: Plasma levels of corticosterone. D: Plasma levels of testosterone. The black bar represents the dark period (night). Differences between groups were statistically analyzed by Kruskal-Wallis test and followed by Mann-Whitney test. \*, P < 0.0209 in all cases.

gene in the female rat kidney. As reported for the rats, the mouse kidney also exhibits a sexual dimorphism in the expression of OAT gene. Although the hormonal regulation of this gene has been poorly studied in both liver and kidney of male and female mice, it has been clearly shown that the endogenous production of testosterone was responsible for a decrease in the renal expression of OAT gene (Levillain et al., 2007). Interestingly, in contrast to the rats, ovariectomy altered neither the level of OAT protein nor OAT activity in the murine kidney (Levillain et al., 2007). Estrogens do not control the renal expression of OAT in the mouse kidney. Consequently, testosterone naturally controls and down-regulates the expression of

Orchidectomy Upregulates While Testosterone Treatment Downregulates

adrenalectomy may afford a model to resolve this point.

appears to be negatively regulated by the AR *in vivo*.

California, USA for reading and improving the paper.

diseases as reported for prostate cancer.

**5. Acknowledgements** 

ISSN 0264-6021

**6. References** 

the Expression of Ornithine Aminotransferase Gene in the Mouse Kidney 129

testosterone treatment decreased by 2.6-fold the weight of the female adrenal gland which did not differ from that of the control male adrenal gland (Bastida et al, 2007). The plasma level of corticosterone was 2.8-fold higher in the females as compared to the males (Bastida et al, 2007). These results led us to hypothesize that testosterone may control the plasma level of corticosterone which may secondarily regulate the expression of OAT gene in the mouse kidney. For this reason the plasma level of corticosterone were measured in the different groups of mice. Our results did show a trend for the plasma levels of corticosterone to be higher in orchidectomized mice than in untreated male mice and in castrated male mice treated with testosterone, however, with a low sample number and high standard error the difference did not reach statistical significance. Contrarily, when analyzing the time course effect of testosterone 8 hrs after the hormonal treatment the plasma level of corticosterone and testosterone were both very high suggesting that testosterone does not control the rate of synthesis and release of corticosterone. The presence of GRE and ARE in the promoter of the mouse OAT gene offers new insight on the hormonal regulation of this gene. Further experiments utilizing male mice subjected to both orchidectomy and

Hormonal regulation of OAT gene by steroids can impact human disease model, such as prostate cancer (PCa). Androgen receptor (AR) targets have been recently identified in a model of androgen-dependent (LNCaP) and a model of castrate-resistant (C4-2B) human PCa cell lines (Jariwala et al., 2007) with OAT expressed at a higher level in the castrate resistant (C4-2B) cells. Expression of OAT was repressed by siRNA in C4-2B cells demonstrating that OAT was an AR regulated gene, although it did not respond to dihydrotestosterone (DHT). Furthermore, the expression of OAT was analyzed in primary PCa tumors and primary PCa tumors after 3 months of androgen ablation therapy. OAT

In conclusion, in the mouse kidney, testosterone down-regulates the expression OAT gene at the transcriptional level. This finding may have interesting consequences for human

The authors are indebted to Dr Joseph Satriano, Department of Medicine, University of California, San Diego, and Veterans Administration San Diego Healthcare System, La Jolla,

Alonso, E. & Rubio, V. (1989). Participation of ornithine aminotransferase in the synthesis

Bastida, C.M.; Cremades, A.; Castells, M.T.; Lopez-Contreras, A.J.; Lopez-Garcia, C.;

*and Metabolism,* Vol.292, No.4, pp. E1010-1017, ISSN 0193-1849

and catabolism of ornithine in mice. *Biochemical Journal,* Vol.259, No.1, pp. 131-138,

Sanchez-Mas, J. & Penafiel, R. (2007). Sexual dimorphism of ornithine decarboxylase in the mouse adrenal: influence of polyamine deprivation on catecholamine and corticoid levels. *American Journal of Physiology - Endocrinology* 

OAT gene in the male mouse kidney. Furthermore, orchidectomy completely abolished the sexual dimorphism of the renal expression of OAT gene (Levillain et al., 2007). The present findings confirm these statements. Moreover, our results show that testosterone negatively regulated the expression of OAT gene at the transcriptional level. Indeed, the levels of OAT mRNA, protein, and enzyme activity were markedly increased following castration. The castrated male mice were killed either 11 days or 18 days after the surgery to completely deplete the endogenous pool of testosterone. Plasma testosterone levels showed that all mice were testosterone-free as soon as 11 days after castration (Levillain et al., 2005).

