- Email: [email protected]
The effects of hypoxanthine, xanthine oxidase and hyperoxia on the accumulation of bilirubin and albumin in young rat brain
Early Human Development. 30
171
(I 992) Ill- 117
Elsevier Scientific Publishers Ireland Ltd. EHD 01315
The effects of hypoxanthine, xanthine oxidase and hyperoxia on the accumulation of bilirubin and albumin in young rat brain Thor Willy Ruud Hansena, Jan Peter Poulsenb and Dag Bratlid’ aDepartments of Pediatrics and Pediatric Research, hlnstitute for Surgical Research and ‘Department of Pediatrics Rikshospitalet, University of Oslo, Oslo (Norway)
(Received 12 February 1992; revision received 7 April 1992; accepted 22 April 1992)
Summary
Hyperoxia has been suggested as a risk factor for kernicterus. The toxicity of hyperoxia may be mediated by free radicals. We investigated the effects of free radicals, formed by the hypoxanthine/xanthine oxidase system, with and without additional hyperoxia, on the accumulation of bilirubin and albumin in rat brain. Hypoxanthine was infused for 60 min into retrograde carotid catheters in awake, young, male SPRD rats. After 30 min the infusion was briefly interrupted to inject xanthine oxidase 1 U/kg through the same catheter. Group I (controls) received 0.9% NaCl in lieu of hypoxantine/xanthine oxidase. Groups I and II breathed room air at all times, while group III breathed 90% Oz. After 60 min all groups received a bolus dose of ‘251-albumin through a peripheral venous catheter, followed by bilirubin 25 mg/kg for 5 min, then bilirubin 35 mg/kg for 55 min. There were no significant differences between the groups as regards serum bilirubin, serum albumin, brain bilirubin, or brain albumin. Neither during normoxic nor hyperoxic conditions did the hypoxanthine/xanthine oxidase system increase the accumulation of bilirubin or albumin in rat brain. Key words: bilirubin; albumin; brain; bilirubin encephalopathy; xanthine; xanthine oxidase; free radicals; rat
hyperoxia; hypo-
Introduction
Bilirubin encephalopathy continues to be seen in the newborn nursery [3,5]. Several factors such as hyperosmolality, hypercarbia, increased brain blood flow, Correspondence to: Thor Willy Ruud Hansen, Department of Pediatrics, Rikshospitalet,
Norway. 037%3782/92/$05.00 0 1992 Elsevier Scientific Publishers Ireland Ltd Printed and Published in Ireland
N-0027 Oslo,
172
bilirubin binding competitors and reduced albumin binding capacity increase the accumulation of bilirubin in brain [7,9,13,15,22,24,26,34,41], and thus presumably increase the risk for bilirubin encephalopathy. Based on an association between pontosubicular necrosis and kemicterus in human neonates [1,2], hyperoxia has been suggested as a risk factor for bilirubin, encephalopathy [25]. However, we recently showed that hyperoxia alone did not increase the accumulation of bilirubin or albumin in rat brain [ 181. Hypoxanthine may play a key role in diseases thought to be caused by hypoxia/ hyperoxia conditions [36]. Hypoxanthine is a potential oxygen free radical generator in the presence of the enzyme xanthine oxidase (EC 1.1.3.22) [16,30]. Oxygen radicals formed through the hypoxanthine/xanthine oxidase system may be a cause of the often dramatic tissue injury seen in the reoxygenation period after hypoxia. This so called ‘oxygen paradox’ has been difficult to explain. The damage could be caused by a temporary production of oxygen radicals at a rate greater than could be accomodated by the body’s defence mechanisms. Other possible explanatory mechanisms for oxygen toxicity include generation of superoxide by leukocytes [4]. The aim herein was to investigate whether free radicals formed by hypoxanthineixanthine oxidase administered to the cerebral circulation would influence the accumulation of bilirubin and/or albumin in rat brain. Further, combining this radical generating system with hyperoxia could theoretically augment the damage. Materials and Methods Male Sprague-Dawley rats were from Dyrlmge Mollegaards Avlslaboratorium, Ll. Skensved, Denmark. At the time of the infusion they weighed 87 f 9 g (mean f SD). Bilirubin, BSA, hypoxanthine and xanthine oxidase (grade III, activity l-2 U/mg protein) were from Sigma Chemical Co., St. Louis, MO, USA. 1251-labelled human serum albumin (HSA, spec. act. 110 Mbq/mmol) was from Institutt for Energiteknikk, Kjeller, Norway. Other reagents, of analytical grade, were from standard commercial suppliers. Polyethylene tubing (800/l 10/200/100, I.D. 0.58 mm, E.D. 0.96 mm) was from Portex, Hythe, England. Preparation of reagents
Bilirubin was dissolved in 0.1 N NaOH, stabilized with bovine diluted with phosphate buffer (0.055 mol/l) to a concentration pH = 8) [7]. Hypoxanthine was dissolved in Ringer acetate (pH tration 10 mmol/l. Xanthine oxidase was diluted with 0.9% NaCl, 0.5 U/ml.
