- Email: [email protected]
Acute effects of a nucleoside analog dideoxyinosine (DDI) on the pancreas
JOURNAL
OF SURGICAL
RESEARCH
63,610-614
(1992)
Acute Effects of a Nucleoside Analog Dideoxyinosine on the Pancreas
(DDI)
ISTO H. NORDBACK, M.D.,* JEAN L. OLSON, M.D.,? RICHARD E. CHAISSON, M.D., ANDJOHN L. CAMERON, M.D.,FACS**’ Departments
of *Surgery,
tPathology,
and Medicine,
The Johns Hopkins
University,
Baltimore,
Maryland
21205
Submitted for publication July 8, 1991 MATERIALS Dideoxyinosine (DDI, Videx) is a recently developed nucleoside analog with activity against the human immunodeficiency virus. A significant number of patients with AIDS or AIDS-related complex treated with DDI have developed acute pancreatitis. This study was performed to investigate the acute effects of DDI on the pancreas utilizing an isolated ex vivo perfused canine pancreas preparation. Control preparations remained normal throughout a 4-hr perfusion period. The addition of 12.5 mg of DDI to the perfusate (ca. 100 Mmol/ liter) did not induce any changes in the preparation. The addition of 62.5 mg of DDI to the perfusate (ca. 500 pmol/liter) did not induce changes in the gross appearance, weight gain, or amylase activity. However, the arterial pressure and the oxygen consumption of the preparation decreased significantly after the administration of DDI. The amount of zymogen in the acinar cells also decreased, as evaluated by electron microscopy. Protein secretion increased temporarily, probably as a result of acinar cell emptying (increased secretion without new synthesis). Water and bicarbonate secretion were also increased during the fourth perfusion 0 1992 Academic Press, Inc. hour.
Experimental
INTRODUCTION
Dideoxyinosine (DDI, Videx) is a recently developed nucleoside analog with activity against the human immunodeficiency virus [ 11. DDI appears to be less toxic than other nucleoside analogs used for treating AIDS, such as 3’-azido-3’-deoxythymidine [ 11. However, acute pancreatitis has been observed in association with DDI treatment in 6-14% of patients included in phase I trials [2, 31. The pathogenesis of DDI-associated pancreatitis is unknown. Since the effects of DDI on the pancreas are not yet known, this study was performed to investigate the acute effects of DDI on the pancreas in an isolated ex viva perfused canine pancreas preparation. ’ To whom requests for reprints should be addressed. 0022-4&x04/92 $4.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form
METHODS
Preparation
Adult mongrel dogs weighing between 18 and 23 kg were anesthetized with sodium pentobarbital (Nembutal, Abbott Laboratories, Chicago, IL). The pancreas with a cuff of duodenum was removed and placed in a perfusion circuit in a humified chamber. The preparation was perfused through the splenic and superior mesenteric arteries at a constant flow (17 ml/100 g of tissue/ min) as previously described [4]. Briefly, the perfusate was collected from the portal vein, oxygenated in an infant Harvey oxygenator (C. R. Bard, Inc., Santa Ana, CA), and pumped back into the preparation by a roller pump (Watson-Marlow Ltd., Falmouth, England). The perfusate consisted of 400 ml of autologous blood, 100 ml of Ringer’s lactate solution, 0.5 g of glucose, 2.5 g of human albumin, and 14 mEq of sodium bicarbonate. An additional l-2 mEq of sodium bicarbonate was given when necessary to keep the pH of the perfusate in the physiologic range (7.35-7.55). The temperature of the perfusate was maintained at 37°C by a heat exchanger. The perfusion consisted of an initial 30-min stabilization period, followed by a 4-hr study perfusion. During the 4-hr study perfusion the preparation was monitored continuously and recorded hourly for gross appearance, weight gain, and arterial blood pressure. Venous samples were taken after the 30-min stabilization period (T,), and hourly thereafter to assay for hematocrit, hemoglobin, pH, blood gases, glucose level, and amylase activity. Arterial samples were taken at the same intervals to determine hemoglobin, pH, and blood gases. Pancreatic juice was collected hourly via a 16-gauge cannula in the pancreatic duct to assay for total proteins, bicarbonate, sodium, potassium, and chloride. Tissue specimens were taken from the pancreas for light and electron microscopy before and after the 4-hr perfusion. Oxygen consumption was calculated as previously described [5]. Paired and unpaired Student’s t test and Mann-Whittney U test were used to calculate the significance of the differences.
610 Inc. reserved.
AND
NORDBACK
ET AL.: DDI
oJ mmHq
611
PANCREAS
1 STABILIZATION
FIG. 1. Arterial pressure of a DDI 500 preparation.
