Streptozotocin (STZ) is often used to induce diabetes in animal models. However, morbidity associated with STZ and its ability to induce diabetes vary with different dosages among different animal species, including nonhuman primates. To find an optimal dose of STZ that would cause diabetes with minimal toxicity, we compared low and high doses of STZ. Male cynomolgus monkeys (3–6 years old) were given a single dose of 100 mg/kg (high dose, 4 animals) or 55 mg/kg (low dose, 20 animals) of STZ. Blood glucose levels, intravenous glucose tolerance test (IVGTT), pancreatic biopsies, liver function tests (LFTs), liver biopsies, kidney function tests, and kidney biopsies were performed periodically. Animals from both groups developed diabetes within 24 h after administration of STZ. Serum C-peptide levels in both groups decreased from 2 to 8 ng/mL before STZ to between 0.01 and 0.6 ng/mL after STZ. Animals with the high dose of STZ developed transient vomiting within minutes after injection. During the first week after STZ injection, high-dose animals developed elevated LFTs, BUN and creatinine. In contrast, low-dose animals had normal liver and kidney function tests. Histological analysis showed that animals given the high dose of STZ developed marked steatosis of the liver and tubular injury in the kidneys, whereas animals given the low dose of STZ had normal-looking liver and kidney histology. The pancreatic islets in both groups were indistinguishable by immunoperoxidase staining for insulin, and showed either no insulin-positive cells or rare insulin-positive cells. Glucagon staining was normal. Over time, low-dose diabetic monkeys remained persistently hyperglycemic with negligible C-peptide stimulation by intravenous glucose. We conclude that low-dose STZ at 55 mg/mL successfully induces diabetes in cynomolgus monkeys with minimal liver and kidney toxicity.
A number of reports have described successful methods for inducing insulin-dependent diabetes in nonhuman primates (1–3). These methods include total pancreatectomy or treatment with streptozotocin (STZ) at a variety of different doses. Streptozotocin, derived from Streptomyces achromogenes, is toxic to beta cells. The preferential death of beta cells is due to its uptake in these cells via the GLUT-2 glucose transporter.
Both pancreatectomy and STZ induce diabetes, but they can also cause animal morbidity and mortality. Pancreatectomy often leads to abnormal exocrine function requiring daily supplements of oral pancreatic enzymes (4). In STZ-injected animals complications include liver and kidney toxicity (5). To identify a STZ dose that induces diabetes without kidney and liver toxicity, we tested two doses of STZ in cynomolgus monkeys.
Materials and Methods
Male cynomolgus monkeys (Macaca fascilularis) were 3–6 years of age and weighed 3.1–9 kg (Charles River/Biomedical Resource, Inc, Houston, TX) and were fasted overnight. The next morning the animals were anesthetized with intramuscular ketamine 10–15 mg/kg and hydrated with 50–60 cc of normal saline (NS). The two doses of STZ (100 mg/kg or 55 mg/kg) were diluted with 10 cc of NS and given rapidly i.v. immediately after dilution. Additional hydration with NS 100–150 cc was given.
Animal care after streptozotocin injection
Diabetic animals were treated with 2–3 injections of insulin per day (6–15 units a day). The blood glucose levels were checked 2–3 times a day. A number of tests were performed on a weekly basis, including a complete blood count (CBC), electrolytes, creatinine (CRE), blood urea nitrogen (BUN), SGOT, alkaline phosphatase, and bilirubin. All animals in both groups were given intravenous normal saline weekly. Serum C-peptide levels were measured by radioimmunoassay (Human C-peptide RIA Kit, Linco research, Inc., St. Charles, Mo). The assay has a 90% cross reactivity with cynomolgus monkey C-peptide. Statistical analysis was done by Wilcoxon Signed Rank Test or anova.
Isolation of donor Islets
Distal pancreatectomy was performed on the donor animal on the day of islet transplantation. The distal pancreas was mobilized medially until the superior mesenteric vessels were encountered. At this point the splenic artery and vein were isolated and ligated. A clamp was placed across the pancreas at the level of the superior mesenteric vessels and the distal pancreas was cut free and removed from the field. A distal pancreatectomy removes approximately 70% of the pancreas. The pancreas was then perfused via the pancreatic duct with Universal of Wisconsin (UW) solution (warm ischemia time less then 5 min) and transported on ice to the JDRF islet isolation facility at Joslin Diabetes Center (cold ischemia time less then an hour). The pancreas was distended with Liberase HI (Roche Biochemicals, Indianapolis, IN) and was incubated at 37 °C in a static digestion chamber for 45 min. The digested tissue was collected and applied to a three-layer discontinuous Euroficoll gradient (densities of 1.112, 1096 and 1.060). The pancreatic tissue was bottom-loaded with the 1.112 layer and centrifuged at 900 × g for 22 min at 4 °C. Three 50-μL samples were stained with dithizone and counted to assess total islet cell yield. Samples were also taken for DNA content, insulin staining, viability and histology. Finally, the total number of islets were calculated as islet equivalent (IEQ) with an average diameter of 150 μm per islet. A total of 7000 IEQ/kg were then transplanted (day 0) into the portal vein in a 10-mL volume of 0.9% sodium chloride.
