Tyrosine phosphorylation regulates manganese superoxide dismutase (MnSOD) RNA-binding protein activity and MnSOD protein expression.

All cells tested contain a cytosolic protein that binds to a defined region in the 3' untranslated region of manganese superoxide dismutase (MnSOD) RNA; both the MnSOD RNA-binding protein (MnSOD-BP) and the cis element are required for efficient translation of MnSOD RNA [Chung, D. J., Wright, A. E., and Clerch, L. B. (1997) Biochemistry 37, 16298-16306]. This study was designed to test the hypothesis that MnSOD-BP activity is regulated by phosphorylation. When cell extracts from whole rat lung or a rat lung fibroblast cell line, RFL-6, were treated in vitro with a protein tyrosine phosphatase, there was a 4-fold increase in MnSOD-BP activity indicating that MnSOD-BP activity was upregulated by tyrosine dephosphorylation. RFL-6 cells treated in cell culture with herbimycin A or genistein, inhibitors of protein tyrosine kinase, had significantly more MnSOD-BP activity than cells treated with diluent. In RFL-6 cells treated with herbimycin A, the increase in MnSOD-BP activity was associated with an increase in the level of MnSOD protein without a change in MnSOD mRNA concentration. We propose that the modulation of MnSOD protein expression by the tyrosine phosphorylation state of MnSOD-BP is a potential therapeutic target for increasing MnSOD activity during periods of oxidative stress.

Rat lung peroxiredoxins I and II are differentially regulated during development and by hyperoxia.

Peroxiredoxin I (Prx I) and peroxiredoxin II (Prx II) are found in abundance in the cytoplasm of cells and catalyze the reduction of hydrogen peroxide with the use of electrons provided by thioredoxin. Here we examined Prx I and Prx II expression in rat lung during perinatal development and in response to hyperoxia. Prx I protein increased during late gestation and after birth fell to adult levels; conversely, Prx I mRNA increased after birth. Prx II protein concentration was unchanged in the perinatal period, but Prx II mRNA increased after birth. In response to hyperoxia begun on postnatal day 4, there was no change in Prx II expression; however, Prx I mRNA, protein, and enzymatic activity increased significantly. These data show that 1) Prx I and Prx II are developmentally regulated at the level of translational efficiency and 2) Prx I, but not Prx II, is inducible and is upregulated during the late-gestational preparation for the oxidative stress experienced by the lung at birth and during exposure to hyperoxia in the neonatal period.

Retinoic acid receptor-beta: an endogenous inhibitor of the perinatal formation of pulmonary alveoli.

Pulmonary alveoli are formed, in part, by subdivision (septation) of the gas-exchange saccules of the immature lung. Septation is developmentally regulated, and failure to septate at the appropriate time is not followed by delayed spontaneous septation. We report retinoic acid receptor (RAR) beta knockout mice exhibit premature septation; in addition, they form alveoli twice as fast as wild-type mice during the period of septation but at the same rate as wild-type mice thereafter. Consistent with the perinatal effect of RARbeta knockout, RARbeta agonist treatment of newborn rats impairs septation. These results 1) identify RARbeta as the first recognized endogenous signaling that inhibits septation, 2) demonstrate premature onset of septation may be induced, and 3) show the molecular signaling regulating alveolus formation differs during and after the period of septation. Suppressing perinatal RARbeta signaling by RARbeta antagonists may offer a novel, nonsurgical, means of preventing, or remediating, failed septation in prematurely born children.

Post-transcriptional regulation of lung antioxidant enzyme gene expression.

It is an honor, and indeed fitting, to have a chapter on pulmonary oxygen toxicity included in a Festschrift for Dan Gilbert, whose contributions to the free radical theory of oxygen toxicity have been a catalyst to the last half-century of investigation in this field. There is cellular damage that results in pulmonary edema and even death if the increase in reactive oxygen species produced in the lung during exposure to hyperoxia is not counterbalanced by an increase in the cell's antioxidant defense systems. In this chapter experimental evidence will substantiate the importance of post-transcriptional regulation of antioxidant enzyme gene expression in animal models of pulmonary oxygen toxicity and tolerance to hyperoxia with special emphasis given to the role of manganese superoxide dismutase (MnSOD) synthesis, specific activity, and RNA half-life and to a proposed function of a MnSOD RNA-binding protein as a positive regulator in the control of translational efficiency.

