Inducible Arginase 1 Deficiency in Mice Leads to Hyperargininemia and Altered Amino Acid Metabolism

Arginase deficiency is a rare autosomal recessive disorder resulting from a loss of the liver arginase isoform, arginase 1 (ARG1), which is the final step in the urea cycle for detoxifying ammonia. ARG1 deficiency leads to hyperargininemia, characterized by progressive neurological impairment, persistent growth retardation and infrequent episodes of hyperammonemia. Using the Cre/loxP-directed conditional gene knockout system, we generated an inducible Arg1-deficient mouse model by crossing “floxed” Arg1 mice with CreER^T2 mice. The resulting mice (Arg-Cre) die about two weeks after tamoxifen administration regardless of the starting age of inducing the knockout. These treated mice were nearly devoid of Arg1 mRNA, protein and liver arginase activity, and exhibited symptoms of hyperammonemia. Plasma amino acid analysis revealed pronounced hyperargininemia and significant alterations in amino acid and guanidino compound metabolism, including increased citrulline and guanidinoacetic acid. Despite no alteration in ornithine levels, concentrations of other amino acids such as proline and the branched-chain amino acids were reduced. In summary, we have generated and characterized an inducible Arg1-deficient mouse model exhibiting several pathologic manifestations of hyperargininemia. This model should prove useful for exploring potential treatment options of ARG1 deficiency.     

Polyamine homeostasis in arginase knockout mice

The role of ornithine decarboxylase (ODC) in polyamine metabolism has long been established, but the exact source of ornithine has always been unclear. The arginase enzymes are capable of producing ornithine for the production of polyamines and may hold important regulatory functions in the maintenance of this pathway. Utilizing our unique set of arginase single and double knockout mice, we analyzed polyamine levels in the livers, brains, kidneys, and small intestines of the mice at 2 wk of age, the latest timepoint at which all of them are still alive, to determine whether tissue polyamine levels were altered in response to a disruption of arginase I (AI) and II (AII) enzymatic activity. Whereas putrescine was minimally increased in the liver and kidneys from the AII knockout mice, spermidine and spermine were maintained. ODC activity was not greatly altered in the knockout animals and did not correlate with the fluctuations in putrescine. mRNA levels of ornithine aminotransferase (OAT), antizyme 1 (AZ1), and spermidine/spermine-N¹-acetyltransferase (SSAT) were also measured and only minor alterations were seen, most notably an increase in OAT expression seen in the liver of AI knockout and double knockout mice. It appears that putrescine catabolism may be affected in the liver when AI is disrupted and ornithine levels are highly reduced. These results suggest that endogenous arginase-derived ornithine may not directly contribute to polyamine homeostasis in mice. Alternate sources such as diet may provide sufficient polyamines for maintenance in mammalian tissues. ornithine; putrescine; spermidine; spermine; decarboxylase     

Enteral arginase II provides ornithine for citrulline synthesis

The synthesis of citrulline from arginine in the small intestine depends on the provision of ornithine. To test the hypothesis that arginase II plays a central role in the supply of ornithine for citrulline synthesis, the contribution of dietary arginine, glutamine, and proline was determined by utilizing multitracer stable isotope protocols in arginase II knockout (AII(-/-)) and wild-type (WT) mice. The lack of arginase II resulted in a lower citrulline rate of appearance (121 vs. 137 μmol·kg(-1)·h(-1)) due to a reduced availability of ornithine; ornithine supplementation was able to restore the rate of citrulline production in AII(-/-) to levels comparable with WT mice. There were significant differences in the utilization of dietary citrulline precursors. The contribution of dietary arginine to the synthesis of citrulline was reduced from 45 to 10 μmol·kg(-1)·h(-1) due to the lack of arginase II. No enteral utilization of arginine was observed in AII(-/-) mice (WT = 25 μmol·kg(-1)·h(-1)), and the contribution of dietary arginine through plasma ornithine was reduced in the transgenic mice (20 vs. 13 μmol·kg(-1)·h(-1)). Dietary glutamine and proline utilization were greater in AII(-/-) than in WT mice (20 vs. 13 and 1.4 vs. 3.7 μmol·kg(-1)·h(-1), respectively). Most of the contribution of glutamine and proline was enteral rather than through plasma ornithine. The arginase isoform present in the small intestinal mucosa has the role of providing ornithine for citrulline synthesis. The lack of arginase II results in a greater contribution of plasma ornithine and dietary glutamine and proline to the synthesis of citrulline.     

