Dolphins as animal models for type 2 diabetes

General and Comparative Endocrinology 170 (2011) 193–199

Contents lists available at ScienceDirect

General and Comparative Endocrinology journal homepage: www.elsevier.com/locate/ygcen

Dolphins as animal models for type 2 diabetes: Sustained, post-prandial hyperglycemia and hyperinsulinemia Stephanie Venn-Watson ⇑, Kevin Carlin, Sam Ridgway National Marine Mammal Foundation, 2240 Shelter Island Drive Suite 200, San Diego, CA 92106, USA

a r t i c l e

i n f o

Article history: Received 4 June 2010 Revised 4 October 2010 Accepted 6 October 2010 Available online 15 October 2010 Keywords: Bottlenose dolphin Comparative animal model Diabetes Glucagon Hemochromatosis High-protein diets Insulin resistance

a b s t r a c t There is currently no known natural animal model that fully complements type 2 diabetes in humans. Criteria for a true natural animal model include the presence of a fasting hyperglycemia, evidence of insulin resistance, and pathologies matching that reported in humans. To investigate the bottlenose dolphin (Tursiops truncatus) as a comparative model for type 2 diabetes in humans, hourly plasma and urine chemistry changes, including glucose, were analyzed among five healthy, adult dolphins for 24 h following ingestion of 2.5–3.5 kg of mackerel or 2–3 L of 10% dextrose in ionosol. Fasting and 2 h post-prandial insulin levels were also determined among five adult dolphins to assess the presence of hyperinsulinemia. Finally, a case-control study compared insulin and glucagon levels among dolphins with and without iron overload, a condition associated with insulin resistance in humans. Both protein and dextrose meals caused significant increases in plasma glucose during the 0–5 h post-prandial period; dolphins fed dextrose demonstrated a sustained hyperglycemia lasting 5–10 h. Fasting plasma insulin levels among healthy dolphins mimicked those found in humans with some insulin resistance. Dolphins with hemochromatosis had higher post-prandial plasma insulin levels compared to controls. We conclude that bottlenose dolphins can demonstrate metabolic responses consistent with type 2 diabetes, specifically sustained hyperglycemia and hyperinsulinemia. Understanding more about how and why dolphins have a diabetes-like metabolism may provide new research avenues for diabetes in humans. Published by Elsevier Inc.

1. Introduction According to the World Health Organization, approximately 5% of all deaths globally are due to diabetes. Without action, deaths from diabetes are likely to increase by more than 50% over the next 10 years (World Health Organization and Diabetes, 2009). Of people with diabetes, 90–95% have type 2 diabetes, previously called insulin-resistant or adult onset diabetes. A single, natural species has not been identified that fully complements type 2 diabetes in humans (Cefalu, 2006; Kaplan and Wagner, 2006). Discovery of such a model may lead to novel means of preventing, treating, and curing this disease for people. When comparing common bottlenose dolphins (Tursiops truncatus) that are fasted overnight to those recently fed, overnight fasted dolphins demonstrate changes in platelet counts and serum chemistries that mimic those of people with type 2 diabetes (VennWatson and Ridgway, 2007). Changes which parallel those found in people with type 2 diabetes include sustained increases in glucose; increased platelets (Sterner et al., 1998); increased serum gamma-glutamyl transpeptidase (Andre et al., 2006) and alkaline

⇑ Corresponding author. Fax: +1 619 553 5068. E-mail address: [email protected] (S. Venn-Watson). 0016-6480/$ - see front matter Published by Elsevier Inc. doi:10.1016/j.ygcen.2010.10.005

phosphatase (Maxwell et al., 1986); decreased serum uric acid (Nan et al., 2007); and shifts toward a metabolic acidodic state (Androgue et al., 1982). The key difference found between dolphins and people with diabetes, however, is that dolphins appear to turn a diabetes-like state on and off with overnight fasting and daily feeding, respectively. Criteria for diabetes in humans includes random blood glucose levels greater than 200 mg/dl, fasting glucose levels greater than 126 mg/dl, and/or glucose levels greater than 200 mg/dl following an oral glucose tolerance test (Alberti and Zimmet, 1998). Type 2 is typically differentiated from type 1 diabetes from patient history, but insulin resistance and accompanying hyperinsulinemia in type 2 diabetes may be detected. Criteria have been suggested to assess whether or not an animal model fully complements type 2 diabetes in humans (Cefalu, 2006; Kaplan and Wagner, 2006). These criteria include demonstrated sustained post-prandial hyperglycemia, disease complications similar to that seen in people with type 2 diabetes, and a shared evolutionary history of disease. Current animal models that have provided insight into type 2 diabetes include mice, desert rodents, cats, pigs, and old world monkeys (Bellinger et al., 2006; Henson and O’Brien, 2006; Neubauer and Kulkarni, 2006; Shafrir et al., 2006; Wagner et al., 2006). While these animal models mimic parts of type 2 diabetes, none meet all the criteria.

