A. Preparation of the experimental models Healthy adult mongrel dogs of both sexes

Transcription

1 Gastroenterologia Japonica Copyright by The Japanese Society of Gastroenterology Vol 26, No Printed in Japan Experimental study on the pathogenesis of acute acalculous cholecystitis, with special reference to the roles of microcirculatory disturbances, free radicals and membrane-bound phospholipase A2 Hiroki TAOKA First Department of Surgery, Mie University School of Medicine, Mie, Japan Summary: To elucidate the pathogenesis of acute acalculous cholecystitis, the gallbladder was subjected to ischemia-reperfusion by simultaneously occluding the middle hepatic artery and the superior mesenteric vein in dogs, and the degree of inflammation and biochemical changes in the gallbladder mucosa were studied by varying the duration of ischemia or reperfusion Ischemia alone did not induce cholecystitis either macroscopically and histologically, although it increased phospholipase A2 (PLA2) activity, content of lipid peroxide, and superoxide dismutase (SOD) activity in the mucosa with prolongation of the ischemic time Cholecystitis was produced in all animals by 4-min ischemia followed by 90-min reperfusion as the shortest ischemia and reperfusion times In this model, prolongation of the ischemic time increased the area of mucosal inflammation horizontally with increases of the PLA2 activity, content of lipid peroxide, and SOD activity, whereas by prolonging the reperfusion time the inflammation area spread deeper vertically toward the serosal side with significant increase in the mucosal PLA2 activity, content of lipid peroxide, and SOD activity These results revealed that ischemia-reperfusion plays an important role in the pathogenesis of acute acalculous cholecystitis, causing the generation of free radicals and the activation of membrane-bound PLA2 Gastroenterol Jpn 1991;26: Key words: ischemia-reperfusion injury; lipid peroxide; radical scavenger; superoxide dismutase Introduction Acute acalculous cholecystitis a'2, which usually occurs after operation or trauma, is known to have a sudden onset and easily causes necrosis or perforation of the gallbladder, resulting in a serious outcome Since the majority of the patients with this disease are associated with systemic vascular diseases 3 such as atherosclerosis, diabetes mellitus and heart diseases, microcirculatory disturbance in the gallbladder has been considered to play an important role in the pathogenesis However, the precise pathogenesis remains unsolved Recently, free radical reactions and lipid peroxidation have assumed great importance in the pathophysiology of microcirculatory disturbances, and furthermore the activation of membrane-bound phospholipase A2 (PLA2) has been recognized to play a key role in the development of cell injury during ischemia The purpose of this experimental study was to clarify the pathogenesis of acute acalculous cholecystitis, induction of which was attempted in adult mongrel dogs by exposing the gallbladder to ischemia-reperfusion, paying special interest to the roles of free radicals and membrane-bound PLA2 Materials and Methods A Preparation of the experimental models Healthy adult mongrel dogs of both sexes Received November 12, 1990 Accepted May 17, 1991 Address for correspondence: Hiroki Taoka, MD, First Department of Surgery, Mie University School of Medicine, 2-174, Edobashi, Tsucity, Mie 14, Japan

2 634 H Taoka Vol 26, No Table 1 Experimental materials Experiments n Groups (n) I 2 Outline of the experiments Used vessels Ischemic Time Reperfusion Time I-a (10) MHA 4 min 90 min I-b (1) MHA + SMV 4 min 90 min II-a (26) MHA + SMV 4 min - 7 days 0 II 86 II-b (30*) MHA + SMV 30 min - 90 min 90 min II-c (30*) MHV + SMV 4 min 90 min ~ 72 hours II1"* 26 III-a () Infusion of the physiological saline III-b (7) " SOD III-c (7) " CAT III-d (7) " SOD + CAT *: Of the 30 dogs, 1 dogs in Group II-b and Group II-c were identical to the dogs in Group I-b **: The dogs prepared in Group I-b were used as the basic model in Experiment II1 MHA: middle hepatic artery, SMV: superior mesenteric vein weighing about 1 kg were used Prior to the experiments, they were maintained on dog food for at least 2 weeks at the Institute of Laboratory Animals, Mie University School of Medicine After fasting for 24 hours, they were intubated endotracheally, laparotomized by an abdominal midline incision, and used for the following examinations under intravenous anesthesia with Nembutal (2 mg/kg body weight) All