serum albumin and of 2.6 mg/ml (final = 7), final concenfinal concentration
Preparation of the rat model
One day prior to the infusions the rats were anesthetized with fluanizone/fentanyl. A polyethylene catheter was placed retrograde in the right common carotid artery and the free end brought out through the skin on the back of the neck. During the infusion the unanesthetized rats were kept in wire restrainers. The control group and one of the treatment groups breathed room air at all times. The rats in the second treatment group were placed inside a transparent polyethylene bag
173
with a volume of approximately 7.5 1, through which oxygen, was passed at a rate of 10 Ymin, giving an FiOz = 0.9. After drawing baseline blood samples, the rats in the two treatment groups were infused with hypoxanthine 10 mmol/l at 0.1 ml/mm for 30 min, at which time the infusion was briefly interrupted to inject xanthine oxidase 1 U/kg. The hypoxanthine infusion was then continued at 0.05 ml/min for another 30 min. Control rats were infused at corresponding rates with normal saline. All rats were then given a bolus dose of 750 KBq 12’I-HSA. After this bilirubin was infused at 25 mg/kg for 5 min, then at 35 mg/kg for the next 55 min, after which the rats were killed with a bolus dose of saturated potassium chloride. The thorax was opened and a catheter introduced into the ascending aorta, the descending aorta was clamped and the jugular veins were opened, after which the brain vasculature was flushed in situ with 30 ml of cold, normal saline at a rate of 6 ml/min in order to remove as much blood as possible from the brain vessels [7]. Blood sampling and analyses Mixed arterial/venous blood samples were drawn from the cut tip of the tail at 0, 60, 75 and 120 min. Serum bilirubin was measured with a diazo method [33]. Serum ‘free’ bilirubin was estimated with the peroxidase method [21]. Serum albumin was measured with the bromocresol purple method [27]. Hematocrit was measured with microhematocrit tubes. Blood gases were measured on an AVL 945 blood gas analyzer (AVL Biomedical Instruments, Schaflhausen, Switzerland). Serum osmolality was measured with a vapor pressure osmometer (Wescor Model 51OOC, Wescor, Logan, UT, USA). Estimation of brain bilirubin and albumin content The brains were removed from the skulls, stripped of their meningeal coverings and rinsed in cold water. Each brain was then divided in two halves along the sagittal fissure and each half was weighed. For estimation of brain bilirubin uptake one half brain was homogenized in a glass/teflon homogenizer and bilirubin content was determined by chloroform extraction followed by diazotization [6]. For estimation of albumin uptake gamma activity in the other half brain and in a lo-p1 serum sample drawn from the same rat immediately before sacrifice were counted in a Packard 5220 Auto Gamma Scintillation Spectrometer. The specific activity of albumin was computed after analysis of the serum albumin value, after which the brain albumin content could be calculated. In order to counteract possible systematic right-left concentration differences due to unequal blood flow caused by the carotid catheter, alternating right and left halves were used for the determination of brain bilirubin and albumin. Statistical methods Two tailed t-tests were used to compare the control group and the two treatment groups. All values are presented as mean f standard error of the mean (S.E.M.). The level of significance was chosen as P < 0.05. Results
Serum osmolality and blood pH remained normal throughout the experiment and
0 120 75 120 75 120 0 75 120 0 75 120 120 120
Hematocrit (%)
39.4 30.9 189.77 241.4 4.1 3.3 42.2 35.7 34.4 6.9 6.6 6.3 1.8 91.7
2.5 2.3 16.7 31.3 1.4 1.6 4.0 2.5 2.8 0.5 0.6 1.1 0.5 18.6
36.9 31.8 216.3 286.0 5.8 8.3 42.3 34.3 33.4 7.7 7.1 6.6 1.2 85.9
2.0 2.4 47.3 20.3 2.5 3.3 3.2 2.0 3.0 0.5 0.2 0.8 0.4 12.8
S.E.M.
Mean
Mean
S.E.M.
Normoxia (n = 7)
Controls (n = 7)
All statistical comparisons were done with two-tailed r-tests. ?? P < 0.01 versus control. ?? *P < 0.02 versus normoxia. tP < 0.02 versus control and normoxia.
Brain bilirubin (pglg) Brain albumin (pglg)
PO* kP,
Albumin (gQ
Free bilirubin (nmohl)
Bilirubin (Fmolil)
Sampling time (mm)
Parameter
35.4 27.3 205.7 224.3 3.3 1.8 37.3 32.7 28.6 7.2 16.4**** 12.2t 1.9 93.1
Mean
3.2 3.3 18.0 26.5 0.7 1.0 2.1 1.2 1.7 0.8 3.0 1.6 0.6 12.3
S.E.M.
Hyperoxia (n = 7)
Blood, serum and brain parameters in young rats subjected to normoxia and saline infusion (controls), normoxia with infusion of hypoxanthine plus xanthine oxidase (normoxia), or hyperoxia (Fio, = 0.9) with infusion of hypoxanthine plus xanthine oxidase (hyperoxia), followed by infusion of ‘251-albumin and bilirubin. S.E.M., standard error of the mean.