194

S. Venn-Watson et al. / General and Comparative Endocrinology 170 (2011) 193–199

2.2. Protein and dextrose feeding studies

Physiological similarities identified only in primates and cetaceans (toothed whales, porpoises, and dolphins) exist that support a shared drive for common glucose metabolism. Of all terrestrial and aquatic animals tested, only primates and cetaceans have red blood cells that are ‘extraordinarily’ permeable to glucose (Craik et al., 1998). Humans and bottlenose dolphins also share high encephalization quotients (EQ) Marino, 1998, a measurement of actual brain size compared to the expected brain size given the body mass (Jerison, 1973). EQs for humans, bottlenose dolphins, and chimpanzees are 7.4, 5.3, and 2.5, respectively. Given that large brains require readily available blood glucose in order to function, it is hypothesized that humans and dolphins have high blood glucose carrying capacity in their blood to support their shared large brain size (Goodwin, 1956). Dolphins and humans are susceptible to hemochromatosis, also called iron overload, which is responsive to phlebotomy treatment (Johnson et al., 2009). Ferritin is the true measure of iron in blood, and high blood ferritin is reflected in iron overload. In humans, the severity of insulin resistance is associated with increasing serum ferritin levels and hemochromatosis (Wrede et al., 2006). Further, the presence of diabetes mellitus has been identified as a primary risk factor for increased ferritin levels in humans with chronic hepatitis C (D’Souza et al., 2005). To further assess the bottlenose dolphin as a model for type 2 diabetes in humans, we analyzed hourly urine and plasma glucose changes among healthy dolphins fed either a high protein fish meal or dextrose. We also measured fasting plasma insulin levels among healthy dolphins and compared fasting and 2 h post-prandial insulin levels among dolphins with and without hemochromatosis. Our study results were compared to those found in humans with and without insulin resistance and type 2 diabetes.

Six healthy adult bottlenose dolphins were included in nine feeding trials. Animals were fasted, fed 2–3 L 10% dextrose in ionosol, or fed 4.5–5.4 kg Spanish mackerel (Table 1). Health was determined by behavior, appetite, and clinical blood values (Ridgway et al., 1970). All animals were fasted 12 h overnight before the start of each study. The animals were taken from their seawater habitat in a fleece-lined transport sling. They were placed on their sides on a soft rubber pad. An initial blood sample was taken from the central vessels of the fluke. A standard 8–14 Fr. clinical urinary catheter with inflatable cuff was inserted through the urethra into the urinary bladder. The cuff was inflated with 15–20 ml of normal saline to retain the catheter in the bladder throughout the study. An initial urine sample was taken and the collection end of the catheter was pulled through a small central hole in the transport sling and connected to a length of tubing. The dolphin was then lifted into a transport container that had been partially filled with seawater (34 parts per thousand salinity) from the dolphin’s home pool. The level of seawater in the container was sufficient to keep the water surface just above the level of the dolphin’s eyes. During the experiment, seawater was poured or sponged over the dorsal portion of the animal to prevent drying. The catheter tubing was connected through a port in the dolphin transport container and inserted into a urine collection vessel. 2.2.1. Sample collection Blood and urine sample collections were conducted every hour over a desired 24–25 h. In some cases, issues related to catheter placement or animal comfort did not enable studies to persist for the full 24–25 h; the shortest studies (n = 2) lasted 16–17 h. For blood collection, the dolphins flukes were carefully lifted free of the water for obtaining blood samples of approximately 20 ml using a sterile 20 gauge disposable hypodermic needle. At hourly intervals, the urine collection vessel was emptied, measured, and an aliquot saved in a standard urinalysis vial.

2. Methods 2.1. Ethical statement Dolphins involved in this study were part of the U.S. Navy Mammal Program (MMP) population. The MMP is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International and adheres to the national standards of the United States Public Health Service Policy on the Humane Care and Use of Laboratory Animals and the Animal Welfare Act. As required by the Department of Defense, the MMP’s animal care and use program is routinely reviewed by an Institutional Animal Care and Use Committee (IACUC) and the Navy Bureau of Medicine. The protein and dextrose feeding studies were conducted in 1970 and followed the rules applicable to animal research at that time; to minimize unnecessary duplication of effort, the authors used raw, archived datasets from these studies to address our current study hypothesis. The study protocol used to determine fasting and post-prandial plasma insulin measurements among the current MMP dolphin population was reviewed and approved by the MMP IACUC and the Navy Bureau of medicine.

