SEARCH

SEARCH BY CITATION

Keywords:

  • obesity;
  • insulin resistance;
  • glucose;
  • diabetes;
  • islets;
  • morphology;
  • immuno-staining;
  • morphometry;
  • hormones

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

This article presents biochemical data on the BSB mouse model of multigenic obesity indicating increased percentage body fat, increased fasting plasma insulin, and increased insulin resistance in male and female obese mice compared with lean controls. Plasma glucose was significantly increased only in male obese mice. Morphological and morphometrical analyses of pancreatic islets showed increased islet size and number in all obese mice compared with lean controls. Immuno-staining results for insulin-positive islet cells showed greater levels of insulin in male and female obese versus lean mice, while the percent or proportion of insulin immuno-staining, as expected, was not significantly different between obese and lean. The percent or proportion of immuno-staining for islet glucagon and somatostatin showed reduced staining in islets from obese compared with lean mice. The significance of these findings shows, for the first time, the morphologic appearance of pancreatic islets and the quantitative distribution of the three major islet cell hormonal populations in BSB obese mice. The correlation between this descriptive information and physiological data might lend insights to the cause of obesity-related diabetes. Anat Rec, 2010. © 2009 Wiley-Liss, Inc.

Human obesity, a complex genetic and environmental trait, is known to be a risk factor for Type II diabetes, and is associated with coronary atherosclerosis, hypertension, altered plasma lipoprotein levels, gallbladder disease, cancer, and increased mortality (Lew and Garfinkel,1979; Brindly and Rolland,1989; Brindly et al.,1993; Shonfeld-Warden and Warden,1997). The BSB mouse, a hybrid animal model of spontaneous multigenic obesity independent of diet, is the backcross progeny obtained by crossing C57BL/6J x Mus spretus F1 females with C57BL/6J males (Fisler et al.,1993; Warden et al.,1993,1995). The resulting backcross progeny vary in body fat from 1% to >50% (Fisler et al.,1993). Obese BSB mice mimic most cases of human obesity in that they display basal hyperinsulinemia, glucose intolerance, hyperglycemia, hyperlipidemia, and increased lean body mass (Fisler et al.,1993).

Obesity in humans and rodents has been associated with alterations in pancreatic islet morphology. One consistent effect seen in obese humans and rodents is an increase in the number of islets and enlargement of islets because of hyperplasia and/or hypertrophy of islet β cells (Ogilvie,1933; Wissler et al.,1949; Bleisch et al.,1952; Gepts et al.,1960; Hellman et al.,1961; Boquist,1972; Findlay et al.,1973; Lavine et al.,1977; Coleman,1978; Hayek and Woodside,1979; Lucocq and Findlay,1981; Gap et al.,1983; Tassava et al.,1992; Tomita et al.,1992; Figueroa and Taberner,1994; Edvell and Lindstrom,1998; Suzuki et al.,1999; Bonner-Weir,2000; Montanya et al.,2000; Hong et al.,2002). Concomitant increase in insulin content has been described in islets of obese animals (Malaisse et al.,1968; Genuth,1969).

No morphological description of pancreatic islets and islet cells has been reported for the recently described mouse model of obesity (BSB). Nor are there quantitative data on insulin-containing β cells, glucagon-containing α cells or somatostatin-containing δ cells in pancreatic islets of BSB mice compared with lean controls.

It has been reported that islet-cell hormones act in a paracrine fashion, that is, one of these hormones may act to stimulate or depress the release of another islet hormone within the same islet (Kadowaki et al.,1979; Unger and Orci,1983; Pipeleers, 1984; Weir and Bonner-Weir,1990). In light of the latter information, it would be important to know if there are any changes in islet cell populations occurring in BSB obesity. Such results may shed some light on the cause(s) of insulin resistance, glucose intolerance, hyperglycemia, and hyperinsulinism in these animals.

The present report utilizes histological, immunocytochemical, and morphometrical analyses to describe differences in islet morphology between lean and obese BSB mice. Results of this study should provide basic information regarding the relationship between obesity and endocrine islet cell content which could have an impact on the understanding of physiological defects in obesity.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Production, Maintenance, and Phenotypes of BSB Mice

Breeding pairs of C57BL/6J (B6) were obtained from the Jackson Laboratory (Bar Harbor, ME) and Mus spretus from Dr. Michael Potter (NIH, Bethesda, MD). Backcross progeny [(female B6 x male Mus spretus) female F1 x male B6] are designated BSB mice. For a detailed description of the BSB mouse model see Fisler et al., (1993).