TABLE I
175
did not differ between the groups. These results are not presented. The remaining results are presented in Table I. The reduction in hematocrit and serum albumin over time represents sampling-related blood loss. Discussion The results herein show that the hypoxanthine/xanthine oxidase system did not increase the accumulation of bilirubin or albumin in young rat brain. Nor were the results influenced by the addition of hyperoxia (Fi@ = 0.9). In this model therefore, the blood-brain barrier was apparently not opened and bilirubin was not displaced from its binding to albumin. We have not made any attempts to correct for bilirubin or albumin remaining within the cerebral vasculature [ 191. On the basis of previous studies, we assume that the amount of albumin actually entering the brain substance prop,er was negligible [20]. Oxygen radicals are created in many processes where oxygen is involved. Free radicals are highly reactive, injuring cell membranes by peroxidation of unsaturated fatty acids. The term ‘ischemia-reperfusion injury’ has been introduced for this phenomenon [3 11. During ischemia-reperfusion conditions oxygen radicals formed by the hypoxanthine/xanthine oxidase system will injure intestines [17,37], lungs [35,39] and small-vessel walls [14,40]. Direct infusion of hypoxanthine plus xanthine oxidase into cerebral tissue opens the blood-brain barrier, as shown by subsequent extravasation of Evans blue [ 111. However, the term ‘ischemia-reperfusion injury’ could be misleading, because during the reperfusion process there are many uncontrolled factors in addition to hypoxanthine washout. A better term for this phenomenon may be ‘posthypoxic-reoxygenation injury’, since the damage may not occur during the hypoxic period, but rather when normal or hyper oxygenation is subsequently reestablished. The damage to the blood-brain barrier after asphyxia, which has been shown to increase the accumulation of bilirubin and albumin in brain [12,28,29], could be a result of this phenomenon. The lack of effect of the hypoxanthine/xanthine oxidase system on brain accumulation of bilirubin and albumin during normoxia and hyperoxia as reported herein, appears to contradict the findings discussed above. Bilirubin may be a scavenger of free radicals [23,38], but as hypoxanthine and xanthine oxidase were administered first, their toxic effects ought to have been established before bilirubin and.albumin were given. The putative association between hyperoxic injury and bilirubin encephalopathy [1,2,25] may not be one of cause and effect. Thus, Buonocore et al. found a relative deficiency of oxygen radical scavenging enzymes in neonates with hyperbilirubinemia and suggested that this deficiency caused increased hemolysis (and thereby increased production of bilirubin) by rendering the deficient red cells more vulnerable to hemolysis [8]. This type of association would not have been tested in the present experimental design. The possible role of leukocyte-generated superoxide was also not subject to testing herein. Such a study might be of interest given the apparent ability of bilirubin to inhibit production of superoxide by leukocytes in vitro [32]. In conclusion, hypoxanthine plus xanthine oxidase administered to the arterial side of the cerebral circulation in young rats during normoxia and hyperoxia did not
176
increase the accumulation of bilirubin or albumin in brain. Thus we have not been able to substantiate the speculation that hyperoxia may increase the risk of bilirubin encephalopathy. Further studies in other species are needed in order to validate these findings. Acknowledgements Dr Poulsen was supported by a grant from The Norwegian Cancer Society. References 1 Ahdab-Barmada, A., Moossy, J. and Painter, M. (1980): Pontosubicular necrosis and hyperoxemia. Pediatrics 66, 840-847. 2 AhdabBarmada, M. (1981): Neonatal kemicterus: Neuropathologic diagnosis. In: Hyperbilirubinemia of the Newborn. Report of the Eighty-Fifth Ross conference on Pediatric Research, pp. 2-8. Editors: R.L. Levine and M.J. Maisels. Ross Laboratories, Columbus, OH. Ahdab-Barmada, M. and Moossy, J. (1984): The neuropathology of kernicterus in the premature neonate: diagnostic problems. J. Neuropathol. Exp. Neurol., 43, 45-56. Babior, B.M., Kipnes, R.S. and Cumette, J.T. (1973): Biological defense mechanisms. The production by leukocytes of superoxide - a potential bactericidal agent. J. Clin. Invest., 52, 741-744. Ballowitx, L. (1980): Bilirubin encephalopathy: Changing concepts. Brain. Dev., 2, 219-227. Bratlid, D. and Winsnes, A. (1971): Determination of conjugated and unconjugated bilirubin by methods based on direct spectrophotometry and chloroform extraction. A reappraisal. Stand. J. Clin. Lab. Invest., 28, 41-48. 7 Bratlid, D., Cashore, W.J. and Oh, W. (1983): Effect of serum hyperosmolality on opening of bloodbrain barrier for bilirubin in rat brain. Pediatrics, 71, 909-912. 8 Buonocore, G., Talluri, B., De Biase, L., Giorli, M., Bagnoli F. and Bracci, R. (1983): The role of antioxidant erythrocyte enzyme activities in the development of neonatal hemolysis and hyperbilirubinemia. Biol. Neonate, 44, 372-373A. 9 Burgess, G.H., Oh, W., Bratlid, D., Brubakk A.-M., Cashore W.J. and Stonestreet, B.S. (1985): The effects of brain blood flow on brain bilirubin deposition in newborn piglets. Pediatr. Res., 19, 691-696. 10 Cashore, W.J., Horwich, A., Karotkin, E.H. and Oh, W. (1977): Influence of gestational age and clinical status on bilirubin-binding capacity in newborn infants. Am. J. Dis. Child., 131, 898-901. 11 Chan, P.H., Schmidley, J.W., Fishman, R.A. and Longar S.M. (1984): Brain injury, edema and vascular permeability changes induced by oxygen-derived free radicals. Neurology, 34, 315-320. 12 Chen H.-C., Lin C.-S. and Lein I.-N. (1966): Ultrastructural studies in experimental kemicterus. Am. J. Pathol., 48, 683-711. 13 Day, R. (1947): Kemicterus problem: experimental in vivo and in vitro staining of brain tissue with bilirubin. Am. J. Dis. Child., 73, 241-242. 14 Del Maestro, R.F., Thawn, H.H., Bjdrk, J., Planker, M. and Arfors K.E. (1980): Free radicals as mediators of tissue injury. Acta Physiol. Stand., 492, 43-57. 15 Ebbesen, F., Foged, N. and Brodersen, R. (1986): Reduced albumin binding of MADDS - a measure of bilirubin binding in sick children. Acta Paediatr. Stand., 75, 550-554. 16 Fridovich, I. (1970): Quantitative aspects of the production of superoxide anion radical by milk xanthine oxidase. J. Biol. Chem., 245, 4053-4057. 17 Granger, D.N., Rutili, G. and McCord, J.M. (1981): Superoxide radicals in feline intestinal ischemia. Gastroenterology, 81, 22-29. 18 Hansen T.W.R., Odden J.-P. and Bratlid, D. (1987): Effects of hyperoxia on entry of bilirubin and albumin into rat brain. J. Perinatol., 7, 217-220. 19 Hansen T.W.R. and Bratlid, D. (1989): Cerebral blood volumes in young rats without and with in situ flushing of cerebral vasculature. Implications for in vivo studies of brain substance uptake. Biol. Neonate, 56, 15-21.
