General and Comparative Endocrinology 214 (2015) 17–23
Contents lists available at ScienceDirect
General and Comparative Endocrinology journal homepage: www.elsevier.com/locate/ygcen
Histomorphology of the bottlenose dolphin (Tursiops truncatus) pancreas and association of increasing islet b-cell size with chronic hypercholesterolemia Kathleen M. Colegrove a,⇑, Stephanie Venn-Watson b a b
Zoological Pathology Program, College of Veterinary Medicine, University of Illinois at Urbana-Champaign, LUMC, Building 101, Room 0745, Maywood, IL 60153, USA Translational Medicine and Research Program, National Marine Mammal Foundation, 2240 Shelter Island Suite 200, San Diego, CA 92106, USA
a r t i c l e
i n f o
Article history: Received 26 July 2014 Revised 12 December 2014 Accepted 14 February 2015 Available online 5 March 2015 Keywords: Bottlenose dolphin Diabetes Hypercholesterolemia Insulin resistance Islet cells Pancreas
a b s t r a c t Bottlenose dolphins (Tursiops truncatus) can develop metabolic states mimicking prediabetes, including hyperinsulinemia, hyperlipidemia, elevated glucose, and fatty liver disease. Little is known, however, about dolphin pancreatic histomorphology. Distribution and area of islets, a, b, and d cells were evaluated in pancreatic tissue from 22 dolphins (mean age 25.7 years, range 0–51). Associations of these measurements were evaluated by sex, age, percent high glucose and lipids during the last year of life, and presence or absence of fatty liver disease and islet cell vacuolation. The most common pancreatic lesions identified were exocrine pancreas fibrosis (63.6%) and mild islet cell vacuolation (47.4%); there was no evidence of insulitis or amyloid deposition, changes commonly associated with type 2 diabetes. Dolphin islet architecture appears to be most similar to the pig, where a and b cells are localized to the central or periphery of the islet, respectively, or are well dispersed throughout the islet. Unlike pigs, large islets (greater than 10,000 lm2) were common in dolphins, similar to that found in humans. A positive linear association was identified between dolphin age and islet area average, supporting a compensatory response similar to other species. The strongest finding in this study was a positive linear association between islet size, specifically b-cells, and percent blood samples with high cholesterol (greater than 280 mg/dl, R2 = 0.57). This study is the most comprehensive assessment of the dolphin pancreas to date and may help direct future studies, including associations between chronic hypercholesterolemia and b-cell size. Ó 2015 Elsevier Inc. All rights reserved.
1. Introduction Bottlenose dolphins are susceptible to developing clinically relevant insulin resistance-associated metabolic perturbations, including dyslipidemia, fatty liver disease, iron overload, and chronic inflammation (Venn-Watson et al., 2008; Johnson et al., 2009; Venn-Watson et al., 2011a,c, 2012). Due to multiple common metabolic changes in dolphins and people with insulin resistance, there is a need to better understand these syndromes in dolphins (Venn-Watson et al., 2007, 2011a, 2013; Schivo et al., 2013; Schermerhorn, 2013). In humans that develop type 2 diabetes (T2D), pancreatic islet b-cell changes occur along a spectrum, starting with compensatory
Abbreviations: IAPP, islet amyloid polypeptide; MMP, Navy Marine Mammal Program; TD2, type 2 diabetes. ⇑ Corresponding author. E-mail addresses: [email protected] (K.M. Colegrove), stephanie. [email protected] (S. Venn-Watson). http://dx.doi.org/10.1016/j.ygcen.2015.02.023 0016-6480/Ó 2015 Elsevier Inc. All rights reserved.