After weaning at 21 days, mice were individually housed in plastic cages with free access to food (Purina Rodent Chow, Ralston Purina Co., St. Louis, MO) and water. All mice were exposed to a standard light-dark cycle of 12 h with room temperature of 22°C. All animals were housed and cared for under conditions meeting National Institutes of Health standards as stated in the Guide for Care and Use of Laboratory Animals and American Association for Accreditation of Laboratory Animal Care accreditation standards. All animal use was conducted according to Institutional Animal Care and Use Committee-approved protocols.

A total of 24 BSB mice, ∼18 months old, were used in all phases of this study. They were selected as the six most obese males and six most obese females matched for age with six lean males and six lean females from a backcross of 215 mice.

Prior to sacrifice, mice were fasted for ∼15 h before collection of blood through the retroorbital sinus within 90 s of initiating ether anesthesia at ∼3 h into the light phase of the diurnal cycle (1,000 h). Blood samples were collected in iced EDTA and plasma kept at −70°C until analyzed in duplicate for glucose using a glucose analyzer (Yellow Springs Instruments, Yellow Springs, OH) and insulin by double antibody radioimmunoassay (ICN Biomedicals, Costa Mesa, CA). An index of insulin resistance was calculated from fasting plasma glucose and insulin using the Homeostatic Model Assessment (HOMA) of insulin resistance (HOMA-IR) that was developed for humans (Matthews et al., 1985) but is now used extensively in rodent models (Herbach et al.,2007; Chen et al.,2008; Schaalan et al., in press). The HOMA-IR was calculated using the formula: fasting plasma insulin (ng/ml) × fasting plasma glucose (mmol/l)/(22.5 × 0.0417) (Chen et al.,2008). Body composition was based on carcass remaining after removal of liver, kidneys, and spleen for DNA isolation (data not reported here). The carcass was dried to constant weight at 90°C, homogenized, and aliquots taken for extraction of lipid in a Soxhlet apparatus. Carcass lipid was determined gravimetrically.

Morphological and Immunohistochemical Methods and Morphometrical Analyses

For each of the above mice, the body and tail (splenic lobe) of the pancreas was obtained through a midline abdominal incision, placed in cold 10% formalin containing phosphate-buffered saline for at least 24 h. Pancreata were stored in 70% alcohol for several days. Following dehydration in graded concentrations of alcohol and xylene, the blocks of tissue were oriented along their long axes parallel to the bottom surface of the embedding mold and embedded in paraplast. Orientation of the specimens in this manner allowed for histological sections to consist of continuous body and tail segments of pancreas. The embedding method also allowed for maximizing the number of sections containing enough pancreatic islets for histological and immunohistochemical study. Serial sections were cut at 5 μm. For each pancreatic sample, three separate serial sections (500 μm apart) were stained with hematoxylin-eosin (H&E) and examined morphologically. The total number of islets/section was determined manually by microscopic inspection of each pancreatic section. Bioquant image analysis software (Bioquant Image Analysis Corporation, Nashville, TN) was used to measure the mean section area (mm2) of pancreata in all H&E-stained sections. These results provided the mean number of islets/pancreatic section (mm2) comparing lean male to obese male and lean female to obese female mice.

Unstained sections adjacent to those above were immunostained for insulin, glucagons, and somatostatin using pre-diluted (1:200) polyclonal antibodies (guinea pig anti-swine insulin, rabbit anti-human glucagons, and somatostatin), and the peroxidase-labelled streptavidiin biotin method (primary antibodies and LSAB 2 kit obtained from DakoCytomation, Carpinteria, CA). Controls consisted of substituting primary antibodies with nonimmune sera. In addition, positive controls were used with Dakocytomation-provided slides of human pancreatic sections.

For each immunostained section (blindly examined on coded slides), islets were located by visually scanning from left to right and vice-versa until the first 10 immunostained islets were observed and measured morphometrically as described below. Care was taken to make certain that the same islet was not measured twice in the same section. Three levels of sections/animal were examined (i.e., 30 islets/hormone/animal). Measurements of tangential sections through islets were avoided wherever possible. Rarely, in some sections, where less than 10 islets were observed, more sections were examined to reach a total of 30 islets.