20
21 22 23 24 25
26 27 28 29 30 31 32
33 34 35
36 37
38 39
40 41
Hansen T.W.R., (dyasaeter, S., Stiris, T. and Bratlid, D. (1989): Effects of sultisoxazole, hypercarbia and hyperosmolality on entry of bilirubin and albumin into young rat brain regions. Biol. Neonate, 56, 22-30. Jacobsen, J. and Wennberg, R.P. (1974): Determination of unbound bilirubin in the serum of newborns. Clin. Chem., 20, 783-789. Kapitulnik, J., Horner-Mibashan, R., Blondheim, S.H., Kaufmann, N.A. and Russell, A. (1975): Increase in bilirubin-binding aftinity of serum with age of infant. J. Pediatr., 86, 442-445. Kaul, R., Kaul, H.K. and Kirshna Murti, CR. (1980): An alternative pathway for bilirubin catabolism. FEBS Lett., 111, 240-242. Lending, M., Slobody, L.B. and Mestem, J. (1966): The relationship of hypercapnia to the production of kernicterus. Develop. Med. Child. Neural., 9, 145-151. Levine, R.L. (1981): Bilirubin and the blood-brain barrier. In: Hyperbilirubinemia of the Newborn. Report of the Eighty-Fifth Ross conference on Pediatric Research, pp. 125-129. Editors: R.L. Levine and M.J. Maisels. Ross Laboratories, Columbus, OH. Levine, R.L., Fredericks, W.R. and Rapoport, S.I. (1982): Entry of bilirubin into the brain due to opening of the blood-brain barrier. Pediatrics, 69, 255-259. Louderback, A., Meaky, E.H. and Taylor, N.A. (1968): A new dye-binding technic using bromcresol purple for determination of albumin in serum. Clin. Chem., 14, 793-794. Lou, H.C., Tweed, W.A., Johnson, G., Jones, M. and Lassen N.A. (1977): Breakdown of bloodbrain barrier in kemicterus. Lancet, i, 1062-1063. Lucey, J.F., Hibbard, E., Behrman, R.E., Esquivel de Gallardo F.G. and Windle, W.F. (1964): Kernicterus in asphyxiated newborn monkeys. Exp. Neurol., 9, 43-58. McCord, J.M. and Fridovich, I. (1968): The reduction of cytochrome C by milk xanthine oxidase. J. Biol. Chem., 243, 553-560. McCord, J.M. (1985): Oxygen derived free radicals in post-ischemic tissue injury. N. En@. J. Med., 312, 159-163. Nakamura, H., Uetani, Y., Kmoura, M., Takada, S., Sano, K. and Matsuo, T. (1987): Inhibitory action of bilirubin on superoxide production by polymorphonuclear leukocytes. Biol. Neonate, 52, 273-278. Nosslin, B. (1960): The direct diazo reaction of bile pigments in serum. Experimental and clinical studies. Scan. J. Clin. Lab. Invest., 12 (Suppl.), 49. Robinson, P.J. and Rapoport, S.I. (1987): Binding effect of albumin on uptake of bilirubin by brain. Pediatrics, 79, 553-558. Saugstad, O.D., Hallman, M., Abraham, J., Cochrane, C.G., Epstein, B. and Gluck, L. (1984): Hypoxanthine and oxygen induced lung injury: a basic mechanism of tissue damage. Pediatr. Res., 18, 501-504. Saugstad, O.D. (1988): Hypoxanthine as an Indicator of Hypoxia: Its Role in Health and Disease through Free Radical Production. Pediatr. Res., 23, 143-150. Schoenberg, M.H., Fredholm, B.B., Haglund, K., Jung, H., Sellin, D., Younes, M. and Schildberg, F.W. (1985): Studies on the oxygen radical mechanism involved in the small intestinal reperfusion damage. Acta Physiol. Stand., 124, 581-589. Stocker, R., Yamamoto, Y., McDonagh, A.F., Glazer, A.N. and Ames B.N. (1987): Bilirubin is an antioxidant of possible physiological importance. Science, 235, 1043-1046. Tate, R.M., Morris, H.B., Schroeder, W.B. and Repine, J.E. (1984): Oxygen metabolites stimulate thromboxane production and vasoconstriction in isolated saline perfused rabbit lungs. J. Clin. Invest., 74, 608-613. Wei, E.P., Christman, C.W., Kontos, H.A. and Povlishock J.T. (1985): Effects of oxygen radicals on cerebral arterioles. Am. J. Physiol., 248, H157-H162. Wooley, P.V. and Hunter, M. (1970): Effect of anions on bilirubin-albumin in the presence of oleate and salicylate. Arch. Biochem., 140, 197-209.
