I. Morioka

Statin treatment increases formation of carbon monoxide and bilirubin in mice: a novel mechanism of in vivo antioxidant protection

Heme oxygenase (HO) has a central role in cellular antioxidant defences and vascular protection, and it may mediate pleiotropic actions of drugs used in cardiovascular therapy. We investigated whether long-term use of statins upregulates HO activity and increases carbon monoxide (CO) and bilirubin levels in vivo. Adult FvB mice were given atorvastatin or rosuvastatin (5 mg/kg) daily by i.p. injections for 1, 2, or 3 weeks. HO activity, tissue CO, bilirubin, and antioxidant levels, total plasma bilirubin, and carboxyhemoglobin (COHb) were measured. Fold changes in heart HO activity significantly increased after 1, 2, and 3 weeks of atorvastatin (1.24 +/- 0.06 (p < or = 0.05); 1.29 +/- 0.26 (p < or = 0.03); 1.33 +/- 0.08 (p < 0.01), respectively) and 2 and 3 weeks of rosuvastatin (1.23 +/- 0.20 (p < or = 0.03); 1.63 +/- 0.42 (p < 0.01), respectively). Heart tissue CO and COHb levels also increased after 3 weeks with atorvastatin (1.30 +/- 0.24 (p < or = 0.05); 1.92 +/- 0.17 (p < or = 0.001), respectively) and rosuvastatin (1.47 +/- 0.13 (p < or = 0.004); 1.63 +/- 0.12 (p < or = 0.001), respectively). Significant increases in heart antioxidant levels were observed after statin treatment and corroborated by heart bilirubin content elevations. Antioxidant level increases were abolished by treatment with an HO inhibitor. These findings suggest that the induction of HO and the production of its products, CO and bilirubin, may be a mechanism by which statins exert antioxidant actions and confer cardioprotection in vivo.

Systemic Effects of Orally-Administered Zinc and Tin (IV) Metalloporphyrins on Heme Oxygenase Expression in Mice

Some metalloporphyrins (Mps) inhibit heme oxygenase (HO), the rate-limiting enzyme in the production of bilirubin, and are potential compounds for the treatment of neonatal jaundice. We studied the safety and efficacy of Mps following oral administration. Adult HO-1-luc reporter mice were administered 30 μmol/kg body weight of tin mesoporphyrin (SnMP), zinc bis glycol deuteroporphyrin (ZnBG), or zinc protoporphyrin (ZnPP), or vehicle by oral gavage. Bilirubin production was measured as total body carbon monoxide (CO) excretion (VeCO). HO activity was quantitated via CO measurements by gas chromatography. HO-1 protein was determined by Western blot. HO-1 transcription levels were assessed by in vivo bioluminescence imaging. A significant 28% decrease in bilirubin production occurred within 3h of SnMP treatment and persisted beyond 48h. Bilirubin production decreased 15% and 9% by 3h after administration of ZnBG and ZnPP, respectively, but returned to baseline within 48h. Maximal inhibition of liver, spleen, and intestine HO activity was seen at 3h with inhibitory effects decreasing in the order: SnMP ≥ ZnBG ≥ ZnPP. After SnMP treatment, HO-1 transcription increased 5.7-fold after 24h. Furthermore, liver and spleen HO-1 protein significantly increased 3.7- and 2.0-fold, respectively, after 24h. HO-1 transcription and protein were not affected in ZnBG- or ZnPP-treated mice. We conclude that the three Mps are absorbed at different rates in the mouse and affect bilirubin production and HO-1 expression in a tissue- and time-dependent manner.

The effectiveness of oral tin mesoporphyrin prophylaxis in reducing bilirubin production after an oral heme load in a transgenic mouse model.