For each islet, the mean islet area (MIA) in μm2 and the mean hormonal area in μm2 of cells immunostained for insulin, glucagon, and somatostatin (MHA-i, g, s), were determined by use of Image-Pro image analysis software (Media Cybernetics, Silver Springs, MD). From the latter data, the mean percent of hormone-stained cell area/islet area (MPA-i, g, s) was calculated.

Mixed model analyses were used to determine the effect of obesity and sex on islet size, cell area, and the percent of the islet comprised of immunostained cells. Data are presented as means ± SE. All reported P values are two-sided with a 5% level of significance. Statistical analyses were carried out on SAS version 8.0 (SAS Institute, Cary, NC) or using JMP (SAS Institute, Cary, NC).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Table 1 presents the age, body composition, fasting plasma glucose and insulin, and the calculated HOMA-IR of the four groups examined in this study. The obese male and female groups had, respectively, 2.7 and 3.5 fold greater percentage of body fat than did their lean controls. Obesity resulted in significantly increased fasting plasma insulin in both males and females, whereas plasma glucose was significantly increased only in males. The HOMA-IR index was significantly increased in the obese mice. The obese male and female groups had, respectively, 6.1 and 4.0 fold greater insulin resistance than did their lean controls.

Table 1. Body composition, fasting plasma glucose and insulin, and insulin resistance as measured by HOMA-IR in BSB mice
 Gender2-way ANOVA P-values
MaleFemale
LeanaObeseaLeanaObeseaOverall modelEffect of obesityEffect of genderInteraction
  • Data are means ± standard error.

  • a

    Obesity status.

  • There are six mice per group except bN = 5 and cN = 4, because of inadequate sample.

Age (days)558 ± 12556 ± 8556 ± 7569 ± 9NSNSNSNS
Body weight (g)28.8 ± 1.668.5 ± 5.322.7 ± 1.050.2 ± 2.9<0.0001<0.00010.001NS
Carcass fat (%)18.2 ± 3.449.1 ± 4.512.2 ± 3.242.8 ± 7.9<0.0001<0.0001NSNS
Plasma glucose (mmol/l)4.0 ± 0.4b7.8 ± 0.64.4 ± 0.5c5.6 ± 0.7b0.0030.0007NS0.04
Plasma insulin (ng/ml)1.8 ± 0.6b5.4 ± 1.01.1 ± 0.13.6 ± 0.50.00050.0001NSNS
HOMA-IR7.1 ± 1.8b43.6 ± 6.95.3 ± 1.3c21.2 ± 4.3<0.0001<0.00010.03NS

Light microscopic analysis of pancreatic islets from male and female lean BSB mice showed typical profile images of circular, oval or elongated shapes surrounded by exocrine pancreatic tissue (Fig. 1A). A thin connective sheath surrounded each islet. Numerous blood capillaries permeated the entire islet. No inflammatory or fibrotic changes were noted in any of the islets examined. Islets from obese mice showed marked changes compared with the lean animals. The most marked morphological change in obese mice (male and female) was an increase in size of islets compared with lean mice (Fig. 1B,C). This observation was corroborated with morphometrical data (P < 0.001 for islet number; P < 0.001 for islet size). The morphology of islet cells, connective tissue sheaths and blood capillaries in obese mouse islets were comparable with that seen in lean mice. A careful subjective histological analysis of islet β cells showed no differences in cell diameter comparing lean and obese mice.

thumbnail image

Figure 1. Pancreatic islets stained with hematoxylin-eosin. (A) Representative pancreatic islet from a lean male BSB mouse. (B) Pancreatic islets from a lean female BSB mouse. (C) Pancreatic islets from an obese female BSB mouse. Note the marked increase in size of islets in the obese BSB. Not shown in this figure is an increase in total number of islets. Scale bars = 100 μm.

Download figure to PowerPoint

Immunostaining procedures to localize insulin, glucagons, and somatostatin showed (in lean and obese mice, male or female) a central position for insulin-stained β cells (Fig. 2A,B), while glucagon-stained α cells and somatostatin-stained γ cells were peripherally placed (Fig. 3). Delta cells appeared to have dendritic shape morphology in both lean and obese mice islets.

thumbnail image

Figure 2. Pancreatic islets from a lean (A) and obese (B) female BSB immunstained for insulin. Note that the largely central staining for this hormone is similar in both lean and obese BSB mice. Scale bars = 100 μm.

