Effect of FK506 on Insulin
Secretion
in Normal
Dogs
Sheryl Strasser, Rodolfo Alejandro, E. Timothy Shapiro, Camilo Ricordi, S. Todo, and Daniel H. Mintz In this report, we describe the effect of FK506 on glucose-mediated insulin release in normal dogs. Dogs were placed into one of two groups, group 1 dogs received FK506 (1 mgl kg/d orally) for 2 weeks, and group 2 dogs received FK506 at the same dose, but for 4 weeks. Following the treatment period, both groups of dogs were allowed a recovery period during which time no FK506 was administered. Intravenous glucose tolerance tests (IVGlT) were performed (0.5 mg/ kg IV) before FK506 treatment, after 2 or 4 weeks of treatment, and following the recovery periods. Complete blood cell counts (CBC) and serum chemistries were also obtained at these times. Following FK506 treatment for either 2 or 4 weeks, the dogs experienced a delay in glucose disappearance in response to IV glucose injection. Insulin secretion during IVGll was unchanged in dogs treated for only 2 weeks, but was significantly decreased in dogs treated for 4 weeks. Following the recovery period, glucose disappearance during IVGll returned to normal in dogs that were treated for 2 weeks, and was more rapid than normal in dogs that had been treated for 4 weeks. Insulin secretion after the recovery period remained unchanged in group 1 dogs, but continued to be significantly reduced in group 2 dogs that had received FK506 for 4 weeks. No significant change was detected in the CBCs or serum chemistries. Copyright 0 1992 by W. B. Saunders Company
R
ECENT recognition has been given to FKS06 following the demonstration that it enables “rescue” of liver allografts from intractable rejection when conventional measures have been unsuccessful.’ Clinical trials using FK506 as the primary immunosuppressive agent for transplantation of liver, kidney, and thoracic organs’,’ and, sporadically, as a treatment regimen for autoimmune disorders,4 have also led to favorable results. As clinical experience with FK506 broadens, the profile of side effects and possible toxicities becomes clearer.” One such adverse effect that has been identified in experimental transplant animal models treated with FK506 is a derangement of glucose metabolism. This inconsistently reported abnormality varies from elevated fasting blood glucose levels in rodents’ to frank insulin-dependent diabetes mellitus in baboons.6 Due to the relatively recent development of this drug, few studies have been performed to characterize the effects of FK506 on carbohydrate metabolism in the normal individual. Therefore, to evaluate the effects of FK506 on glucose metabolism in the absense of organ allografts, we administered intravenous glucose tolerance tests (IVGTT) to normal dogs treated with FK506. In addition, we examined the effect of the drug on overall well-being of the animal and on hematologic indices and serum chemistries. MATERIALS
AND METHODS
Animals Eleven mongrels
beagles (Marshall (Motsinger Farm,
Farms, North Rose, NY) and two St. Joseph, IL), weighing between 9
From the University of Miami School of Medicine, Diabetes Research Institute, Miami, FL; and the Department of Surgery, University Health Center of Pittsburgh, University of Pittsburgh, Pittsburgh, PA. Supported by Project Grants No. DK25802 and DK29961, and the Diabetes Research Institute Foundation. Address reprint requests to Daniel H. Mintz, MD, Diabetes Research Institute (RI34). University of Miami School of Medicine, PO Box 016960 Miami, FL 33101. Copyright 0 1992 by WB. Saunders Company 0026-0495/92/4101-OOI2$03.00/0 64
and 14 kg and of both sexes, were studied. The dogs were fed dog chow (Hi-Pro, Purina Feeds, St Louis, MO) and allowed free access to water. Physical examinations were performed daily to assess body condition and attitude. AI1 procedures relating to animal studies were performed in accordance with the guidelines set forth by the National Institutes of Health and the University of Miami.
Ekperimental
Protocol
Dogs were randomly assigned to one of three groups. The first group of dogs (N = 5) was treated with FK506 for 2 weeks and the second group (N = 5) was treated for 4 weeks (Table 1). A third group (N = 3) served as controls and received no FK506. Subsequent to the treatment periods, the first and second groups of dogs were allowed a recovery period that was equal in length to their respective treatment periods (ie, dogs that were treated with FK506 for 2 or 4 weeks, were allowed a 2- or 4-week recovery period, respectively). During the recovery period, the dogs received no FK.506. All animals underwent an IVGW on three separate occasions: (1) baseline evaluation, (2) after treatment with FK506 for 2 or 4 weeks and (3) following the 2- or 4-week recovery period. After an over 1 minute overnight fast, glucose (0.5 g/kg) was administered through a percutaneously placed venous catheter. Peripheral venous blood samples were collected from a different vein before and 5, 10, 15, 20, 30, and 60 minutes after completion of the injection. Dogs were held unrestrained in their individual runs during the study. Complete blood cell counts (CBC) and serum chemistries (SMA-20) were also determined at that time. FK506 powder, a 20% solid dispersion in hydroxy-propyl-methyl cellulose and polyoxyethylated hydrogenated castor oil (Fujisawa Pharmaceuticals, Osaka, Japan), was placed in gelatin capsules and administered orally (1 mgikgld). Atropine was administered to prevent gastrointestinal side effects of FK506. All dogs received 2 mg subcutaneously twice a day during the first week. This dose was reduced to once a day during the second week. Dogs treated with FK506 for 4 weeks received 1 mg atropine subcutaneously once a day during the third week of treatment; no atropine was given during the fourth week. To control for possible effects of atropine. the third group of dogs (N = 3) received an identical dose of atropine. This group of animals underwent an IVGm at baseline, after 2 weeks of atropine treatment and 1 week after discontinuation of the atropine treatments. CBCs and serum chemistries were also obtained at those times.