Experimental
AND
PERIOD
STUDY
TO
PERFUSION
Notice a rapid decrease in the arterial
Protocol
These prepaGroup I (Controls) (four preparations). were managed as described above. Group II (DDI 100) (fourpreparations). These preparations were managed identically to those in Group I, except that 12.5 mg of DDI (BMY 40900 for oral solution, Bristol-Meyers Co., Evansville) in 10 ml of saline was added to the perfusate at T,. This dose resulted in approximately 100 PM DDI concentration in the perfusate. These prepGroup III (DDI 500) (fourpreparations). arations were managed identically to those in Group I, except that 62.5 mg of DDI in 50 ml of saline was added to the perfusate at T,. This dose resulted in approximately 500 PM DDI concentration in the perfusate. rations
RESULTS The control preparations were normal in gross appearance and gained minimal weight (10 k 3 g), and the amylase activity in the perfusate remained normal (1091 f 422 U/dl) after 4 hr of perfusion. The DDI 100 and DDI 500 preparations also remained normal in gross appearance. Weight gain (5 k 3 and 7 I 3 g) and amylase activities in the perfusate (1183 +- 160 and 630 +- 590 U/dl) were not increased compared to the control preparations. In the control preparations the blood pressure slowly decreased from a mean of 37 mm Hg to a level of 24 mm Hg during the first 2 hr of the 4-hr perfusion. No change was observed in the DDI 100 preparations as compared to the control preparations. In the DDI 500 preparations the blood pressure decreased rapidly after addition of DDI to the perfusate (Fig. 1). The blood pressure remained lower than in the control preparations throughout the 4-hr perfusion (Table 1). In the control preparations and in the DDI 100 preparations oxygen consumption remained stable through-
-->
pressure after addition
of DDI to the perfusate.
out the 4-hr perfusion, being in the range of 0.30 ml/ min/lOO g of tissue. In the DDI 500 preparations a significant reduction in oxygen consumption was observed (Table 2). The secretory rate (water secretion) of the pancreas decreased slightly during the 4-hr perfusion in the control preparations (Table 3). Water secretion did not decrease in the DDI 100 and DDI 500 preparations. In the DDI 500 preparations water secretion was increased as compared to the control preparations during the fourth perfusion hour (Table 3). The bicarbonate secretion remained stable in both the control and DDI 100 preparations. In the DDI 500 preparations bicarbonate secretion was increased (Table 4). Total protein secretion decreased slightly in the control preparations during the perfusion. This decrease did not occur in the DDI 100 or DDI 500 preparations (Table 5). In the DDI 500 preparations protein secretion rate was significantly increased as compared to the control preparations (Table 5). The secretion rates of sodium, potassium, and chloride remained unchanged in the DDI 100 and DDI 500
TABLE The Arterial
Pressure of the Pancreas Preparations during the 4 hr of Perfusion
Time 0-j
Controls (72 = 4)
0 1 2 3 4
37 f 15 26k 8 17+ 7 19f 5 27f 7
Note. *P ** P *** P
1
mm Hg, mean f SD. < 0.09, as compared to controls. < 0.05, as compared to controls. < 0.01, as compared to controls.
DDI 100 (n = 4) 60 31 20 17 21
+ 3 * 9 f 3 f 2 AZ3
DDI 500 (n = 4) 44 11 11 11 12
+ 8 i 2*** -c 2’ k 1** + 2*
612
JOURNAL
OF SURGICAL
TABLE The Oxygen Preparations Time (hr) 0 1 2 3 4
VOL.