Download figure to PowerPoint

thumbnail image

Figure 3. Nearly adjacent tissue sections of pancreatic islets from female obese BSB mice indicating hematoxylin-eosin staining (A), immunostaining for glucagon within α cells (B), and immunostaining for somatostatin within γ cells (C). Note the peripheral distribution of both glucagon and somatostatin in these islets. Also note the dendritic shape of γ cells. Scale bars = 100 μm.

Download figure to PowerPoint

Morphometrical analysis of immunostained islet cells (Table 2) indicated significant differences between lean and obese mice. The MIA for all insulin-, glucagon-, and somatostatin-stained islets was significantly increased in both male and female obese mice. Data on insulin staining showed that the mean hormonal area (MHA-i) of obese mice (male or female) were significantly greater than that found in lean mice. The mean percent of hormone-stained β cell area (MPA-i) was not significantly different between the two groups.

Table 2. Mean islet area (MIA), hormonal area (MHA), and percent of hormone-stained cell area (MPA) for insulin (i), glucagon (g), and somatostatin (s) immunostained cells and islet density in BSB mice
 Gender2-way ANOVA P-values
MaleFemale
LeanaObeseaLeanaObeseaOverall modelEffect of obesityEffect of genderInteraction
  • Data are means ± standard error, N = 4 for insulin, N = 5 female, and N = 6 male for glucagon and somatostatin.

  • a

    Obesity status.

MIA-i (μm2)22,558 ± 5,72558,999 ± 1,31821,604 ± 4,56236,810 ± 9,3170.0050.01NSNS
MHA-i (μm2)12,663 ± 3,21830,197 ± 8,72611,749 ± 3,02621,989 ± 6,3800.050.05NSNS
MPA-i (%)55.2 ± 3.952.1 ± 6.653.1 ± 3.557.7 ± 4.8NSNSNSNS
MIA-g (μm2)25,345 ± 2,52357,529 ± 7,80724,320 ± 2,43144,490 ± 10,8130.0050.002NSNS
MHA-g (μm2)1,210 ± 200955 ± 2032,224 ± 4542,090 ± 3500.05NS0.002NS
MPA-g (%)5.0 ± 0.72.0 ± 0.28.2 ± 0.95.8 ± 1.60.0040.0050.002NS
MIA-s (μm2)25,320 ± 2,45156,951 ± 3,40125,323 ± 2,68544,180 ± 10,6710.0010.001NSNS
MHA-s (μm2)348 ± 142359 ± 93553 ± 103474 ± 106NSNSNSNS
MPA-s (%)1.6 ± 0.30.7 ± 0.22.3 ± 0.41.4 ± 0.30.050.0020.05NS
Islet density (#islets/section)9.84 ± 0.919.33 ± 1.910.32 ± 1.414.20 ± 1.90.0020.001NSNS

In islets stained for glucagon (Table 2), the mean hormone-stained cell area (MHA-g) was significantly greater in female than male in both lean and obese groups. There was no effect of obesity on this measurement. The mean percent of hormone-stained α cell area (MPA-g) was significantly reduced in male and female obese compared with lean controls. In addition, female lean and obese mice had significantly greater MPA-g than the male counterparts.

In somatostatin-stained γ cells (Table 2), there was no effect of gender or obesity on the mean hormonal area (MHA-s). The percent of hormone-stained cell area (MPA-s) was significantly less in males than females and significantly less in obese compared with lean mice.

Islet density, the average number of islets/tissue-section/mouse, was significantly greater in obese relative to lean mice but did not differ by gender.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

The genome of BSB mice includes Mus spretus genes on a background of B6 genes. The strong effect of genetic background on obesity and diabetes was established by studies showing that either ob/ob or db/db mutations on the B6 background resulted in severe obesity and insulin resistance, but only mild hyperglycemia, whereas these mutations on the C57BL/KsJ background resulted in modest obesity but severe diabetes (Hummel et al.,1972, Coleman and Hummel,1973). Thus, B6 mice are resilient to the effects of leptin deficiency and obesity on carbohydrate metabolism.

No mice in this study were diabetic, using a fasting blood glucose of >8.8 mmol/l (160 mg/dl) for female mice and >10.5 mmol/l (190 mg/dl) for male mice as the cut-off for diabetes (Herbach et al.,2007). We have not retained BSB mice for more than ∼11/2 years and, although there has never been a clear indication that obese BSB mice develop overt diabetes, one male mouse (excluded from this study) from the cross reported here became severely obese after weaning, but was lean at sacrifice with more islets than either lean or obese groups. Our data showing increased insulin resistance in the obese mice, without overt diabetes, are consistent with resilience of the B6 background to diabetes.

