Chem. Res. Toxicol. 2000, 13, 1051-1055
1051
Endotoxin (Lipopolysaccharide)-Induced Nitric Oxide Production in 2,3,7,8-Tetrachlorodibenzo-p-dioxin-Treated Fischer Rats: Detection of Nitrosyl Hemoproteins by EPR Spectroscopy Richard E. Glover,† Dori R. Germolec,‡ Rachel Patterson,‡ Nigel J. Walker,§ George W. Lucier,§ and Ronald P. Mason*,† Laboratory of Pharmacology and Chemistry, Laboratory of Toxicology, and Laboratory of Computational Biology and Risk Analysis, National Institute of Environmental Health Science, National Institutes of Health, P.O. Box 12233, Research Triangle Park, North Carolina 27709 Received June 19, 2000
Electron paramagnetic resonance (EPR) spectroscopy was used to study the effects of 2,3,7,8tetrachlorodibenzo-p-dioxin (TCDD) on endotoxin (lipopolysaccharide)-induced nitric oxide (NO) production in Fischer rats. We found that rats treated with 50 µg/kg TCDD had increased sensitivity to endotoxin, resulting in an approximately 2-fold increase in the level of NO production detected as nitrosylhemoglobin (HbNO) in venous blood. At lower concentrations (e5 µg/kg), TCDD did not affect the endotoxin-induced NO production. The TNF-R serum concentration was found to parallel that of NO. TCDD alone did not induce the production of detectable HbNO or TNF-R. We found that TCDD induced a dose-dependent increase in the EPR signal intensity of (Fe3+) low-spin methemoprotein complexes found in the liver and kidney. These species with EPR resonance at g ) 2.43, 2.26, and 1.92 are attributed to low-spin Fe3+ in cytochromes P450 and P420. Our data confirm previous studies that have shown that TCDD induces a dose-dependent increase in the production of some cytochrome P450 enzymes. However, in rats that were subsequently challenged with endotoxin, a smaller increase in the EPR intensity of these species was observed. The decrease in the low-spin Fe3+ cytochrome P450 EPR signal in endotoxin-challenged rats could be due to one or more of the following occurring: (1) cytochrome destruction, (2) reduction of the ferric to the ESR-silent ferrous oxidation state of cytochromes by nitric oxide, and/or (3) formation of ferrous nitrosyl cytochrome complexes that contribute, in part, to the characteristic five-coordinate nitrosyl hemoprotein triplet also observed in these tissues. Since low concentrations of endotoxin can leak from the gut lumen into the systemic circulation, this investigation explores the possibility that endotoxin interaction with TCDD may be, in part, responsible for the effects of TCDD observed in these tissues.
Introduction Polychlorinated dibenzo-p-dioxins (PCDDs)1 and polychlorinated dibenzofurans (PCDFs) have received increasing attention as toxic environmental pollutants. 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD), which serves as a prototype for PCDDs and PCDFs, is an undesired byproduct formed during the incineration of chlorinecontaining wastes, synthesis of chlorophenol-based products, and chlorine bleaching of paper pulp. TCDD is perhaps one of the most toxic man-made chemicals to * To whom correspondence should be addressed: Laboratory of Pharmacology and Chemistry, National Institute of Environmental Health Sciences, National Institutes of Health, P.O. Box 12233, Research Triangle Park, NC 27709. Telephone: (919) 541-3910. Fax: (919) 541-1043. E-mail:
[email protected]. † Laboratory of Pharmacology and Chemistry. ‡ Laboratory of Toxicology. § Laboratory of Computational Biology and Risk Analysis. 1 Abbreviations: PCDD, polychlorinated dibenzo-p-dioxin; PCDF, polychlorinated dibenzofuran; TCDD, 2,3,7,8-tetrachlorodibenzo-pdioxin; EPR, electron paramagnetic resonance; LPS, lipopolysaccharide; HbNO, nitrosylhemoglobin.
date. The exact mechanism(s) by which TCDD induces its toxicity remains to be fully elucidated. A potentially important determination of susceptibility to the harmful effects of TCDD may be the presence or absence of underlying inflammation that may influence the pathogenic outcome of chemical exposure (1). The contribution of endogenous endotoxin in TCDD toxicity has been proposed previously. Clark and Taylor demonstrated that endotoxin was a contributing factor in TCDD-induced cachexia and lethality (2). They showed that a single dose of 350 µg/kg TCDD induced a significant decrease in body weight in endotoxin-responsive C57BL/6 mice, while the same dose of TCDD did not affect the body weights of the endotoxin-nonresponsive C3H/Hej mice (2). TCDD exposure can either increase or decrease sensitivity to endotoxins, depending on the species used as a model for toxicity. The cytokines TNF-R and IL-1 are important mediators in host defense, and are implicated in the immunological effects of TCDD in rats and mice (3, 4). High levels of these cytokines can suppress appetite and gluconeogenesis, leading to the hypoglyce-
10.1021/tx000128u CCC: $19.00 © 2000 American Chemical Society Published on Web 09/22/2000
1052
Chem. Res. Toxicol., Vol. 13, No. 10, 2000
Glover et al.