Intravenous glucose tolerance test (IVGTT)
An IVGTT was performed at various times during the study. Food was withheld for 15 h before testing. A 22-gauge intravenous catheter was placed and 0.5 g/kg of glucose in a 25% glucose solution was injected over 1 min. Blood samples were collected at 0, 1, 3, 5, 10, 15, 20, 25, 30, 45 and 60 min post glucose injection. Serum C-peptide levels were measured by radioimmunoassay.
Pancreatic and kidney biopsies were obtained at different times during the study. After sacrifice, complete autopsies were done, including tissue samples from pancreases, livers and kidneys. Tissues were fixed in 10% formalin, embedded in paraffin, sectioned at 4–5 μm, and stained with hematoxylin and eosin (H&E) (6). Pancreases were stained by indirect immunoperoxidase immunohistochemistry for insulin and glucagon (6). To enumerate the number of insulin-staining cells per islet in streptozotocin-treated animals, 50 islets at 100–200× magnification from all diabetic animals were scored for the number of insulin-staining cells.
Table 1 summarizes the analysis of the 24 monkeys (4 with the high dose and 20 with the low dose of STZ) treated in this series. The table shows blood glucose and C-peptide levels taken 2 days after STZ injections. CRE and SGOT levels are shown at 5–7 days and 5–9 days, respectively, after the STZ.
Table 1. : Laboratory tests after streptozotocin
STZ dose (mg/kg)
Results from monkeys treated with 100 and 55mg/kg STZ: Fasting blood glucose and C-peptide levels were assayed on day 2. Creatinine and SGOT were tested on days 5–7 and 5–9, respectively. NR = normal range. Statistical significance calculated by the Wilcoxon Signed Rank Test.
(NR=40–100mg/dL) Day 2
(NR=0.1–1.5mg/dL) Day 5–7
(NR=2–8ng/mL) Day 2
(NR=9–150U/L) Day 5–9
Following STZ injection, all animals in both STZ dose groups developed insulin-dependent diabetes within 24 h, with fasting blood glucose levels > 400 mg/dL, and loss of body weight (up to 20%). There was no statistical difference in the blood sugars after STZ injection between the high- and low-dose groups. Serum C-peptide levels from both experimental groups decreased from 2 to 8 ng/mL before STZ injection to between 0.01 and 0.6 ng/mL after STZ injection. Again, there was no statistical difference between C-peptide levels in the high- and low-dose STZ groups. These findings show that both doses induce insulin-dependent diabetes in these animals and were confirmed by histological and immunohistochemical staining for insulin of the pancreas postmortem or from pancreatic biopsies (Figure 1a,b). C-peptide levels remained less then 0.6 ng/mL in all animals up to 1 year. Measurements were taken at 1, 2 weeks and 2, 4, 9 and 12 months. C-peptide levels did not normalize in any animal.
Very few islets showed insulin staining as early as 3 days after STZ injection (Figure 1a, representative islets from 15 different STZ-treated animals), but they showed a normal pattern of staining for glucagon (Figure 1bC, D & E). In contrast, before STZ injection, islets showed a normal pattern of insulin and glucagon staining (Figure 1b, top panels, A & B). The dramatically decreased number of insulin-staining cells in the 15 islets in Figure 1(a) from streptozotocin-treated animals is representative and diagnostic of the insulin staining of streptozotocin-treated animals. Normal islets have more than 10 insulin-staining cells per islet, and many have in excess of 30–40 insulin-positive cells. The number of insulin-staining islets in 50 islets from diabetic animals treated with a low dose of streptozotocin was counted. As shown in Figure 1(c), diabetic animals had only small numbers of insulin-staining cells (typically less than 5). Although some regeneration of beta cells could occur, this pattern of insulin staining was seen in diabetic animals even a year after streptozotocin treatment.
To confirm the persistence of the diabetic state after streptozotocin, intravenous glucose tolerance tests were performed the day before STZ injection and after STZ (from 14 to 360 days). As shown in (Figure 2), essentially no C-peptide response occurred in low-dose STZ-treated animals (p = 0.11, by anova). Thus, C-peptide release by isolated insulin-positive cells in low-dose STZ-treated animals could not be detected by IVGTT.