0.31 0.29 0.26 0.25 0.29
f + k f f
0.06 0.05 0.04 0.02 0.10
The Bicarbonate during
DDI 500 (n = 4) 0.27 0.25 0.23 0.22 0.21
-+ 0.10 f 0.04 A 0.04 ” 0.04 f 0.04**
Time (hr) o-1 1-2 2-3 3-4
Secretion Rate of the Pancreas the 4 hr of Perfusion
Controls (n = 4) 246 58 56 27
preparations as compared to the control preparations. Hemoglobin, hematocrit, and glucose concentration in the perfusate did not differ among the three groups. Light and electron microscopic examination of control and DDI 100 preparations demonstrated no changes during the 4-hr perfusion. In the DDI 500 preparations light microscopy demonstrated a generalized reduction in the amount of zymogen granules in the acinar cells. Electron microscopy of the DDI 500 preparations also demonstrated the reduction in zymogen granules. In addition, large heterogenous granules, possibly representing secondary lysosomes, were noted (Fig. 2) in scattered acinar cells. Lipid droplets were increased in number. The mitochondria contained increased numbers of matrical densities (Fig. 2, inset) which may reflect increased divalent cations. DISCUSSION
The isolated canine pancreas preparation has been utilized in our laboratory to study the pathogenesis of acute pancreatitis. In this preparation five stimuli have been demonstrated to initiate acute pancreatitis during the 4-hr perfusion [4, 6-91. In this study the isolated pancreas preparation was exposed to DDI in an attempt
DDI 100 (n = 4)
f 263 + 28 + 55 + 15
129 82 48 60
3
The Hourly Secretory Rate (Water Secretion) Pancreas during the 4 hr of Perfusion
o-1 l-2 2-3 3-4
Controls (n = 4)
DDI 100 (n = 4)
9.0 3.1 1.9 1.6
5.2 4.2 3.2 3.2
+ + -t +
8.3 1.3 0.6 0.6”
+- 2.3 zk 2.6 k 2.9 + 3.3
+ 125 k 36 + 28 t 56
DDI 500 (n = 4) 2522 277 166 203
+ 4934 + 381 k 171 f 157**
to identify the short-term effects on pancreatic physiology and morphology. In clinical studies the administration of DDI and the occurrence of pancreatitis have been related to the dose of DDI. The majority of patients who have developed acute pancreatitis have received 19.3 mg, or more, of DDI per day per kilogram of body weight, resulting in a plasma concentration of over 20 pmol/liter [2]. During treatment with DDI plasma concentrations over 75 pmol/liter have been described [2]. In our study the calculated concentrations of DDI in the perfusate were 100 and 500 pmol/liter, at doses of 12.5 and 62.5 mg, respectively. These high doses of DDI did not induce acute pancreatitis in our pancreas preparation during the 4-hr perfusion. In addition, pancreatic physiology and ultrastructure remained unchanged with the lower concentration of DDI. The higher concentration resulted in changes in both physiology and acinar cell ultrastructure. The arterial pressure decreased rapidly after the administration of high dose DDI. Because the flow rate was constant, the decreased arterial pressure reflects decreased capillary resistance. The oxygen consumption of the preparation also decreased. The secretion of proteins increased. Because oxygen consumption did not increase, but actually decreased, this increased protein
TABLE
Time (hr)
4
Note. nmol/min/lOO g of tissue, mean k SD. ** P < 0.05, as compared to controls.
Note. ml/min/lOO g of tissue, mean + SD. ** P < 0.05, as compared to controls.
TABLE
1992
TABLE
DDI 100 (n = 4)
-c 0.01 + 0.04 * 0.07 -c 0.06 + 0.03
53, NO. 6, DECEMBER
2
Consumption of the Pancreas during the 4 hr of Perfusion
Controls (n = 4) 0.30 0.30 0.29 0.26 0.28
RESEARCH:
The Protein Secretion Rate of the Pancreas during the 4 hr of Perfusion
of the DDI 500 (n = 4)
12.0 5.4 4.1 5.1
?I k 2 It
19.9 4.9 2.4 1.9***
Note. pl/min/lOO g of tissue, mean f SD. # P < 0.09, as compared to secretory rate during the first hour. *** P < 0.01, as compared to controls.
5
Time (hr)
Controis (n = 4)
o-1 l-2 2-3 3-4
404 -t 392 94 + 52 52k 4 39 + 20*
DDI 100 (n = 4) 383 -t 261+ 194 k 153 -c
167 145 183 172
DDI 500 (n = 4) 374 184 143 248
+ 348 + 78** + 56** k 163**
Note. ag/min/lOO g of tissue, mean k SD. I P < 0.09, as compared to the protein secretion rate during the first hour. ** P < 0.05, as compared to controls.
NORDBACK
ET AL.: DDI
AND
613
PANCREAS
FIG. 2. Electron micrograph of acinar cells from a DDI 500 preparation demonstrating large, heterogenous granules (G) and dispersed lipid droplets (L) (x4000 uranyl acetate and lead citrate). (Inset) Electron micrograph of a mitochondrion from an acinar cell of a DDI 500 preparation demonstrating numerous densities within the matrix (X26,700 Uranyl acetate and lead citrate).
