Materials and Methods

Transcription

1 GASTROENTEROLOGY Copyright 1966 by The Williams & Wilkins Co. Vo!' SO, No.6 Printed in U.S.A. THE EFFECT OF BILE DUCT LIGATION, BILE DUCT CANNULATION, AND HYPOTHERMIA ON a-naphthylisothiocyanate-induced HYPERBILIRUBINEMIA AND CHOLESTASIS IN RATS ROBERT J. ROBERTS, B.PHARM., AND GABRIEL L. PLAA, PH.D. Department of Pharmacology, College of Medicine, The University of Iowa, Iowa City, Iowa The ability of a-naphthylisothiocyanate (ANIT) to induce in the rat an organ response resembling that seen in human biliary cirrhosis has been well documented and has prompted its use in numerous investigations. 1-3 The cirrhotic response, however, is preceded by an acute lesion characterized by hyperbilirubinemia and cholestasis. This hepatotoxic effect can occur following a single oral dose of ANIT.4 The etiology of the biochemical changes and cholestasis resulting from ANIT administration is as yet unknown, although a number of theories have been proposed. 1-5 Most observers have failed to find morphological evidence of an anatomical obstruction of the bile passages, and this has led some investigators to designate the eventual cirrhosis as being free of biliary obstruction. 6 Others have suggested that an obstruction of the biliary pathways is involved in the pathological picture of ANIT administration. 1, 2, 5 Previous workers have studied various biochemical parameters following ANIT treatment, in addition to the morphological examination of the liver. In a comparative study Phillips and Steiner 7 have shown that there seem to be, species differences in the various ANIT responses. These authors seem to have ruled out hemolysis as a primary cause of hyperbilirubinemia. Moran et al. 8 have also attempted, by administer- Received October 15, Accepted February 8, Address requests for reprints to: Dr. Gabriel L. Plaa, Department of Pharmacology, College of Medicine, University of Iowa, Iowa City, Iowa This work was supported by Research Grant AM and Training Grant 5TI-GM-141 from the United States Public Health Service. ing cortisone, to alter the morphological responses leading to cirrhosis. Recent work in our laboratory9,10 has shown that in the mouse the ANIT-induced hyperbilirubinemia and cessation of bile flow can be potentiated or inhibited by pretreatment with agents which affect microsomal metabolism. The purpose of the following study was to attempt to alter the hyperbilirubinemic and bile flow responses to ANIT in the rat by subjecting the test animals to various environmental and surgical manipulations for the purpose of gaining a better understanding of the mechanism involved in ANIT -induced hyperbilirubinemia and cholestasis. Materials and Methods Male Simonsen rats weighing 250 to 400 g were used throughout the study. Animals were treated with AN IT (K and K Laboratories, Plainview,' N. Y.) suspended in 1% carboxymethyl cellulose (CMC) of sufficient volume to deliver a 300 mg/kg oral dosage in 5 mljkg. Control animals received an equivalent treatment with 1% CMC only. The animals were housed individually; food and water were withheld during all experimental procedures. Randomization was employed both in the selection and in the treatment of animals in each experiment. The control and treated groups of animals were run simultaneously in all experiments. Bile duct ligations and bile duct cannulations were carried out just prior to the administration of AN IT or CMC. Sham operations were also completed at this time as a control procedure. These procedures were carried out under pentobarbital anesthesia (45 mg/kg, ip). When it was discovered that ANIT caused disturbances in temperature regulation, several experiments were carried out under hypothermic conditions. Hypothermia could be induced by placing the rats in a rapidly moving air current (25 C) created by an electric oscillating fan. 768