dogs had been examined under intravenous anesthesia within 24 hr after the beginning of the experiment After 24 hours, they were given free access to food and water and examination of gallbladder wall was performed by sacrificing them 1 Preliminary study The supplying arteries to the liver, examined in 20 dogs, were the left hepatic, right hepatic, and middle hepatic arteries, all of which originated from the common hepatic artery The cystic artery originated form the middle hepatic artery in all animals examined In a preliminary study, the cystic artery or the arterial and portal systems flowing into the gallbladder were blocked to explore a method of achieving complete gallbladder ischemia, as follows The cystic artery alone was blocked by ligation in Group 1 (n=), the middle hepatic artery was blocked in Group 2 (n=), the cystic artery and main portal vein were simultaneously blocked in Group 3 (n=), and the middle hepatic artery and superior mesenteric vein were simultaneously blocked in Group 4 (n=) As the control group, normal dogs with sham operation without any blocking blood supplies to the gallbladder and liver were used The hydrogen electrode of a hydrogen gas clearance tissue flow-meter (DHM3001, M-T Technical Research, Tokyo, Japan) was inserted into the mucosa of the body of the gallbladder from the peritoneal side, and the tissue blood flow of the gallbladder mucosa was measured in each group 2 Ischemia-reperfusion study Three experiments were designed to elucidate the pathogenesis of acute acalculous cholecystitis (Table 1) According to the results of a preliminary study, Experiment I was performed to determine whether Group I-b could be used as the basic model of a experimental cholecystitis or not as compared with Group I-a The purpose of Experiment II was to investigate the degree of inflammation of the gallbladder by changing the ischemia or reperfusion times In Group II-a, only the ischemic times were changed In Group II-b, the ischemic time was varied with a fixed reperfusion time In Group II-c, the ischemic time was fixed with a varied reperfusion time Using Group I-b in Experiment I as a basic model, various radical scavengers were infused into the gastroduodenal artery to compare the effects of these agents on the development of cholecystitis in Experiment III

3 October 1991 Pathogenesis of acute acalculous cholecystitis 63 and superior mesenteric vein In Group II-a, the ischemic time was varied from 4 min (n=) to 60 rain (n=4), 90 rain (n=4), 180 min (n=), 24 hr (n=), and 7 days (n=3) without reperfusion In Group II-b, the ischemic time was varied from 30 min (n=) to 4 rain (n=1), 60 min (n=), and 90 min (n=) with a fixed reperfusion time of 90 min In Group II-c, the reperfusion time was varied from 90 min (n=1) to 180 min (n=), 24 hr (n=), and 72 hr (n=) following a fixed ischemic time of 4 min The dogs with 4-rain ischemia in Group II-b and those with 90-min reperfusion in Group II-c were identical to Group I-b in Experiment I Fig 1 Experimental model a: vascular occluder on the middle hepatic artery b: vascular occluder on the superior mesenteric vein c: arterial cannula for injection of radical scavengers into the gastroduodenal artery d: The right gastric artery was ligated and cut a Experiment I The gastroduodenal artery and the right gastric artery were transected after ligation very carefully in order not to injure the common bile duct and the cystic duct After sufficient devascularization of the hepatoduodenal ligament, two vascular occluders (Bioresearch Center) were applied to the middle hepatic artery and superior mesenteric vein, taking care not to injure them (Fig 1) The middle hepatic artery alone was occluded for 4 min, followed by reperfusion for 90 min in Group I-a (n=10), while both the middle hepatic artery and superior mesenteric vein were simultaneously blocked for 4 min, followed by 90-min reperfusion in Group I-b (n= 1) No significant changes in the serum GOT and GPT levels were observed even after these procedures, and thus the effects on the liver were considered to be negligible b Experiment II The degree of inflammation of the gallbladder was studied by changing the ischemia or reperfusion times of both the middle hepatic artery c Experiment III Using a model corresponding to Group I-b in Experiment I, physiological saline and various radical scavengers were infused through a 4 Fr arterial catheter inserted into the gastroduodenal artery during reperfusion to compare the effects of these agents