A region in the 3' UTR of MnSOD RNA enhances translation of a heterologous RNA.

The studies reported in this paper were designed to test the hypothesis that a cis element located in the 3' UTR of manganese superoxide dismutase (MnSOD) RNA, designated MnSOD-response element (MnSOD-RE), is a translational enhancer in vivo. NIH/3T3 cells were transfected with a posttranscriptional reporter construct in which MnSOD-RE was placed 3' of the coding region of chloramphenicol acetyltransferase (CAT); this construct is designated CAT-RMS. Transient transfection of CAT-RMS did not change the concentration of CAT mRNA but increased CAT activity by approximately 400% compared to a control construct, CAT-V, which contains approximately the same size of non-MnSOD 3' UTR sequence. Transfection of CAT-RMS had no effect on endogenous MnSOD protein, mRNA, or MnSOD RNA-binding protein activity. Because of its ability to increase translation of a heterologous RNA, MnSOD-RE may be useful in designing expression vectors for in vitro expression systems and in vivo gene therapy.

Estrogen regulates angiotensin AT1 receptor expression via cytosolic proteins that bind to the 5' leader sequence of the receptor mRNA.

Two of the most highly recognized factors implicated in the pathogenesis of hypertension, atherosclerosis, congestive heart failure and associated cardiovascular disease are the renin angiotensin system (RAS) and estrogen. A major effect of estrogen results from its influence on the RAS. Beta-estradiol (E2) replacement in ovariectomized (OVX) rats significantly decreased type 1 angiotensin (AT1) receptor expression in the pituitary and adrenal, whereas it significantly increased receptor expression in the uterus when compared to OVX controls. Additional evidence demonstrated an important influence of estrogen on a recently discovered post-transcriptional mechanism for regulating expression of the AT1 receptor. This mechanism consists of cytosolic RNA binding proteins (BPs) that recognize the 5' leader sequence (5'LS) of the receptor mRNA. The activities of these 5'LS BPs were modulated by estrogen in an inverse manner to AT1 receptor regulation. Moreover, in vitro translation assays in wheat germ lysates suggested that the 5'LS BPs inhibited AT1 receptor translation. Our data therefore indicate that hormonal regulation of AT1 receptors involves modulation of 5'LS BPs by estrogen. These findings may in part account for the observed protective effects of estrogen on cardiovascular disease.

Regulation of lung manganese superoxide dismutase: species variation in response to lipopolysaccharide.

Lipopolysaccharide (LPS) treatment increases survival of rats, but not of mice, during hyperoxia. Manganese superoxide dismutase (Mn SOD) in the lung plays a critical role in LPS-induced tolerance to hyperoxia in rats. Therefore, we now compared the response of lung Mn SOD with treatment of mice and rats with LPS. LPS treatment of rats increased Mn SOD activity and protein concentration, did not change its specific activity, increased Mn SOD mRNA concentration 35-fold, and elevated Mn SOD synthesis 50% without changing general protein synthesis. LPS treatment of mice did not alter any of these parameters except for a 16-fold increase in Mn SOD mRNA concentration. Mn SOD translational efficiency (synthesis/mRNA concentration) was diminished 93% in rat lung and 76% in mouse lung by treatment with LPS. However, the absolute translational efficiency was twofold higher in lungs of LPS-treated rats than in lungs of LPS-treated mice. The failure of LPS to raise Mn SOD activity in mouse lungs is due, at least in part, to a smaller increase in Mn SOD mRNA and lower translational efficiency in LPS-treated mice than in LPS-treated rats.

Gene expression of cellular retinoid-binding proteins: modulation by retinoic acid and dexamethasone in postnatal rat lung.