enlargement of b-cells during early stages of prediabetes to b-cell dysfunction as the disease progresses to early T2D, and, finally, b-cell failure and insulin dependence with advanced T2D (Turner, 1998; Bonner-Weir, 2000; Butler et al., 2003a,b; Yoon et al., 2003; Prentki and Nolan, 2006; Bonner-Weir and O’Brien, 2008). Causes of b-cell dysfunction in humans continue to be investigated including the roles of chronic hyperglycemia, hyperlipidemia, and amyloid deposition (Lorenzo et al., 1994; Prentki et al., 2002; Bonner-Weir and O’Brien, 2008). Other pancreatic changes, including islet cell vacuolation and fibrosis, have been associated with prediabetes and T2D states in mammals (Carpenter and Lazarow, 1967; Clark et al., 1988; Jansson et al., 1995). Understanding islet architecture, b-cell area, and other pancreatic lesions in non-human mammals, including non-human primates, mice, rats, and cats, has provided comparative value to help understand pancreatic processes in humans (Lutz and Rand, 1995; Pick et al., 1998; Hubbard et al., 2002; Butler et al., 2003a,b). There are few natural animal models for insulin resistance, and there is a need to understand more about the dolphin
18
K.M. Colegrove, S. Venn-Watson / General and Comparative Endocrinology 214 (2015) 17–23
pancreas to assess its relevance to human health studies (Cefalu, 2006; Kaplan and Wagner, 2006). The aims of this investigation were to provide a general overview of the normal histologic features of the dolphin pancreas (including distribution and area of islets and a, b and d cells) and to assess potential differences in these features among dolphins with and without high glucose, cholesterol, and triglycerides. Findings were compared to those reported in other mammals, with a focus on islet changes related to insulin resistance and metabolic syndrome. 2. Materials and methods 2.1. Study animals and sample collection Post-mortem, opportunistically collected pancreatic tissue from a total of 22 (11 males and 11 females) bottlenose dolphins (Tursiops truncatus) from the Navy Marine Mammal Program (MMP) were evaluated for this study. Ages ranged from neonate to 51 years. All dolphins died between 1993 and 2011 due to acute or chronic disease conditions unrelated to pancreatic or hepatic disease. Gross necropsies were completed within 12 h of death. For each animal, representative sections from all major tissues including one to three sections of pancreas were sampled and placed in 10% neutral buffered formalin. Tissues were processed routinely and paraffin blocks were sectioned at 5 lm and stained with hematoxylin and eosin for histologic examination. Dolphins in this population participate in a vigilant preventive medicine program, including routine, behavior-trained (voluntary) blood sample collection to test for complete blood cell counts and clinical serum chemistries, including glucose, cholesterol, and triglycerides as previously described (Venn-Watson et al., 2007). These health data were collected throughout their lifetime. MMP dolphins live in open ocean enclosures and have the lowest mortality (2–4%) and highest annual survival rate (0.99) of any published data from either wild or managed dolphin populations (Venn-Watson et al., 2011b). Eleven (50%) MMP dolphins in this study were previously diagnosed with hepatic hemosiderosis, six (27.3%) with fatty liver disease, and ten (45.5%) with hepatitis based upon routine histologic evaluation. One dolphin, Dolphin A, a 43-year old geriatric male, had a history of had multiple abnormalities related to insulin regulation. Dolphins in this study set and the methodology for blood sample collection and testing have been previously described (Venn-Watson et al., 2007).
two step polymer horse radish peroxidase (HRP) detection system with the Super Enhancer™ Reagent (Biogenex, San Ramon, California USA;HK518-06K) and Poly HRP Reagent (Biogenex, San Ramon, California USA; HK519-06K). Sections were pretreated with hydrogen peroxide (insulin, glucagon, and somatostatin), citrate buffer (pH 6.0, glucagon only), and trypsin (insulin only). Pancreas sections were incubated with polyclonal antibodies against insulin (BioGenex, San Ramon, California USA; PU029-UP; 1:150), glucagon (Dako, Capinteria, California, USA; AO566, 1:600), and somatostatin (Dako, Capinteria, California, USA; AO566; 1:800). Positive controls were canine pancreas sections and negative controls were dolphin and canine pancreas sections stained without the primary antibody. Staining was detected using diaminobenzidine and counter stained with hematoxylin. Pancreas sections were also stained with Masson’s trichrome to highlight collagen and Prussian blue to screen for iron deposition.