Metabolism,
Vol41,
No
1
(January),
1992: pp 64-67
65
FK506 AND INSULIN SECRETION IN NORMAL DOGS
Table 1. Signalment and Clinical Parameters for Normal Dogs Treated With FK5g6 for 2 or 4 Weeks
Dog
Breed
Sex
1
Mongrel
?
FK506
Initial
% Weight
Gastrointestinal
Therapy
Weight (kg)
LOSS
Abnormalities
F
3
days
13.2
6
++++
12.5
4
++++
9.2
2
+
2
Mongrel
F
4 days
3
Beagle
F
2 weeks
4
Beagle
F
2 weeks
9.0
0
5
Beagle
M
2weeks
14.0
2
6
Beagle
F
2 weeks
10.0
7
7
Beagle
F
2 weeks
10.7
0
6
Beagle
M
4weeks
13.0
12
++
9
Beagle
M
4 weeks
13.5
15
++
10
Beagle
F
4 weeks
9.0
0
.P -g
I I
16
0
A
1
4
+ ++ + + 04
‘
a
10
12
14
16
18
+ Initial Dog Weight (kg)
Analytical Techniques Blood samples were collected into tubes containing 1.2 mg/mL NaEDTA and 500 kIU/mL Trasylol and immediately placed on ice. Following centrifugation and separation, plasma samples for immunoreactive insulin determination were stored at -70°C until assayed.’ Plasma glucose levels were measured immediately (Glucose Analyzer II, Beckman Instruments, Fullerton, CA). Data Analysis All results are expressed as the mean 2 SEM. Areas under the glucose and insulin concentration curves were calculated by the trapezoidal rule. Group means were compared using repeated measures ANOVA and the Student-Newman-Keuls t test (primer of Eiostatisfics:The Program, New York, NY, McGraw-Hill, 1988). Differences were considered significant if the corresponding P value was less than or equal to .05.
Fig 1. Relationship between weight loss 1% of initial body weight) and average initial body weight (0.8 to 10 kg; 0.10 to 12 kg; A, 12 to 14 kg; A, 14 to 16 kg; 0, 16 to 16 kg, N = 19) in dogs treated wfth FK506 (r = .96, P < .Ol).
to IVGIT in dogs who received atropine alone (data not shown). Blood glucose concentrations in response to IVG’IT before, during, and after treatment with FK506 for two weeks are shown in Fig 2A. FK506 treatment was associated with a delayed disappearance of glucose, which reached significance at 15 and 20 minutes (10.7 + 0.8 mmol/L v
RESULTS
Animals The hvo mongrels (Table l), developed intussusceptions on days 3 and 4 after FK506 treatment, respectively, and were killed. They were excluded from analysis of the data. Dogs that received atropine only (group 3) were clinically normal and had normal CBC and serum chemistry values (SMA-20) throughout the experiment. Physical examination of the dogs receiving FK506 (group 1 and 2) showed listlessness, loss of appetite, diarrhea, and weight loss (Table 1). The weight loss was greater in the larger dogs (Fig 1) and when FK506 was given for 4 rather than 2 weeks. Calculations of weight loss from this study, pooled with similar data from other ongoing experiments with FK506 (unpublished data), are shown in Fig 1. The larger dogs lost a significantly greater percent of their initial body weight during treatment with FK506 (r = .96, P < .Ol). However, after the drug was discontinued, all dogs regained their baseline weights and were clinically normal so that persistent effects of FK506 could be detected independent of weight loss. The CBC and serum chemistries (SMA-20) showed no significant changes during or following therapy with FK506. Peripheral Glucose and Insulin Concentrations No significant change from baseline values were observed in peripheral glucose or insulin concentrations in response
15
10
5
*
* OB 0
10
20
30
40
50
60
TIME (min) Fig 2. Glucose concentrations (mean + SEM) in response to IVGTT (0.5 mg/ kg IV) in normal dogs treeted with FKXt9 for (A) 2 weeks and (El 4 weeks; 0, Pre-FK: 0. on FK: n. off FK. lSignffkrantly different from control, P < .05.