output may simply be an emptying of the acinar cells of their zymogen stores. This is supported by the microscopic finding that the amount of zymogen decreased in the acinar cells in the DDI 500 preparations. Furthermore, a decrease in the metabolic state of acinar cells has been reported to result in the release of zymogen, because the retaining of zymogen intracellularly is an energy requiring phenomenon [lo]. Structural changes were not observed in the ductal cells. Water and bicarbonate secretion, a function of the ductal cells, were slightly increased during the fourth hour of perfusion after the high dose of DDI, as compared to the controls. Previous studies [6] have demonstrated that the isolated pancreas model is very refractory to ischemia, as demonstrated by decreasing flow and/or oxygen content of the perfusate. Thus, the changes observed after high dose DDI are most likely caused by the direct influence of DDI on the acinar and ductal cells, and not through its influence on capillary resistance and arterial perfusion pressure. The changes in pancreatic physiology and ultrastructure that took place after high dose DDI did not result in the typical injury response seen in this preparation when pancreatitis is initiated by the five disperate stimuli [4, 6-91, that is, edema of the pancreas, weight gain of the preparation, and increased amylase activity in the per-
fusate. One possible reason why is that DDI, administered to the perfusate as a solution for oral clinical use, would be ineffective in our preparation. This is, however, an unlikely explanation, since DDI is water soluble and is easily absorbed into the circulation unmetabolized, has a half-life of approximately 12 hr, and is not metabolized until it moves intracellularly [2, 31. Obviously the studies presented here represent only acute short-term exposure of DDI to the pancreas. Perhaps long-term exposure is necessary before pancreatitis is initiated. Nevertheless because the pathogenesis of pancreatitis remains largely unknown, and since DDI treatment clearly is associated with this disorder, the acute effects of DDI on acinar cell physiology and ultrastructure, as demonstrated in this study, are of interest. ACKNOWLEDGMENTS This work was supported by NIH Grant ROl DK 32435. The Emil Aaltonen Foundation, The Paulo Foundation, and Tampere Tuberculous Foundation provided financial support for Dr. Nordback. Mr. Frederick Gilliam provided excellent technical assistance.
REFERENCES 1.
Yarchoan, R., Mitsuya, H., Thomas, R. V., Pluda, J. M., Hartman, N. R., Perno, C. F., Marczyk, K. S., Allain, J. P., Johns, D. J., and Broder, S. In viva activity against HIV and favourable
614
JOURNAL
OF SURGICAL
RESEARCH:
toxicity profile of 2’,3’-dideoxyinosine. Science 245: 412-415, 1989. Lambert, J. S., Seidlin, M., Reichman, R. C., Plank, C. S., Laverty, M., Morse, G. D., Knupp, C., McLaren, C., Pettinelli, C., Valentine, F. T., and Dolin, R. 2’,3’-Dideoxyinosine (dd1) in patients with the acquired immunodeficiency syndrome or AIDSrelated complex. A phase I trial. N. Engl. J. Med. 322: 13331340, 1990. Cooley, T. P., Kunches, L. M., Saunders, C. A., Ritter, J. K., Perkins, C. J., McLaren, C., McCaffrey, R. P., and Liebman, H. A. Once-daily administration of 2’,3’-dideoxyinosine (dd1) in patients with the acquired immunodeficiency syndrome or AIDS-related complex: Results of a phase I trial. N. Engl. J. Med. 322: 1341-1345, 1990. Saharia, P., Margolis, S., Zuidema, G. D., and Cameron, J. L. Acute pancreatitis with hyperlipemia: Studies with an isolated perfused canine pancreas. Surgery 82: 60-67,1977. Pascal, J. P., Roux, P., Vaysse, N., Lacroix, A., Martinel, C., and
VOL.
6.
7.
8.
9.
10.
53, NO. 6, DECEMBER
1992
Ribet, A. Respiratory exchanges and acid-base balance during perfusion of ex vivo isolated pancreas. Dig. Dis. 21: 381-388, 1976. Broe, P. J., Zuidema, G. D., and Cameron, J. L. The role of ischemia in acute pancreatitis: Studies with an isolated perfused canine pancreas. Surgery 91: 377-382, 1982. Broe, P. J., and Cameron, J. L. Experimental gallstonepancreatitis. Pathogenesis and response to different treatment modalities. Ann. Surg. 195: 566-573, 1982. Clemens, J. A., Olson, J., and Cameron, J. L. Cerulein induced pancreatitis in the ex vivo isolated perfused canine pancreas. Surgery 109: 515-522, 1991. Nordback, I. H., MacGowan, S., Potter, J. J., and Cameron, J. L. The role of acetaldehyde in the pathogenesis of acute alcoholic pancreatitis. Ann. Surg., in press. Bauduin, H., Colin, M., and Dumont, J. E. Energy sources for protein synthesis and enzymatic secretion in rat pancreas in vitro. Biochim. Biophys. Acta 174: 722-733,1969.