2 June 1966 ANIT-INDUCED HYPERBILIRUBINEMIA 7GB Hypothermia was further facilitated by shaving the animals prior to exposure to the circulating air. In other experiments, unshaven animals were maintained in a thermostatically controlled incubation chamber set at an ambient temperature of 30 C. Following the various procedures, three responses were measured consecutively 24 hr after the administration of AN IT or CMC. Rectal temperature measurements were first taken using a rectal temperature probe in conjunction with a Yellow Springs Tele-Thermometer. The presence or absence of bile flow was then determined under ether anesthesia by exposing a small portion of the bile duct approximately 1 em above the hilum and inserting polyethylene tubing (PE-lO) via a small cut made with microscissors. The tip of the cannula was advanced to the hilum. Because the presence or absence of bile flow was the criterion used for studying the state of bile flow, cannulations which failed to demonstrate bile flow were routinely and carefully evaluated by aspiration or recannulation, or both. Only those which still failed to flow within 5 to 10 min were classified as "blocked." No attempt was made to quantitatively evaluate the flow of bile in those not showing complete cessation of bile flow. Total plasma bilirubin was then determined by the method of Ferro and Ham" on blood samples obtained from the abdominal aorta by needle puncture. Restraint of the animals was accomplished by the use of Bollman type restraining cages (C. H. Stoelting Company, Chicago). For statistical analysis, the Student's t-test 12 was used to test the null hypothesis in the case of the parametric data. The non parametric data were compared by the Fisher exact probability method.'" The method described by Goldstein" was used to compute the best-fitting line and correlation coefficient for the bilirubin level vs. rectal temperature data shown in figure 1. Results T able 1 summarizes the results obtained in a pilot experiment in which rectal temperature and bile flow were measured in both control and ANIT-treated animals subjected to several combinations of environmental treatments. It was impractical to run all animals concurrently. Therefore, three or four rats from each treatment group were run simultaneously so that all treatments were represented in each experiment. The controls were run concurrently with the ANIT -treated ani- 4.0 RECTAL TEMPERATURE. FIG. 1. Effect of temperature on response of rats to «-naphthylisothiocyanate (ANIT). The total plasma bilirubin (mg/ loo ml of plasma ) and rectal temperature (DC) were determined 24 hr after ANIT administration (300 mg/ kg, p.o.). Each point indicates the response obtained from a single animal. The line represents a calculated linear regression line. A correlation coefficient of r = 0.8 was found between rectal temperature and plasma bilirubin level, which 18 statistically significant at the P < 0.05 level. mals. This experiment was then replicated three times and the data were pooled. Group IV showed that ANIT caused cessation of bile flow within 24 hr, as expected. This group also showed that if ANITtreated rats were subjected to cold air, a mild hypothermia was induced. All rat s in this pilot study were unshaven. This hypothermia could be enhanced by restraining the rats during the 24-hr post ANIT period (group III). Restraint also appeared to cause an elevation in temperature in ANIT -treated groups incubated at 30 C (group VI). The normal rat temperature was considered to be 36.5 C under our experimental conditions. When the incubator temperature was increased to 35 C, the mortality 24 hr after ANIT was significantly enhanced. Hypothermia, however, blocked the ANIT effect on bile flow (group III vs. IV or V). A critical rectal temperature seemed to exist at around 31 C. Below this tem-..