ET AL
66
8.4 2 0.4 mmol/L and 8.9 + 1.0 mmol/L v 4.6 2 0.3 mmol/L, respectively, P < .OS). Two weeks after discontinuation of FK506, glucose concentrations in response to IV glucose returned to normal. Dogs treated with FK.506 for 4 weeks (Fig 2B) had glucose levels that were significantly higher at 20 and 30 minutes when compared with the baseline evaluation (11.0 & 0.8 mmol/L v 8.5 ? 0.8 mmol/L and 8.3 ? 0.6 mmol/L v 5.6 ? 0.7 mmol/L, respectively, P < .05). After discontinuation of FK506 for 4 weeks, glucose levels following IV glucose challenge returned to or declined to below normal levels. At the 15- and 20-minute time points, the glucose concentrations were significantly less than before drug treatment (7.0 t 1.3 v 10.8 ? l.Ommol/Land 5.4 * 1.0 v 8.4 + 0.8 mmol/L, respectively, P < .05). Peripheral insulin concentrations in response to IVGTT before, during, and after treatment with FK506 for 2 and 4 weeks are shown in Fig 3. After 2 weeks of treatment with FK506, the peak plasma insulin concentration in response to IV glucose was reduced at 10 minutes compared with pretreatment levels (134 ? 26 pmol/L v 259 ? 99 pmol/L, respectively), but this difference did not reach statistical significance (Fig 3A). Following discontinuation of the drug, the insulin levels in response to IV glucose returned to pretreatment levels.
I
p(O.05
L
1 I
p(O.05
I
1
i PE
On
Of‘
Pre
On
Off
Fig 4. Incremental areas under the glucose and insulin concentration curvas in response to IVGlT (0.5 mg/kg IV) expressed as a percent of the pretreatment area in normal dogs (N = 3) treated with FK506 for 4 weeks (0. glucose areas; q, insulin areas).
By contrast, insulin levels in response to IVG’IT after 4 weeks of treatment with FK506 (Fig 3B) were significantly decreased compared with pretreatment levels 10 and 15 minutes following administration of glucose (164 c 47 r’ 376 it 77 pmol/L and 139 t 58 v 388 + 87 pmol/L, respectively, P < .05). Moreover, in contrast to animals treated for only 2 weeks (Fig 3A), insulin responses to IV glucose following recovery from 4 weeks of FK506 therapy remained significantly depressed. Incremental areas under the glucose and insulin concentration curves during IVGTT in dogs treated with FK506 for 4 weeks are shown in Fig 4. After 4 weeks of FK506 therapy, the incremental glucose area tended to be higher than normal, but was not significantly different from baseline. However, following discontinuation of the drug for 4 weeks, the area under the glucose curve was diminished significantly compared with both the baseline and treatment areas (55% ? 13% v 100% ? 14% and 128% + 13%, respectively, P < .05). The incremental area under the insulin concentration curve was significantly decreased during FK506 treatment and remained significantly reduced 4 weeks after drug withdrawal (49% 2 23% and 47% * 11% 1~100% ? 22%, respectively, P < .05).
400 ? a E 4 5
KL.
DISCUSSION
/
300 * 200
*
z
4/
z
*
100 ”
* \I
A\.
0
I
0
lb
2-O
i0
-
io
SO
$0
TIME (min) Fig 3. Insulin concentrations (mean 2 SEM) in response to IVGIT (0.5 mg/kg IV) in normal dogs treeted wfth PK506 for (A) 2 weeks and (6) 4 weeks. 0, Pre FK; 0, on FK; n , off FK. *Significantly different from contr0l.P < .05.
Species-specific vulnerability to the toxic effects of FK506 have been well documented.8.Y In this experiment, we may have observed an intraspecies susceptability to the drug. The mongrel dogs in this study (Table 1, N = 2) were rapidly and severely debilitated by FK506 and developed intussusceptions, a complication that was not encountered in beagle dogs similarly treated (N = 19). Breed susceptibility to toxic effects of various drugs has been shown in veterinary medicine.“’ However, concentration of the genetic pool, which is the consequence of purpose-breeding to obtain purebred animals, like the beagles used in this study, may contribute in some way to the ability of these dogs to tolerate FK506. This intraspecies peculiarity requires further investigation for an adequate explanation.