Animals in the high-dose group developed transient vomiting within minutes after STZ injection. Within the first 12 h, animals in both groups became transiently hypoglycemic, presumably due to the insulin release from dying beta cells. Kidney function tests in the high-dose-treated animals showed increased levels of BUN (25–50 mg/dL) and CRE (1.8–6.1 mg/dL) (Table 1). In contrast, animals given the low dose of STZ had kidney function within the normal range (BUN 8–25 mg/dL, and CRE 0.7–1.8 mg/dL) (Table 1). Liver function tests in the high-dose animals showed elevated liver function tests with total bilirubin levels greater than 1 mg/dL and with SGOT levels of 179–310 U/L (Table 1). In contrast, the liver function tests were normal in the low-dose animals (total bilirubin levels 0.0–1.0 mg/dL, and SGOT levels 42–145 U/L) (Table 1). The creatinine and SGOT levels were statistically significantly different (p < 0.001) between the high and low STZ dose groups. Biopsies of the liver and kidney from high-dose animals showed severe hepatic steatosis and acute renal tubular injury (Figure 3, left panels). In contrast, biopsies after the low dose of STZ showed no steatosis or acute tubular injury (Figure 3, right panels). These histological findings confirm that high-dose STZ induces acute tissue injury in both the liver and the kidney.
Figure 4 shows the blood-sugar levels of one monkey treated with 55 mg/kg STZ that also underwent islet transplantation. The animal became normoglycemic for 10 days after islet transplantation, then rejected its islet graft, and again became hyperglycemic. C-peptide levels pre-STZ were 3.63 ng/mg, and after STZ on the day of transplant were 0.18 ng/mL. During islet graft function the levels varied from 1.06 to 1.91 ng/mL. On the day the animal was sacrificed (day 45), the C-peptide levels were 0.2 ng/mL. Histology of the host pancreas showed negative insulin staining in the islets and positive staining for glucagon, similar to that found in Figure 1(a and b). Histology of the liver (20 sections) showed no detectable islets. In our experience, monkeys with surviving islet allografts have detectable islets in one of five sections of liver (6).
Different strategies to induce diabetes in nonhuman primates have been reported. Pancreatectomy (4) or STZ at different doses, from 30 mg/kg up to 150 mg/kg, have been used (7,8). Litwak et al. tried to induce diabetes in cynomolgus monkeys using a dose of 30 mg/kg and found that it did not reliably cause diabetes (7). Only three of eight animals became mildly hyperglycemic and the rest had to be given a second dose of STZ (7). In another study, Pitkin and Reynolds examined the diabetogenic effects of STZ in rhesus monkeys at three different doses (30, 45 and 60 mg/kg) (5). They found that animals given 30 mg/kg had normal glucose tolerance. Five out of seven animals given a dose of 45 mg/kg developed hyperglycemia, and two out of two animals given 60 mg/kg developed hyperglycemia (5). Stegall et al. reported that 80% of cynomolgus monkeys became diabetic after 50 mg/kg STZ and remained hyperglycemic for up to 1 year (1). These differences indicate that there are variations between different species of nonhuman primates in their response to STZ.
Our results using 55 mg/kg of STZ showed that we could induce diabetes in 100% of our cynomolgus monkeys for up to 1 year without any evidence of regeneration of the beta cells. Successful islet transplantation can be performed in low-dose-treated animals (Figure 4). In contrast to our findings, Thieriault et al. found that they needed 150 mg/kg STZ to induce diabetes in cynomolgus monkeys (8). Within the first 72 h after this high dose, they observed acute elevation in hepatic enzymes and BUN. However, these numbers normalized within 72 h (8). One of the reasons for the discrepancy between our results and those of the above study may be that they used younger monkeys, 6–8 months old (1.0–2.0 kg), and we and others have used older monkeys, 3–6 years old (3–9 kg). Young monkeys may be more resistant to the beta-cell STZ toxicity or need larger doses to cause diabetes. Another variable that may influence the results is the way in which STZ is administered. Streptozotocin has a short half-life (about 10 min) and therefore requires rapid infusion.
It should be emphasized that the use of STZ at either of the doses we tested does not render the monkeys absolutely C-peptide negative, as would be the case after total pancreatectomy. We have not found this to be a problem for our experiments that are designed to test the survival of allogeneic islet transplants, because the degree of hyperglycemia is so severe and the level of C-peptide is so low in the absence of a successful transplant. However, the low level of C-peptide might make it difficult to interpret the results of experiments designed for other purposes. For example, sensitive testing of the function of marginal numbers of islets might be difficult in animals that still produce small amounts of C-peptide on their own (6).
We successfully induced stable diabetes using a low dose of STZ in over 20 male cynomolgus monkeys (3–6 years old) with minimal toxicity to the kidney or liver. Low-dose streptozotocin at 55 mg/kg can be used as a satisfactory protocol for the induction of diabetes for preclinical transplantation experiments.
The authors thank Luba Zachachin for expert assistance with tissue processing. This project was supported by the Juvenile Diabetes Research Foundation International Center for Islet transplantation at Harvard Medical School and by the NIH grant PO1 DK53087.