171
(I 992) Ill- 117
Elsevier Scientific Publishers Ireland Ltd. EHD 01315
The effects of hypoxanthine, xanthine oxidase and hyperoxia on the accumulation of bilirubin and albumin in young rat brain Thor Willy Ruud Hansena, Jan Peter Poulsenb and Dag Bratlid’ aDepartments of Pediatrics and Pediatric Research, hlnstitute for Surgical Research and ‘Department of Pediatrics Rikshospitalet, University of Oslo, Oslo (Norway)
(Received 12 February 1992; revision received 7 April 1992; accepted 22 April 1992)
Summary
Hyperoxia has been suggested as a risk factor for kernicterus. The toxicity of hyperoxia may be mediated by free radicals. We investigated the effects of free radicals, formed by the hypoxanthine/xanthine oxidase system, with and without additional hyperoxia, on the accumulation of bilirubin and albumin in rat brain. Hypoxanthine was infused for 60 min into retrograde carotid catheters in awake, young, male SPRD rats. After 30 min the infusion was briefly interrupted to inject xanthine oxidase 1 U/kg through the same catheter. Group I (controls) received 0.9% NaCl in lieu of hypoxantine/xanthine oxidase. Groups I and II breathed room air at all times, while group III breathed 90% Oz. After 60 min all groups received a bolus dose of ‘251-albumin through a peripheral venous catheter, followed by bilirubin 25 mg/kg for 5 min, then bilirubin 35 mg/kg for 55 min. There were no significant differences between the groups as regards serum bilirubin, serum albumin, brain bilirubin, or brain albumin. Neither during normoxic nor hyperoxic conditions did the hypoxanthine/xanthine oxidase system increase the accumulation of bilirubin or albumin in rat brain. Key words: bilirubin; albumin; brain; bilirubin encephalopathy; xanthine; xanthine oxidase; free radicals; rat
hyperoxia; hypo-
Introduction
Bilirubin encephalopathy continues to be seen in the newborn nursery [3,5]. Several factors such as hyperosmolality, hypercarbia, increased brain blood flow, Correspondence to: Thor Willy Ruud Hansen, Department of Pediatrics, Rikshospitalet,
Norway. 037%3782/92/$05.00 0 1992 Elsevier Scientific Publishers Ireland Ltd Printed and Published in Ireland
N-0027 Oslo,
172
bilirubin binding competitors and reduced albumin binding capacity increase the accumulation of bilirubin in brain [7,9,13,15,22,24,26,34,41], and thus presumably increase the risk for bilirubin encephalopathy. Based on an association between pontosubicular necrosis and kemicterus in human neonates [1,2], hyperoxia has been suggested as a risk factor for bilirubin, encephalopathy [25]. However, we recently showed that hyperoxia alone did not increase the accumulation of bilirubin or albumin in rat brain [ 181. Hypoxanthine may play a key role in diseases thought to be caused by hypoxia/ hyperoxia conditions [36]. Hypoxanthine is a potential oxygen free radical generator in the presence of the enzyme xanthine oxidase (EC 1.1.3.22) [16,30]. Oxygen radicals formed through the hypoxanthine/xanthine oxidase system may be a cause of the often dramatic tissue injury seen in the reoxygenation period after hypoxia. This so called ‘oxygen paradox’ has been difficult to explain. The damage could be caused by a temporary production of oxygen radicals at a rate greater than could be accomodated by the body’s defence mechanisms. Other possible explanatory mechanisms for oxygen toxicity include generation of superoxide by leukocytes [4]. The aim herein was to investigate whether free radicals formed by hypoxanthineixanthine oxidase administered to the cerebral circulation would influence the accumulation of bilirubin and/or albumin in rat brain. Further, combining this radical generating system with hyperoxia could theoretically augment the damage. Materials and Methods Male Sprague-Dawley rats were from Dyrlmge Mollegaards Avlslaboratorium, Ll. Skensved, Denmark. At the time of the infusion they weighed 87 f 9 g (mean f SD). Bilirubin, BSA, hypoxanthine and xanthine oxidase (grade III, activity l-2 U/mg protein) were from Sigma Chemical Co., St. Louis, MO, USA. 1251-labelled human serum albumin (HSA, spec. act. 110 Mbq/mmol) was from Institutt for Energiteknikk, Kjeller, Norway. Other reagents, of analytical grade, were from standard commercial suppliers. Polyethylene tubing (800/l 10/200/100, I.D. 0.58 mm, E.D. 0.96 mm) was from Portex, Hythe, England. Preparation of reagents
Bilirubin was dissolved in 0.1 N NaOH, stabilized with bovine diluted with phosphate buffer (0.055 mol/l) to a concentration pH = 8) [7]. Hypoxanthine was dissolved in Ringer acetate (pH tration 10 mmol/l. Xanthine oxidase was diluted with 0.9% NaCl, 0.5 U/ml.