2.4. Islet evaluation Islet area of the pancreas (N = 22) and the areas of a (N = 19), b (N = 19), and d (N = 17) cells were evaluated using the Bersoft Image Measurement Program version 8.16 (Bersoft Software & Technology, Bridgewater, Nova Scotia, Canada), by a single pathologist blinded to the age, sex, and status of the animal. Images were generated using the Olympus DP 71 camera. For each animal, one section of pancreas was evaluated. Severely autolyzed sections were excluded from evaluation due to lack of cell and staining detail and or excessive background staining. For each pancreas section, three randomly selected, non-autolyzed regions were evaluated for islet area and endocrine cell area on 4X fields. Regions were randomly selected using a grid overlay system and random number generator (Microsoft Excel 2010) to eliminate bias. Islet areas were calculated using the elliptical and free form area measurement tools. For each section, the average number of islets per section, average individual islet area, average total islet area, and average islet area as a percentage of total pancreatic area were tabulated. Areas of a, b, and d cells were digitally quantified using a color-based threshold function using the same software, similar to previously described (Meier et al., 2012). For each section, the average total endocrine cell area and average endocrine cell area as a percentage of total pancreatic area were quantified.
2.5. Statistical analysis 2.2. Histologic grading Histologic pancreatic lesions were subjectively graded by a single pathologist blinded to animal age, sex, cause of death, and histologically diagnosed liver lesions. Lesions were evaluated in areas where autolysis was minimal or did not affect interpretation. The severity of each pancreatic lesion was graded on a scale of 0–3 where 0 = absence of lesion, 1 = less than 10% of tissue area affected, 2 = 10–40% of tissue area affected, and 3 = greater than 40% of tissue area affected (Newman et al., 2006). Insulitis, vacuolation, and amyloid deposition were evaluated in the islets. The following lesions in the exocrine portion of the pancreas were graded: neutrophilic inflammation, lymphocytic inflammation, pancreatic necrosis, fat necrosis, lobular atrophy, fibrosis (on Masson’s trichrome stained sections), edema, nodular lobular hyperplasia, and ductular hyperplasia. 2.3. Immunohistochemistry and histochemical staining Immunohistochemical stains were used to evaluate the distribution of a, b, and d cells in the dolphin pancreas utilizing a
Data were analyzed using SAS version 9.2 (SAS Incorporated, Cary, North Carolina). Differences in total number of islets, islet areas, islet area as a percentage of total area, and a, b, and d cell area were evaluated by sex; age at death; percent blood samples with high glucose (greater than 140 mg/dl), triglyceride greater than 138 mg/dl), or cholesterol (greater than 280 mg/dl) during the last 12 months of life (up to month before death); age category (less than 20 years old = young, and greater than or equal to 20 = old); b-cell vacuolation (yes/no); hepatitis (yes/no); fatty liver disease (yes/no); and hemosiderosis (yes/no). Clinically high levels in this study were based upon this dolphin population’s healthy normal reference ranges involving 1100 routine, non-clinical samples (Venn-Watson et al., 2007). Differences by continuous age and percent high blood values were assessed using simple linear regression; and by sex, age, islet vacuolation, hepatitis, fatty liver disease, and hemosiderosis using Wilcoxon rank-sum test. Due to our findings of increasing average area of islet with age, a general linear model including sex as a covariate was applied to assess potential differences by sex. Significance was defined as a P value less than 0.05.