In rats, septation of gas-exchange saccules occurs during the first 2 postnatal weeks; dexamethasone (DEX) treatment irreversibly impairs septation, and treatment with all-trans retinoic acid (RA) prevents the DEX-induced inhibition of septation. Cellular retinoic acid-binding protein I (CRABP I) and cellular retinol-binding protein I (CRBP I) are important modulators of the cellular metabolism of retinoids. In the present study, therefore, we measured the mRNA concentration of CRABP I and CRBP I in lungs of neonatal rats. In untreated rats, CRABP I and CRBP I mRNA peaked at postnatal d 8, indicating that CRABP I and CRBP I are developmentally regulated at least in part at a pretranslational level during lung septation. Daily treatment of 3- to 8-d-old rats with RA (500 microg/kg/d) had no effect on the level of CRABP I mRNA; treatment with DEX (0.25 microg/d) from d 4 to 8 caused a decrease in CRABP I mRNA that was not prevented by concomitant treatment with RA. These findings suggest that a decrease in CRABP I expression may be important in the DEX-induced block of septation but not in the prevention by RA of DEX-induced inhibition of septation. RA treatment caused an increase of CRBP I mRNA; conversely, treatment with DEX caused a decrease in CRBP I mRNA that was prevented by concomitant treatment with RA. These data suggest CRBP I may play a role in RA-induced septation, in the inhibition of septation caused by DEX, and in the ability of RA to prevent DEX-blocked septation.

The 3' untranslated region of manganese superoxide dismutase RNA contains a translational enhancer element.

A redox-sensitive protein that binds to the 3' untranslated region (UTR) of manganese superoxide dismutase (MnSOD) RNA has been described previously [Fazzone, H., Wangner, A., and Clerch, L. B. (1993) J. Clin. Invest. 92, 1278-1281; Chung, D. J., and Clerch, L. B. (1997) Am. J. Physiol. 16, L714-L719]. In the present study, cross-competition gel retardation and RNase H assays were used to identify a 41-base region located 111 bases downstream of the stop codon as the 3' UTR cis element involved in protein binding. The base sequence of this region is approximately 75% conserved among the 3' UTRs of rat, mouse, cow, and human MnSOD mRNAs at approximately the same distance downstream of the stop codon. The role of this protein-binding region in RNA translation was assessed in an in vitro rabbit reticulocyte lysate system. Translation of MnSOD RNA from which the 3' UTR element was deleted decreased 60% compared with translation of MnSOD RNA containing the 3' UTR cis element. In the presence of a specific competitor oligoribonucleotide that inhibits MnSOD RNA protein-binding activity, translation of MnSOD RNA containing the 3' UTR was decreased by 65%. Thus, both the cis element and RNA protein-binding activity were required for more efficient translation of the MnSOD. An analysis of ribosomal profiles suggests the MnSOD RNA-binding protein participates in the formation of the translation initiation complex. When MnSOD RNA-binding activity was inhibited, initiation complex formation was decreased by 50%. From the data obtained in this study, we propose that the 3' UTR cis element of MnSOD through its interaction with MnSOD RNA-binding protein may function as a translational enhancer.

Molecular mechanisms of antioxidant enzyme expression in lung during exposure to and recovery from hyperoxia.

Manganese superoxide dismutase (MnSOD) activity falls approximately 50% in lung during 48 h of exposure of adult rats to > 95% O2 (L. B. Clerch and D. Massaro. J. Clin. Invest. 91: 499-508, 1993). We now show that hyperoxia also decreased MnSOD activity in lungs of adult baboons, making the phenomenon potentially more important to humans. In rats, a decrease in lung MnSOD activity during an initial 48 h of exposure to > 95% O2 and its increase during an immediately subsequent 24 h in air were due to decreases and increases, respectively, in MnSOD specific activity and synthesis rate; the latter was due to altered translational efficiency. The concentration in the lung of copper-zinc superoxide dismutase mRNA, catalase mRNA, and glutathione peroxidase mRNA, unchanged during the initial 48 h of exposure to O2, rose approximately twofold during reexposure to O2 after 24 h in air. The demonstration that the fall in MnSOD activity is translationally and posttranslationally regulated during the initial exposure to hyperoxia suggests that gene transfer to increase MnSOD activity in hyperoxic lungs may also require therapy that maintains translational efficiency and MnSOD specific activity.