OF SURGICAL
RESEARCH
63,610-614
(1992)
Acute Effects of a Nucleoside Analog Dideoxyinosine on the Pancreas
(DDI)
ISTO H. NORDBACK, M.D.,* JEAN L. OLSON, M.D.,? RICHARD E. CHAISSON, M.D., ANDJOHN L. CAMERON, M.D.,FACS**’ Departments
of *Surgery,
tPathology,
and Medicine,
The Johns Hopkins
University,
Baltimore,
Maryland
21205
Submitted for publication July 8, 1991 MATERIALS Dideoxyinosine (DDI, Videx) is a recently developed nucleoside analog with activity against the human immunodeficiency virus. A significant number of patients with AIDS or AIDS-related complex treated with DDI have developed acute pancreatitis. This study was performed to investigate the acute effects of DDI on the pancreas utilizing an isolated ex vivo perfused canine pancreas preparation. Control preparations remained normal throughout a 4-hr perfusion period. The addition of 12.5 mg of DDI to the perfusate (ca. 100 Mmol/ liter) did not induce any changes in the preparation. The addition of 62.5 mg of DDI to the perfusate (ca. 500 pmol/liter) did not induce changes in the gross appearance, weight gain, or amylase activity. However, the arterial pressure and the oxygen consumption of the preparation decreased significantly after the administration of DDI. The amount of zymogen in the acinar cells also decreased, as evaluated by electron microscopy. Protein secretion increased temporarily, probably as a result of acinar cell emptying (increased secretion without new synthesis). Water and bicarbonate secretion were also increased during the fourth perfusion 0 1992 Academic Press, Inc. hour.
Experimental
INTRODUCTION
Dideoxyinosine (DDI, Videx) is a recently developed nucleoside analog with activity against the human immunodeficiency virus [ 11. DDI appears to be less toxic than other nucleoside analogs used for treating AIDS, such as 3’-azido-3’-deoxythymidine [ 11. However, acute pancreatitis has been observed in association with DDI treatment in 6-14% of patients included in phase I trials [2, 31. The pathogenesis of DDI-associated pancreatitis is unknown. Since the effects of DDI on the pancreas are not yet known, this study was performed to investigate the acute effects of DDI on the pancreas in an isolated ex viva perfused canine pancreas preparation. ’ To whom requests for reprints should be addressed. 0022-4&x04/92 $4.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form
METHODS
Preparation
Adult mongrel dogs weighing between 18 and 23 kg were anesthetized with sodium pentobarbital (Nembutal, Abbott Laboratories, Chicago, IL). The pancreas with a cuff of duodenum was removed and placed in a perfusion circuit in a humified chamber. The preparation was perfused through the splenic and superior mesenteric arteries at a constant flow (17 ml/100 g of tissue/ min) as previously described [4]. Briefly, the perfusate was collected from the portal vein, oxygenated in an infant Harvey oxygenator (C. R. Bard, Inc., Santa Ana, CA), and pumped back into the preparation by a roller pump (Watson-Marlow Ltd., Falmouth, England). The perfusate consisted of 400 ml of autologous blood, 100 ml of Ringer’s lactate solution, 0.5 g of glucose, 2.5 g of human albumin, and 14 mEq of sodium bicarbonate. An additional l-2 mEq of sodium bicarbonate was given when necessary to keep the pH of the perfusate in the physiologic range (7.35-7.55). The temperature of the perfusate was maintained at 37°C by a heat exchanger. The perfusion consisted of an initial 30-min stabilization period, followed by a 4-hr study perfusion. During the 4-hr study perfusion the preparation was monitored continuously and recorded hourly for gross appearance, weight gain, and arterial blood pressure. Venous samples were taken after the 30-min stabilization period (T,), and hourly thereafter to assay for hematocrit, hemoglobin, pH, blood gases, glucose level, and amylase activity. Arterial samples were taken at the same intervals to determine hemoglobin, pH, and blood gases. Pancreatic juice was collected hourly via a 16-gauge cannula in the pancreatic duct to assay for total proteins, bicarbonate, sodium, potassium, and chloride. Tissue specimens were taken from the pancreas for light and electron microscopy before and after the 4-hr perfusion. Oxygen consumption was calculated as previously described [5]. Paired and unpaired Student’s t test and Mann-Whittney U test were used to calculate the significance of the differences.
610 Inc. reserved.
AND
NORDBACK
ET AL.: DDI
oJ mmHq
611
PANCREAS
1 STABILIZATION
FIG. 1. Arterial pressure of a DDI 500 preparation.