Background: Neonatal jaundice is commonly encountered and rarely associated with morbidity and mortality. Nonetheless, infants with glucose-6-phosphate dehydrogenase deficiency often have hemolysis (a heme load) caused by an environmental oxidant trigger, thus increasing their risk for serious morbidity. The use of tin mesoporphyrin (SnMP) has been proposed for interdicting the development of severe hyperbilirubinemia in a variety of conditions. Objectives: We studied the in vivo effects of prophylactic oral SnMP on heme oxygenase (HO) activity and bilirubin production, as indexed by the excretion rate of carbon monoxide (VeCO), following a subsequent oral heme load. Methods: Adult mice were exposed serially to heme and assessed for in vivo bilirubin production rates, HO-1 transcription and protein, and HO activity. The effect of prophylaxis with a single oral dose of SnMP prior to an oral heme load was assessed by measuring VeCOand tissue HO activities. Results: After serial heme exposures, VeCO, HO-1 transcription and protein, and liver and spleen HO activities increased incrementally. After pretreatment with oral SnMP, bilirubin production decreased in response to an oral heme load. Also, heme-mediated increases in liver, spleen, and intestine HO activities were significantly dampened. Conclusions: A single oral dose of SnMP results in durable inhibition of bilirubin production and HO activity for at least 24 h in a mouse model of oral heme loading. Further studies are needed to fully elucidate the duration of this protection against hyperbilirubinemia due to a delayed heme load and any long-term consequences of prophylaxis with SnMP on HO-1 transcription and HO-1 protein.

280 DOSE-DEPENDENT EFFECTS OF TIN MESOPORPHYRIN ON HEME OXYGENASE ACTIVITY INHIBITION IN NEWBORN MICE.

Tin mesoporphyrin (SnMP), a synthetic heme analog, potently inhibits heme oxygenase (HO), the rate-limiting enzyme in the production of bilirubin, and thus is being studied for use in the chemoprevention of jaundice. We have previously shown that 30 μM SnMP is orally absorbed by adult mice and results in significant inhibition of HO activity in liver, spleen, and brain. Because we are interested in the use of these compounds for treating neonatal jaundice, we determined the efficacy of various doses of SnMP on inhibiting HO enzyme activity in newborn mice. SnMP at doses of 30, 7.5, 3.75, or 1 μM or vehicle was administered by direct gastric injection to 1-week-old HO-1-luc transgenic mice and then monitored for 7 d. After 3 h, 24 h, and 7 d following treatment with SnMP, mice were sacrificed and liver, spleen, and brain harvested. Tissues were sonicated in 9 parts phosphate buffer and then incubated in the dark for 15 min at 37°C with 50 μM methemalbumin and 1.5 mM NADPH in septum-sealed vials. Blank reactions contained no NADPH. Vial headspace CO was quantitated by gas chromatography. HO activity left (% of control) in tissues at 3 h, 24 h, and 7 d after SnMP treatment is as follows: (p ≤ .05 in bold compared to control levels): We found that SnMP significantly inhibited liver HO activity at all doses and times. Spleen HO activity was inhibited at 30, 7.5, and 3.75 μM SnMP for all times. Brain HO activity was inhibited by 30 μM SnMP for all times, and only after 3 h of treatment at doses of 7.5 and 3.75 μM. Furthermore, no inhibition in spleen and brain HO activity was observed at 1 μM SnMP. We therefore conclude that SnMP is not only orally absorbed by the neonatal mouse, but also affects tissue HO activity in a dose- and time-dependent manner. Use of SnMP doses ≤ 7.5 μM is effective in long-term inhibition of liver and spleen HO activity without affecting brain HO activity. This work was study was supported by National Institutes of Health grants #HL68703 and HD58013.