Kidney extracts from spontaneously hypertensive rats (SHR) have greater dopamine 1A receptor RNA-binding activity than extracts from normotensive Wistar-Kyoto (WKY) rats.

Rat kidney extracts contain a 52 kDa protein that binds to the 3' untranslated region of the dopamine 1A (D1A) receptor mRNA at a 243 base-long cis element starting at the stop codon and ending approximately 220 bases upstream of an AUUUA-rich region. The D1A receptor RNA-binding protein (D1A-BP) is redox-sensitive; free sulfhydryl groups on the protein are required for binding. Kidney extracts from SHR have significantly more D1A-BP activity than extracts from WKY rats. When kidney extracts were tested for binding to an 80-base RNA containing four AUUUA repeats, there was also greater binding activity in extracts from SHR. These increases are at least partly specific because there was no difference in catalase RNA-binding protein activity between the two rat strains. These data suggest D1A-BP and AUUUA-binding protein may play a role in posttranscriptional regulation of the D1A receptor in the hypertensive rat.

RNA in polysomes is an inhibitor of manganese superoxide dismutase RNA-binding protein activity.

A redox-sensitive protein in rat lung binds to the 3'-untranslated region (3'-UTR) of manganese superoxide dismutase (Mn-SOD) mRNA; the activity of this Mn-SOD RNA-binding protein (Mn-SOD-BP) is greater in 12,000-g supernatant fractions (S12) from neonates than in S12 from adults (H. Fazzone, A. Wangner, and L. B. Clerch. J. Clin. Invest. 92: 1278-1281, 1993). To determine the mechanism underlying this developmental difference, lung subcellular fractions were tested for their effect on Mn-SOD-BP activity. Protein in the 130,000-g supernatant (S130) of lung extracts bound the 3'-UTR. However, the developmental difference in binding was not present in S130. The 130,000-g pellet (P130) did not bind the 3'-UTR; rather, it contained an inhibitor of Mn-SOD-BP activity. Addition of P130 to S130 decreased RNA binding in a dose-dependent manner. Furthermore, adult P130 was a more potent inhibitor of RNA-binding activity than neonatal P130. These data indicate that the developmental difference in Mn-SOD-BP activity is due, in part, to an inhibitor in P130. Biochemical characterization revealed that the inhibitor is an RNA that may participate in the posttranscriptional control of Mn-SOD gene expression.

Increased manganese superoxide dismutase activity, protein, and mRNA levels and concurrent induction of tumor necrosis factor alpha in radiation-initiated Syrian hamster cells.

The levels of the antioxidant enzyme manganese superoxide dismutase (Mn-SOD) are frequently decreased in tumor cells and increased in normal cells upon treatment with ionizing radiation. We studied Mn-SOD at different stages during the neoplastic conversion of radiation-initiated Syrian hamster embryo HDR-3 cells. Mn-SOD activity and the concentration of Mn-SOD protein and mRNA increased gradually during the malignant transformation of HDR-3 cells after radiation exposure; fully neoplastic cells showed greater Mn-SOD levels than preneoplastic and normal 84-3 cells. Inhibitors of superoxide (SO) anion production (thenoyltrifluoroacetone and rotenone) decreased the concentration of Mn-SOD mRNA, raising the possibility that the generation of SO radicals participated in the upregulation of Mn-SOD in cells transformed by exposure to radiation. We observed an increase in the concentration of tumor necrosis factor alpha (TNF alpha) in HDR-3 cells relative to mock-irradiated cells. Together with the observation that TNF alpha stimulates the production of SO by mitochondria and increases the level of Mn-SOD mRNA in other experimental systems, our results suggest that as normal 84-3 cells undergo malignant transformation induced by ionizing radiation they produce TNF alpha, to which the cells respond by increasing the concentration of Mn-SOD mRNA and protein and the activity of the enzyme.

Pertussis toxin alters the concentration and turnover of manganese superoxide dismutase in rat lung.