19
K.M. Colegrove, S. Venn-Watson / General and Comparative Endocrinology 214 (2015) 17–23
3. Results 3.1. Islet architecture The dolphin pancreas is similar to the human pancreas and the pancreas of most mammalian species. The pancreas is divided into lobules separated by bands of variably thick fibrous connective tissue. Lobules contain many exocrine acini and variable numbers of randomly distributed Islets of Langerhans (Fig. 1A). Within lobules are intercalated ducts and interlobular ducts. Islets vary greatly in size among individual animals and are composed of cords of polygonal cells bounded by a by a thin fibrous tissue septae. Most animals had small, 2–5 cell clusters of endocrine cells outside of well-defined larger islets, scattered among exocrine acini. Very young animals (0–1 year old) had numerous small clusters of endocrine cells within lobules and well defined islets were small to medium in size (Fig. 1B). Rarely, adult animals had rare very large islets. Occasional small islets were found within interlobular connective tissue, and small clusters of endocrine cells could be found adjacent to small caliber blood vessels. Most islets had an irregular outline, however, small numbers of islets were oval. Few lesions were noted in the pancreas of the examined dolphins (Table 1). There was no evidence of insulitis or islet amyloid deposition in any of the examined dolphins. Mild islet cell vacuolation was noted in 47.4% of the dolphins examined. In these animals, islet cells were slightly enlarged, rounded and had moderate amounts of granular to vacuolated cytoplasm with increased amounts of cytoplasmic clear space (Fig. 2). Occasionally cytoplasmic herniations were observed in the nucleus of vacuolated islet cells. There were no associations between islet vacuolation and incidence of hemosiderosis, hepatitis, or fatty liver disease; as well as by age, sex, and percent samples with high glucose, triglycerides, or cholesterol. There was no evidence of iron deposition in any of the dolphins on Prussian blue stained sections of pancreas. Within the exocrine pancreas, fibrosis (63.6%) was the most common lesion among those evaluated. Connective tissue was most commonly noted in the periductular interstitium and was not associated with parenchymal damage. There were no associations between the presence of fibrosis and incidence of hemosiderosis, hepatitis, or fatty liver disease; as well as by age, sex, and percent samples with high glucose, triglycerides, or cholesterol.
3.2. Islet size Among all dolphins evaluated in this study, the mean ± SD number of islets per section was 32 ± 28 (range, 9–134). The mean
Table 1 Number of bottlenose dolphins (Tursiops truncatus) with pancreatic lesions in each grade category. A total of 22 dolphins were evaluated. Histologic Lesion
Grade 1
Grade 2
Grade 3
Total
Islet lesions Insulitis Islet cell vacuolation Islet amyloidosis
0 9 0
0 0 0
0 0 0
0 9 0
Exocrine pancreatic lesions Neutrophilic inflammation Lymphocytic inflammation Pancreatic necrosis Peri-pancreatic fat necrosis Edema Fibrosis Lobular atrophy Nodular hyperplasia Pancreatic ductular hyperplasia
0 5 1 0 3 7 2 0 1
0 0 0 0 0 7 1 0 0
0 0 0 0 0 0 0 0 0
0 0 0 0 0 14 3 0 1
Fig. 2. Hematoxylin and eosin stained section of pancreas from an adult bottlenose dolphin (Tursiops truncatus) Bar = 20 lm. Islet cells are mildly enlarged, rounded, and have stippled to vacuolated cytoplasm. Several cells (arrows) have nuclei with eosinophilic cytoplasmic herniations.
area of individual islets was 7595 ± 2941 lm2 (range, 2984– 13,598 lm2) and mean total islet area was 221,944 ± 157,178 lm2 (range, 29,346–617,091 lm2). The mean islet area as a percentage of total pancreatic area was 1.9 ± 1.6% (range 0–7%). There was no significant difference between sexes for total number of islets, average islet area and total islet area per section (Table 2). Islets were significantly smaller (P = 0.03) and had a
Fig. 1. Hematoxylin and eosin stained section of pancreas from bottlenose dolphins (Tursiops truncatus). A. Pancreas from an adult dolphin. There are multiple small, medium, and large sized islets within lobules (arrows). Fine bands of fibrous connective tissue separate lobules and pancreatic ducts are surrounded by small amounts of fibrous connective tissue. Bar = 100 lm B. Pancreas from a dolphin less than a year of age. There are numerous small islets and clusters of endocrine cells scattered throughout the lobules (arrows). Bar = 50 lm.
20
K.M. Colegrove, S. Venn-Watson / General and Comparative Endocrinology 214 (2015) 17–23
Table 2 Comparisons of pancreatic measurements in bottlenose dolphins (Tursiops truncatus) by sex, age, and presence of fatty liver disease, hemosiderosis, and hepatitis.