Enzymes of ornithine metabolism in adult and developing rat intestine.

The levels of 11 enzymes, most of them involved in the metabolism of ornithine, were measured in whole upper intestine, or in duodenum, small intestine and colon of adult rats. The developmental formations in small intestine of arginase, ornithine aminotransferase, and ornithine transcarbamylase were compared with those in liver. Changes with age (late gestation of adult) of the intestinal activities of pyrroline-5-carboxylate reductase, proline oxidase and glutamyl transpeptidase are also described. The results suggest that the proximal part of the intestine is well endowed with enzymes involved in the conversion of ornithine to proline as well as to citrulline. Fetal intestine is rich in proline oxidase and pyrroline-5-carboxylate reductase. The peak levels of ornithine aminotransferase found in intestine in the first 3 postnatal weeks were higher than seen in any other rat tissue. Some of the properties of arginase, ornithine aminotransferase and pyrroline-5-carboxylate reductase in small intestine were compared with those in liver. Isozymes of arginase in small intestine differed from those in liver; the kinetic properties of ornithine aminotransferase were similar in the two tissues. In intestine of 14-day-old rats, the ornithine aminotransferase reaction was reversible, forming ornithine from pyrroline-5-carboxylate. The intestinal pyrroline-5-carboxylate reductase was cold-labile as was the hepatic enzyme in rat.     

The kinetics of inhibition of Vigna catjang cotyledon and buffalo liver arginase by L-proline and branched-chain amino acids

The effect of proline, isoleucine, leucine, valine, lysine and ornithine under standard physiological conditions, on purified Vigna catjang cotyledon and buffalo liver arginases was studied. The results showed that V. catjang cotyledon arginase is inhibited by proline at a lower concentration than buffalo liver arginase and the inhibition was found to be linear competitive for both enzymes. Buffalo liver arginase was more sensitive to inhibition by branched-chain amino acids than V. catjang cotyledon. Leucine, lysine, ornithine and valine are competitive inhibitors while isoleucine is a mixed type of inhibitor of liver arginase. We have also studied the effect of manganese concentration which acts as a cofactor and leads to activation of arginase. The optimum Mn2+ concentration for Vigna catjang cotyledon arginase is 0.6 mM and liver arginase is 2.0 mM. The preincubation period required for liver arginase is 20 min at 55 degrees C, the preincubation period and temperature required for activation of cotyledon arginase was found to be 8 min at 35 degrees C. The function of cotyledon arginase in polyamine biosynthesis and a possible role of branched chain amino acids in hydrolysis of arginine in liver are discussed.     

Changes in arginase isoenzymes pattern in human hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is one of the most common tumors worldwide affecting preferentially patients with liver cirrhosis. The studies were performed on tissues obtained during surgery from 50 patients with HCC, 40 with liver cirrhosis and 40 control livers. It was found that arginase activity in HCC was nearly 5- and 15-fold lower than in cirrhotic and normal livers, respectively. Isoenzymes AI (so-called liver-type arginase) and AII (extrahepatic arginase) were identified by Western blotting in all studied tissues, however the amount of AI, as well as the expression of AI-mRNA were lower in HCC, in comparison with normal liver, and those of AII were significantly higher. Since HCC is arginine-dependent, and arginine is essential for cells growth, the decrease of AI may preserve this amino acid within tumor cells. Concurrently, the rise of AII can increase the level of polyamines, compounds crucial for cells proliferation. Thus, both arginase isoenzymes seem to participate in liver cancerogenesis.     