FK506 AND INSULIN SECRETION IN NORMAL DOGS
67
In this study, we found that administration of FKS06 to normal beagles resulted in significant alterations in glucose disappearance rates following IV glucose injection. We have reported, as well, that glucagon-mediated insulin release is substantially inhibited at 2 weeks and nearly completely abolished after 4 weeks of FK506 therapy.” Thus, although we evaluated glucose metabolism here based on the response to a single secretagogue, we were able to find evidence to support at least two possible mechanisms that may contribute either separately or together to derangements in glucose homeostasis in FK506treated normal dogs. One mechanism implicated in the FKS06-induced glucose tolerance abnormalities is impaired insulin secretion. Despite normal fasting blood glucose levels before, during, and after the study, the response to IV glucose showed an impairment of insulin secretion. Specifically, 4 weeks of FKS06 treatment followed by a 4-week recovery period during which time the dogs returned to baseline weights and health status, revealed a persistent reduction in insulin secretion in response to IV glucose. The notion that this could be due to a direct p-cell effect is seemingly corroborated by in vitro dose-response studies of whole, cultured rodent and human islets,“.” which identify an FK506 related inhibition of glucose-mediated insulin secretion. A second possible mechanism responsible for the observed glucose intolerance is an alteration in insulin action. An apparent insensitivity to insulin’s action was not the focus of attention in this study; however, evidence in
support of this mechanism is based on the following observation. Both groups of dogs experienced significant improvement in IV glucose disappearance following discontinuation of the drug despite either (1) unchanged insulin concentrations, as in dogs that had been treated for only 2 weeks, or (2) significantly diminished insulin concentrations, as seen in dogs that had been treated for 4 weeks. Specific testing methodology, such as the euglycemic clamp or frequently sampled IV GTT,14 may prove useful in evaluating this particular drug-induced effect. Until this is done, the caveat exists that these alterations in insulin response during the recovery periods are driven by the glucose levels, and not the reverse. Based on the derangements we observed in this preliminary study, FK506, at 1 mg/kg/d orally, is tolerated in the beagle dog, but causes significant abnormalities in insulin secretion and glucose utilization that persist as long as the animal is receiving the drug. These abnormalities tend to reverse following discontinuation of the drug; however, normal insulin secretion was not re-established in animals that received FK506 for 4 weeks. Thus, our findings require further confirmation under conditions that control for weight gain and weight loss cycle in this and other species, but the observations reported here are similar to our findings in cyclosporine-treated beagles.” ACKNOWLEDGMENT
Special thanks are extended to the Diabetes Research Institute staff for dedicated technical and secretarial assistance.
REFERENCES 1. Fung J, Todo S, Jain A, et al: Conversion from cyclosporin to FK506 in liver allograft recipients with cyclosporin-related complications. Transplant Proc 22:6-12, 1990 2. Todo S, Fung JJ, Demetris AJ, Jain A, et al: Early trials with FK506 as primary treatment in liver transplantation. Transplant Proc 22:13-16,199O (suppl 1) 3. Todo S, Fung J, Starzl T, et al: Liver, kidney and thoracic organ transplantation under FK506. Am J Surg 212:295-307, 1990 4. McCauley J, Bronsther 0, Fung J, et al: Treatment of cyclosporin-induced heamolytic uraemic syndrome with FK506. Lancet 2:1516,1989 5. Nalesnik MA, Todo S, Murase N, et al: Toxicology of FK506 in the Lewis rat. Transplant Proc 19:89-92,1987 6. Thiru S, Collier D StJ, Calne R: Pathological studies in canine and baboon renal allograft recipients immunosuppressed with FK506. Transplant Proc 19:98-99,1987 7. Heding LG: Determination of total serum insulin (IRI) in insulin-treated diabetic patients. Diabetologia 8:260-266,1972 8. Todo S, Ueda Y, Demetris J, et al: Immunnosuppression of canine, monkey and baboon allografts by FK506: With special reference to synergism with other drugs and to tolerance induction. Surgery 104:239-249,198s
9. Collier D, Calne R, Thiru S, et al: FK506 in experimental renal allografts in dogs and primates. Transplant Proc 20:226-228, 1989 10. Paul AJ, Tranquilli WJ, Seward RL, et al: Clinical observations in collies given ivermectin orally. Am J Vet Res 48:684-685, 1987 11. Strasser S, Alejandro R, Ricordi C, et al: The effect of FK.506 on canine pancreatic islet cell function in beagle dogs. Transplant Proc 23:756-757, 1991 12. Tze W, Tai J, Cheung and rat islets. Transplantation
S: In vitro effects of FK506 on human 491172-l 174, 1990
13. Carroll P, Boschero A, Li M, et al: Effect of the immunosuppressant FK506 on glucose-induced insulin secretion from adult rat islets of Langerhans. Transplantation 51:275-278, 1991 14. Bergman RN, Prager R, Volund A, et al: Equivalence of the insulin sensitivity index in man derived by the minmal model method and the euglycemic glucose clamp. J Clin Invest 79:790800,1987 15. Alejandro R, Feldman E, Bloom A, et al: Effects of cyclosporin on insulin and C-peptide secretion in healthy beagles. Diabetes 38:698-703,1989
Secretion
in Normal
Dogs