*
Pancreatic
Sex
Age (years)
Measurements
Female (n = 9)
Male (n = 9)
Total # of islets Total Average islet area (lm2) 103 Average area per islet (lm2) 103 Total average islet area as % pancreas area
22 ± 12 169 ± 113 8.0 ± 3.6 1.0 ± 0.8
31 ± 16 246 ± 175 7.7 ± 2.1 1.9 ± 1.1
Contents lists available at ScienceDirect
General and Comparative Endocrinology journal homepage: www.elsevier.com/locate/ygcen
Histomorphology of the bottlenose dolphin (Tursiops truncatus) pancreas and association of increasing islet b-cell size with chronic hypercholesterolemia Kathleen M. Colegrove a,⇑, Stephanie Venn-Watson b a b
Zoological Pathology Program, College of Veterinary Medicine, University of Illinois at Urbana-Champaign, LUMC, Building 101, Room 0745, Maywood, IL 60153, USA Translational Medicine and Research Program, National Marine Mammal Foundation, 2240 Shelter Island Suite 200, San Diego, CA 92106, USA
a r t i c l e
i n f o
Article history: Received 26 July 2014 Revised 12 December 2014 Accepted 14 February 2015 Available online 5 March 2015 Keywords: Bottlenose dolphin Diabetes Hypercholesterolemia Insulin resistance Islet cells Pancreas
a b s t r a c t Bottlenose dolphins (Tursiops truncatus) can develop metabolic states mimicking prediabetes, including hyperinsulinemia, hyperlipidemia, elevated glucose, and fatty liver disease. Little is known, however, about dolphin pancreatic histomorphology. Distribution and area of islets, a, b, and d cells were evaluated in pancreatic tissue from 22 dolphins (mean age 25.7 years, range 0–51). Associations of these measurements were evaluated by sex, age, percent high glucose and lipids during the last year of life, and presence or absence of fatty liver disease and islet cell vacuolation. The most common pancreatic lesions identified were exocrine pancreas fibrosis (63.6%) and mild islet cell vacuolation (47.4%); there was no evidence of insulitis or amyloid deposition, changes commonly associated with type 2 diabetes. Dolphin islet architecture appears to be most similar to the pig, where a and b cells are localized to the central or periphery of the islet, respectively, or are well dispersed throughout the islet. Unlike pigs, large islets (greater than 10,000 lm2) were common in dolphins, similar to that found in humans. A positive linear association was identified between dolphin age and islet area average, supporting a compensatory response similar to other species. The strongest finding in this study was a positive linear association between islet size, specifically b-cells, and percent blood samples with high cholesterol (greater than 280 mg/dl, R2 = 0.57). This study is the most comprehensive assessment of the dolphin pancreas to date and may help direct future studies, including associations between chronic hypercholesterolemia and b-cell size. Ó 2015 Elsevier Inc. All rights reserved.
1. Introduction Bottlenose dolphins are susceptible to developing clinically relevant insulin resistance-associated metabolic perturbations, including dyslipidemia, fatty liver disease, iron overload, and chronic inflammation (Venn-Watson et al., 2008; Johnson et al., 2009; Venn-Watson et al., 2011a,c, 2012). Due to multiple common metabolic changes in dolphins and people with insulin resistance, there is a need to better understand these syndromes in dolphins (Venn-Watson et al., 2007, 2011a, 2013; Schivo et al., 2013; Schermerhorn, 2013). In humans that develop type 2 diabetes (T2D), pancreatic islet b-cell changes occur along a spectrum, starting with compensatory
Abbreviations: IAPP, islet amyloid polypeptide; MMP, Navy Marine Mammal Program; TD2, type 2 diabetes. ⇑ Corresponding author. E-mail addresses: [email protected] (K.M. Colegrove), stephanie. [email protected] (S. Venn-Watson). http://dx.doi.org/10.1016/j.ygcen.2015.02.023 0016-6480/Ó 2015 Elsevier Inc. All rights reserved.