mia and the wasting syndrome associated with TCDD exposure. As yet, no increase in the level of TNF-R or IL-1 has been reported in TCDD-challenged animals without coexposure to endotoxin or another antigen (3). However, Fan et al. (3) have shown that a single high dose of TCDD (60 µg/kg) alone was able to significantly increase the level of expression of mRNA for both TNF-R and IL-1 in livers of Sprague-Dawley rats. It is thought that low concentrations of endotoxin that leak from the gut lumen into the systemic circulation may contribute to the pathology of TCDD toxicity. As such, endotoxin is used in models of TCDD-induced disease. In this study, we used electron paramagnetic resonance (EPR) spectroscopy to study NO production and monitor changes in cytochrome P450 and P420 concentrations induced by both TCDD and endotoxin.
Materials and Methods The National Institute of Environmental Health Sciences Institute Review Board approved this study. Animal Preparations. Female F344 rats (Charles River Laboratories, Raleigh, NC), 8 weeks of age, received a single dose of 0, 0.5, 5, or 50 µg/kg TCDD (Accustandard, New Haven, CT) in corn oil vehicle via oral gavage. Seven days later, rats were injected intravenously with 0.1 mg/kg lipopolysaccharide (LPS) (endotoxin; Escherichia coli Serotype 0127:B8, Sigma, St. Louis, MO). Four hours postinjection, the rats were anesthetized by CO2 narcosis, and approximately 400 µL of venous blood was collected via retro-orbital bleeding. Blood was transferred into EPR quartz tubes (3 mm i.d. × 4 mm o.d.), frozen immediately in liquid nitrogen, and stored at -70 °C prior to EPR analysis. Serum samples for cytokine analysis were collected via cardiac puncture and stored at -70 °C prior to ELISA analysis. Animals were then euthanized via decapitation per NIH Guidelines for the Humane Sacrifice of Laboratory Rodents. The median lobe of the liver was removed and a portion stored in 10% neutral buffered formalin for histological analysis. The remaining liver tissue and excised kidneys were quickly frozen in liquid nitrogen and stored separately at -80 °C for later analysis. Cytokine Analysis by the ELISA. Serum from each individual rat was assayed for circulating TNF-R or IL-6 by ELISA (BioSource, Camarillo, CA). Briefly, eight-well strips coated with the appropriate anti-cytokine antibody were incubated with 100 µL of serum or standard and 100 µL of diluent buffer for 2 h at 37 °C. Following incubation, the plates were washed four times in PBS-Tween and incubated for 1 h at 37 °C with 100 µL of matching biotinylated antibody. The plates were washed as described above, incubated with 100 µL of streptavidin-HRP for 45 min at 37 °C, and washed again. Color development was initiated by the addition of 100 µL of 3,3′,5,5′tetramethylbenzidine. The plates were incubated for 20 min, and the reaction was stopped by the addition of 100 µL of stop solution. The absorbance was read at 450 nm using a Dynex Technologies (Chantilly, VA) microplate reader. Kidney and Liver Preparation. The liver or kidney was forced into a 1 mL Monoject syringe (Sherwood Medical, St. Louis, MO) to a volume of 0.5 mL. This was then immediately frozen in liquid N2 to form an icicle. The tissue was removed from the syringe by cutting the end of the syringe, then partially thawing the surface and carefully plunging the icicle into fresh liquid N2 using the syringe plunger. The tissues were directly transferred into the EPR finger dewar containing liquid N2 ready for EPR analysis. EPR Analysis. All EPR measurements were carried out on a Bruker ESP300 spectrometer operated at 9.5 GHz with 100 kHZ modulation frequency at liquid nitrogen temperatures with samples held in a quartz finger dewar. Unless otherwise stated, typical spectrometer settings were as follows: power, 20 mW; modulation amplitude, 4 G; time constant, 1.31 s; scan time,
Figure 1. Typical EPR spectra obtained from venous blood of rats treated with (A) vehicle, (B) 50 µg/kg TCDD, (C) 0.1 mg/kg LPS, and (D) 50 µg/kg TCDD and 0.1 mg/kg LPS as described in Materials and Methods. The characteristic HbNO spectrum is shown in spectrum C; a weak triplet indicative of HbNO in spectra A and B around g ) 2.011 can be seen. The broad component at g ≈ 2.06 in spectra A and B can be attributed to Cu2+ from the plasma protein ceruloplasmin. 1342 s; and scan range, 400 G. The EPR signal from Cr3+ in MgO was used as a g value marker (g ) 1.9800 ( 0.0006) (5).