Sheryl Strasser, Rodolfo Alejandro, E. Timothy Shapiro, Camilo Ricordi, S. Todo, and Daniel H. Mintz In this report, we describe the effect of FK506 on glucose-mediated insulin release in normal dogs. Dogs were placed into one of two groups, group 1 dogs received FK506 (1 mgl kg/d orally) for 2 weeks, and group 2 dogs received FK506 at the same dose, but for 4 weeks. Following the treatment period, both groups of dogs were allowed a recovery period during which time no FK506 was administered. Intravenous glucose tolerance tests (IVGlT) were performed (0.5 mg/ kg IV) before FK506 treatment, after 2 or 4 weeks of treatment, and following the recovery periods. Complete blood cell counts (CBC) and serum chemistries were also obtained at these times. Following FK506 treatment for either 2 or 4 weeks, the dogs experienced a delay in glucose disappearance in response to IV glucose injection. Insulin secretion during IVGll was unchanged in dogs treated for only 2 weeks, but was significantly decreased in dogs treated for 4 weeks. Following the recovery period, glucose disappearance during IVGll returned to normal in dogs that were treated for 2 weeks, and was more rapid than normal in dogs that had been treated for 4 weeks. Insulin secretion after the recovery period remained unchanged in group 1 dogs, but continued to be significantly reduced in group 2 dogs that had received FK506 for 4 weeks. No significant change was detected in the CBCs or serum chemistries. Copyright 0 1992 by W. B. Saunders Company
R
ECENT recognition has been given to FKS06 following the demonstration that it enables “rescue” of liver allografts from intractable rejection when conventional measures have been unsuccessful.’ Clinical trials using FK506 as the primary immunosuppressive agent for transplantation of liver, kidney, and thoracic organs’,’ and, sporadically, as a treatment regimen for autoimmune disorders,4 have also led to favorable results. As clinical experience with FK506 broadens, the profile of side effects and possible toxicities becomes clearer.” One such adverse effect that has been identified in experimental transplant animal models treated with FK506 is a derangement of glucose metabolism. This inconsistently reported abnormality varies from elevated fasting blood glucose levels in rodents’ to frank insulin-dependent diabetes mellitus in baboons.6 Due to the relatively recent development of this drug, few studies have been performed to characterize the effects of FK506 on carbohydrate metabolism in the normal individual. Therefore, to evaluate the effects of FK506 on glucose metabolism in the absense of organ allografts, we administered intravenous glucose tolerance tests (IVGTT) to normal dogs treated with FK506. In addition, we examined the effect of the drug on overall well-being of the animal and on hematologic indices and serum chemistries. MATERIALS
AND METHODS
Animals Eleven mongrels
beagles (Marshall (Motsinger Farm,
Farms, North Rose, NY) and two St. Joseph, IL), weighing between 9
From the University of Miami School of Medicine, Diabetes Research Institute, Miami, FL; and the Department of Surgery, University Health Center of Pittsburgh, University of Pittsburgh, Pittsburgh, PA. Supported by Project Grants No. DK25802 and DK29961, and the Diabetes Research Institute Foundation. Address reprint requests to Daniel H. Mintz, MD, Diabetes Research Institute (RI34). University of Miami School of Medicine, PO Box 016960 Miami, FL 33101. Copyright 0 1992 by WB. Saunders Company 0026-0495/92/4101-OOI2$03.00/0 64
and 14 kg and of both sexes, were studied. The dogs were fed dog chow (Hi-Pro, Purina Feeds, St Louis, MO) and allowed free access to water. Physical examinations were performed daily to assess body condition and attitude. AI1 procedures relating to animal studies were performed in accordance with the guidelines set forth by the National Institutes of Health and the University of Miami.
Ekperimental
Protocol
Dogs were randomly assigned to one of three groups. The first group of dogs (N = 5) was treated with FK506 for 2 weeks and the second group (N = 5) was treated for 4 weeks (Table 1). A third group (N = 3) served as controls and received no FK506. Subsequent to the treatment periods, the first and second groups of dogs were allowed a recovery period that was equal in length to their respective treatment periods (ie, dogs that were treated with FK506 for 2 or 4 weeks, were allowed a 2- or 4-week recovery period, respectively). During the recovery period, the dogs received no FK.506. All animals underwent an IVGW on three separate occasions: (1) baseline evaluation, (2) after treatment with FK506 for 2 or 4 weeks and (3) following the 2- or 4-week recovery period. After an over 1 minute overnight fast, glucose (0.5 g/kg) was administered through a percutaneously placed venous catheter. Peripheral venous blood samples were collected from a different vein before and 5, 10, 15, 20, 30, and 60 minutes after completion of the injection. Dogs were held unrestrained in their individual runs during the study. Complete blood cell counts (CBC) and serum chemistries (SMA-20) were also determined at that time. FK506 powder, a 20% solid dispersion in hydroxy-propyl-methyl cellulose and polyoxyethylated hydrogenated castor oil (Fujisawa Pharmaceuticals, Osaka, Japan), was placed in gelatin capsules and administered orally (1 mgikgld). Atropine was administered to prevent gastrointestinal side effects of FK506. All dogs received 2 mg subcutaneously twice a day during the first week. This dose was reduced to once a day during the second week. Dogs treated with FK506 for 4 weeks received 1 mg atropine subcutaneously once a day during the third week of treatment; no atropine was given during the fourth week. To control for possible effects of atropine. the third group of dogs (N = 3) received an identical dose of atropine. This group of animals underwent an IVGm at baseline, after 2 weeks of atropine treatment and 1 week after discontinuation of the atropine treatments. CBCs and serum chemistries were also obtained at those times.