on the development of cholecystitis Amount of physiological saline (0 ml), superoxide dismutase (SOD, Nippon Kayaku Tokyo, Japan) or 0,000 U/kg body weight/0 ml physiological saline in Group III-b (n=7), catalase (CAT) (Sigma St Louis, MO) of 100,000 U/kg body weight/0 ml physiological saline in Group II-c (n=7), and a mixture of SOD (0,000 U/kg body weight/2 ml physiological saline) and CAT (100,000 U/kg body weight/2 ml physiological saline) in Group III-d (n=7) were infused in Group III-a (n=) B Examination and assays 1 Gross appearance and stereoscopic microscopy Immediately after sacrificing the animals, the gallbladder was removed with a part of its liver bed to prevent it from damage and was opened longitudinally along the surface facing the abdominal cavity The gallbladder mucosa was grossly examined and the horizontal extent of cholecystitis was calculated by the following formula for the inflammatory index (II) II = Inflammation area of gallbladder mucosa x 100 Total area of gallbladder mucosa

4 636 H Taoka Vol 26, No Under a stereoscopic microscope, microstructures of the gallbladder showing gross inflammation were examined at a magnification of x Light microscopy Gallbladders obtained at sacrifice were fixed with 10% formalin, embedded in paraffin, sectioned and stained with hematoxylin-eosin (HE) solution for histological examination Furthermore, the vertical extent of inflammation in the gallbladder wall was evaluated using the following -Grade scale, according to a modification of Lancy's classification4; Grade 1: normal mucosa, Grade 2: inflammatory cell infiltration within the mucosa, leaving the mucosal structure relatively intact, Grade 3: inflammatory cell infiltration to the proper muscular layer and disturbance of the papillary structure of the mucosa, Grade 4: marked inflammatory cell infiltration extending to the subserosal tissue with mild loss and degeneration of the mucosa, Grade : marked inflammatory cell infiltration extending to the serosa with severe loss and degeneration of the mucosa 3 Biochemical evaluation of the gallbladder mucosa The area of gallbladder mucosa showing gross inflammation were scratched off with a glass slide, and then placed in various buffers or distilled water, homogenized in a Hiscotron (NS10, Nippon Irika), and centrifuged (20PR-, Hitachi, Tokyo, Japan) at,000 rpm at -~ for 30 min The supernatant obtained was used for the assays of PLA2 activity, content of lipid peroxide and SOD activity 1 PLA2 activity: A buffer used for assay was 01 M glycine and 00 M borate containing 2 mm CaCI2 The PLA2 activity in the supernatant was measured by a radiochemical method using lecithin labeled with 14C at the/3 position of the fatty acid as the substrate 2 Lipid peroxide: The content of lipid peroxide was measured by the thiobarbituric acid reaction 6 using 00 M phosphate buffer 3 SOD activity: The mucosa was homogenized with distilled water The SOD activity in the su- pernatant was measured with an electron spin response (ESR) spectrometer (JES-RE3X, JOEL) by the spin trap method 7 using,- dimethyl- 1-1-pyrroline-l-oxide (Labotech) The PLA2 activity, content of lipid peroxide and SOD activity in the supernatant ware all corrected for the protein content determined by Lowry's method s C Statistical analyses All results were expressed as the mean+standard deviation (M+SD), and the differences were examined by Scheffe's multiple comparison test, defining P<00 as significance Results 1 Preliminary study The tissue blood flow of the gallbladder mucosa was ml/min/100 g in the normal control group Even when the cystic artery alone (Group 1) or the middle hepatic artery alone (Group 2) was blocked, the tissue blood flow showed ml/min/100 g and ml/min/100 g, respectively, suggesting compensatory blood supply to the gallbladder from the portal vein However, the mucosal blood flow became zero when both the cystic artery and main portal vein were simultaneously blocked (Group 3) (Fig 2) Moreover, when the middle hepatic artery and superior mesenteric vein were simultaneously blocked with maintaining the flow to the liver from the gastrosplenic veins, the mucosal blood flow became approximately zero within the range of measurement (Group 4) In addition, all dogs with simultaneously blocked cystic arteries and main portal veins died within 12 hrs Therefore, in the subsequent experiments dogs with simultaneous blocking of the middle hepatic artery and superior mesenteric vein, all of which survived for more than 7 days, were used to achieve gallbladder ischemia 2 Ischemia-reperfusion study a Experiment I (Table 2) 1) Inflammation of the gallbladder Inflammation of the gallbladder mucosa was