Experimental
AND
PERIOD
STUDY
TO
PERFUSION
Notice a rapid decrease in the arterial
Protocol
These prepaGroup I (Controls) (four preparations). were managed as described above. Group II (DDI 100) (fourpreparations). These preparations were managed identically to those in Group I, except that 12.5 mg of DDI (BMY 40900 for oral solution, Bristol-Meyers Co., Evansville) in 10 ml of saline was added to the perfusate at T,. This dose resulted in approximately 100 PM DDI concentration in the perfusate. These prepGroup III (DDI 500) (fourpreparations). arations were managed identically to those in Group I, except that 62.5 mg of DDI in 50 ml of saline was added to the perfusate at T,. This dose resulted in approximately 500 PM DDI concentration in the perfusate. rations
RESULTS The control preparations were normal in gross appearance and gained minimal weight (10 k 3 g), and the amylase activity in the perfusate remained normal (1091 f 422 U/dl) after 4 hr of perfusion. The DDI 100 and DDI 500 preparations also remained normal in gross appearance. Weight gain (5 k 3 and 7 I 3 g) and amylase activities in the perfusate (1183 +- 160 and 630 +- 590 U/dl) were not increased compared to the control preparations. In the control preparations the blood pressure slowly decreased from a mean of 37 mm Hg to a level of 24 mm Hg during the first 2 hr of the 4-hr perfusion. No change was observed in the DDI 100 preparations as compared to the control preparations. In the DDI 500 preparations the blood pressure decreased rapidly after addition of DDI to the perfusate (Fig. 1). The blood pressure remained lower than in the control preparations throughout the 4-hr perfusion (Table 1). In the control preparations and in the DDI 100 preparations oxygen consumption remained stable through-
-->
pressure after addition
of DDI to the perfusate.
out the 4-hr perfusion, being in the range of 0.30 ml/ min/lOO g of tissue. In the DDI 500 preparations a significant reduction in oxygen consumption was observed (Table 2). The secretory rate (water secretion) of the pancreas decreased slightly during the 4-hr perfusion in the control preparations (Table 3). Water secretion did not decrease in the DDI 100 and DDI 500 preparations. In the DDI 500 preparations water secretion was increased as compared to the control preparations during the fourth perfusion hour (Table 3). The bicarbonate secretion remained stable in both the control and DDI 100 preparations. In the DDI 500 preparations bicarbonate secretion was increased (Table 4). Total protein secretion decreased slightly in the control preparations during the perfusion. This decrease did not occur in the DDI 100 or DDI 500 preparations (Table 5). In the DDI 500 preparations protein secretion rate was significantly increased as compared to the control preparations (Table 5). The secretion rates of sodium, potassium, and chloride remained unchanged in the DDI 100 and DDI 500
TABLE The Arterial
Pressure of the Pancreas Preparations during the 4 hr of Perfusion
Time 0-j
Controls (72 = 4)
0 1 2 3 4
37 f 15 26k 8 17+ 7 19f 5 27f 7
Note. *P ** P *** P
1
mm Hg, mean f SD. < 0.09, as compared to controls. < 0.05, as compared to controls. < 0.01, as compared to controls.
DDI 100 (n = 4) 60 31 20 17 21
+ 3 * 9 f 3 f 2 AZ3
DDI 500 (n = 4) 44 11 11 11 12
+ 8 i 2*** -c 2’ k 1** + 2*
612
JOURNAL
OF SURGICAL
TABLE The Oxygen Preparations Time (hr) 0 1 2 3 4
VOL.