serum albumin and of 2.6 mg/ml (final = 7), final concenfinal concentration
Preparation of the rat model
One day prior to the infusions the rats were anesthetized with fluanizone/fentanyl. A polyethylene catheter was placed retrograde in the right common carotid artery and the free end brought out through the skin on the back of the neck. During the infusion the unanesthetized rats were kept in wire restrainers. The control group and one of the treatment groups breathed room air at all times. The rats in the second treatment group were placed inside a transparent polyethylene bag
173
with a volume of approximately 7.5 1, through which oxygen, was passed at a rate of 10 Ymin, giving an FiOz = 0.9. After drawing baseline blood samples, the rats in the two treatment groups were infused with hypoxanthine 10 mmol/l at 0.1 ml/mm for 30 min, at which time the infusion was briefly interrupted to inject xanthine oxidase 1 U/kg. The hypoxanthine infusion was then continued at 0.05 ml/min for another 30 min. Control rats were infused at corresponding rates with normal saline. All rats were then given a bolus dose of 750 KBq 12’I-HSA. After this bilirubin was infused at 25 mg/kg for 5 min, then at 35 mg/kg for the next 55 min, after which the rats were killed with a bolus dose of saturated potassium chloride. The thorax was opened and a catheter introduced into the ascending aorta, the descending aorta was clamped and the jugular veins were opened, after which the brain vasculature was flushed in situ with 30 ml of cold, normal saline at a rate of 6 ml/min in order to remove as much blood as possible from the brain vessels [7]. Blood sampling and analyses Mixed arterial/venous blood samples were drawn from the cut tip of the tail at 0, 60, 75 and 120 min. Serum bilirubin was measured with a diazo method [33]. Serum ‘free’ bilirubin was estimated with the peroxidase method [21]. Serum albumin was measured with the bromocresol purple method [27]. Hematocrit was measured with microhematocrit tubes. Blood gases were measured on an AVL 945 blood gas analyzer (AVL Biomedical Instruments, Schaflhausen, Switzerland). Serum osmolality was measured with a vapor pressure osmometer (Wescor Model 51OOC, Wescor, Logan, UT, USA). Estimation of brain bilirubin and albumin content The brains were removed from the skulls, stripped of their meningeal coverings and rinsed in cold water. Each brain was then divided in two halves along the sagittal fissure and each half was weighed. For estimation of brain bilirubin uptake one half brain was homogenized in a glass/teflon homogenizer and bilirubin content was determined by chloroform extraction followed by diazotization [6]. For estimation of albumin uptake gamma activity in the other half brain and in a lo-p1 serum sample drawn from the same rat immediately before sacrifice were counted in a Packard 5220 Auto Gamma Scintillation Spectrometer. The specific activity of albumin was computed after analysis of the serum albumin value, after which the brain albumin content could be calculated. In order to counteract possible systematic right-left concentration differences due to unequal blood flow caused by the carotid catheter, alternating right and left halves were used for the determination of brain bilirubin and albumin. Statistical methods Two tailed t-tests were used to compare the control group and the two treatment groups. All values are presented as mean f standard error of the mean (S.E.M.). The level of significance was chosen as P < 0.05. Results
Serum osmolality and blood pH remained normal throughout the experiment and
0 120 75 120 75 120 0 75 120 0 75 120 120 120
Hematocrit (%)
39.4 30.9 189.77 241.4 4.1 3.3 42.2 35.7 34.4 6.9 6.6 6.3 1.8 91.7
2.5 2.3 16.7 31.3 1.4 1.6 4.0 2.5 2.8 0.5 0.6 1.1 0.5 18.6
36.9 31.8 216.3 286.0 5.8 8.3 42.3 34.3 33.4 7.7 7.1 6.6 1.2 85.9
2.0 2.4 47.3 20.3 2.5 3.3 3.2 2.0 3.0 0.5 0.2 0.8 0.4 12.8
S.E.M.
Mean
Mean
S.E.M.
Normoxia (n = 7)
Controls (n = 7)
All statistical comparisons were done with two-tailed r-tests. ?? P < 0.01 versus control. ?? *P < 0.02 versus normoxia. tP < 0.02 versus control and normoxia.
Brain bilirubin (pglg) Brain albumin (pglg)
PO* kP,
Albumin (gQ
Free bilirubin (nmohl)
Bilirubin (Fmolil)
Sampling time (mm)
Parameter
35.4 27.3 205.7 224.3 3.3 1.8 37.3 32.7 28.6 7.2 16.4**** 12.2t 1.9 93.1
Mean
3.2 3.3 18.0 26.5 0.7 1.0 2.1 1.2 1.7 0.8 3.0 1.6 0.6 12.3
S.E.M.
Hyperoxia (n = 7)
Blood, serum and brain parameters in young rats subjected to normoxia and saline infusion (controls), normoxia with infusion of hypoxanthine plus xanthine oxidase (normoxia), or hyperoxia (Fio, = 0.9) with infusion of hypoxanthine plus xanthine oxidase (hyperoxia), followed by infusion of ‘251-albumin and bilirubin. S.E.M., standard error of the mean.