Enzymes of orithine metabolism in adult developing rat intestine

The levels of 11 enzymes, most of them involved in the metabolism of orithine, were measured in whole upper intestine, or in duodenum, small intestine and colon of adult rats. The developmental formations in small intestine of arginase, orithine aminotransferase, and orithine transcarbamylase were compared with those in liver. Changes with age (late gestation to adult) of the intestinal activities of pyrroline-5-carboxylate reductase, proline oxidase and glutamyl transpeptidase are also described.The results suggests that the proximal part of the intestine is well endowed with enzymes involved in the conversion of ornithine to proline as well as to citrulline. Fetal intestine is rich in proline oxidase and pyrroline-5-carboxylate reductase. The peak levels of ornithine aminotraferase found in intestine in the first 3 postnatal weeks were higher than seen in any other rat tissue.Some of the properties of arginase, ornithine aminotransferase and pyrroline-5-carboxylate reductase in small intestine were compared with those in liver. Isozymes of arginase in small intestine differed from those in liver; the kinetic properties of ornithine aminotransferase were similar in the two tissues. In intestine of 14-day-old rats, the orithine aminotransferase reaction was reversible, forming ornithine from pyrroline-5-carboxylate. The intestinal pyrroline-5-carboxylate reductase was cold-labile as was the hepatic enzyme in rat.     

CRP2 transcript expression pattern in embryonic chick limb

We are using the model of the developing mouse embryo to elucidate the pattern of arginase expression in mammalian cells in normal animals and in arginase I (AI) deficiency during development by digoxigenin-labeled RNA in situ hybridization. Our goal is to understand the regulation of these isozymes, with the expectation that this knowledge will help patients suffering from AI deficiency. We found that AI mRNA was widely and strongly expressed in the normal developing mouse embryo; in contrast, a relatively strong AII mRNA signal was found only in the intestine. In the AI knockout mouse embryo, no AII overexpression was found. These results indicated that arginases are needed in mouse embryonic development and AI is the principal form required. The strong AI expression in the peripheral nervous system suggests that the pathogenesis of the neurological retardation in AI deficiency may be conditioned by AI deficiency in the nervous system during embryonic development.     

The kinetics of inhibition of Vigna catjang cotyledon and buffalo liver arginase by L-proline and branched-chain amino acids

The effect of proline, isoleucine, leucine, valine, lysine and ornithine under standard physiological conditions, on purified Vigna catjang cotyledon and buffalo liver arginases was studied. The results showed that V. catjang cotyledon arginase is inhibited by proline at a lower concentration than buffalo liver arginase and the inhibition was found to be linear competitive for both enzymes. Buffalo liver arginase was more sensitive to inhibition by branched-chain amino acids than V. catjang cotyledon. Leucine, lysine, ornithine and valine are competitive inhibitors while isoleucine is a mixed type of inhibitor of liver arginase. We have also studied the effect of manganese concentration which acts as a cofactor and leads to activation of arginase. The optimum Mn^{2 +} concentration for Vigna catjang cotyledon arginase is 0.6 mM and liver arginase is 2.0 mM. The preincubation period required for liver arginase is 20 min at 55°C, the preincubation period and temperature required for activation of cotyledon arginase was found to be 8 min at 35°C. The function of cotyledon arginase in polyamine biosynthesis and a possible role of branched chain amino acids in hydrolysis of arginine in liver are discussed.     

Amino acid metabolizing enzymes in rat submaxillary gland, normal or neoplastic, and in pancreas.

The activities of 12 enzymes, many related to ornithine metabolism, were measured in rat submaxillary gland, submaxillary gland tumors and pancreas. In submaxillary gland, the activities of arginase, ornithine aminotransferase, pyrroline-5-carboxylate reductase and glutamine synthetase were high, but no ornithine transcarbamylase or proline oxidase could be detected. In the fetal submaxillary gland, arginase was at almost adult levels while ornithine aminotransferase reached 50% of its adult value postnatally. Submaxillary tumors deviated from their cognate tissue by lower levels of amino acid metabolizing enzymes and by high concentrations of thymidine kinase. In pancreas, none of the pyrroline-5-carboxylate metabolizing enzymes were as high as in either liver or submaxillary gland. The outstanding activities were those of gamma-glutamyl transpeptidase and glutamate dehydrogenase. Although arginase activities in submaxillary gland and pancreas were quantitatively similar, they differed qualitatively: submaxillary gland contained the same variant as liver while the pancreatic isozymes resembled those of other nonhepatic tissues.     