3 770 ROBERTS 'AND PLAA Vol. 50, No.6 TABLE 1. Effect of various treatments on body temperature and bile flow a Treatments b ANIT Control Group A B C D E Rectal temperature blocked/ Rectal temperature, blocked/ total total I X X X 26.0 ± 1.1 0/ ± 1.6 0/9 II X X X 26.5 ± 0.7 0/ ± 1.2 0/9 III X X 26.5 ± 0.7 3/ ± 1.8 1/9 IV X 31.5 ± 1.5' 10/12' 36.5 ± O.ld 0/9 V X 37.5 ± 0.8' 10/ ± 0.1 0/9 VI X X 38.0 ± 0.1 7/ ± /9 a Rectal temperature and bile flow determinations were conducted 24 hr after the administration of a-naphthylisothiocyanate (ANIT) (300 mg/kg, p.o.). Rectal temperatures are expressed as mean value ± SE. determinations are expressed as the number of animals showing cessation of flow per total number of animals examined by direct bile duct cannulation. b Animals were subjected to the indicated treatments during the 24-hr period prior to analysis: A, Bile duct cannulation just prior to ANIT administration; B, sham operation just prior to ANIT administration; C, restraint using individual restrainer; D, exposure to circulating room air created by an electric fan; E, maintained in 30 C incubator. C Significantly different from ANIT group III. d Significantly different from control group III. Significantly different from ANIT group IV. f Significantly different from control group V. 'C 'C perature the incidence of cessation of bile flow diminished; above this value, the incidence increased. In the process of determining bile flow it was observed that debris was present in the bile duct in about 75% of the ANIT-treated animals (excluding those made hypothermic) but never in the controls. In approximately 95% of these ANIT-treated animals, removal of this threadlike material, from the bile duct or cannula, or from both, failed to cause a flow of bile. In about 5%, removal did result in a flow of bile; these rats were classified as having bile flow. Microscopic evidence' of "bile plugs" following ANIT has been reported previously by other laboratories. 2, 14 Another interesting finding was the apparent hypothermia induced in control rats subjected to both restraint and cold air (group III). This effect seemed to be enhanced by superimposing an abdominal surgical procedure (groups I and II). Because of this effect we eliminated the use of the restrainers in the following experiments. Because the pilot study had shown an obvious influence of temperature on ANIT activity, a second study was designed to better assess this phenomenon. In this study all rats for one particular environment group were run simultaneously. Three experiments were required to obtain the data for. all three environmental conditions. Table 2 shows the effect of three environmental temperature conditions on rectal temperature, plasma bilirubin levels, and bile flow in control and ANIT-treated animals, both groups unrestrained. As can be seen, both hypothermia and hyperthermia were induced in the ANIT -treated groups but not in the control animals. The ANIT-induced hyperbilirubinemia and cessation of bile flow appeared to be temperature-dependent. This relationship between the response and body temperature is even more apparent in figure 1. Here the plasma bilirubin responses of the individual rats treated with ANIT are plotted against their respective rectal temperatures. A positive correlation was found to exist. The correlation coefficient was 0.8, which is statistically significant at the P < 0.05 level.

4 June 1966 A.NIT-INDUCED HYPERBILIRUBINEMIA 771 The effect of bile duct cannulation on ANIT -induced hyperthermia, cholestasis, and hyperbilirubinemia is shown in table 3. In this experiment the bile cannula was run under the skin of the back and exteriorized behind the head of the rat. The bile was allowed to flow freely to the outside for the 24 hi'; restraint was not necessary. The animals were maintained in a 30 C ambient environment, and all the rats were run simultaneously. In the sham-operated group, a laparotomy was performed and the bile duct cleared of connective tissue, but not cannulated. Both groups were given ANIT. It can be seen that cannulation significantly protected the rats from TABLE 2. Effect of enviromnent on rectal temperature, plasma bilirubin level, and bile flow following a-naphthylisothiocyanate (ANIT) administration G Environment Rectal tem- Plasma (no. perature bilirubin blocked/ no. treated) "C mg/ioo ml Cold temperature b Control ± ± 0.2 0/ 10 ANIT ± 2.0' 