Metabolism,
Vol41,
No
1
(January),
1992: pp 64-67
65
FK506 AND INSULIN SECRETION IN NORMAL DOGS
Table 1. Signalment and Clinical Parameters for Normal Dogs Treated With FK5g6 for 2 or 4 Weeks
Dog
Breed
Sex
1
Mongrel
?
FK506
Initial
% Weight
Gastrointestinal
Therapy
Weight (kg)
LOSS
Abnormalities
F
3
days
13.2
6
++++
12.5
4
++++
9.2
2
+
2
Mongrel
F
4 days
3
Beagle
F
2 weeks
4
Beagle
F
2 weeks
9.0
0
5
Beagle
M
2weeks
14.0
2
6
Beagle
F
2 weeks
10.0
7
7
Beagle
F
2 weeks
10.7
0
6
Beagle
M
4weeks
13.0
12
++
9
Beagle
M
4 weeks
13.5
15
++
10
Beagle
F
4 weeks
9.0
0
.P -g
I I
16
0
A
1
4
+ ++ + + 04
‘
a
10
12
14
16
18
+ Initial Dog Weight (kg)
Analytical Techniques Blood samples were collected into tubes containing 1.2 mg/mL NaEDTA and 500 kIU/mL Trasylol and immediately placed on ice. Following centrifugation and separation, plasma samples for immunoreactive insulin determination were stored at -70°C until assayed.’ Plasma glucose levels were measured immediately (Glucose Analyzer II, Beckman Instruments, Fullerton, CA). Data Analysis All results are expressed as the mean 2 SEM. Areas under the glucose and insulin concentration curves were calculated by the trapezoidal rule. Group means were compared using repeated measures ANOVA and the Student-Newman-Keuls t test (primer of Eiostatisfics:The Program, New York, NY, McGraw-Hill, 1988). Differences were considered significant if the corresponding P value was less than or equal to .05.
Fig 1. Relationship between weight loss 1% of initial body weight) and average initial body weight (0.8 to 10 kg; 0.10 to 12 kg; A, 12 to 14 kg; A, 14 to 16 kg; 0, 16 to 16 kg, N = 19) in dogs treated wfth FK506 (r = .96, P < .Ol).
to IVGIT in dogs who received atropine alone (data not shown). Blood glucose concentrations in response to IVG’IT before, during, and after treatment with FK506 for two weeks are shown in Fig 2A. FK506 treatment was associated with a delayed disappearance of glucose, which reached significance at 15 and 20 minutes (10.7 + 0.8 mmol/L v
RESULTS
Animals The hvo mongrels (Table l), developed intussusceptions on days 3 and 4 after FK506 treatment, respectively, and were killed. They were excluded from analysis of the data. Dogs that received atropine only (group 3) were clinically normal and had normal CBC and serum chemistry values (SMA-20) throughout the experiment. Physical examination of the dogs receiving FK506 (group 1 and 2) showed listlessness, loss of appetite, diarrhea, and weight loss (Table 1). The weight loss was greater in the larger dogs (Fig 1) and when FK506 was given for 4 rather than 2 weeks. Calculations of weight loss from this study, pooled with similar data from other ongoing experiments with FK506 (unpublished data), are shown in Fig 1. The larger dogs lost a significantly greater percent of their initial body weight during treatment with FK506 (r = .96, P < .Ol). However, after the drug was discontinued, all dogs regained their baseline weights and were clinically normal so that persistent effects of FK506 could be detected independent of weight loss. The CBC and serum chemistries (SMA-20) showed no significant changes during or following therapy with FK506. Peripheral Glucose and Insulin Concentrations No significant change from baseline values were observed in peripheral glucose or insulin concentrations in response
15
10
5
*
* OB 0
10
20
30
40
50
60
TIME (min) Fig 2. Glucose concentrations (mean + SEM) in response to IVGTT (0.5 mg/ kg IV) in normal dogs treeted with FKXt9 for (A) 2 weeks and (El 4 weeks; 0, Pre-FK: 0. on FK: n. off FK. lSignffkrantly different from control, P < .05.