238 METALLOPORPHYRIN INHIBITION OF IN VITRO MOUSE HEME OXYGENASE ISOZYME ACTIVITY.

Most of the CO excreted by mammals is a byproduct of the degradation of heme to form bilirubin. The rate-limiting enzyme in this catabolic pathway is heme oxygenase (HO), which can be inhibited by synthetic metalloporphyrin derivatives of heme (Mps). Therefore, Mps are being studied for use in the suppression of hyperbilirubinemia. There are two forms of HO which are present in differing amounts and proportions in various tissues. In the mouse brain, the predominant isozyme is the constitutive HO-2, whereas, in the mouse spleen, it is the inducible HO-1. We studied the effects of 4 Mps [tin mesoporphyrin (SnMP), chromium mesoporphyrin (CrMP), zinc protoporphyrin (ZnPP), and zinc deuteroporphyrin bis glycol (ZnBG)] on these isozymes using the spleen and brain as sources of HO-1 and HO-2, respectively. HO activity was determined by measuring CO produced in the vial headspace by gas chromatography after tissue sonicates were incubated in the dark with varying concentrations of each porphyrin (n ≥ 5). The concentration of each porphyrin needed to inhibit CO production by 50% (I50) was then determined. Statistical significance was calculated using a two-tailed unpaired t-test (p ≤ .05). All 4 Mps showed greater inhibitory potency (lower I50) towards HO-2 (brain) versus HO-1 (spleen). Both SnMP and ZnPP were found to exhibit the greatest specificity between isozymes (≈2x). The same order of potency was observed in both isozymes with SnMP showing the highest and ZnPP the lowest. CrMP lacked significant differentiation in I50 between isozymes (*) and when compared to ZnBG for HO-1 (†). All other comparisons were significant. Of note, the I50s of CrMP and ZnPP for both isozymes were significantly different from that previously found in the rat. We conclude that there is a wide spectrum of inhibitory potency and a fair level of selectivity among Mps with the constitutive HO-2 being more susceptible to inhibition than HO-1. These findings should be considered when comparing studies of Mps performed in rat and mouse models and in developing selective inhibitors of HO-1 for therapeutic purposes.

INHIBITION OF HEME OXYGENASE ACTIVITY FOLLOWING REPEATED HEME LOADS BY TIN MESOPORPHYRIN IN NEWBORN MICE.: 73

Heme oxygenase (HO) is the rate-limiting enzyme in the degradation of heme and produces equimolar amounts of carbon monoxide (CO), iron, and bilirubin. Because excess bilirubin production due to increased heme loads (ie, hemolysis) can lead to the development of jaundice, the use of competitive HO inhibitors, such as tin mesoporphyrin (SnMP), may be an ideal treatment strategy for the prevention of pathologic neonatal jaundice (or hyperbilirubinemia). Our objective was to determine the effect of SnMP on HO activity following heme loading in newborn mice. Seven-day-old mice were administered 30 μM hemin (H) by subcutaneous (SQ) injection on day 1. On the second day, mice were administered with 30 or 3.75 μM SnMP (SnMP30 or SnMP3.75, respectively) or vehicle (V) by direct intragastric injections. On day 3, V or a second H load was injected SQ again. On the fourth day, mice were sacrificed and liver, spleen, and brain harvested for quantitation of HO activity using gas chromatography. %HO activity compared to the H-V-V Group (mean ± SD) for each group (n = 4-6 per group) is shown in the table below: In the H-SnMP30-V group, HO activity was significantly inhibited in all tissues. In the H-SnMP3.75-V group, liver and spleen were inhibited at the same level as that of the H-SnMP30-V group, but without inhibition of brain HO activity. Following a second heme load (H-V-H group), HO activity increased 36%, 27%, and 15% in liver, spleen, and brain, respectively. In the H-SnMP30-H group, HO activity was still significantly inhibited in all tissues. In contrast, for the H-SnMP3.75-H group, HO activity was no longer inhibited. We conclude the administration of 30 and 3.75 μM SnMP was similarly potent after a single heme load. However, after a second heme load, only SnMP at a dose of 30 μM was effective in inhibiting not only liver HO activity but also that of the brain. These findings show that low doses of SnMP do not exert long-term inhibition of HO activity in the context of repeated heme loads. This work was supported by National Institutes of Health grants #HL68703 and HD58013.