0.31 0.29 0.26 0.25 0.29
f + k f f
0.06 0.05 0.04 0.02 0.10
The Bicarbonate during
DDI 500 (n = 4) 0.27 0.25 0.23 0.22 0.21
-+ 0.10 f 0.04 A 0.04 ” 0.04 f 0.04**
Time (hr) o-1 1-2 2-3 3-4
Secretion Rate of the Pancreas the 4 hr of Perfusion
Controls (n = 4) 246 58 56 27
preparations as compared to the control preparations. Hemoglobin, hematocrit, and glucose concentration in the perfusate did not differ among the three groups. Light and electron microscopic examination of control and DDI 100 preparations demonstrated no changes during the 4-hr perfusion. In the DDI 500 preparations light microscopy demonstrated a generalized reduction in the amount of zymogen granules in the acinar cells. Electron microscopy of the DDI 500 preparations also demonstrated the reduction in zymogen granules. In addition, large heterogenous granules, possibly representing secondary lysosomes, were noted (Fig. 2) in scattered acinar cells. Lipid droplets were increased in number. The mitochondria contained increased numbers of matrical densities (Fig. 2, inset) which may reflect increased divalent cations. DISCUSSION
The isolated canine pancreas preparation has been utilized in our laboratory to study the pathogenesis of acute pancreatitis. In this preparation five stimuli have been demonstrated to initiate acute pancreatitis during the 4-hr perfusion [4, 6-91. In this study the isolated pancreas preparation was exposed to DDI in an attempt
DDI 100 (n = 4)
f 263 + 28 + 55 + 15
129 82 48 60
3
The Hourly Secretory Rate (Water Secretion) Pancreas during the 4 hr of Perfusion
o-1 l-2 2-3 3-4
Controls (n = 4)
DDI 100 (n = 4)
9.0 3.1 1.9 1.6
5.2 4.2 3.2 3.2
+ + -t +
8.3 1.3 0.6 0.6”
+- 2.3 zk 2.6 k 2.9 + 3.3
+ 125 k 36 + 28 t 56
DDI 500 (n = 4) 2522 277 166 203
+ 4934 + 381 k 171 f 157**
to identify the short-term effects on pancreatic physiology and morphology. In clinical studies the administration of DDI and the occurrence of pancreatitis have been related to the dose of DDI. The majority of patients who have developed acute pancreatitis have received 19.3 mg, or more, of DDI per day per kilogram of body weight, resulting in a plasma concentration of over 20 pmol/liter [2]. During treatment with DDI plasma concentrations over 75 pmol/liter have been described [2]. In our study the calculated concentrations of DDI in the perfusate were 100 and 500 pmol/liter, at doses of 12.5 and 62.5 mg, respectively. These high doses of DDI did not induce acute pancreatitis in our pancreas preparation during the 4-hr perfusion. In addition, pancreatic physiology and ultrastructure remained unchanged with the lower concentration of DDI. The higher concentration resulted in changes in both physiology and acinar cell ultrastructure. The arterial pressure decreased rapidly after the administration of high dose DDI. Because the flow rate was constant, the decreased arterial pressure reflects decreased capillary resistance. The oxygen consumption of the preparation also decreased. The secretion of proteins increased. Because oxygen consumption did not increase, but actually decreased, this increased protein
TABLE
Time (hr)
4
Note. nmol/min/lOO g of tissue, mean k SD. ** P < 0.05, as compared to controls.
Note. ml/min/lOO g of tissue, mean + SD. ** P < 0.05, as compared to controls.
TABLE
1992
TABLE
DDI 100 (n = 4)
-c 0.01 + 0.04 * 0.07 -c 0.06 + 0.03
53, NO. 6, DECEMBER
2
Consumption of the Pancreas during the 4 hr of Perfusion
Controls (n = 4) 0.30 0.30 0.29 0.26 0.28
RESEARCH:
The Protein Secretion Rate of the Pancreas during the 4 hr of Perfusion
of the DDI 500 (n = 4)
12.0 5.4 4.1 5.1
?I k 2 It
19.9 4.9 2.4 1.9***
Note. pl/min/lOO g of tissue, mean f SD. # P < 0.09, as compared to secretory rate during the first hour. *** P < 0.01, as compared to controls.
5
Time (hr)
Controis (n = 4)
o-1 l-2 2-3 3-4
404 -t 392 94 + 52 52k 4 39 + 20*
DDI 100 (n = 4) 383 -t 261+ 194 k 153 -c
167 145 183 172
DDI 500 (n = 4) 374 184 143 248
+ 348 + 78** + 56** k 163**
Note. ag/min/lOO g of tissue, mean k SD. I P < 0.09, as compared to the protein secretion rate during the first hour. ** P < 0.05, as compared to controls.
NORDBACK
ET AL.: DDI
AND
613
PANCREAS
FIG. 2. Electron micrograph of acinar cells from a DDI 500 preparation demonstrating large, heterogenous granules (G) and dispersed lipid droplets (L) (x4000 uranyl acetate and lead citrate). (Inset) Electron micrograph of a mitochondrion from an acinar cell of a DDI 500 preparation demonstrating numerous densities within the matrix (X26,700 Uranyl acetate and lead citrate).