2.2.2. Blood and urine diagnostics Study plasma and urine variables included urea (mg/dl), sodium (mEq/L), chloride (mEq/L), potassium (mEq/L), and osmolality (mOsm/kg). Plasma and urine glucose (mg/dl) were measured during the fasting studies and the study involving seawater with protein ingestion. Specimens were stored at 4 °C. Periodically, plasma was collected from each heparinized blood tube after cells had settled. Plasma and urine were transported to a clinical laboratory experienced with dolphin specimens (Bioscience Laboratories, Van Nuys, California) for analysis. The following methodologies were employed: plasma urea, Autoanalyzer (Technicon SMA-12, Technicon Corp. Ardaley, NY); urine urea, modified urease and Berthelot method; sodium and potassium, flame photometry (Instrumentation Laboratory, Inc., Waterman, MS.); chloride, Buchler-Cotlove chloridometer (Buchler Instruments, Inc., Fort Lee, NJ). Urine and plasma osmolality were measured by freezing point

Table 1 Descriptions of nine feeding studies among six adult bottlenose dolphins (Tursiops truncatus). Study number

Study type

Total hours

Study date

Animal ID

Plasma samples (#)

Urine samples (#)

1 2 3 4 3 6 7 8 9

Fasting Fasting 10% Dextrose in ionosol G (3 L) 10% Dextrose in ionosol G (2 L) Mackerel (10 lb) Mackerel (11 lb) Mackerel: (10 lb) Mackerel: (10 lb) Mackerel: (12 lb)

16 23 17 25 24 24 23 24 23

01 25 01 24 13 13 16 26 25

A B C D E F F F B

16 23 13 17 15 16 16 16 17

15 23 14 17 24 26 26 26 15

May November May September March March October November September


depression with a Fiske osmometer (Advanced Instruments Inc., Norwood, MA). 2.2.3. Excretion rates and urine-to-plasma osmolality ratios Hourly excretion rates of solutes were calculated as mEq/ min = (total urinary solute/60). Urine to plasma osmolality ratio = urine osmolality/plasma osmolality. 2.2.4. Statistics conducted in 2009 Data were analyzed using SASÒ software (Release 9.2; SAS Institute Inc., Cary, NC). Within the dextrose and mackerel study groups, changes in urine and plasma values over time were assessed by comparing mean values among the following time categories: 0–5 h, >5–10 h, >10–15 h, and >15 h; a one-sided analysis of variance with a general linear model was used with post-hoc comparisons among each of the time categories (PROC GLM; CLASS TIMECATEGORY; MODEL [serum and urine variables] = TIMECATEGORY; MEANS TIMECATEGORY/SCHEFFE; BY STUDY_TYPE). 2.3. Fasting and 2h post-prandial plasma insulin and glucagon levels and hemochromatosis case-control study

195

ric = greater than 40 years) using a one-way analysis of variance (PROC GLM; CLASS animal age_category; MODEL fast_insulin post_insulin post_glucagon = animal age_category). Post-hoc Scheffe’s tests were run for intergroup comparisons. P value less than 0.05 was defined as significant. For the hemochromatosis case-control study, five healthy bottlenose dolphins (controls, described above) and three dolphins with hemochromatosis (cases) were included in the study. All cases have been previously described in the literature as having hemochromatosis, including high serum iron levels (>300 lg/dl) for greater than 5 years and histopathologic confirmation of disease with liver biopsies (Venn-Watson et al., 2008; Johnson et al., 2009). All controls had no evidence of hemochromatosis, including normal serum iron levels and normal aminotransferases for at least 1 year previous and up to the study period. Mean values for fasting plasma insulin and 2 h post-prandial insulin and glucagon were compared among cases and controls using a one-way analysis of variance (PROC GLM; CLASS case_control; MODEL fast_insulin post_insulin post_glucagon = case_control). P value less than 0.05 was defined as significant. 3. Results