enlargement of b-cells during early stages of prediabetes to b-cell dysfunction as the disease progresses to early T2D, and, finally, b-cell failure and insulin dependence with advanced T2D (Turner, 1998; Bonner-Weir, 2000; Butler et al., 2003a,b; Yoon et al., 2003; Prentki and Nolan, 2006; Bonner-Weir and O’Brien, 2008). Causes of b-cell dysfunction in humans continue to be investigated including the roles of chronic hyperglycemia, hyperlipidemia, and amyloid deposition (Lorenzo et al., 1994; Prentki et al., 2002; Bonner-Weir and O’Brien, 2008). Other pancreatic changes, including islet cell vacuolation and fibrosis, have been associated with prediabetes and T2D states in mammals (Carpenter and Lazarow, 1967; Clark et al., 1988; Jansson et al., 1995). Understanding islet architecture, b-cell area, and other pancreatic lesions in non-human mammals, including non-human primates, mice, rats, and cats, has provided comparative value to help understand pancreatic processes in humans (Lutz and Rand, 1995; Pick et al., 1998; Hubbard et al., 2002; Butler et al., 2003a,b). There are few natural animal models for insulin resistance, and there is a need to understand more about the dolphin
18
K.M. Colegrove, S. Venn-Watson / General and Comparative Endocrinology 214 (2015) 17–23
pancreas to assess its relevance to human health studies (Cefalu, 2006; Kaplan and Wagner, 2006). The aims of this investigation were to provide a general overview of the normal histologic features of the dolphin pancreas (including distribution and area of islets and a, b and d cells) and to assess potential differences in these features among dolphins with and without high glucose, cholesterol, and triglycerides. Findings were compared to those reported in other mammals, with a focus on islet changes related to insulin resistance and metabolic syndrome. 2. Materials and methods 2.1. Study animals and sample collection Post-mortem, opportunistically collected pancreatic tissue from a total of 22 (11 males and 11 females) bottlenose dolphins (Tursiops truncatus) from the Navy Marine Mammal Program (MMP) were evaluated for this study. Ages ranged from neonate to 51 years. All dolphins died between 1993 and 2011 due to acute or chronic disease conditions unrelated to pancreatic or hepatic disease. Gross necropsies were completed within 12 h of death. For each animal, representative sections from all major tissues including one to three sections of pancreas were sampled and placed in 10% neutral buffered formalin. Tissues were processed routinely and paraffin blocks were sectioned at 5 lm and stained with hematoxylin and eosin for histologic examination. Dolphins in this population participate in a vigilant preventive medicine program, including routine, behavior-trained (voluntary) blood sample collection to test for complete blood cell counts and clinical serum chemistries, including glucose, cholesterol, and triglycerides as previously described (Venn-Watson et al., 2007). These health data were collected throughout their lifetime. MMP dolphins live in open ocean enclosures and have the lowest mortality (2–4%) and highest annual survival rate (0.99) of any published data from either wild or managed dolphin populations (Venn-Watson et al., 2011b). Eleven (50%) MMP dolphins in this study were previously diagnosed with hepatic hemosiderosis, six (27.3%) with fatty liver disease, and ten (45.5%) with hepatitis based upon routine histologic evaluation. One dolphin, Dolphin A, a 43-year old geriatric male, had a history of had multiple abnormalities related to insulin regulation. Dolphins in this study set and the methodology for blood sample collection and testing have been previously described (Venn-Watson et al., 2007).
two step polymer horse radish peroxidase (HRP) detection system with the Super Enhancer™ Reagent (Biogenex, San Ramon, California USA;HK518-06K) and Poly HRP Reagent (Biogenex, San Ramon, California USA; HK519-06K). Sections were pretreated with hydrogen peroxide (insulin, glucagon, and somatostatin), citrate buffer (pH 6.0, glucagon only), and trypsin (insulin only). Pancreas sections were incubated with polyclonal antibodies against insulin (BioGenex, San Ramon, California USA; PU029-UP; 1:150), glucagon (Dako, Capinteria, California, USA; AO566, 1:600), and somatostatin (Dako, Capinteria, California, USA; AO566; 1:800). Positive controls were canine pancreas sections and negative controls were dolphin and canine pancreas sections stained without the primary antibody. Staining was detected using diaminobenzidine and counter stained with hematoxylin. Pancreas sections were also stained with Masson’s trichrome to highlight collagen and Prussian blue to screen for iron deposition.