On the possible significance of the transamidination reaction in evolution

The origin of the various muscle phosphagens during evolution is considered in the context of the need to conserve ornithine for the synthesis of proline for connective tissue necessary for structural strength and flexibility and/or a complicated musculature. In each phosphagen, arginine is known to have contributed its amidine moiety thus maintaining the function of the phosphagen and setting free the proline precursor ornithine. Tissues from an earthworm, a starfish and a sea-squirt have been found to contain the enzymes arginase, ornithine aminotransferase and pyrroline-5-carboxylate reductase which are necessary to convert arginine to proline. For each of the animals studied analysis for the relevant free amino acids and for the characteristic amino acids (Pro, Oh-Pro, Oh-Lys, Gly) of collagen are presented. The amino acid composition of the diet of the sea-squirt Pyura stolonifera and of the starfish Coscinasterias calamaria is presented along with the level of the phosphagen kinases of the animals studied. The significance of the experimental results is discussed in connection with the importance of the transamidination reaction.     

Effects of congenital hyperammonemia on the cerebral and hepatic levels of the intermediates of energy metabolism in spf mice.

Sparse-fur (spf) mutant mice with X-linked ornithine transcarbamylase (OTC) deficiency were examined for hyperammonemia and its effect on energy metabolism. We compared the levels of ammonia, glutamine, glutamate and some of the intermediates of energy metabolism in the brain and liver of spf mice with those of control mice. In spf mice we observed significant increases in ammonia, glutamine, alpha-ketoglutarate and glucose with a significant decrease in ATP, glutamate and pyruvate in both brain and liver. The redox states of the brain and liver were also altered in spf mice. The results suggest that many of the metabolic alterations seen in spf mice could be due to the elevated ammonia levels. The spf mouse may, therefore, be an ideal model for the study of the neurotoxic effects of ammonia in chronic hyperammonemic syndromes.     

Induction of arginase I and II in bleomycin-induced fibrosis of mouse lung

Arginase, which hydrolyzes arginine to urea and ornithine, is a precursor for the synthesis of polyamines and proline, which is abundant in collagen. The supply of proline can be a crucial factor in the process of lung fibrosis. We investigated the induction of arginine metabolic enzymes in bleomycin-induced mouse lung fibrosis. Histological studies and quantification of lung hydroxyproline showed that lung fibrosis develops in up to 14 days after bleomycin treatment. Under these conditions, collagen I mRNA was induced gradually in up to 15 days, and the content of hydroxyproline reached a maximum at 10 days. Arginase I mRNA was undetectable before bleomycin treatment but was induced 5-10 days after this treatment. Arginase I protein was induced at 7 days and remained little changed for up to 10 days and decreased at 14 days. On the other hand, arginase II mRNA that was detectable before treatment was increased gradually for up to 10 days and decreased at 14 days. Arginase II protein began to increase at day 5, increased for up to 10 days, and was decreased at day 14. mRNAs for cationic amino acid transporter-2 and ornithine decarboxylase were induced in a manner similar to that seen with collagen I mRNA. Immunohistochemical analysis showed that arginase I is induced in macrophages, whereas arginase II is induced in various cell types, including macrophages and myofibroblasts, and roughly colocalizes with the collagen-specific chaperone heat shock protein 47. Our findings suggest that arginine metabolic enzymes play an important role in the development of lung fibrosis, at least in mice.     

Metabolism of ornithine in human gingival tissue

The behavior of two enzymes of the ornithine pathway, leading to the formation of proline and, eventually, of collagen, arginase and ornithine oxo-acid aminotransferase has been investigated in normal and inflamed gingival tissue. Both enzymatic activities show a statistically significant decrease in pathological samples as compared to normal ones. The data on arginase activity may be in agreement with the already documented low level of urea in pathological gingival fluid, while a decrease of the ornithine aminotransferase activity could be linked to the phenomenon of gingival retraction, i.e. the lack of complete regeneration of gingival tissue usually observed in chronically inflamed subjects, that would be reasonably parallel to a decreased proline/collagen synthesis.     