0.6 ± 0.3 2/ Room temperature Control ± ± 0.3 0/ 10 ANIT ± 2.4' 1.5 ± 0.3' 7/10< Incubation temperature Control ± ±0.1 0/ 10 ANIT ± 0.5' 3.2 ±0.7' 7/7' a Rectal temperature, plasma bilirubin level, and bile flow determinations were conducted 24 hi' after the administration of ANIT (300 mg/ kg p.o.). Results are expressed as mean v alues ± SE. determinations are expressed as the number showing cessation of bile flow per total number examined by direct bile duct cannulation. b Cold was induced by shaving and exposing the animals to circulating room air. All other treatments utilized unshaven animals., Significantly different from the corresponding control group. TABLE 3. Effect of bile du el cannulation on a-naphthylisothiocyanate (ALYIT) -induced hyperthermia, hyperbilirubinemia, and cholestasis Treated b Rectal Plasma. (no. temperature bilirubin blocked/ no. treated ) I C mg/ I OO IItI Sham operation + ANIT ± ± 0.3 5/ 6 Cannulation + ANIT ± 0.1' 0.2 ± 0.1' 0/ 6' a All analyses were completed 24 hi' foll owing ANIT administration. The results are expressed as mean values ± SE. determinations are expressed as the number showing cessat ion of bile flow per total number examined by direct bile duct cannulation. b Animals were cannulated or sham-operated just prior to ANIT administration (300 mg/kg, p.o.), following which they were incubated at 30 C for the duration of the experiment. There were six animals per treatment group., Significantly different from the group having sham operation + ANIT. ANIT -induced hyperthermia, cessation of bile flow, and hyperbilirubinemia. Table 4 shows the effect of bile duct ligation on the ANIT -induced responses. In these experiments the bile duct was ligated about 0.5 cm from the hilum just prior to ANIT administration. The sham-operated groups underwent laparotomy and clearing of the duct. Twenty-four hours after ANIT or CMC the rats were anesthetized and the proximal portion of the bile duct was cannulated to measure bile flow. Two experiments were performed, one at room temperature and one at 30 C; both groups were unshaven. Bile duct ligation significantly protected the ANIT -treated animals from cessation of bile flow at both temperatures. This procedure did not abolish the poikilothermic responses to ANIT, but it did cause a significant reduction in the hypothermia induced. In the incubation experiment, where rectal temperat ures were maintained at 37 C or above, ANIT treatment in the bile duct-ligated rats caused the bilirubin levels to rise significantly above those levels produced by bile duct ligation alone. In the room temperature

5 772 ROBERTS AND PLAA Vol. 50, No.6 TABLE 4. Effect of bile duct ligation on rectal temperature, plasma bilirubin level, and bile flow following a-naphthylisothiocyanate (ANIT) administration a Trea tmen t lj Rectal temperature Plasma bilirubin (no. blocked/ no. treated) C mg/100 ml Cold environment BDL + CMC...., ± ± 0.6 0/7 Sham operation + CMC.,....., ± ± 0.1 0/9 BDL + ANIT ± 1.0' 3.8 ± 0.6 1/8 Sham operation + ANIT ± ± 0.2 0/ Incubation environment Sham operation + ANIT ± 0.2d 1.0 ± 0.2' 7/9 1 BDL + ANIT ± 0.2" 5.0 ± O.4 k 1/10 BDL + CMC ± ± 0.2 0/9 Sham operation + CMC ± ± 0.2 0/10 a Analyses of rectal temperature, plasma bilirubin, and bile flow were conducted 24 hr after ANIT administration. Results are expressed as mean values ± SE. determinations are expressed as the number showing cessation of bile flow per total number examined by direct bile duct cannulation. b Bile duct ligations (BDL) and sham operations were performed just prior to ANIT administration (300 mg/kg, p.o.). The cold environment involved exposing the animals to circulating room temperature air. The incubation environment was created by conducting the experiment in a constant temperature chamber held at 30 C. CMC, Carboxymethylcellulose., Significantly different from BDL + CMC and sham operation + ANIT. d Significantly different from sham operation + CMC., Significantly different from sham operation + CMC. 1 Significantly different from BDL + ANIT. g Significantly different from BDL + CMC. h Significantly different from BDL + CMC. experiment, bile duct ligation seemed to prevent the hyperbilirubinemic response of ANIT; however, these animals were hypothermic. Discussion The data presented illustrate a number of different aspects in the action of ANIT. In addition to its effect on plasma bilirubin and bile flow, ANIT