2.3.1. Study populations Five healthy bottlenose dolphins were included in the fasting and post-prandial study. Three (60%) were female, and the animals were aged 8, 16, 27, 31, and 48 years. A total of 30 samples were collected from the five control animals, a result of weekly, controlled, 2 h post-prandial sampling over 6 weeks. For the hemochromatosis case-control study, triplicate fasting and 2 h post-prandial samples were included from the five healthy bottlenose dolphins and three adult dolphins with hemochromatosis confirmed by live biopsy histopathology. 2.3.2. Sample collection Two hour post-prandial blood samples were collected a mean of 137 min (median 113 min) following initial morning meals once a week for 6 weeks for each of the five healthy, control animals. Fasted blood samples were collected following >12 h of overnight fasting. Blood samples were collected by venipuncture from dolphins trained to voluntarily present their tails (caudal flukes) for sampling by veterinary technicians. Up to 10 ml of whole blood were collected using 20- or 21-gauge, 1.5-inch needles in blood collection tubes (Vacutainer Systems, Becton Dickinson, Rutherford, NJ) containing ethylenediaminetetraacetic acid (EDTA). Blood was spun in a Fisher Scientific AccuSpin 3 benchtop centrifuge (Pittsburgh, PA) at 3000 RPM for 10 min and separated prior to storage. Plasma samples were stored in a Revco ultra-low freezer (Asheville, NC) at 80 °C until submission for analysis of insulin levels. 2.3.3. Insulin diagnostics Plasma samples were analyzed at Esoterix Inc. Laboratory Services in Calabasas Hills, California. Insulin was measured by twosite immunochemiluminometric assay (ICMA). Insulin present in the plasma was complexed between the bead-bound antibody and the liquid-phase labeled antibodies utilized in the assay. Bound complexes were measured in luminometers (Ciba Corning Diagnostics Corp. MAGIC Lite Analyzer II, Medfield, MA; Berthold Technologies AutoLumatPlus Model LB953, Oak Ridge, TN). The relative light units generated were directly proportional to insulin concentration in the plasma. Plasma insulin levels were reported in lIU/ml units.

3.1. Protein and dextrose feeding studies Following ingestion of 2–3 L of 10% dextrose in ionosol, dolphin plasma glucose levels were significantly higher during 0–5 h and 5–10 h compared to 10–15 h and 15–24 h (P = 0.0002, Fig. 1). Urine flow rate was significantly higher during the first 5 h after ingestion (2.8 ± 2.2 ml/min) compared to 15–24 h after ingestion (1.0 ± 0.4 ml/min P = 0.01) (Table 2). Similar to trends found with plasma glucose, plasma chloride levels were significantly higher during 0–5 h and 5–10 h compared to 10–15 h and 15–24 h (P < 0.0001). After ingestion of 4.5–5.5 kg of Spanish mackerel, plasma glucose levels were significantly higher during 0–5 h compared to 5–10 h, 10–15 h, and 15–24 h (P = 0.002, Table 3). Urine flow rate was significantly higher during 0–5 h and 5–10 h after ingestion (3.3 ± 1.6, 3.4 ± 1.5 ml/min, respectively) compared to 10–15 and 15–24 h after ingestion (2.4 ± 1.4, 1.0 ± 0.5 ml/min, respectively). Urine concentrations of sodium, chloride, potassium, and urea were highest during the first 10 h after ingestion. 3.2. Fasting and 2 h post-prandial plasma insulin and glucagon levels Fasted, healthy dolphins had mean ± SEM fasting insulin levels of 12 ± 3.6 lIU/ml. Among the five healthy control animals there were significant differences in mean 2 h post-prandial plasma insulin and glucagon values when comparing animals by age category (P = 0.005); the two young dolphins were more likely to have lower 2 h post-prandial glucagon compared to two adults and one geriatric dolphin (53.4 ± 11.3, 110 ± 17.7, 127.5 ± 10 pg/ml, respectively). The geriatric dolphin was more likely to have higher 2 h post-prandial insulin compared the two adult and two young dolphins (14.3 ± 2.3, 6.4 ± 1.4, 4.8 ± 1.1 lIU/ml, respectively). While there was no significant difference in mean 2 h post-prandial insulin by sex, controlling for age, (males = 8.9 and females = 6.9 lIU/ ml; P = 0.36), males were significantly more likely to have higher 2 h post-prandial glucagon compared to females (males = 145.9, females = 77.1 pg/ml; P = 0.001). 3.3. Hemochromatosis case-control study

2.3.4. Statistics Mean, standard deviation, median and range values were determined for fasting plasma insulin and 2 h post-prandial insulin and glucagon. Mean values were compared by animal and age category (young = less than 17 years; adult = 17–32 years; geriat-

There were no significant differences in mean age and sex distribution among cases and controls (mean case age = 32.9 y (SD = 7.8), mean control age = 25.1 y (SD = 14.8), P = 0.32; percent case females = 50%, percent control females = 80% P = 0.52). When

196


Fig. 1. Plasma glucose (mg/dl) by hour for bottlenose dolphins (Tursiops truncatus) that were (a) fasted (last 24 h of a 36 h fast), (b) tube fed 2–3 L 10% dextrose in ionosol, or (c) fed 10–12 lb Spanish mackerel.