5 October 1991 Pathogenesis of acute acalculous cholecystitis 637 B 4 -"'T", tim (min) D Fig 2 The blood flow of the gallbladder mucosa was measured by using a hydrogen gas clearance tissue flow-meter The hydrogen gas clearance curves of typical cases A: Control; a normal dog without vascular occlusion B: Group 1; a dog with occlusion of the cystic artery alone C: Group 2; a dog with occlusion of the middle hepatic artery alone D: Group 3; a dog with both the cystic artery and the main portal vein occluded ~,, tim (lin), o, o, 9 Table 2 Inflammatory index (II) and PLA2, lipid peroxide, SOD in the mucosa of gallbladder in Experiment I lipid peroxide Group n II (%) PLA2 (U/mgprot) (n mol/mgprot) SOD (U/mgprot) I-a _ _+8 I-b 1 243_ _+083 **++ 18_+ 10 **++ 180_+70 **++ control 0 130_ _ _+ 109 **P<001 vs control, ++P<001 vs I-a Group I-a: Middle hepatic artery was interrupted for 4 minutes, followed by 90-minute reperfusion Group I-b: Both the middle hepatic artery and superior mesenteric vein were interrupted for 4 minutes, followed by 90-minute reperfusion control: normal dogs not grossly evident in the dogs of Group I-a, subjected to 4-min ischemia and 90-min reperfusion by blocking of the middle hepatic artery alone However, apparent cholecystitis accompanied by congestion and edema with II of 243_+922% was observed in all dogs in Group I-b, subjected to 4-min ischemia and 90-min reperfusion by the simultaneous blocking of the middle hepatic artery and superior mesenteric vein Stereoscopic microscopy of the gallbladder revealed disturbance of the fine papillary structures in the sites of gross inflammation as compared with the normal mucosa Histologically, inflammatory cells infiltrated into the mucosa, and mild edema was observed (Grade 2) 2) Biochemical changes in the gallbladder mucosa The PLA2 and SOD activities in the gallbladder mucosa were and U/ mgprot (units omitted hereafter) in Group I-a, which were not significantly different from and , respectively, in the normal control group However, they increased signifi-

6 638 H Taoka Vol 26, No Table 3 PLA2, lipid peroxide, and SOD in the mucosa of gallbladder in Group II-a with interruption of both middle hepatic artery and superior mesenteric vein for various times without reperfusion in Experiment II Group Ischemic Time n II (%) PLA2 (U/mgprot) (n lipid mol/mgprot) peroxide SOD (U/mgprot) II-a 4 min 0 142_+03 03_ _ min _ _+043 1,82_ min _ _ , min _ ,01 _ hours 0 482_+096** 320_+086** 1622_ days _+18** 260_+063** 123_+67 control _ _+ 109 **P<001, vs control, control; normal dogs cantly to and , respectively, in Group I-b with significant differences (P<001) as compared with those in the control group The content of lipid peroxide increased in both groups, and especially is was 18_+10 n mol/mgprot (units omitted hereafter) in Group I-b, being significantly higher (P<001) than 036_+023 in the control group The PLA2 activity, content of lipid peroxide, and SOD activity were all significantly different (P<001) between Groups I-a and I-b b Experiment II (Tables 3, 4, and ) The degree of inflammation and the biochemical changes in the gallbladder were studied by changing the ischemia and reperfusion times of the middle hepatic artery and superior mesenteric vein 1) Inflammation of the gallbladder (1) Area of inflammation (Inflammatory Index; II) Group II-a: When the ischemic time was varied from 4 min to 60 min, 90 min, 180 min, 24 hr and 7 days without reperfusion, white discolorations of the mucosa were grossly observed without any signs of inflammation even after prolonged ischemia However, the fine papillary structures of the mucosa disappeared, and the gallbladder wall thinned with time Group II-b: When the ischemic time was varied from 30 min to 4 rain, 60 min, and 90 min with the reperfusion time fixed at 90 min, the area of inflammation gradually increased with prolongation of the ischemic time, demonstrating the II of 243+_922% at 4 rain, 6868_+1021% at 60 min, and % at 90 rain, although an apparent inflammation was not observed at 30 min, showing the II of (Table 4, Fig 3) Group II-c: Even when the reperfusion time was increased from 90 rain to 180 min, 24 hr, and 72 hr with the ischemic time fixed at 4 min, the II of the gallbladder remained within the range of %, showing 631_+ 1867% even after 72-hr reperfusion (Table, Fig 