apparently interferes with temperature control in rats. Under the usual laboratory conditions this effect is reflected as a moderate drop in rectal temperature. However, under extreme environmental conditions (shaving and rapid air exchange) the hypothermia becomes quite pronounced. Conversely, when ANITtreated rats are maintained in a high ambient temperature, hyperthermia is induced. The exact etiology of the ANIT-induced poikilothermia is unclear. It was observed that the ANIT-treated animals under hypothermic conditions became lethargic and had difficulty in moving about. However, an inability of the animal to adapt to extreme environmental conditions simply because of inactivity as a result of illness does not seem to justify the degree of hypothermia observed. A more satisfactory explanation would seem to be an interference with those compensatory physiological mechanisms involved in temperature control. Also, since bile duct ligation alone failed to produce the hypothermic or hyperthermic response, it seems reasonable to conclude that cessation of bile flow itself, and the resulting hyperbilirubinemia, can be eliminated as the cause of the loss of control over body temperature seen following ANIT treatment. The poikilothermic effect of ANIT permitted us to demonstrate that the other responses to ANIT treatment-hyperbilirubinemia and cholestasis-are temperature-dependent. The data in figure 1 show that the degree of hyperbilirubinemia

6 June 1966 ANIT-INDUCED HYPERBILIRUBINEMIA 773 doubles when the rectal temperature increases by 10 C. Since the method of determining bile flow permitted us to measure only the presence or absence of bile flow, a quantitative comparison of bile flow vs. rectal temperature could not be made. However, the data presented do show that below a critical temperature of about 31 C the cholestatic response to ANIT cannot be elicited. This effect of temperature on hepatic response, and the added finding that ANIT itself modifies temperature control, might account for the relatively large variability in response to ANIT that is observed in rats and mice The temperature dependence of the hepatotoxic effect of ANIT is also reminiscent of the protective effect of hypothermia on carbon tetrachloride-induced liver injuryy A close relationship seems to exist between the number of animals showing cessation of bile flow and the extent of the hyperbilirubinemia. In each instance of marked hyperbilirubinemia, cessation of bile flow was also present. On the other hand, there does seem to be a hyperbilirubinemic phase which might be independent of bile flow. Bile duct-ligated animals receiving ANIT and maintained at 30 C had higher bilirubin levels than the control bile duct-ligated rats, although bile flow was still present. There exist several possible explanations for this higher level of plasma bilirubin in the ANITtreated, bile duct-ligated animals. It may be due to an increased production of bilirubin subsequent to ANIT treatment or to an effect on bilirubin storage and distribution. It may also represent an inhibitory effect of ANIT on the excretion of bilirubin by nonhepatic routes such as the kidneys. The over-all experimental procedures did not provide for an evaluation of kidney function or bilirubin synthesis, or for the detection of diminished bile flow. Therefore, before a causal relationship between bile flow and hyperbilirubinemia can be stated definitely, more conclusive evidence is needed. The presence of a normal enterohepatic pathway seems necessary for some of the actions of ANIT. Bile duct ligation prior to treatment prevented cessation of bile flow, even in animals with elevated rectal temperatures. Therefore, cholestasis may be due to an action exerted on the larger bile ducts, and ligation may be protective by preventing the distribution of ANIT to these areas. The bile cannulation studies are consistent with such an interpretation, since cannulation prior to treatment prevented both the hyperbilirubinemic and the cholestatic responses in normothermic animals. The protective effect of cannulation might be attributed either to removal of ANIT from the animal by preventing enterohepatic recirculation, or to prevention of the accumulation of an inflammatory exudate in the bile ducts. The observation that surgical interruption of enterohepatic pathways affects ANIT activity could also be taken as evidence for involvement of