TABLE I
175
did not differ between the groups. These results are not presented. The remaining results are presented in Table I. The reduction in hematocrit and serum albumin over time represents sampling-related blood loss. Discussion The results herein show that the hypoxanthine/xanthine oxidase system did not increase the accumulation of bilirubin or albumin in young rat brain. Nor were the results influenced by the addition of hyperoxia (Fi@ = 0.9). In this model therefore, the blood-brain barrier was apparently not opened and bilirubin was not displaced from its binding to albumin. We have not made any attempts to correct for bilirubin or albumin remaining within the cerebral vasculature [ 191. On the basis of previous studies, we assume that the amount of albumin actually entering the brain substance prop,er was negligible [20]. Oxygen radicals are created in many processes where oxygen is involved. Free radicals are highly reactive, injuring cell membranes by peroxidation of unsaturated fatty acids. The term ‘ischemia-reperfusion injury’ has been introduced for this phenomenon [3 11. During ischemia-reperfusion conditions oxygen radicals formed by the hypoxanthine/xanthine oxidase system will injure intestines [17,37], lungs [35,39] and small-vessel walls [14,40]. Direct infusion of hypoxanthine plus xanthine oxidase into cerebral tissue opens the blood-brain barrier, as shown by subsequent extravasation of Evans blue [ 111. However, the term ‘ischemia-reperfusion injury’ could be misleading, because during the reperfusion process there are many uncontrolled factors in addition to hypoxanthine washout. A better term for this phenomenon may be ‘posthypoxic-reoxygenation injury’, since the damage may not occur during the hypoxic period, but rather when normal or hyper oxygenation is subsequently reestablished. The damage to the blood-brain barrier after asphyxia, which has been shown to increase the accumulation of bilirubin and albumin in brain [12,28,29], could be a result of this phenomenon. The lack of effect of the hypoxanthine/xanthine oxidase system on brain accumulation of bilirubin and albumin during normoxia and hyperoxia as reported herein, appears to contradict the findings discussed above. Bilirubin may be a scavenger of free radicals [23,38], but as hypoxanthine and xanthine oxidase were administered first, their toxic effects ought to have been established before bilirubin and.albumin were given. The putative association between hyperoxic injury and bilirubin encephalopathy [1,2,25] may not be one of cause and effect. Thus, Buonocore et al. found a relative deficiency of oxygen radical scavenging enzymes in neonates with hyperbilirubinemia and suggested that this deficiency caused increased hemolysis (and thereby increased production of bilirubin) by rendering the deficient red cells more vulnerable to hemolysis [8]. This type of association would not have been tested in the present experimental design. The possible role of leukocyte-generated superoxide was also not subject to testing herein. Such a study might be of interest given the apparent ability of bilirubin to inhibit production of superoxide by leukocytes in vitro [32]. In conclusion, hypoxanthine plus xanthine oxidase administered to the arterial side of the cerebral circulation in young rats during normoxia and hyperoxia did not
176
increase the accumulation of bilirubin or albumin in brain. Thus we have not been able to substantiate the speculation that hyperoxia may increase the risk of bilirubin encephalopathy. Further studies in other species are needed in order to validate these findings. Acknowledgements Dr Poulsen was supported by a grant from The Norwegian Cancer Society. References 1 Ahdab-Barmada, A., Moossy, J. and Painter, M. (1980): Pontosubicular necrosis and hyperoxemia. Pediatrics 66, 840-847. 2 AhdabBarmada, M. (1981): Neonatal kemicterus: Neuropathologic diagnosis. In: Hyperbilirubinemia of the Newborn. Report of the Eighty-Fifth Ross conference on Pediatric Research, pp. 2-8. Editors: R.L. Levine and M.J. Maisels. Ross Laboratories, Columbus, OH. Ahdab-Barmada, M. and Moossy, J. (1984): The neuropathology of kernicterus in the premature neonate: diagnostic problems. J. Neuropathol. Exp. Neurol., 43, 45-56. Babior, B.M., Kipnes, R.S. and Cumette, J.T. (1973): Biological defense mechanisms. The production by leukocytes of superoxide - a potential bactericidal agent. J. Clin. Invest., 52, 741-744. Ballowitx, L. (1980): Bilirubin encephalopathy: Changing concepts. Brain. Dev., 2, 219-227. Bratlid, D. and Winsnes, A. (1971): Determination of conjugated and unconjugated bilirubin by methods based on direct spectrophotometry and chloroform extraction. A reappraisal. Stand. J. Clin. Lab. Invest., 28, 41-48. 7 Bratlid, D., Cashore, W.J. and Oh, W. (1983): Effect of serum hyperosmolality on opening of bloodbrain barrier for bilirubin in rat brain. Pediatrics, 71, 909-912. 8 Buonocore, G., Talluri, B., De Biase, L., Giorli, M., Bagnoli F. and Bracci, R. (1983): The role of antioxidant erythrocyte enzyme activities in the development of neonatal hemolysis and hyperbilirubinemia. Biol. Neonate, 44, 372-373A. 9 Burgess, G.H., Oh, W., Bratlid, D., Brubakk A.-M., Cashore W.J. and Stonestreet, B.S. (1985): The effects of brain blood flow on brain bilirubin deposition in newborn piglets. Pediatr. Res., 19, 691-696. 10 Cashore, W.J., Horwich, A., Karotkin, E.H. and Oh, W. (1977): Influence of gestational age and clinical status on bilirubin-binding capacity in newborn infants. Am. J. Dis. Child., 131, 898-901. 11 Chan, P.H., Schmidley, J.W., Fishman, R.A. and Longar S.M. (1984): Brain injury, edema and vascular permeability changes induced by oxygen-derived free radicals. Neurology, 34, 315-320. 12 Chen H.-C., Lin C.-S. and Lein I.-N. (1966): Ultrastructural studies in experimental kemicterus. Am. J. Pathol., 48, 683-711. 13 Day, R. (1947): Kemicterus problem: experimental in vivo and in vitro staining of brain tissue with bilirubin. Am. J. Dis. Child., 73, 241-242. 14 Del Maestro, R.F., Thawn, H.H., Bjdrk, J., Planker, M. and Arfors K.E. (1980): Free radicals as mediators of tissue injury. Acta Physiol. Stand., 492, 43-57. 15 Ebbesen, F., Foged, N. and Brodersen, R. (1986): Reduced albumin binding of MADDS - a measure of bilirubin binding in sick children. Acta Paediatr. Stand., 75, 550-554. 16 Fridovich, I. (1970): Quantitative aspects of the production of superoxide anion radical by milk xanthine oxidase. J. Biol. Chem., 245, 4053-4057. 17 Granger, D.N., Rutili, G. and McCord, J.M. (1981): Superoxide radicals in feline intestinal ischemia. Gastroenterology, 81, 22-29. 18 Hansen T.W.R., Odden J.-P. and Bratlid, D. (1987): Effects of hyperoxia on entry of bilirubin and albumin into rat brain. J. Perinatol., 7, 217-220. 19 Hansen T.W.R. and Bratlid, D. (1989): Cerebral blood volumes in young rats without and with in situ flushing of cerebral vasculature. Implications for in vivo studies of brain substance uptake. Biol. Neonate, 56, 15-21.