To further explore the mechanismes involved in the regulation of OAT gene by testosterone in the mouse kidney, experiments were conducted to test the consequence of testosterone replacement on the expression of OAT gene at the transcriptional, translational, and posttranslational levels. The dose of testosterone injected was calculated to reach a physiological plasma level of testosterone. The efficiency of testosterone to lower the level of OAT mRNA in the mouse kidney was confirmed. These results demonstrate that testosterone regulates OAT gene expression at the transcriptional level. The regulation of OAT gene at the transcriptional level has been also reported in the rat tissues (Mueckler & Pitot, 1983; Mueckler et al., 1983, 1984). The levels of OAT protein and OAT activity were also diminished by a single injection of testosterone, however, this decrease was less pronounced than that of OAT mRNA. This difference may be explained by the long half-life of OAT protein. The half-life of OAT protein is estimated at about 48 hrs (Ip et al., 1974). The timecourse study of the effect of testosterone on the expression of OAT gene in castrated male mice also supports this timeline with the level of OAT mRNA decreasing more rapidly than OAT activity. This experiment also revealed a lag-time of 8 hrs or more from testosterone administration before detecting significant changes in the level of OAT mRNA. Similar results were observed with injection of testosterone to female mice (paper in preparation). This delay from administration to mRNA expression suggests that several genomic and/or non-genomic events may take place in the signaling cascade of testosterone (Heinlein & Chang, 2002). One hypothesis is that a target molecule such as the eukaryotic initiation factor eIF4-E is rate-limiting for OAT translation (Fagan et al., 1991). However, in our expriments, the level of eIF4-E mRNA was altered neither by orchidectomy nor by testosterone replacement.

In order to explain a direct genomic action of testosterone, an attempt was made to identify ARE on the promoter of the mouse OAT gene. We carried out an *in silico* search for homology to the consensus ARE motifs by selecting either the left half-site 5'-AGAACA-3' or the right half-site 5'-TGTTCT-3' motifs by search computational using MatInspector (http://www.genomatix.de/) (Beato et al., 1989; Fabre et al., 1994; Roche et al., 1992; Wang et al., 2007; Merkulov & Merkulova, 2009). Several ARE (5'-AGAACAnnnTGTTCT-3') and glucocorticoid responsive element (GRE, 5'-GGTACAnnnTGTTCT-3') which share the same sequence 5'-TGTTCT-3' (Nelson et al., 1999, Verrijdt et al., 2003) were identified on the promoter of the mouse OAT gene. Interestingly, four GRE have been found in the human OAT gene (Zintz & Inana, 1990). Further experiments will be required to determine the functionality of these elements in response to hormone administration.

Recently, it has been reported that, in Swiss CD1 mice, sex hormones influence the weight of the adrenal gland and the plasma level of corticosterone (Bastida et al, 2007). Indeed, the weight of the female adrenal gland was 2.8-fold higher than that of the male (Bastida et al, 2007). Orchidectomy enhanced the weight of the male adrenal gland by 2.1-fold whereas testosterone treatment decreased by 2.6-fold the weight of the female adrenal gland which did not differ from that of the control male adrenal gland (Bastida et al, 2007). The plasma level of corticosterone was 2.8-fold higher in the females as compared to the males (Bastida et al, 2007). These results led us to hypothesize that testosterone may control the plasma level of corticosterone which may secondarily regulate the expression of OAT gene in the mouse kidney. For this reason the plasma level of corticosterone were measured in the different groups of mice. Our results did show a trend for the plasma levels of corticosterone to be higher in orchidectomized mice than in untreated male mice and in castrated male mice treated with testosterone, however, with a low sample number and high standard error the difference did not reach statistical significance. Contrarily, when analyzing the time course effect of testosterone 8 hrs after the hormonal treatment the plasma level of corticosterone and testosterone were both very high suggesting that testosterone does not control the rate of synthesis and release of corticosterone. The presence of GRE and ARE in the promoter of the mouse OAT gene offers new insight on the hormonal regulation of this gene. Further experiments utilizing male mice subjected to both orchidectomy and adrenalectomy may afford a model to resolve this point.

Hormonal regulation of OAT gene by steroids can impact human disease model, such as prostate cancer (PCa). Androgen receptor (AR) targets have been recently identified in a model of androgen-dependent (LNCaP) and a model of castrate-resistant (C4-2B) human PCa cell lines (Jariwala et al., 2007) with OAT expressed at a higher level in the castrate resistant (C4-2B) cells. Expression of OAT was repressed by siRNA in C4-2B cells demonstrating that OAT was an AR regulated gene, although it did not respond to dihydrotestosterone (DHT). Furthermore, the expression of OAT was analyzed in primary PCa tumors and primary PCa tumors after 3 months of androgen ablation therapy. OAT appears to be negatively regulated by the AR *in vivo*.

In conclusion, in the mouse kidney, testosterone down-regulates the expression OAT gene at the transcriptional level. This finding may have interesting consequences for human diseases as reported for prostate cancer.