Results The EPR spectrum of venous blood obtained from female Fischer rats challenged with various doses of TCDD followed by 0.1 mg/kg LPS exhibited the characteristic signal of the HbNO complex indicative of NO production in venous blood (Figure 1) (6). As is known, administration of 0.1 mg/kg LPS results in the formation of HbNO as demonstrated in Figures 1 and 2a (7). In contrast, we found that TCDD alone lacked the ability to cause an increase in the level of HbNO over the range of doses studied. In fact, a small decrease in the relative intensity of the HbNO signal, though not significant, could be observed up to 5 µg/kg TCDD (Figure 2A). However, at a relatively high pretreatment dose of 50 µg/kg TCDD, a synergistic effect could be observed with LPS, resulting in an approximately 2-fold increase in the measured level of HbNO as compared to LPS alone. In addition, the concentration of TNF-R in the serum paralleled that of HbNO in its response to both LPS and pretreatment with TCDD (Figure 2B). The EPR spectra of livers and kidneys from normal rats shown in panels a and b of Figure 3 both exhibited signals of a free radical of unknown origin at g ≈ 2.0 and a reduced iron-sulfur complex at g ≈ 1.93. Some of the six Mn2+ peaks can also be seen at g ≈ 2.10, 2.04, 1.98, and 1.92. These signals were in all liver and kidney samples, and the g ≈ 2.0 free radical signal overlaps that of nitrosyl hemoprotein. Interestingly, the intensity of the
Endotoxin-Induced NO in TCDD-Treated Rats
Chem. Res. Toxicol., Vol. 13, No. 10, 2000 1053
Figure 2. (A) Graph of the double integral of the EPR spectra obtained in venous blood from rats treated with TCDD or TCDD and LPS. The EPR spectra were double-integrated over 400 G to give the total concentration of both HbNO and the Cu2+ from the plasma protein ceruloplasmin. The difference between nonand LPS-treated rats represents the relative HbNO concentrations induced by LPS, assuming the concentration of ceruloplasmin is independent of LPS. (B) Chart of the TNF-R concentrations obtained in venous blood from control and TCDDtreated rats treated with 0.1 mg/kg LPS. TCDD alone did not induce the production of TNF-R, and all values in rats not receiving LPS were below the detection limits of the ELISA (3 pg/mL). Each point represents the mean ( the standard error (n ) 5).
g ≈ 2.0 signal in the liver was relatively constant within each dose group (data not shown), which is in marked contrast to the high variability reported in venous blood (8). We did not attempt to measure this signal in LPSdosed rats as it is obscured by the partial overlap with the HbNO EPR signal. The identity of the free radical responsible for the g ≈ 2.0 EPR signal is elusive due to the lack of structural information. The status of the liver and kidney cytochromes P450 and P420 (P450/P420) was investigated by recording EPR spectra over a scan range of 1000 G (Figures 4 and 5). The known EPR signals of low-spin ferric cytochromes P450/P420 have similar resonance peaks at g ) 2.43, 2.25, and 1.92 (9). We observed these cytochrome species at g ) 2.43, 2.26, and 1.92 in both the livers and kidneys of control and TCDD-treated rats. In addition, the first five resonance peaks (the sixth peak is out of spectral range) of Mn2+ at g ) 2.16, 2.10, 2.04, 1.98, and 1.92 are observable. The relative intensity of the EPR signal at g ≈ 2.26 from ferric low-spin P450/P420 increased in a TCDD dose-dependent manner in both the livers and kidneys (Figure 6). The relative intensity of the g ≈ 2.26
Figure 3. Typical EPR spectra obtained from the livers (a) and kidneys (b) of rats treated with (A) vehicle, (B) 50 µg/kg TCDD, and (C) 50 µg/kg TCDD and 0.1 mg/kg LPS as described in Materials and Methods. Spectra D show the characteristic fivecoordinate difference spectra of NO nitrosyl hemoprotein complex obtained by subtraction of spectra B from spectra C.
EPR signal was approximately 3-4-fold greater in the liver than in the kidney. The ferric low-spin cytochrome P450/P420 concentration was decreased in the livers of
1054
Chem. Res. Toxicol., Vol. 13, No. 10, 2000
Glover et al.
Figure 6. Graph of the TCDD dose dependency of the g ) 2.26 low-spin ferric cytochrome P450/P420 peak intensities in the liver and kidney. Each point represents the mean ( the standard error (n ) 5). Figure 4. Typical EPR (1000 G) spectra obtained from the kidneys of Fischer rats. The EPR spectra are from the kidneys of rats treated with (A) vehicle, (B) 50 µg/kg TCDD, (C) 0.1 mg/ kg LPS, and (D) 50 µg/kg TCDD and 0.1 mg/kg LPS.
Figure 5. Typical EPR (1000 G) spectra obtained from the livers of Fischer rats. The EPR spectra are from the livers of rats treated with (A) vehicle, (B) 50 µg/kg TCDD, (C) 0.1 mg/kg LPS, and (D) 50 µg/kg TCDD and 0.1 mg/kg LPS.
rats exposed to 0.1 mg/kg LPS 4 h prior to euthanasia as compared to the equivalent saline control group with
little effect of LPS on the kidney concentration of ferric low-spin P450/P420.
Discussion It is hypothesized that bacterial toxins such as endotoxin can modulate the pathogenic outcome of exposure to TCDD (1). In this study, we show that NO production parallels that of TNF-R in rats challenged with both TCDD and endotoxin. When rats were treated with a TCDD concentration of 50 µg/kg, there was an approximately 2- and 3-fold increase in endotoxin-induced NO and TNF-R production, respectively. A TCDD concentration of