4) (2) Depth of inflammation The depth of inflammation of the gallbladder was examined histologically In Group II-a, even when the ischemic time was prolonged, inflammatory cell infiltration and edema were scarcely noted in the gallbladder, mucosa, although the gallbladder wall became thin due to necrosis or exfoliation of the mucosal epithelium In Group II-b, the II increased with prolongation of the ischemic time as mentioned above, but the depth of inflammation was relatively unremarkable and it was Grade 3 (the inflammatory cell infiltration limited to the proper muscular layer) in the majority of animals In Group II-c, the vertical spread of inflammation became more remarkable with prolongation of the reperfusion time Inflammatory cell infiltration and edema were limited at the mucosa (Grade 2) after 90-min reperfusion, but inflammation extended to the proper mus-

7 Pathogenesis of acute acalculous cholecystitis October Fig 3 Macroscopic findings of the gallbladder after reperfusion for 90 minutes following ischemia for various times in group II-b in Experiment II Ischemic time: a; 30 min, b: 4 min, c; 60 min, d; 90 min The inflammatory index increased with prolongation of the ischemic time, see Table 3 Table 4 Inflammatory index (II) and PLA2, lipid peroxide, SOD in the mucosa of gallbladder in Group II-b with interruption of both middle hepatic artery and superior mesenteric vein for various times, followed by 90-minute reperfusion in Experiment II Group Ischemic Time n 30 min II-b control 4 min 60 min 90 min 11 (%) 62_ _ _ _ SOD (U/mgprot) peroxide (nlipid mol/mgprot) PLA2 (U/mgprot) _+72** ** 333_+010"* 443_+122** 18_+10** "* 28_+098** _ _+70"* 22_+832** 2783_+836** 83_+ 109 *P<00, **P<001 vs control, control; normal dogs cular layer (Grade 3) after 180-min reperfusion and it deepened to Grades 4 and with increasing reperfusion time to 24 and 72 hr In the animals reperfused for 72 hr, abscess formation was observed together with complete necrosis of the mucosa (Fig ) 2) Biochemical changes in the gallbladder mucosa Group II-a (Table 3): The PLA2 activity in the gallbladder mucosa was after 4-min ischemia alone, which was not significantly different from the control group (130_+092) However, it increased gradually with prolongation of the

8 640 Vol 26, No H Taoka Fig 4 Macroscopic findings of the gallbladder mucosa after reperfusion for various times following 4 minutes of ischemia in group II-c in Experiment I1 Reperfusion time: A; 90 min, B; 180 min, C; 24 hours, D; 72 hours Table Inflammatory index (II) and PLA2, lipid peroxide, SOD in the mucosa of gallbladder in Group II-c with interruption of both middle hepatic artery and superior mesenteric vein for 4 minutes, followed by reperfusion for various times in Experiment II Group II-c Reperfusion Time n II (%) PLA2 (U/mgprot) peroxide (nlipid mol/mgprot) SOD (U/mgprot) 90 min 180 min 1 243_ _ _ _ + 10"* * "* ** 24 hours 72 hours control 636_ _ ** ** 280_+ 106** 17_ ** 1428_+69* *P<00, **P<001, vs control, control; normal dogs ischemic time and became significantly increased to 482_+096 at 24 hr, and 74_+18 at 7 days (P<001) Changes in the content of lipid peroxide were similar to those in the PLA2 activity The SOD activity increased to about twice the normal level, but there was no significant difference Group II-b (Table 4): When the reperfusion time was fixed at 90 min, the PLA2 activity and content of lipid peroxide in the gallbladder mucosa gradually increased with prolongation of the ischemic time, showing significant increases (P< 001) to 27_+083 and 18_+10, respectively,

9 October 1991 Pathogenesis of acute acalculous cholecystitis 641 Fig Histological findings of the gallbladder after reperfusion for various times following 4 minutes of ischemia in group II-c in Experiment I1 Reperfusion time: A; 90 min, B; 180 min, C; 24 hours, D; 72 hours Polymorphonuclear leukocyte infiltration into the gallbladder wall increased with prolongation of the reperfusion time Severe edema and exfoliation of the surface epithelium were found in the groups with reperfusion for 24 hours and 72 hours after 4-min ischemia as compared with the control value, although there were no significant increases after 30 in ischemia The increases in these values were also significant at much longer ischemic times The SOD activity, on the other hand, showed a peak value of 3022_+72 after 30-min ischemia, fell after 4 min of ischemia and increased again with further prolongation of the ischemic time Group II-c (Table ): When the ischemic time was fixed at 4 min, the PLA2 activity, content of lipid peroxide, and SOD activity in the gallbladder mucosa all increased with prolongation of the reperfusion time up to 24 hr, but the values after 72-hr reperfusion were lower than those after 24-hr reperfusion c Experiment II! (Table 6) Since simultaneous blocking of the middle hepatic artery and superior mesenteric vein for 4 min followed by 90-min reperfusion was the shortest ischemia and reperfusion times to produce cholecystitis macroscopically in all animals in Experiment II, animals prepared by this method were used as the basic model in Experiment III 1) Inflammation of the gallbladder In Group III-a with infusion of physiological saline as the control, the II of the gallbladder was 872_+836% This was significantly reduced to 4276_+ 322% and 4371 _+ 91 %, respectively, in Group III-b with SOD infusion alone and Group III-c with CAT infusion alone (P<00) On the other hand, in Group III-d with the infusion of a mixture of SOD+CAT, the II decreased significantly (P<001) to 2240_+794% as compared with Group III-a (Table 6) By

10 642 H Taoka Vol 26, No Table 6 Inflammatory index (II) and PLA2, lipid peroxide, SOD in the mucosa of gallbladder in Experiment Ill with interruption of both middle hepatic artery and superior mesenteric vein for 4 minutes, followed by 90-minute reperfusion lipid peroxide Group n II (%) PLA2 (U/mgprot) (n mol/mgprot) SOD (U/mgprot) Ill-a: saline _ _ _+48 Ill-b: SOD _+322* 214_ "* Ill-c: CAT _+91" 226_ _ _+226 Ill-d: SOD + CAT _+794** 183_ * ** *P<00, **P<001, vs Group III-a In Groups Ill-b, Ill-c, and Ill-d, radical scavengers (SOD, CAT) were infused into the gastroduodenal artrery during reperfusion ATP ~ un~iu A e dd, y~r*pr~ m~nosil inolim u*tkim OXi~le J, ~,~, bypolmnthitm t _ O~ 0~- "~'~V-'~i HzO,~H,O 1<1 Cu*,,u:"]"": 2+ [ I'GSH :P;I REPERFUSION 7 Fe~t'+ PLAt, [ ~ fatty acid I :,,,,gf~,: ~4 ~, o:, H,O I ~ proltlandin --r (I~E,,I~,,)~ am~-alinal/ ~ q I 9169 I \ \ ~ ~ ' ---~, "ITXA,I- 3 ",! o,yllol~ithin ~ " I ~ ~ J ~ leultotorimle ~lmrlmmm~,,w~ i'2~_'~'_'"_3 Fig 6 The biochemical mechanism of ischemia-reperfusion injury stereoscopic microscopy, disturbances of the fine papillary structure of the mucosa were noted in Groups III-b and III-c, as in Group III-a In Group Ill-d, the gallbladder mucosa showed no such changes and appeared similar to the normal gallbladder mucosa Concerning depth of inflammation, the mucosal epithelium was relatively intact in Groups III-b and III-c, as in Group Ill-a, but inflammation was noted a Grade 2 to 3 In Group Ill-d, such changes were not evident, and the appearance of the mucosa was similar to that in the normal gallbladder 2) Biochemical changes in the gallbladder mucosa The PLA2 activity in the gallbladder mucosa J ranged from in the groups with infusion of SOD and/or CAT (Groups III-b, III-c, III-d), which was not significantly different from in Group III-a with infusion of physiological saline The content of lipid peroxide was in Group III-b and in Group Ill-c, showing no difference compared with the value of in Group III-a However, it was significantly lower in Group III-d (08+022, P<00) In groups III-b and Ill-d, in which SOD was infused, the SOD activity was and , respectively The levels were significantly higher (P<001, respectively) than in Group III-a ( ) (Table 6) Discussion With the recent advances in the studies of cell injury due to microcirculatory disturbances, free radical chain reactions of activated oxygen and lipid radicals have emerged as the primary mediators of the cell injury induced by ischemia-reperfusion On the basis of the free radical chain reaction, a lot of attempts have been made to explain the pathophysiology of various diseases, such as the stomach, large intestine, liver, spleen, and brain diseases Release of oxygen-flee radicals from stimulated polynuclear leukocytes and aggravation of ischemic injury by superoxide radical (02-) produced during reperfusion have been demonstrated in acute inflammatory diseases, in which migration of polynuclear leukocytes is observed 9 Isaji et al 1~ reported that polyunsaturated fatty acids produced by the ac-