Table 2 Comparisons of mean selected blood and urine values, by time after feeding challenge, in two bottlenose dolphins (Tursiops truncatus). Variable

2–3 L 10% Dextrose in ionosol solution

P value

>0–5 h (n = 9)

>5–10 h (n = 8)

>10–15 h (n = 10)

>15 h (n = 11)

Urine flow rate (ml/min) Plasma osmolality (mOsm/kg) Urine osmolality (mOsm/kg) Urine:plasma osmolality ratio

2.8 ± 2.2 337 ± 2 1780 ± 159 5.3 ± 0.7

1.9 ± 1.0 333 ± 2 1791 ± 108 5.2 ± 0

1.4 ± 0.5 336 ± 5 1524 ± 82 4.5 ± 0.2

1.0 ± 0.4 339 ± 0 1757 ± 76 5.2 ± 0.2

0.01 0.28 0.02 0.08

Sodium Urine concentration (mEq/L) Excretion (mEq/min) Plasma (mEq/L)

26 ± 23 2.4 ± 0.5 152 ± 4

12 ± 5 2.1 ± 0.6 151 ± 3

13 ± 7 2.8 ± 1.0 156 ± 2

13 ± 15 2.6 ± 2.2 151 ± 5

0.20 0.70 0.11

Chloride Urine concentration (mEq/L) Excretion (mEq/min) Plasma (mEq/L)

32 ± 26 3.5 ± 1.4 109 ± 2

21 ± 4 4.1 ± 2.0 110 ± 2

22 ± 11 4.7 ± 1.2 114 ± 1

19 ± 15 4.0 ± 2.0 115 ± 2

0.41 0.54 4

1,2 < 3,4

2,1 > 4,3

197

S. Venn-Watson et al. / General and Comparative Endocrinology 170 (2011) 193–199 Table 3 Comparisons of mean selected blood and urine values, by time after feeding challenge, in five bottlenose dolphins (Tursiops truncatus). Variable

a

4.5–5.6 kg Spanish Mackerel

P value

Significant comparisons

>0–5 h (n = 34)

>5–10 h (n = 35)

>10–15 h (n = 34)

>15 h (n = 60)

Urine flow rate (ml/min) Plasma osmolality (mOsm/kg) Urine osmolality (mOsm/kg) Urine:plasma osmolality ratio

3.3 ± 1.6 346 ± 12 1669 ± 233 4.9 ± 0.7

3.4 ± 1.5 344 ± 8 1815 ± 180 5.4 ± 0.5

2.4 ± 1.4 342 ± 8 1581 ± 429 4.8 ± 0.5

1.0 ± 0.5 345 ± 13 1566 ± 442 4.5 ± 1.2

3 > 4

Sodium Urine concentration (mEq/L) Excretion (mEq/min) Plasma (mEq/L)

20 ± 16 1.8 ± 1.0 156 ± 6

16 ± 12 1.4 ± 0.7 155 ± 5

11 ± 8 1.3 ± 0.6 154 ± 4

4.8 ± 3.4 1.3 ± 0.8 156 ± 6

3,4

Chloride Urine concentration (mEq/L) Excretion (mEq/min) Plasma (mEq/L)

23 ± 19 2.0 ± 1.1 112 ± 7

24 ± 18 2.0 ± 0.9 112 ± 6

18 ± 12 1.9 ± 0.4 114 ± 6

9±4 2.5 ± 0.9 112 ± 7

4

Potassium Urine concentration (mEq/L) Excretion (mEq/min) Plasma (mEq/L)

17 ± 8 1.4 ± 0.3 3.9 ± 0.3

14 ± 5 1.2 ± 0.4 3.7 ± 0.4

10 ± 3 1.3 ± 0.6 3.8 ± 0.6

7±4 1.8 ± 0.8 3.9 ± 1.1

3.4 4 > 2,3

Urea Urine concentration (mEq/L) Excretion (mEq/min) Plasma (mEq/L)

5.3 ± 2.9 0.04 ± 0.02 47 ± 6

6.7 ± 2.7 0.05 ± 0.02 51 ± 6

4.8 ± 2.4 0.05 ± 0.02 50 ± 10

2.2 ± 1.4 0.04 ± 0.02 52 ± 14

3,4

Glucosea Urine concentration (mg/dl) Excretion (mg/min) Plasma (mg/dl)

19 ± 8 0.2 ± 0.2 160 ± 30

20 ± 11 0.2 ± 0.1 124 ± 31

18 ± 8 0.2 ± 0.2 125 ± 25

48 ± 68 0.8 ± 1.1 120 ± 29

0.15 0.07 0.002

1 > 2,3,4

Only three studies included measurements of glucose.

comparing three adult dolphins with and five adult dolphins without hemochromatosis, controlling for age and sex, there were no significant differences in fasting plasma insulin (least squares means, cases = 12.2 lIU/ml, controls = 8.4 lIU/ml; P = 0.37). Dolphins with hemochromatosis, however, were more likely to have higher 2 h post-prandial insulin and glucagon plasma levels compared to controls (least squares means for insulin = 25 and 7.3 lIU/ml, respectively, P < 0.0001; least squares means for glucagon = 183 and 116 pg/ml, respectively, P < 0.0001) (Fig. 2). 4. Discussion 4.1. Fasting hyperglycemia In our study, bottlenose dolphins developed a sustained hyperglycemia lasting 10 h after ingesting dextrose. This response is a