CHRONIC ADMINISTRATION OF STATINS ON IN VIVO HEME OXYGENASE EXPRESSION: A NOVEL MECHANISM OF ANTIOXIDANT PROTECTION.: 72

Heme oxygenase (HO) is the rate-limiting enzyme in degrading heme to form bilirubin and thus serves as an ideal therapeutic target for preventing neonatal jaundice. Understanding the regulatory pathways of HO is crucial for developing strategies for this disorder. We have previously shown that statins, inhibitors of 3-hydroxyl-3-methylglutaryl coenzyme A (HMG-CoA) reductase, selectively induce the HO-1 isozyme in vivo following a large single oral dose. The objective in this study was to investigate the effects of chronic administration of statins on in vivo induction of HO-1 expression. Adult mice were given atorvastatin or rosuvastatin (5 mg/kg body weight) or vehicle (control) by daily intraperitoneal injections for 1, 2, or 3 weeks. At each time point, mice were sacrificed. HO activity and tissue carbon monoxide (CO) levels in the liver, lung, brain, and heart were measured and then expressed as fold change from control levels. Total serum bilirubin (TSB) and carboxyhemoglobin (COHb) levels were measured after 3 weeks of statin treatment. Heart HO activity significantly increased after 2 and 3 weeks of treatment with atorvastatin (1.24 ± 0.27 [n = 7, p = .04] and 1.27 ± 0.07 [n = 4, p = .0009], respectively) and rosuvastatin (1.20 ± 0.13 [n = 5, p = .02] and 1.48 ± 0.27 [n = 5, p = .01], respectively). After 3 weeks of treatment, similar increases in heart tissue CO and COHb levels with atorvastatin (1.30 ± 0.24 [n = 5, p = .023] and 1.92 ± 0.17 [n = 5, p = .001], respectively) and rosuvastatin (1.47 ± 0.13 [n = 3, p = .01] and 1.63 ± 0.12 [n = 5, p = .001], respectively) were also found. Significant increases in antioxidant capacity in the heart in both atorvastatin- (n = 4) and rosuvastatin- (n = 5) treaqted mice were also observed and were paralleled by elevations in TSB of 1.49 ± 0.15-fold (p = .001) and 1.14 ± 0.14-fold (p = .02) from baseline levels, respectively. We conclude that the induction of HO activity by chronic treatment with atorvastatin and rosuvastatin is primarily in the heart. Subsequent production of CO and bilirubin, bioactive metabolic products of HO-1, is the major mechanism by which statins exert their antioxidant actions on the cardiovascular system and confer cardiac cytoprotection. This work was study was supported in part by National Institutes of Health grant #HD58013 and AstraZeneca grant #1RUSROSU0190.

239 STATINS AS TISSUE-SPECIFIC INDUCERS OF IN VIVO HEME OXYGENASE EXPRESSION.:

Heme oxygenase (HO) is the rate-limiting enzyme in degrading heme to form bilirubin, and thus serves as an ideal therapeutic target for preventing neonatal jaundice. Understanding the regulatory pathways of HO is crucial for developing strategies for this disorder. Statins, inhibitors of 3-hydroxyl-3-methylglutaryl coenzyme A (HMG-CoA) reductase, have been reported to induce HO-1. Our objective was to characterize the effects of statins with different structures and lipophilicities on in vivo induction of HO-1 expression. Adult FVB mice (6-8 weeks old) were orally administered 100 mg/kg body weight of simvastatin, lovastatin, atorvastatin, or rosuvastatin. 24 hours after treatment, mice were sacrificed, and tissues were harvested, sonicated, and then analyzed for HO activity [carbon monoxide (CO) formation] by gas chromatography and HO proteins by Western blot. HO activity was calculated as pmol CO produced/hour/mg fresh weight. HO protein levels were measured by densitometry. All values were then expressed as mean ± SD percent of control levels (*p ≤ .03) as follows: Simvastatin did not affect brain HO activity. Liver HO activity was not induced by atorvastatin or rosuvastatin. Lovastatin increased HO activity in all tissues. Interestingly, all statins significantly increased HO activity in the heart and lung. When HO protein levels were measured in atorvastatin-treated mice, corresponding significant increases in HO-1 protein in the heart (132 ± 39%; n = 6) and lung (136 ± 13%; n = 4) were found, whereas no changes in HO-2 protein levels were observed in either tissue. We conclude that the induction of HO activity is statin- as well as tissue-specific. However, further studies are needed to elucidate the exact mechanism by which statins differentially induce HO expression in various organs. This work was study was supported in part by National Institutes of Health grant #HD58013.