2.4. Islet evaluation Islet area of the pancreas (N = 22) and the areas of a (N = 19), b (N = 19), and d (N = 17) cells were evaluated using the Bersoft Image Measurement Program version 8.16 (Bersoft Software & Technology, Bridgewater, Nova Scotia, Canada), by a single pathologist blinded to the age, sex, and status of the animal. Images were generated using the Olympus DP 71 camera. For each animal, one section of pancreas was evaluated. Severely autolyzed sections were excluded from evaluation due to lack of cell and staining detail and or excessive background staining. For each pancreas section, three randomly selected, non-autolyzed regions were evaluated for islet area and endocrine cell area on 4X fields. Regions were randomly selected using a grid overlay system and random number generator (Microsoft Excel 2010) to eliminate bias. Islet areas were calculated using the elliptical and free form area measurement tools. For each section, the average number of islets per section, average individual islet area, average total islet area, and average islet area as a percentage of total pancreatic area were tabulated. Areas of a, b, and d cells were digitally quantified using a color-based threshold function using the same software, similar to previously described (Meier et al., 2012). For each section, the average total endocrine cell area and average endocrine cell area as a percentage of total pancreatic area were quantified.
2.5. Statistical analysis 2.2. Histologic grading Histologic pancreatic lesions were subjectively graded by a single pathologist blinded to animal age, sex, cause of death, and histologically diagnosed liver lesions. Lesions were evaluated in areas where autolysis was minimal or did not affect interpretation. The severity of each pancreatic lesion was graded on a scale of 0–3 where 0 = absence of lesion, 1 = less than 10% of tissue area affected, 2 = 10–40% of tissue area affected, and 3 = greater than 40% of tissue area affected (Newman et al., 2006). Insulitis, vacuolation, and amyloid deposition were evaluated in the islets. The following lesions in the exocrine portion of the pancreas were graded: neutrophilic inflammation, lymphocytic inflammation, pancreatic necrosis, fat necrosis, lobular atrophy, fibrosis (on Masson’s trichrome stained sections), edema, nodular lobular hyperplasia, and ductular hyperplasia. 2.3. Immunohistochemistry and histochemical staining Immunohistochemical stains were used to evaluate the distribution of a, b, and d cells in the dolphin pancreas utilizing a
Data were analyzed using SAS version 9.2 (SAS Incorporated, Cary, North Carolina). Differences in total number of islets, islet areas, islet area as a percentage of total area, and a, b, and d cell area were evaluated by sex; age at death; percent blood samples with high glucose (greater than 140 mg/dl), triglyceride greater than 138 mg/dl), or cholesterol (greater than 280 mg/dl) during the last 12 months of life (up to month before death); age category (less than 20 years old = young, and greater than or equal to 20 = old); b-cell vacuolation (yes/no); hepatitis (yes/no); fatty liver disease (yes/no); and hemosiderosis (yes/no). Clinically high levels in this study were based upon this dolphin population’s healthy normal reference ranges involving 1100 routine, non-clinical samples (Venn-Watson et al., 2007). Differences by continuous age and percent high blood values were assessed using simple linear regression; and by sex, age, islet vacuolation, hepatitis, fatty liver disease, and hemosiderosis using Wilcoxon rank-sum test. Due to our findings of increasing average area of islet with age, a general linear model including sex as a covariate was applied to assess potential differences by sex. Significance was defined as a P value less than 0.05.