11 October 1991 Pathogenes& of acute acalculous cholecystitis 643 tivation of membrane-bound PLA2 are involved in ischemic cell injury The present study investigated the pathogenesis of acute acalculous cholecystitis from this perspective in the hope of establishing a method of treatment for this disease Experimental acute acalculous cholecystitis was invariably produced by simultaneously blocking the middle hepatic artery and superior mesenteric vein for 4 min followed by reperfusion for 90 min, as the shortest ischemia and reperfusion times The PLA2 activity, content of lipid peroxide, and SOD activity in the gallbladder mucosa were significantly increased in this cholecystitis model compared with the levels in normal control dogs, indicating the involvement of free radicals and membrane-bound PLA2 in gallbladder mucosal damage Mersereau et al 11 examined changes in the gastric mucosa associated with ischemia in vivo by the chamber technique, and revealed that the lesions began with whitish discoloration of the mucosa Watt et al 12 reported that cells on the surface of mucosa showing this whitish discoloration were necrotic Sies 13 noted gradual degeneration of cells due to anoxia during ischemia, in addition to tissue injury by free radicals during reperfusion Therefore, these biochemical changes and cell injuries during ischemia and reperfusion are considered to be closely interrelated In Experiment II, the gallbladder mucosa showed degeneration with whitish discoloration in Group II-a subjected to ischemia alone Histologically, the mucosal epithelium of the gallbladder was necrotic and desquamated, but little inflammatory cell infiltration or edema was observed When the gallbladder was reperfused after ischemia, the development of cholecystitis was grossly and histologically demonstrated In Group II-b (reperfusion time fixed to 90 min), the II increased with prolongation of the ischemic time, but the vertical spread of inflammation of the gallbladder wall was limited to the proper muscular layer in most animals In Group II-c (ischemic time fixed to 4 min), the II did not vary markedly, but the depth of inflammatory change increased with prolongation of the reper- fusion time until the mucosa became necrotic and abscesses formed Haglund et al 14 produced ischemia in the cat small intestine by blocking the superior mesenteric artery for 120 min and observed vertical spread of inflammation with prolongation of reperfusion Hemorrhage and ulceration with nearly complete exfoliation of the villous epithelium were seen after 60-min reperfusion With regard to biochemical findings in the gallbladder mucosa in Experiment II, the PLA2 activity and content of lipid peroxide increased with time in Group II-a, although the SOD activity showed no clear differences according to the duration of ischemia In Group II-b, the PLA2 activity and content of lipid peroxide increased with prolongation of the ischemic time, and the SOD activity showed the peak value after 30-min ischemia, but it was reduced after 4-min ischemia and increased again with further prolongation of the ischemic time In Group II-c, the PLA2 activity, content of lipid peroxide, and SOD activity all increased with prolongation of the reperfusion time until 24 hr These findings suggest the following mechanism of reperfusion injury in ischemic tissues Tissue ischemia, in general, caused a decrease in the production of ATP with an increase in AMP levels by disturbances of oxidative phosphorylation Increased AMP is degraded in a hypoxic condition into hypoxanthine or xanthine, from which 027 and H202 are produced by the action of xanthine oxidase These are reduced by transitional metal ions (Fe 2+, Cu +) into HO, which promotes the lipid peroxidation responsible for the oxidative injury of ischemic tissues during reperfusion Membrane-bound PLA2 is present on the surface of the cell membrane of various organs as well as the organelles, and is activated by various stimulations such as ischemia, with release of polyunsaturated fatty acids including arachidonic acid A part of these polyunsaturated fatty acids are peroxidized by enzymes including cyclooxygenase into endoperoxides (PGE2, PGI2), and thereafter various prostanoids with cytoprotective or inflammation-enhancing actions are synthesized However, most polyunsatu-