Treatment of rats with pertussis toxin (PTX) decreases the activity of manganese superoxide dismutase (Mn-SOD) in the lung and results in oxygen toxicity in air (L. B. Clerch, G. Neithardt, U. Spencer, J. A. Melendez, G. D. Massaro, and D. Massaro. J. Clin. Invest. 93: 2482-2489, 1994). To examine aspects of the mechanism of the PTX-induced fall in Mn-SOD activity, we injected rats with PTX (50 micrograms/kg), killed the rats 72 h later, and measured the activity, concentration, specific activity, and turnover of Mn-SOD in the lung. Treatment with PTX caused an approximately 50% fall in Mn-SOD activity and Mn-SOD concentration but no change in Mn-SOD specific activity. PTX also caused an increase in Mn-SOD mRNA concentration, a fall in Mn-SOD synthesis, and an increase in the half-life of Mn-SOD and general proteins. We conclude the PTX-induced low concentration of Mn-SOD is due to a decrease of translational efficiency. We suggest that, under normoxic conditions, signal transduction via heterotrimeric guanine nucleotide binding proteins regulates the expression of Mn-SOD at the level of translation and Mn-SOD degradation.

Early divergent lung antioxidant enzyme expression in response to lipopolysaccharide.

Lipopolysaccharide (LPS), a component of the cell wall of Gram-negative bacteria, interacts with eukaryotic cells, causing changes in gene expression and a rapid increase in the formation of superoxide and H2O2. We now report that, within 6 h of treating rats with LPS, there was a divergent response in the lung of manganese superoxide dismutase (Mn-SOD) and catalase expression without a change in expression of copper-zinc superoxide dismutase or glutathione peroxidase. The activity and mRNA concentration of Mn-SOD increased during the time catalase mRNA concentration and activity decreased. These divergent changes and activation of nuclear factor-kappa B (NF-kappa B) were preceded by a fall, 1 h after LPS treatment, in the RNA binding activity of two redox-sensitive proteins: Mn-SOD RNA binding protein and catalase RNA binding protein. The rapid onset of these changes, the bacteriostatic properties of H2O2, and the signaling function of NF-kappa B suggest that the divergent expression of Mn-SOD and catalase is a coordinated component of the acute phase reaction to bacterial invasion.

Evidence that glutathione peroxidase RNA and manganese superoxide dismutase RNA bind the same protein.

Rat lung extract contains protein that binds to a cis element in the 3' untranslated region of glutathione peroxidase (GPx) mRNA; this region is located 200 bases downstream of the stop codon and 77 bases downstream of the selenocysteine insertion sequence. GPx mRNA-binding protein (GPx-BP) has the following characteristics in common with Mn superoxide dismutase mRNA-binding protein (MnSOD-BP): 1. RNA-binding activity is redox-sensitive; free sulfhydryl groups on the protein are required for binding. 2. RNA-binding activity is enriched in a 130,000 x g supernatant fraction and is inhibited by RNA in the polysomal fraction. 3. The MnSOD and GPx cis elements compete with each other for protein binding. 4. UV crosslinking studies with each probe reveal a [32P]-labeled protein of the same apparent molecular mass, 56 kDa. These observations provide evidence that MnSOD and GPx RNAs bind the same protein.

Dinucleotide-binding site of bovine liver catalase mimics a catalase mRNA-binding protein domain.

Rat lung contains protein that interacts with catalase mRNA to form specific, redox-sensitive RNA-protein complexes. The studies in this report were aimed at determining whether catalase protein binds its own RNA. We found that rat catalase RNA binds NADPH-depleted bovine liver catalase (Sigma) but does not bind bovine liver catalase in the presence of NADPH. Complex formation between liver catalase and catalase RNA is competitively eliminated by a CA dinucleotide repeat. These data suggest the dinucleotide binding site of catalase mimics or is homologous to a catalase RNA-binding protein domain. These findings support the hypothesis that a class of RNA-binding proteins may have evolved from (di)nucleotide binding enzymes (M. W. Hentze. Trends Biol. Sci. 19: 101, 1994). When bovine liver catalase from two other commercial sources (Calbiochem and Boehringer) was used, we could not detect binding to catalase RNA. We have not yet been able to identify the basis for this difference. Thus the physiological importance of our observation of the NADPH-sensitive protein binding to catalase RNA cannot be assessed at this time.