Fig. 2. Mean fasting and 2 h post-prandial plasma insulin (lIU/ml) and 2 h postprandial glucagon (pg/ml 10) in triplicate among five healthy adult dolphins and three dolphins with hemochromatosis. indicates values that were significantly higher compared to control dolphins.

hallmark of diabetes mellitus in humans. For people with type 1 diabetes, hyperglycemia is a result of an autoimmune disorder that attacks pancreatic b-cells and prevents insulin production (Mathis et al., 2001). In comparison, type 2 diabetes is caused by an inappropriate response to insulin by fat, liver, and muscle cells, also called insulin resistance. The result of insulin resistance is increased triglycerides from fat, non-suppressed gluconeogenesis by the liver, and decreased glucose uptake by muscle. As plasma glucose levels rise, more insulin is released by the pancreas (Buchanan, 2003). Eventually, sustained hyperglycemia may result from pancreatitis and inefficient release of insulin (Cefalu, 1994). In our study, ingestion of mackerel, a very high protein, very low carbohydrate meal caused increased plasma glucose that lasted up to 5 h. Plasma glucose levels, however, did not reach those as high or sustained as those found in dolphins fed dextrose. These increases are undoubtedly due to gluconeogenesis triggered by glucagon stimulation of the liver and may be considered a normal, healthy response for dolphins with high glucose demands. High-protein diets affect glucose metabolism in humans and rats (Rossetti et al., 1989), and ingestion of high-protein diets (50% protein) by obese humans have demonstrated improved type 2 diabetes without detriments to cardiovascular health (McAuley et al., 2005). Humans fed high-protein diets (55% protein, 30% fat, 15% carbohydrate) over 5 weeks have demonstrated sustained, 4fold increases in glucagon over baseline and demonstrated improved glucose metabolism (Gannon et al., 2003). There are concerns, however, regarding the negative impact of long-term highprotein diets on renal health, especially for people with diabetes and associated nephropathy. Further research is needed to determine how dolphins metabolize a diet of 73% protein, 24% fat, and 3% carbohydrate (Venn-Watson and Ridgway, 2007)without harming renal function. While diabetic ketoacidosis (DKA) is not solely associated with type 1 diabetes, it is more often associated with people with type 1 compared to type 2 diabetes (Newton and Raskin, 2004). The feline model is currently considered one of the best animal models for

198


type 2 diabetes, however, cats with diabetes develop ketoacidosis (Henson and O’Brien, 2006). It is interesting to note that dolphins do not develop DKA; in fact, ketones have been demonstrated to decrease between 24 and 72 h of fasting (Ridgway, 1972). The combination of a sustained hyperglycemia following ingestion of protein or dextrose and the lack of ketoacidosis support that dolphins may be an important model for type 2 diabetes. 4.2. Fasting and post-prandial insulin and glucagon Normal fasting plasma insulin levels in humans is less than 11 lIU/ml (Cefalu, 1994). Human fasting insulin levels between 11 and 30 lIU/ml are consistent with some insulin resistance, while levels greater than 30 lIU/ml indicate insulin resistance. We report mean ± SEM fasting insulin levels of 12 ± 3.6 lIU/ml among healthy dolphins, indicating that dolphins may have underlying insulin resistance, or at least a greater need for higher postprandial insulin plasma levels compared to humans without diabetes. Scientists believe that insulin resistance in humans may have evolved in our ancestral primates during the last ice age (Miller and Colagiuri, 1994). During this time, our ancestors changed from a high carbohydrate, low protein diet to a low carbohydrate, highprotein diet. This change enabled genetic selection of insulin resistance to help maintain blood sugar levels needed for large brains. When humans returned to increasingly high carbohydrate diets, insulin resistance became a pathologic condition and led to type 2 diabetes. Approximately 55 million years ago, the dolphin was a terrestrial mammal that evolved to live completely in the marine environment (Thewissen and Madar, 1999). The closest terrestrial relatives of dolphins are artiodactyls, including cows, pigs, and camels. Most of these relatives are strict herbivores, and none are strict carnivores. As such, it may be assumed that the terrestrial ancestor of dolphins ate a high carbohydrate, low protein diet. Similar to the evolutionary path of humans during the ice age, dolphins changed to a high protein, low carbohydrate diet when they moved to the ocean. Because dolphins, too, have developed large brains with high demands for readily available glucose, they may have also selected for insulin resistance to maintain high blood sugar levels. Unlike humans, however, dolphins continue to eat low carbohydrate diets and may continue to use insulin resistance to their advantage. 4.3. Hemochromatosis and insulin resistance Dolphins with hemochromatosis had 2 h post-prandial insulin levels significantly higher than healthy controls, suggestive of a pathologic form of insulin resistance. In humans, elevated serum ferritin levels have been associated with insulin resistance (Wrede et al., 2006) and diabetes is a primary risk factor for increased ferritin levels in patients with chronic hepatitis C (Lecube et al., 2004). Follow on studies are needed to assess whether post-prandial hyperinsulinemia in dolphins with hemochromatosis is due to hepatic and pancreatic tissue damage from excess iron deposits or if iron deposition increased because of insulin resistance. It is interesting to note that all case dolphins were in a post-treatment phase of hemochromatosis when this study was conducted; thus, these animals still demonstrated post-prandial hyperinsulinemia despite successful treatment of high iron levels. This suggests that either long-term damage occurred from previous tissue iron deposition or that insulin resistance is a pre-existing and continuing condition in these animals. The present study had several limitations, including the use of five different adult bottlenose dolphins for nine feeding studies; while dolphins may have responded differently to the three differ-