269 ORAL ADMINISTRATION OF METALLOPORPHYRINS AND ITS SYSTEMIC EFFECTS ON HEME OXYGENASE EXPRESSION

Metalloporphyrins (Mps) inhibit heme oxygenase (HO), the rate-limiting enzyme in the production of bilirubin, and are potential compounds for use in the chemoprevention of neonatal jaundice. . Our object was to monitor the effects of selected Mps on in vivo heme degradation and HO-1 gene transcription, and in vitro HO enzyme activity and protein levels after oral administration. 30 μM of tin mesoporphyrin (SnMP), zinc bis glycol porphyrin (ZnBG), zinc protoporphyrin (ZnPP), or vehicle was orally administered to adult HO-1-luc transgenic mice. In vivo heme degradation rates were measured as total body CO excretion levels (VeCO) by GC. In vivo HO-1 transcription levels were assessed by bioluminescence imaging (BLI). In vitro HO activity was quantitated via CO measurements by GC. HO-1 protein levels were determined by Western blot. VeCO decreased 28% within 3 h of SnMP treatment and lasted beyond 48 h. Administration of ZnBG and ZnPP decreased VeCO 15% and 9%, respectively, by 3 h. HO-1 transcription increased only in SnMP-treated mice (6.5-fold) after 24 h. % inhibition of tissue HO activities are shown below: (Table) HO activity was maximally inhibited in liver, spleen and intestine within 3 h, with the greatest inhibition in SnMP mice, then in ZnBG, with the least in ZnPP mice. Furthermore, in SnMP mice, liver and spleen HO-1 protein levels significantly increased 3.5- and 2.0-fold, respectively, after 24 h. HO-1 protein was minimally affected in ZnBG- or ZnPP-mice. These results are in contrast to our previous findings in neonatal rat pups, where ZnPP was not orally absorbable. Therefore, we conclude that orally administered Mps are absorbed variably and affect HO expression in a species- as well as in a tissue- and time-dependent manner.

267 ATTENUATION OF HEME INDUCTION FOLLOWING EXPOSURE TO TIN MESOPORPHYRIN IN MICE

Heme oxygenase (HO) is the rate-limiting enzyme in the degradation of heme and produces equimolar amounts of carbon monoxide (CO), iron, and bilirubin. HO-1 can be induced by several conditions and compounds, including heme. Tin mesoporphyrin (SnMP), a competitive inhibitor of HO, is currently in clinical trials for use in the prevention of neonatal jaundice. Our objective was to determine the effect of SnMP on heme induction following heme exposure. In vivo heme degradation rates were measured by quantitating total body CO excretion (VeCO) in adult male mice (6-8 wk old) by gas chromatography (GC). After baseline VeCO were established, mice were administered 30-μM of hemin or SnMP or vehicle (VEH) by oral gavage (1st exposure). After 3 h of VeCO monitoring, mice were gavaged with 30-μM hemin or VEH (2nd exposure). VeCO levels were then followed until peak levels were reached. 24 h later, VeCO and liver, spleen, and intestine HO activity were quantitated by GC. All values are expressed as fold change (mean±SD) from control (*p≤0.05). (Table) In the VEH-heme group, VeCO and liver HO activity significantly increased 2.8- and 1.8-fold following heme treatment, respectively. In the heme-heme group, VeCO increased 3.8- and 7.5-fold and following the 1st and 2nd heme exposure. Spleen and liver HO activities were 1.9- and 1.8-fold higher than controls, respectively. In mice pre-treated with SnMP, VeCO was 20% inhibited and did not increase following heme treatment. Spleen HO activity was inhibited by 40%. We conclude that successive exposures to heme does not alter VeCO rates, and that SnMP treatment attenuates heme-mediated increases in VeCO and HO activity. These findings suggest that SnMP is a potent inhibitor of HO with an effect lasting at least 24 h, which may be advantageous in the context of jaundice caused by an ongoing hemolytic process.