Polyamine synthesis from proline in the developing porcine placenta

Polyamines (putrescine, spermidine, and spermine) are essential for placental growth and angiogenesis. However, little is known about polyamine synthesis in the porcine placenta during conceptus development. The present study was conducted to test the hypothesis that arginine and proline are the major sources of ornithine for placental polyamine production in pigs. Placentae, amniotic fluid, and allantoic fluid were obtained from gilts on Days 20, 30, 35, 40, 45, 50, 60, 90, and 110 of the 114-day gestation (n = 6 per day). Placentae as well as amniotic and allantoic fluids were analyzed for arginase, proline oxidase, ornithine aminotransferase (OAT), ornithine decarboxylase (ODC), proline transport, concentrations of amino acids and polyamines, and polyamine synthesis using established radiochemical and chromatographic methods. Neither arginase activity nor conversion of arginine into polyamines was detected in the porcine placenta. In contrast, both proline and ornithine were converted into putrescine, spermidine, and spermine in placental tissue throughout pregnancy. The activities of proline oxidase, OAT, and ODC as well as proline transport, polyamine synthesis from proline, and polyamine concentrations increased markedly between Days 20 and 40 of gestation, declined between Days 40 and 90 of gestation, and remained at the reduced level through Day 110 of gestation. Proline oxidase and OAT, but not arginase, were present in allantoic and amniotic fluids for the production of ornithine (the immediate substrate for polyamine synthesis). The activities of these two enzymes as well as the concentrations of ornithine and total polyamines in fetal fluids were highest at Day 40 but lowest at Days 20, 90, and 110 of gestation. These results indicate that proline is the major amino acid for polyamine synthesis in the porcine placenta and that the activity of this synthetic pathway is maximal during early pregnancy, when placental growth is most rapid. Our novel findings provide a new base of information for future studies to define the role of proline in fetoplacental growth and development.     

Host liver changes in leukemic mice: effect of alpha-difluoromethyl ornithine, an inhibitor of polyamine synthesis

Elevated levels of polyamines and gamma-Glutamyl Transpeptidase were seen in the liver of P388 leukemia bearing mice. There was also increase in specific activity of beta-Hexosaminidase and the B/A isoenzyme ratio. Administration of a 2% solution of alpha-Difluoromethylornithine (DFMO) immediately after inoculation of tumor cells prevented increases in polyamines in liver, but did not have any effect on gamma-Glutamyl Transpeptidase. Almost normal ratio of beta-Hexosaminidase B to A was maintained during treatment. Endogenous ornithine level was not altered both in treated and untreated mice. However, proline level was elevated in liver of untreated mice and DFMO prevented this increase. Glutathione levels were altered both by leukemia and DFMO in the host liver. The effect of drug was more prominent in the early stages rather than during terminal stages of leukemic growth.     

Competitive interrelationships between lysine and arginine in rat liver under normal conditions and in experimental hyperammonemia

The effects of lysine administration on arginine and ornithine liver levels were studied in normal and urease-treated rats. L-Arginine injections produced a rise in liver arginine with a parallel increase in liver ornithine. Pretreatment with L-lysine resulted in an elevation in liver arginine. Administration of lysine to urease treated rats induced a significant increase in liver arginine content with a parallel drop in ornithine/arginine ratio. A similar decrease in ornithine/arginine ratio due to lysine administration was observed in animals, in which arginine and ornithine levels had been raised by loading with arginine. The mechanism of the lysine effect is most likely by inhibition of liver arginase activity $\text{in vivo}$.     

Expression of inducible nitric oxide synthase and enzymes of arginine metabolism in Fusarium kyushuense-exposed mouse lung.