ent challenges, changes in urine and plasma levels were tracked over time in individual animals, allowing animals to serve as their own controls. The duration of hourly measurements varied from 16 to 25 h post-infusion, which may have limited detectable and comparable impacts of water and protein challenges after 16 h. Most significant results, however, involved the first 5 to 10 h postinfusion. In summary, dolphins are emerging as important models for type 2 diabetes in humans. In addition to producing a sustained, post-prandial hyperglycemia following ingestion of sugars, dolphins have similar disease complications, such as excessive iron, associated with insulin resistance. More research is needed to fully understand the role that research on dolphins can play in understanding, preventing and treating type 2 diabetes in humans. Acknowledgments The authors wish to thank Mr. Christopher Hammell for his assistance entering archived, hand-written research data and to Ms. Risa Daniels for her work with insulin and glucagon measurement methodologies. S.H.R. and S.V.-W. conceived the study hypotheses and designs. S.H.R. implemented the 1975 experimental design, and K.C. implemented the 2009 hemochromatosis study design. S.V.-W. analyzed the data. All authors wrote and edited the manuscript. The experiments in this study complied with the current laws of the country in which the experiments were performed. This work was supported by the U.S. Navy Marine Mammal Program, Biosciences Division, Space and Naval Warfare Systems Centre Pacific. References Alberti, K.G.M.M., Zimmet, P.Z., 1998. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus. Provisional report of a WHO Consultation. Diabetic Med. 15, 539–553. Andre, P., Balkau, B., Born, C., Charles, M.A., Eschwege, E., et al., 2006. Three-year increase of gamma-glutamyltranspeptidase level and development of type 2 diabetes in middle-aged men and women: the DESIR cohort. Diabetologia 49, 2599–2603. Androgue, H.J., Wilson, H., Boyd, A.E., Suki, W.N., Eknoyan, G., 1982. Plasma acid–base patterns in diabetic ketoacidosis. N. Engl. J. Med. 307, 1603– 1610. Bellinger, D.A., Merricks, E.P., Nichols, T.C., 2006. Swine models of type 2 diabetes mellitus: insulin resistance, glucose tolerance, and cardiovascular complications. ILAR J. 47, 243–258. Buchanan, T.A., 2003. Pancreatic beta-cell loss and preservation in type 2 diabetes. Clin. Ther. 25 (Suppl. B), B32–B46. Cefalu, W.T., 1994. Insulin resistance, in: The Medical Management of Diabetes Mellitus, J. C.B. Cefalu, W.T., 2006. Animal models of type 2 diabetes: clinical presentation and pathophysiological relevance to the human condition. ILAR J. 47, 186–198. Craik, J.D., Young, J.D., Chesseman, C.I., 1998. GLUT-1 mediation of rapid glucose transport in dolphin (Tursiops truncatus) red blood cells. Am. J. Physiol. 274, R112–R119. D’Souza, R.F., Feakins, L., Mears, L., Sabin, C.A., Foster, G.R., 2005. Relationship between serum ferritin, hepatic iron staining, diabetes mellitus and fibrosis progression in patients with chronic hepatitis C. Aliment. Pharmacol. Ther. 21, 519–524. Gannon, M.C., Nuttall, F.Q., Saeed, A., Jordan, K., Hoover, H., 2003. An increase in dietary protein improves glucose response in persons with type 2 diabetes. Am. J. Nutr. 78, 734–741. Goodwin, R.F., 1956. The distribution of sugar between red cells and plasma: variations associated with age and species. J. Phyiol. 134, 88–101. Henson, M.S., O’Brien, T.D., 2006. Feline models of type 2 diabetes. ILAR J. 47, 234– 242. Jerison, H.J., 1973. Evolution of the Brain and Intelligence. Academic Press, New York. Johnson, S.P., Venn-Watson, S.K., Cassle, S.E., Smith, C.R., Jensen, E.D., Ridgway, S.H., 2009. Use of phlebotomy treatment in Atlantic bottlenose dolphins with iron overload. J. Am. Vet. Med. Assoc. 235, 194–200. Kaplan, J.R., Wagner, J.D., 2006. Type 2 diabetes – an introduction to the development and use of animal models. ILAR J. 47, 181–185. Lecube, A.C., Hernandez, J., Genesca, J.I., Esteban, L., Simo, R., 2004. Diabetes is the main factor accounting for the high ferritin levels detected in chronic hepatitis C virus infection. Diabetes Care 27, 2669–2675.