output may simply be an emptying of the acinar cells of their zymogen stores. This is supported by the microscopic finding that the amount of zymogen decreased in the acinar cells in the DDI 500 preparations. Furthermore, a decrease in the metabolic state of acinar cells has been reported to result in the release of zymogen, because the retaining of zymogen intracellularly is an energy requiring phenomenon [lo]. Structural changes were not observed in the ductal cells. Water and bicarbonate secretion, a function of the ductal cells, were slightly increased during the fourth hour of perfusion after the high dose of DDI, as compared to the controls. Previous studies [6] have demonstrated that the isolated pancreas model is very refractory to ischemia, as demonstrated by decreasing flow and/or oxygen content of the perfusate. Thus, the changes observed after high dose DDI are most likely caused by the direct influence of DDI on the acinar and ductal cells, and not through its influence on capillary resistance and arterial perfusion pressure. The changes in pancreatic physiology and ultrastructure that took place after high dose DDI did not result in the typical injury response seen in this preparation when pancreatitis is initiated by the five disperate stimuli [4, 6-91, that is, edema of the pancreas, weight gain of the preparation, and increased amylase activity in the per-
fusate. One possible reason why is that DDI, administered to the perfusate as a solution for oral clinical use, would be ineffective in our preparation. This is, however, an unlikely explanation, since DDI is water soluble and is easily absorbed into the circulation unmetabolized, has a half-life of approximately 12 hr, and is not metabolized until it moves intracellularly [2, 31. Obviously the studies presented here represent only acute short-term exposure of DDI to the pancreas. Perhaps long-term exposure is necessary before pancreatitis is initiated. Nevertheless because the pathogenesis of pancreatitis remains largely unknown, and since DDI treatment clearly is associated with this disorder, the acute effects of DDI on acinar cell physiology and ultrastructure, as demonstrated in this study, are of interest. ACKNOWLEDGMENTS This work was supported by NIH Grant ROl DK 32435. The Emil Aaltonen Foundation, The Paulo Foundation, and Tampere Tuberculous Foundation provided financial support for Dr. Nordback. Mr. Frederick Gilliam provided excellent technical assistance.
REFERENCES 1.
Yarchoan, R., Mitsuya, H., Thomas, R. V., Pluda, J. M., Hartman, N. R., Perno, C. F., Marczyk, K. S., Allain, J. P., Johns, D. J., and Broder, S. In viva activity against HIV and favourable
614
JOURNAL
OF SURGICAL
RESEARCH:
toxicity profile of 2’,3’-dideoxyinosine. Science 245: 412-415, 1989. Lambert, J. S., Seidlin, M., Reichman, R. C., Plank, C. S., Laverty, M., Morse, G. D., Knupp, C., McLaren, C., Pettinelli, C., Valentine, F. T., and Dolin, R. 2’,3’-Dideoxyinosine (dd1) in patients with the acquired immunodeficiency syndrome or AIDSrelated complex. A phase I trial. N. Engl. J. Med. 322: 13331340, 1990. Cooley, T. P., Kunches, L. M., Saunders, C. A., Ritter, J. K., Perkins, C. J., McLaren, C., McCaffrey, R. P., and Liebman, H. A. Once-daily administration of 2’,3’-dideoxyinosine (dd1) in patients with the acquired immunodeficiency syndrome or AIDS-related complex: Results of a phase I trial. N. Engl. J. Med. 322: 1341-1345, 1990. Saharia, P., Margolis, S., Zuidema, G. D., and Cameron, J. L. Acute pancreatitis with hyperlipemia: Studies with an isolated perfused canine pancreas. Surgery 82: 60-67,1977. Pascal, J. P., Roux, P., Vaysse, N., Lacroix, A., Martinel, C., and
VOL.
6.
7.
8.
9.
10.
53, NO. 6, DECEMBER
1992
Ribet, A. Respiratory exchanges and acid-base balance during perfusion of ex vivo isolated pancreas. Dig. Dis. 21: 381-388, 1976. Broe, P. J., Zuidema, G. D., and Cameron, J. L. The role of ischemia in acute pancreatitis: Studies with an isolated perfused canine pancreas. Surgery 91: 377-382, 1982. Broe, P. J., and Cameron, J. L. Experimental gallstonepancreatitis. Pathogenesis and response to different treatment modalities. Ann. Surg. 195: 566-573, 1982. Clemens, J. A., Olson, J., and Cameron, J. L. Cerulein induced pancreatitis in the ex vivo isolated perfused canine pancreas. Surgery 109: 515-522, 1991. Nordback, I. H., MacGowan, S., Potter, J. J., and Cameron, J. L. The role of acetaldehyde in the pathogenesis of acute alcoholic pancreatitis. Ann. Surg., in press. Bauduin, H., Colin, M., and Dumont, J. E. Energy sources for protein synthesis and enzymatic secretion in rat pancreas in vitro. Biochim. Biophys. Acta 174: 722-733,1969.