20
21 22 23 24 25
26 27 28 29 30 31 32
33 34 35
36 37
38 39
40 41
Hansen T.W.R., (dyasaeter, S., Stiris, T. and Bratlid, D. (1989): Effects of sultisoxazole, hypercarbia and hyperosmolality on entry of bilirubin and albumin into young rat brain regions. Biol. Neonate, 56, 22-30. Jacobsen, J. and Wennberg, R.P. (1974): Determination of unbound bilirubin in the serum of newborns. Clin. Chem., 20, 783-789. Kapitulnik, J., Horner-Mibashan, R., Blondheim, S.H., Kaufmann, N.A. and Russell, A. (1975): Increase in bilirubin-binding aftinity of serum with age of infant. J. Pediatr., 86, 442-445. Kaul, R., Kaul, H.K. and Kirshna Murti, CR. (1980): An alternative pathway for bilirubin catabolism. FEBS Lett., 111, 240-242. Lending, M., Slobody, L.B. and Mestem, J. (1966): The relationship of hypercapnia to the production of kernicterus. Develop. Med. Child. Neural., 9, 145-151. Levine, R.L. (1981): Bilirubin and the blood-brain barrier. In: Hyperbilirubinemia of the Newborn. Report of the Eighty-Fifth Ross conference on Pediatric Research, pp. 125-129. Editors: R.L. Levine and M.J. Maisels. Ross Laboratories, Columbus, OH. Levine, R.L., Fredericks, W.R. and Rapoport, S.I. (1982): Entry of bilirubin into the brain due to opening of the blood-brain barrier. Pediatrics, 69, 255-259. Louderback, A., Meaky, E.H. and Taylor, N.A. (1968): A new dye-binding technic using bromcresol purple for determination of albumin in serum. Clin. Chem., 14, 793-794. Lou, H.C., Tweed, W.A., Johnson, G., Jones, M. and Lassen N.A. (1977): Breakdown of bloodbrain barrier in kemicterus. Lancet, i, 1062-1063. Lucey, J.F., Hibbard, E., Behrman, R.E., Esquivel de Gallardo F.G. and Windle, W.F. (1964): Kernicterus in asphyxiated newborn monkeys. Exp. Neurol., 9, 43-58. McCord, J.M. and Fridovich, I. (1968): The reduction of cytochrome C by milk xanthine oxidase. J. Biol. Chem., 243, 553-560. McCord, J.M. (1985): Oxygen derived free radicals in post-ischemic tissue injury. N. En@. J. Med., 312, 159-163. Nakamura, H., Uetani, Y., Kmoura, M., Takada, S., Sano, K. and Matsuo, T. (1987): Inhibitory action of bilirubin on superoxide production by polymorphonuclear leukocytes. Biol. Neonate, 52, 273-278. Nosslin, B. (1960): The direct diazo reaction of bile pigments in serum. Experimental and clinical studies. Scan. J. Clin. Lab. Invest., 12 (Suppl.), 49. Robinson, P.J. and Rapoport, S.I. (1987): Binding effect of albumin on uptake of bilirubin by brain. Pediatrics, 79, 553-558. Saugstad, O.D., Hallman, M., Abraham, J., Cochrane, C.G., Epstein, B. and Gluck, L. (1984): Hypoxanthine and oxygen induced lung injury: a basic mechanism of tissue damage. Pediatr. Res., 18, 501-504. Saugstad, O.D. (1988): Hypoxanthine as an Indicator of Hypoxia: Its Role in Health and Disease through Free Radical Production. Pediatr. Res., 23, 143-150. Schoenberg, M.H., Fredholm, B.B., Haglund, K., Jung, H., Sellin, D., Younes, M. and Schildberg, F.W. (1985): Studies on the oxygen radical mechanism involved in the small intestinal reperfusion damage. Acta Physiol. Stand., 124, 581-589. Stocker, R., Yamamoto, Y., McDonagh, A.F., Glazer, A.N. and Ames B.N. (1987): Bilirubin is an antioxidant of possible physiological importance. Science, 235, 1043-1046. Tate, R.M., Morris, H.B., Schroeder, W.B. and Repine, J.E. (1984): Oxygen metabolites stimulate thromboxane production and vasoconstriction in isolated saline perfused rabbit lungs. J. Clin. Invest., 74, 608-613. Wei, E.P., Christman, C.W., Kontos, H.A. and Povlishock J.T. (1985): Effects of oxygen radicals on cerebral arterioles. Am. J. Physiol., 248, H157-H162. Wooley, P.V. and Hunter, M. (1970): Effect of anions on bilirubin-albumin in the presence of oleate and salicylate. Arch. Biochem., 140, 197-209.