Female B6 mice have better glucose tolerance and insulin sensitivity than males because of better sensitivity of liver and fat tissues to insulin in the females (Goren et al.,2004). This differential regulation of insulin resistance by gender is likely because of gonadal production of estrogen in females (Vasudevan et al.,2005; Yakar et al.,2006).

We used HOMA-IR calculated from fasting plasma glucose and insulin to estimate insulin resistance. The HOMA-IR values reported for both male (43.6) and female (21.2) obese chow-fed mice in this study are well above that seen in female rats fed a high-fat diet (13.7) (Chen et al.,2008). The HOMA-IR of the obese males was even higher than that of rats made diabetic with streptozotocin (33.7) (Schaalan et al., in press), but was within the range of male B6 Mc4r mutant mice fed a high-fat diet (∼48) (Trevaskis et al.,2008). Among wild-type and Mc4r and Lep mutant mice fed a high-fat diet there was a tight positive relationship between % fat mass and HOMA-IR: wild-type mice with ∼21% fat mass had a mean HOMA-IR of ∼2, Mc4r/Lep double heterozygotes with ∼33% fat mass had a mean HOMA-IR of ∼30, and Mc4r heterozygous males with ∼40% fat mass had a mean HOMA-IR of ∼48 (Trevaskis et al.,2008). Thus, the greater insulin resistance observed in obese BSB mice, particularly obese males, is likely because of the obesity per se, whereas obese females are somewhat protected by estrogen from insulin resistance produced by obesity.

Our findings of obesity-related increased islet size and/or number in male and female BSB mouse model are in agreement with previous reports in obese humans and other obese animal models even though the sources of obesity are varied (Ogilvie,1933; Wissler et al.,1949; Bleisch et al.,1952; Gepts et al.,1960; Hellman et al.,1961; Kennedy and Parker,1963; Boquist,1972; Findlay et al.,1973; Lavine et al.,1977; Coleman,1978; Bray and York,1979; Hayek and Woodside,1979; Lucocq and Findlay,1981; Gap et al.,1983; Starich et al.,1991; Tassava et al.,1992; Tomita et al.,1992; Figueroa and Taberner,1994; Edvell and Lindstrom,1998; Suzuki et al.,1999; Bonner-Weir,2000; Butler et al., 2003; Montanya et al.,2000; Finegood et al.,2001; Unger and Orci,2001; Hong et al.,2002; Hull et al.,2005).

We could not determine the source for increased numbers of islets in our obese model. Others have suggested that new islets can form by proliferation from existing differentiated islet cells (Dor et al.,2004), undifferentiated stem cells (Bonner-Weir and Sharma,2002) or by neogenesis from pancreatic ducts (Bonner-Weir,2000; Bonner-Weir et al.,2000; Paris et al.,2000; Butler et al.,2003; Bonner-Weir et al.,2004).

In contrast to our findings with BSB obese mice, Bock et al. (2003), using stereological methodology, indicated that an increase in islet mass in ob/ob obese mice was because of islet enlargement with no change in islet number compared with control mice. This discrepancy may have been because of mouse strain differences or differences in methodology. Future morphometric analysis to determine possible differences in β cell diameter between lean and obese BSB mice are warranted.

An increase in the quantity of insulin in islets from obese animals has been reported (Stauffacher et al.,1967; Malaisse et al.,1968; Genuth,1969; Dolais-Kitabgi, et al.,1979; Tomita et al.,1984; Figueroa and Taberner,1994; Unger and Orci,2001). These results are in accordance with our findings of increased insulin staining in islets of obese BSB mice. Others have reported that insulin and insulin-like growth factors can stimulate β-cell replication in vitro (Rabinovitch et al., 1982) and in vivo (McEvoy and Hegre,1978; Brown et al.,1981). Recently, there is evidence that β cell hyperplasia in obese and insulin-resistant animals may also be because of the pancreatic homeodomain protein PDX-1 (Kulkarni et al.,2004) or islet neogenesis associated protein (Del Zotto et al.,2004) or some mitogenic growth factor (Flier et al.,2001).