Lung manganese superoxide dismutase protein expression increases in the baboon model of bronchopulmonary dysplasia and is regulated at a posttranscriptional level.

The expression of lung manganese superoxide dismutase (MnSOD) mRNA and protein were examined in a premature baboon model of hyperoxia-induced bronchopulmonary dysplasia (BPD) and BPD superimposed with bacterial infection. When 140-d gestation baboons were delivered by hysterotomy and treated for 16 d with appropriate ventilatory and oxygen support (pro re nada controls), there was an increase in both MnSOD mRNA and protein compared with 140-d or 156-d gestation, nonventilated controls. The concentration of MnSOD protein was also elevated when the prematurely delivered baboons were ventilated with a high fraction of inspired O2 to produce a primate homolog of BPD, but there was a significant decrease in the concentration of MnSOD mRNA in BPD animals compared with pro re nada controls. In the lungs of premature baboons in which Escherichia coli infection was superimposed on hyperoxia-induced BPD, MnSOD mRNA was diminished to approximately the same extent as in BPD alone, but MnSOD protein was significantly increased compared with all other groups. Taken together these data indicate that the premature baboon is capable of mounting an antioxidant response and that increased MnSOD protein expression in BPD and BPD-infected premature baboons is regulated, at least in part, at a posttranscriptional level.

A 3' untranslated region of catalase mRNA composed of a stem-loop and dinucleotide repeat elements binds a 69-kDa redox-sensitive protein.

Rat lung extract contains protein that forms redox-sensitive, specific complexes with a 1130-base catalase cRNA (J. Biol. Chem. 267, 2853-2855, 1992). The present paper reports studies aimed at delimiting the site of protein binding on the RNA and characterizing the protein. A 240-base sequence was identified as the 3' untranslated region of catalase mRNA that binds lung protein in a redox-sensitive manner. Two elements within this 240-base region bind protein; one is a 36-base element that has a computer-predicted stem-loop secondary structure and the other is a CA dinucleotide repeat. Competition studies indicate that both elements are required for specific binding. Cross-competition experiments demonstrated that catalase RNA-binding protein (CAT-BP) is not the iron-responsive element-binding protein. Ultraviolet light-induced cross-linking and two-dimensional electrophoresis showed that CAT-BP has an apparent molecular mass of 69 kDa and appears to be composed of four isoforms. Competition studies indicate that stem-loop cis element is directly involved in binding CAT-BP. In addition to rat, the 69-kDa catalase RNA-binding protein is present in mouse and human fibroblast cell lines.

Pertussis toxin treatment alters manganese superoxide dismutase activity in lung. Evidence for lung oxygen toxicity in air-breathing rats.

Exposure of rats to hyperoxia or to treatment with endotoxin, increases lung manganese superoxide dismutase (MnSOD) gene expression. However, the paths by which these environmental signals are transduced into enhanced MnSOD gene expression are unknown. We now provide evidence that heterotrimeric G proteins are involved in the hyperoxia-induced increase in lung MnSOD gene expression but that pertussis toxin-sensitive G proteins are not involved in the endotoxin-induced elevation of lung MnSOD gene expression. We also show that treating rats with pertussis toxin decreased lung MnSOD activity approximately 50%. This decline in MnSOD activity occurred without a change in the lung activity of copper-zinc SOD, catalase, or glutathione peroxidase. In air-breathing rats, the pertussis toxin-induced decrease in MnSOD activity was associated with the development of lung edema, pleural effusion with a high concentration of protein, and biochemical evidence of lung oxygen toxicity. Compared to air-breathing rats, maintenance of pertussis toxin-treated rats under hypoxic or hyperoxic conditions respectively decreased or increased intrathoracic fluid. Endotoxin treatment elevated lung MnSOD activity and protected pertussis toxin-treated rats from an increase in intrathoracic fluid.