ET AL
66
8.4 2 0.4 mmol/L and 8.9 + 1.0 mmol/L v 4.6 2 0.3 mmol/L, respectively, P < .OS). Two weeks after discontinuation of FK506, glucose concentrations in response to IV glucose returned to normal. Dogs treated with FK.506 for 4 weeks (Fig 2B) had glucose levels that were significantly higher at 20 and 30 minutes when compared with the baseline evaluation (11.0 & 0.8 mmol/L v 8.5 ? 0.8 mmol/L and 8.3 ? 0.6 mmol/L v 5.6 ? 0.7 mmol/L, respectively, P < .05). After discontinuation of FK506 for 4 weeks, glucose levels following IV glucose challenge returned to or declined to below normal levels. At the 15- and 20-minute time points, the glucose concentrations were significantly less than before drug treatment (7.0 t 1.3 v 10.8 ? l.Ommol/Land 5.4 * 1.0 v 8.4 + 0.8 mmol/L, respectively, P < .05). Peripheral insulin concentrations in response to IVGTT before, during, and after treatment with FK506 for 2 and 4 weeks are shown in Fig 3. After 2 weeks of treatment with FK506, the peak plasma insulin concentration in response to IV glucose was reduced at 10 minutes compared with pretreatment levels (134 ? 26 pmol/L v 259 ? 99 pmol/L, respectively), but this difference did not reach statistical significance (Fig 3A). Following discontinuation of the drug, the insulin levels in response to IV glucose returned to pretreatment levels.
I
p(O.05
L
1 I
p(O.05
I
1
i PE
On
Of‘
Pre
On
Off
Fig 4. Incremental areas under the glucose and insulin concentration curvas in response to IVGlT (0.5 mg/kg IV) expressed as a percent of the pretreatment area in normal dogs (N = 3) treated with FK506 for 4 weeks (0. glucose areas; q, insulin areas).
By contrast, insulin levels in response to IVG’IT after 4 weeks of treatment with FK506 (Fig 3B) were significantly decreased compared with pretreatment levels 10 and 15 minutes following administration of glucose (164 c 47 r’ 376 it 77 pmol/L and 139 t 58 v 388 + 87 pmol/L, respectively, P < .05). Moreover, in contrast to animals treated for only 2 weeks (Fig 3A), insulin responses to IV glucose following recovery from 4 weeks of FK506 therapy remained significantly depressed. Incremental areas under the glucose and insulin concentration curves during IVGTT in dogs treated with FK506 for 4 weeks are shown in Fig 4. After 4 weeks of FK506 therapy, the incremental glucose area tended to be higher than normal, but was not significantly different from baseline. However, following discontinuation of the drug for 4 weeks, the area under the glucose curve was diminished significantly compared with both the baseline and treatment areas (55% ? 13% v 100% ? 14% and 128% + 13%, respectively, P < .05). The incremental area under the insulin concentration curve was significantly decreased during FK506 treatment and remained significantly reduced 4 weeks after drug withdrawal (49% 2 23% and 47% * 11% 1~100% ? 22%, respectively, P < .05).
400 ? a E 4 5
KL.
DISCUSSION
/
300 * 200
*
z
4/
z
*
100 ”
* \I
A\.
0
I
0
lb
2-O
i0
-
io
SO
$0
TIME (min) Fig 3. Insulin concentrations (mean 2 SEM) in response to IVGIT (0.5 mg/kg IV) in normal dogs treeted wfth PK506 for (A) 2 weeks and (6) 4 weeks. 0, Pre FK; 0, on FK; n , off FK. *Significantly different from contr0l.P < .05.
Species-specific vulnerability to the toxic effects of FK506 have been well documented.8.Y In this experiment, we may have observed an intraspecies susceptability to the drug. The mongrel dogs in this study (Table 1, N = 2) were rapidly and severely debilitated by FK506 and developed intussusceptions, a complication that was not encountered in beagle dogs similarly treated (N = 19). Breed susceptibility to toxic effects of various drugs has been shown in veterinary medicine.“’ However, concentration of the genetic pool, which is the consequence of purpose-breeding to obtain purebred animals, like the beagles used in this study, may contribute in some way to the ability of these dogs to tolerate FK506. This intraspecies peculiarity requires further investigation for an adequate explanation.