Expression of inducible nitric oxide (NO) synthase (iNOS) and related enzymes of arginine metabolism in the mouse lung exposed to filamentous fungus Fusarium kyushuense was studied by RNA blot, immunoblot, and histological analyses. When mice were exposed intranasally to the fungi only once, no induction of iNOS mRNA was observed. However, when the animals were infected again 6 days after the first exposure, iNOS mRNA was induced, reached a maximum 12-24 h after the exposure, and decreased to an undetectable level at 48 h. mRNAs for cationic amino acid transporter-2 (CAT2) and argininosuccinate synthetase were induced gradually, reached a maximum at 24 h, and decreased at 48 h. Arginase II mRNA increased at 24 h and decreased markedly at 48 h. On the other hand, arginase I mRNA started to increase at 24 h and reached to a much higher level at 48 h. Ornithine decarboxylase and ornithine aminotransferase mRNAs were also induced. Immunoblot analysis showed that iNOS, argininosuccinate synthetase, and arginase I and II proteins were induced with similar kinetics as those of their respective mRNAs. In histological examination, fungal elements were observed in the bronchoalveolar lumen at 3-6 h, decreased at 12 h, and almost disappeared at 48 h. Small granuloma appeared 3 h after the infection and their size increased with time. These results suggest that NO is produced in the mouse lung in response to F. kyushuense exposure and that the NO production is regulated by CAT2, the citrulline-NO cycle, and arginase isoforms. Enhanced synthesis of polyamines and proline (and thus collagen) is also suggested.     

Widespread expression of arginase I in mouse tissues. Biochemical and physiological implications.

Arginase I (AI), the fifth and final enzyme of the urea cycle, detoxifies ammonia as part of the urea cycle. In previous studies from others, AI was not found in extrahepatic tissues except in primate blood cells, and its roles outside the urea cycle have not been well recognized. In this study we undertook an extensive analysis of arginase expression in postnatal mouse tissues by in situ hybridization (ISH) and RT-PCR. We also compared arginase expression patterns with those of ornithine decarboxylase (ODC) and ornithine aminotransferase (OAT). We found that, outside of liver, AI was expressed in many tissues and cells such as the salivary gland, esophagus, stomach, pancreas, thymus, leukocytes, skin, preputial gland, uterus and sympathetic ganglia. The expression was much wider than that of arginase II, which was highly expressed only in the intestine and kidney. Several co-localization patterns of AI, ODC, and OAT have been found: (a) AI was co-localized with ODC alone in some tissues; (b) AI was co-localized with both OAT and ODC in a few tissues; (c) AI was not co-localized with OAT alone in any of the tissues examined; and (d) AI was not co-localized with either ODC or OAT in some tissues. In contrast, AII was not co-localized with either ODC or OAT alone in any of the tissues studied, and co-localization of AII with ODC and OAT was found only in the small intestine. The co-localization patterns of arginase, ODC, and OAT suggested that AI plays different roles in different tissues. The main roles of AI are regulation of arginine concentration by degrading arginine and production of ornithine for polyamine biosynthesis, but AI may not be the principal enzyme for regulating glutamate biosynthesis in tissues and cells.     

Effect of hyperammonemia on brain amino acids in young and adult ferrets

Summary Effects of arginine deficiency and hyperammonemia on the brain concentrations of amino acids and urea cycle enzyme activities in young and adult ferrets were investigated. Only young ferrets developed hyperammonemia and encephalopathy immediately after consuming the arginine-free diet. Brain ornithine and citrulline concentrations in young ferrets fed arginine containing diet were significantly lower than those in adult ferrets. Compared to rats and other animals, young and adult ferrets had lower concentrations of brain glutamic acid and glutamine. Unlike in other species, brain glutamine was not elevated in young, hyperammonemic ferrets. Brain arginase and glutamate dehydrogenase activities were significantly increased in young ferrets fed arginine-free diet. Young ferrets provide a useful animal model for investigating the neurotoxicity of acute hyperammonemia.Abbreviations ACD Arginine-containing diet - AFD Arginine-free diet This work was presented, in part, at the annual meeting of the Midwest Society for Pediatric Research, Chicago, IL, 1991.