19
K.M. Colegrove, S. Venn-Watson / General and Comparative Endocrinology 214 (2015) 17–23
3. Results 3.1. Islet architecture The dolphin pancreas is similar to the human pancreas and the pancreas of most mammalian species. The pancreas is divided into lobules separated by bands of variably thick fibrous connective tissue. Lobules contain many exocrine acini and variable numbers of randomly distributed Islets of Langerhans (Fig. 1A). Within lobules are intercalated ducts and interlobular ducts. Islets vary greatly in size among individual animals and are composed of cords of polygonal cells bounded by a by a thin fibrous tissue septae. Most animals had small, 2–5 cell clusters of endocrine cells outside of well-defined larger islets, scattered among exocrine acini. Very young animals (0–1 year old) had numerous small clusters of endocrine cells within lobules and well defined islets were small to medium in size (Fig. 1B). Rarely, adult animals had rare very large islets. Occasional small islets were found within interlobular connective tissue, and small clusters of endocrine cells could be found adjacent to small caliber blood vessels. Most islets had an irregular outline, however, small numbers of islets were oval. Few lesions were noted in the pancreas of the examined dolphins (Table 1). There was no evidence of insulitis or islet amyloid deposition in any of the examined dolphins. Mild islet cell vacuolation was noted in 47.4% of the dolphins examined. In these animals, islet cells were slightly enlarged, rounded and had moderate amounts of granular to vacuolated cytoplasm with increased amounts of cytoplasmic clear space (Fig. 2). Occasionally cytoplasmic herniations were observed in the nucleus of vacuolated islet cells. There were no associations between islet vacuolation and incidence of hemosiderosis, hepatitis, or fatty liver disease; as well as by age, sex, and percent samples with high glucose, triglycerides, or cholesterol. There was no evidence of iron deposition in any of the dolphins on Prussian blue stained sections of pancreas. Within the exocrine pancreas, fibrosis (63.6%) was the most common lesion among those evaluated. Connective tissue was most commonly noted in the periductular interstitium and was not associated with parenchymal damage. There were no associations between the presence of fibrosis and incidence of hemosiderosis, hepatitis, or fatty liver disease; as well as by age, sex, and percent samples with high glucose, triglycerides, or cholesterol.
3.2. Islet size Among all dolphins evaluated in this study, the mean ± SD number of islets per section was 32 ± 28 (range, 9–134). The mean
Table 1 Number of bottlenose dolphins (Tursiops truncatus) with pancreatic lesions in each grade category. A total of 22 dolphins were evaluated. Histologic Lesion
Grade 1
Grade 2
Grade 3
Total
Islet lesions Insulitis Islet cell vacuolation Islet amyloidosis
0 9 0
0 0 0
0 0 0
0 9 0
Exocrine pancreatic lesions Neutrophilic inflammation Lymphocytic inflammation Pancreatic necrosis Peri-pancreatic fat necrosis Edema Fibrosis Lobular atrophy Nodular hyperplasia Pancreatic ductular hyperplasia
0 5 1 0 3 7 2 0 1
0 0 0 0 0 7 1 0 0
0 0 0 0 0 0 0 0 0
0 0 0 0 0 14 3 0 1
Fig. 2. Hematoxylin and eosin stained section of pancreas from an adult bottlenose dolphin (Tursiops truncatus) Bar = 20 lm. Islet cells are mildly enlarged, rounded, and have stippled to vacuolated cytoplasm. Several cells (arrows) have nuclei with eosinophilic cytoplasmic herniations.
area of individual islets was 7595 ± 2941 lm2 (range, 2984– 13,598 lm2) and mean total islet area was 221,944 ± 157,178 lm2 (range, 29,346–617,091 lm2). The mean islet area as a percentage of total pancreatic area was 1.9 ± 1.6% (range 0–7%). There was no significant difference between sexes for total number of islets, average islet area and total islet area per section (Table 2). Islets were significantly smaller (P = 0.03) and had a
Fig. 1. Hematoxylin and eosin stained section of pancreas from bottlenose dolphins (Tursiops truncatus). A. Pancreas from an adult dolphin. There are multiple small, medium, and large sized islets within lobules (arrows). Fine bands of fibrous connective tissue separate lobules and pancreatic ducts are surrounded by small amounts of fibrous connective tissue. Bar = 100 lm B. Pancreas from a dolphin less than a year of age. There are numerous small islets and clusters of endocrine cells scattered throughout the lobules (arrows). Bar = 50 lm.
20
K.M. Colegrove, S. Venn-Watson / General and Comparative Endocrinology 214 (2015) 17–23
Table 2 Comparisons of pancreatic measurements in bottlenose dolphins (Tursiops truncatus) by sex, age, and presence of fatty liver disease, hemosiderosis, and hepatitis.
*
Pancreatic
Sex
Age (years)
Measurements
Female (n = 9)
Male (n = 9)
Total # of islets Total Average islet area (lm2) 103 Average area per islet (lm2) 103 Total average islet area as % pancreas area
22 ± 12 169 ± 113 8.0 ± 3.6 1.0 ± 0.8
31 ± 16 246 ± 175 7.7 ± 2.1 1.9 ± 1.1