S. Venn-Watson et al. / General and Comparative Endocrinology 170 (2011) 193–199 Marino, L., 1998. A comparison of encephalization between odontocete cetaceans and anthropoid primates. Brain Behav. Evol. 51, 230–238. Mathis, D., Venece, L., Benoist, C., 2001. Beta-cell death during progression to diabetes. Nature 414, 792–798. Maxwell, D.B., Fisher, E.A., Ross-Clunis 3rd, H.A., Estep, H.L., 1986. Serum alkaline phosphatase in diabetes mellitus. J. Am. Coll. Nutr. 5, 55–59. McAuley, K.A., Hopkins, C.M., Smith, K.J., McLay, R.T., Williams, S.M., et al., 2005. Comparison of high-fast and high-protein diets with a high-carbohydrate diet in insulin-resistant obese women. Diabetologia 48, 8–16. Miller, J.C., Colagiuri, S., 1994. The carnivore connection: dietary carbohydrate in the evolution of NIDDM. Diabetologia 37, 1280–1286. Nan, H., Dong, Y., Gao, W., Tuomilehto, J., Qiao, Q., 2007. Diabetes associated with a low serum uric acid level in a general Chinese population. Diabetes Res. Clin. Pract. 76, 68–74. Neubauer, N., Kulkarni, R.N., 2006. Molecular approaches to study control of glucose homeostasis. ILAR J. 47, 199–211. Newton, C.A., Raskin, P., 2004. Diabetes ketoacidosis in type 1 and type 2 diabetes mellitus: clinical and biochemical differences. Arch. Intern. Med. 164, 1925– 1931. Ridgway, S.H., Simpson, J.G., Patton, G.S., Gilmartin, W.G., 1970. Hematologic findings in certain small cetaceans. J. Am. Vet. Med. Assoc. 157, 566–575. Ridgway, S.H., 1972. Mammals of the Sea: Biology and Medicine. Charles C. Thomas, Springfield, IL. p. 747.

199

Rossetti, L., Rothman, D.L., DeFronzo, R.A., Shulman, G.I., 1989. Effect of dietary protein on in vivo insulin action and liver glycogen repletion. Am. J. Physiol. Endocrinol. Metab. 257, E212–E219. Shafrir, E., Ziv, E., Kalman, R., 2006. Nutritionally induced diabetes in desert rodents as models for type 2 diabetes: Acomys cahirinus (spiny mice) and Psammomys obesus (desert gerbil). ILAR J. 47, 212–224. Sterner, G., Carlson, J., Eckberg, G., 1998. Raised platelet levels in diabetes mellitus complicated with nephropathy. J. Intern. Med. 244, 437–441. Thewissen, J.G.M., Madar, S.I., 1999. Ankle morphology of the earliest cetaceans and its implications for the phylogenetic relations among ungulates. Syst. Biol. 48, 21–30. Venn-Watson, S., Ridgway, S.H., 2007. Big brains and blood glucose: common ground for diabetes mellitus in humans and healthy dolphins. Comp. Med. 57, 390–395. Venn-Watson, S., Smith, C.R., Jensen, E.D., 2008. Clinical relevance of elevated transaminases in a bottlenose dolphin (Tursiops truncatus) population. J. Wildlf. Dis. 44, 318–330. Wagner, J.D., Kavanaugh, K., Ward, G.M., Auerbach, B.J., Harwood Jr., H.J., et al., 2006. Old world primate models of type 2 diabetes mellitus. ILAR J. 47, 259–271. World Health Organization, 2009. Diabetes, Fact Sheet No. 312. Wrede, C.E., Buettner, R., Bollheimer, L.C., Scholmerich, J., Palitzsch, K.D., et al., 2006. Association between serum ferritin and the insulin resistance syndrome in a representative population. Eur. J. Endocrinol. 154, 333–340.