a metabolite of ANIT in the various responses. The temperature dependence of ANIT for its effects also points to this possibility. This hypothesis has been previously suggested by other investigators and by our laboratory However, nothing is known about ANIT metabolism, let alone whether a biotransformation product is responsible for its action. Another interesting facet uncovered in these experiments was the observation that restraint of the animals alone lowered body temperature about 5 C in both the control and ANIT-treated animals (table 1). In the control group, surgery enhanced the hypothermia. Since restraint is employed in a variety of studies, these results emphasize the need for caution when drawing conclusions from experiments which involve the use of restraint for relatively long periods of time. Summary The hyperbilirubinemic and cholestatic responses to a-naphthylisothiocyanate (ANIT) in the rat were altered by subjecting test animals to various environmental and surgical manipulations. Studies utilizing hypo- and hyperthermic conditions showed that a positive correlation exists between the rectal temperature and the effects of ANIT. In addition, it was observed that ANIT produced an apparent

7 774 ROBERTS AND PLAA Vol. 50, No.6 poikilothermic response, in that treated rats were unable to maintain normal rectal temperatures. Bile duct ligation inhibited the cholestatic and altered the hypothermic responses to ANIT treatment. Cannulation of the bile duct prior to AN IT administration significantly inhibited the hyperbilirubinemia and cessation of bile flow. These data demonstrate the importance of an intact enterohepatic circulation and normal body temperature in the actions of ANIT. The effect of the various parameters on the ANIT-induced hyperbilirubinemia and cholestasis suggests the involvement of a biotransformation product of ANIT. REFERENCES 1. Goldfarb, S., E. J. Singer, and H. Popper Experimental cholangitis due to alphanaphthyl-isothiocyanate (ANIT). Amer. J. Path. 40: Ungar, H., E. Moran, M. Eisner, and M. Eliakim Rat intrahepatic biliary tract lesions from alpha-naphthylisothiocyanate. Arch. Path. (Chicago) 73: Griffiths, D. B., K. R Rees, and K. P. Sinha Blood and bile composition in experimental biliary cirrhosis. J. Path. Bact. 82: Eliakim, M., M. Eisner, and H. Ungar Experimental intra-hepatic obstructive jaundice following ingestion of alpha-naphthyliso-thiocyanate. Bull. Res. Counc. I srael (Exp. MedJ 8: Steiner, J. W., M. J. Phillips, and C. M. Baglio Electron microscopy of the excretory pathways in the liver in alphanaphthyl isothiocyanate intoxication; a study of intrahepatic cholestasis. Amer. J. Path. 43: McLean, M. R, and Ii. R. Rees Hyperplasia of bile ducts induced by alphanaphthyl-iso-thiocyanatc; experimental biliary cirrhosis free from biliary obstruction. J. Path. Bact. 76: Phillips, M. J., and J. W. Steiner Comparative study of (X-naphthyl isothiocyanate induced liver injury. Lab. Invest. 13: Moran, E., M. Eliakim, and H. Ungar The effect of cortisone on cholangitis induced by a-naphthyl isothiocyanate in rats. Gastroenterology 48 : Roberts, R J., and G. L. Plaa Potentiation and inhibition of alpha-naphthylisothiocyanate-induced hyperbilirubinemia and cholestasis. J. Pharmacol. E xp. Ther. 150: Roberts, R J., and G. L. Plaa Effect of norethandrolone, acetohexamide and Enovid on alpha-naphthylisothiocyanateinduced hyperbilirubinemia and cholestasis. Biochem. Pharmacal. In press. 11. Ferro, P. R, and A. B. Ham A new colorimetric method for determination of total serum bilirubin. II. Micro method. Amer. J. Clin. P ath. 44 : Goldstein, A Biostatistics. The Macmillan Company, New York. 13. Siegel, S Nonparametric statistics for the behavioral sciences. McGraw-Hill Book Company, Inc., New York. 14. Moran, E., and H. Ungar The effect of intermittent administration of alpha-naphthyl isothiocyanate to rats. Amer. J. Path. 44: Garay, E. R, V. Perez, B. Noir, and M. Royer Chronic intoxication induced by alphanaphti)ylisothiocyanate in the rat. Arch. Path. (Chicago) 80: Becker, B. A., and G. L. Plaa Quantitative and temporal delineation of various parameters of liver dysfunction due to (Xnaphthylisothiocyanate. Toxic. App!. Pharmacol. 7: Larson, R E., and G. L. Plaa A correlation of the effects of cervical cordotomy, hypothermia, and catecholamines on carbon tetrachloride-induced hepatic lesions. J. Pharmacal. Exp. T her. 147 :