FK506 AND INSULIN SECRETION IN NORMAL DOGS
67
In this study, we found that administration of FKS06 to normal beagles resulted in significant alterations in glucose disappearance rates following IV glucose injection. We have reported, as well, that glucagon-mediated insulin release is substantially inhibited at 2 weeks and nearly completely abolished after 4 weeks of FK506 therapy.” Thus, although we evaluated glucose metabolism here based on the response to a single secretagogue, we were able to find evidence to support at least two possible mechanisms that may contribute either separately or together to derangements in glucose homeostasis in FK506treated normal dogs. One mechanism implicated in the FKS06-induced glucose tolerance abnormalities is impaired insulin secretion. Despite normal fasting blood glucose levels before, during, and after the study, the response to IV glucose showed an impairment of insulin secretion. Specifically, 4 weeks of FKS06 treatment followed by a 4-week recovery period during which time the dogs returned to baseline weights and health status, revealed a persistent reduction in insulin secretion in response to IV glucose. The notion that this could be due to a direct p-cell effect is seemingly corroborated by in vitro dose-response studies of whole, cultured rodent and human islets,“.” which identify an FK506 related inhibition of glucose-mediated insulin secretion. A second possible mechanism responsible for the observed glucose intolerance is an alteration in insulin action. An apparent insensitivity to insulin’s action was not the focus of attention in this study; however, evidence in
support of this mechanism is based on the following observation. Both groups of dogs experienced significant improvement in IV glucose disappearance following discontinuation of the drug despite either (1) unchanged insulin concentrations, as in dogs that had been treated for only 2 weeks, or (2) significantly diminished insulin concentrations, as seen in dogs that had been treated for 4 weeks. Specific testing methodology, such as the euglycemic clamp or frequently sampled IV GTT,14 may prove useful in evaluating this particular drug-induced effect. Until this is done, the caveat exists that these alterations in insulin response during the recovery periods are driven by the glucose levels, and not the reverse. Based on the derangements we observed in this preliminary study, FK506, at 1 mg/kg/d orally, is tolerated in the beagle dog, but causes significant abnormalities in insulin secretion and glucose utilization that persist as long as the animal is receiving the drug. These abnormalities tend to reverse following discontinuation of the drug; however, normal insulin secretion was not re-established in animals that received FK506 for 4 weeks. Thus, our findings require further confirmation under conditions that control for weight gain and weight loss cycle in this and other species, but the observations reported here are similar to our findings in cyclosporine-treated beagles.” ACKNOWLEDGMENT
Special thanks are extended to the Diabetes Research Institute staff for dedicated technical and secretarial assistance.
REFERENCES 1. Fung J, Todo S, Jain A, et al: Conversion from cyclosporin to FK506 in liver allograft recipients with cyclosporin-related complications. Transplant Proc 22:6-12, 1990 2. Todo S, Fung JJ, Demetris AJ, Jain A, et al: Early trials with FK506 as primary treatment in liver transplantation. Transplant Proc 22:13-16,199O (suppl 1) 3. Todo S, Fung J, Starzl T, et al: Liver, kidney and thoracic organ transplantation under FK506. Am J Surg 212:295-307, 1990 4. McCauley J, Bronsther 0, Fung J, et al: Treatment of cyclosporin-induced heamolytic uraemic syndrome with FK506. Lancet 2:1516,1989 5. Nalesnik MA, Todo S, Murase N, et al: Toxicology of FK506 in the Lewis rat. Transplant Proc 19:89-92,1987 6. Thiru S, Collier D StJ, Calne R: Pathological studies in canine and baboon renal allograft recipients immunosuppressed with FK506. Transplant Proc 19:98-99,1987 7. Heding LG: Determination of total serum insulin (IRI) in insulin-treated diabetic patients. Diabetologia 8:260-266,1972 8. Todo S, Ueda Y, Demetris J, et al: Immunnosuppression of canine, monkey and baboon allografts by FK506: With special reference to synergism with other drugs and to tolerance induction. Surgery 104:239-249,198s
9. Collier D, Calne R, Thiru S, et al: FK506 in experimental renal allografts in dogs and primates. Transplant Proc 20:226-228, 1989 10. Paul AJ, Tranquilli WJ, Seward RL, et al: Clinical observations in collies given ivermectin orally. Am J Vet Res 48:684-685, 1987 11. Strasser S, Alejandro R, Ricordi C, et al: The effect of FK.506 on canine pancreatic islet cell function in beagle dogs. Transplant Proc 23:756-757, 1991 12. Tze W, Tai J, Cheung and rat islets. Transplantation
S: In vitro effects of FK506 on human 491172-l 174, 1990
13. Carroll P, Boschero A, Li M, et al: Effect of the immunosuppressant FK506 on glucose-induced insulin secretion from adult rat islets of Langerhans. Transplantation 51:275-278, 1991 14. Bergman RN, Prager R, Volund A, et al: Equivalence of the insulin sensitivity index in man derived by the minmal model method and the euglycemic glucose clamp. J Clin Invest 79:790800,1987 15. Alejandro R, Feldman E, Bloom A, et al: Effects of cyclosporin on insulin and C-peptide secretion in healthy beagles. Diabetes 38:698-703,1989