12 644 H Taoka Vol 26, No rated fatty acids are immediately auto-oxidized to lipid peroxides, thereby leading to a vicious cycle of free radical chain reactions resulting in cell necrosis In the present study, a high activity of membrane-bound PLA2, which is activated primarily by ischemia, was observed even during reperfusion This may be explained by continuation of the conditions induced by ischemia even after reperfusion and activation of membranebound PLA2 by lipid radicals ~ In the present experiments, there were variable changes in the value of SOD activity Ischemia of the gallbladder in Group II-a increased SOD activity up to twice the normal level, although prolongation of ischemic time did not increase it further Reimer et al 16 reported that free radicals were produced even when arachidonic acid metabolized into prostaglandins and leukotrienes Egan et a117 stated that SOD promoted the synthesis of prostaglandins These findings support our results of high SOD activity during ischemia only Superoxide radical (O2-) is almost always produced in the reperfusion period and SOD is the enzyme to dimutate O2- into H202 In this experiment, the increased SOD levels in Group II-b and II-c was considered to be related to reperfusion injury mainly to protected against free radical toxicity However, the relation between SOD activity and inflammatory index could not be clearly demonstrated, then the effect of radical scavengers (SOD, CAT) was confirmed in Experiment III Moreover, as the involvement of free radicals in the pathogenesis of various diseases has become more evident, drugs for their treatment and prevention based on elimination of free radicals have been developed, and clinical 18 as well as experimental evaluations of such drugs are in progress today The present study revealed that free radicals were eliminated very efficiently by the intraarterial infusion of a mixutre of SOD and CAT, clearly suppressing the development of cholecystitis Abstracts of this study were presented at the 30th and 31st Annual Meeting of the Japanese Society of GastroenteroLogy and the 2th Annual Meeting of the Japan Biliary Association The author expresses his sincere gratitude to Professor Ryuji Mizumoto for the valuable guidance and encouragement throughout the present work The instruction by Assistant professor Yoshifumi Ogura, First Department of Surgery, Mie University School of Medicine and cooperation by members of the department are deeply appreciated References 1 Gleen F: Acute cholecystitis following the surgical treatment of unrelated disease Ann Surg 1947;126: Thomas CG, Womack NA: Acute cholecystitis, its pathogcnesis and repair Arch Surg 192;64: More BM, Mcrdinger WF, Sommers SC, et al: Cholecystitis and stenotic arteriosclerosis Am J Clin Path 1966;4: Lancy ER, lto S: Microscopic analysis of ethanol damage to rat gastric mucosa after treatment with a prostaglandin Gastroenterology 1982;83: Isaji S: Pathogenesis of fat necrosis in acute pancreatitis, with special refercnce to the role of pancreatic phospholipase A2 Mie Med J 198;3: Yagi K: A simple fluorometric assay for lipidperoxide in blood plasma Biochem Med 1976;1: Hiramatsu M, Kouno M: Determination of supcroxide dismutase activity by Electron Spin Resonance Spectrometry using spin trap method JEOL NEWS 1987;23:7-9 8 Lowry OH, Rosenbrough N J, Farr AL, et al: Protein measurement with the Folin phenol reagent J Biol Chem 191;193: Yoshikawa T, Kondo M: Biochemical mechanisms and clinical findings of free radicals Nipponrinsho 1988;46: (in Japanese) 10 lsaji S, Taoka H, Mizumoto R, et al: The role of lipid peroxide and phospholipase A2 in acute pancreatitis, with special interest to the pathogenesis of fat necrosis, respiratory and renal failures In: Tsuchiya M, ed Free radicals in disease Excepta Medica, Amsterdam 1988; Mersereau WA, Hinchey EJ: Effect of gastric acidity on gastric ulceration induced by hemorrhage in the rat, utilizing a gastric chamber technique Gastroenterology 1973;64: Watt J: The mechanism of histamine ulceration in the guinea pig Gastroenterology 199;37: Sies H: Oxidative stress Translation Copyright 1st ed Japan Scientific Societies Press, Tokyo 1987; Haglund U, Arvidsson S, Schoenberg H: Oxyradicals and acute gastrointestinal mucosal damage In: Novelli GP, ed Oxigen Free Radicals in Shock 1st ed Karger publication, Florence 1986; Van Kunk FJG, Handelman GJ, Dratz EA, et al: Consecutive action of phospholipase A2 and glutathione peroxidase is required for reduction of phospholipid hydroperoxide and provides a convenient method to determine peroxide values in membrane J Free Radicals in Bio 198;1: Reimer KA, Jennings RB: Failure of the xanthine oxidase inhibitor allopurinol to limit infarct size after ischemia and reperfusion in the dog Circulation 198;71: Egan KW: Mechanism for irrcversible self-deactivation of prostaglandin synthctase J Biol Chem 1976;21: Atalla SL, Pereyara LH, Mackenzie GH, et al: Influence of oxygen derived free radical scavengers on ischemic liver Transplantation 198;40:84-89