Islet hypertrophy in obese animals (including BSB mice) may reflect hyperinsulinemia as a compensatory response to insulin resistance (Kahn,1994; DeFronzo,1997; Brunning,1997). Pancreatic islets from obese Zucker rats in vitro and in vivo released higher concentrations of insulin in a basal state and in response to glucose compared with islets from lean control animals (Schade and Eaton,1975; Fournier et al.,1992; Milburn et al.,1995). It has been suggested that the hypersecretion of insulin in isolated islets from obese Zucker rats is because of β-cell hyperplasia (Milburn et al.,1995). In addition, insulin hypersecretion has been found to be an early manifestation in leptin-deficient ob/ob mice (Chen and Romsos,1995). Also, it is well documented that it is the obese state that induces hyperinsulinism and increase in beta-cell mass (Pick et al.,1998). Insulin hypersecretion and insulin resistance has also been reported in human obesity (Ferrannini et al.,1997; Camastra et al.,2005). Furthermore, it has been demonstrated that leptin can induce an increase in β-cell mass in vitro (Okuya et al.,2001).

It is interesting to note the proposal of an adipoinsular axis in which there is a dual feedback loop between leptin produced by adipose tissue and insulin produced by pancreatic islets (Kieffer and Habener,2000).

Our findings show that the mean percent of glucagon-stained cell area was reduced in male and female obese animals compared with lean controls. Thus, while islet size increased in obese mice, the amount of glucagon staining did not change with obesity. This finding is in agreement using ob/ob or db/db mice (Baetens et al.,1978). However, it has been shown that glucagon content increased in islets using biochemical and histochemical methodology in male ob/ob mice compared with lean controls (Findlay et al.,1973; Tomita et al.,1984). This difference may have been because of our use of immuncytochemical staining methods which may be more precise than older histochemical methods.

The effect of reduced glucagon staining in islets of obese BSB mice on paracrine influences on other islet cells has not been established. Considering that α and γ are anatomically adjacent to each other may suggest some sort of modulation of glucagon and somatostatin secretion by paracrine effects. It has also been shown that insulin travels via capillaries from the central islet core to the islet periphery and affects the suppression of glucagon secretion (Maruyama et al.,1984). Based on the latter findings, increased insulin secretion in islets of BSB mice may have suppressed glucagon secretion which may, in fact, be responsible for the decreased glucagon staining reported in our paper. Unfortunately, quantitative measurements of glucagon nor somatostatin secretion were not determined for BSB mice.

Our findings that the percent glucagon staining in lean and obese female mice was greater than in male counterparts indicate interesting gender differences. The mechanisms for these gender differences are unknown. Reports have shown that pancreatic glucagon content is higher in lean females compared with lean male mice and that female mice respond differently than male mice to induced hypoglycemia (Bonnevie-Nielson,1980,1982). It has been proposed that the latter effect may have been because of the increase in α-cell mass (Karlsson et al.,2001).

Our findings that γ cells have a dendritic morphology corroborate those who indicated that the cellular process of γ cells penetrate the core of β cells, and suggest that there may be an influence of somatostatin secretion on β cell function (Leiter et al.,1979). In fact, gap junctions have been demonstrated between β cells and non-β cells suggesting the presence of some form of intercellular or paracrine communication (Meda et al.,1984). Another report has determined that islet somatostatin secretion restrains glucagon release from α cells in rats by one paracrine control (Cejvan et al.,2003). In addition, evidence for the inhibition of insulin and glucagon secretion by endogenous somatostatin in rat islets has been reported (Itoh et al.,1980)

Previous reports of change in somatostatin content in islets of obese animals have been reported using biochemical techniques. An increase in somatostatin content of islets in obese mice was demonstrated (Dolais-Kitabgi et al.,1979), while others supporting our data, have reported a decrease in somatostatin in islets of ob/ob mice (Patel et al., 1977; Baetens et al.,1978; Petersson et al., 1979). In another report, there was no change in biochemically detected somatostatin content comparing obese to lean controls (Makino et al., 1979; Tomita et al.,1984). Whether the decrease in the γ cell population of obese BSB mice has any effect on insulin secretion remains unknown. Certainly, considering that BSB mice are hyperinsulinemic indicates little, if any, effect of islet somatostatin on insulin secretion. This is in agreement with the statement that the concentration of somatostatin secreted within the islet may be too small to influence the secretions of other islet cells (Weir and Bonner-Weir,1990).

In conclusion, our study demonstrates the first time the morphology of pancreatic islets as well as a description of islet cell populations using immunocytochemical methods in the BSB mouse model of human obesity. Because obesity plays a major role in the etiology of Type 2 diabetes, it is important to recognize any changes in islet cell structure as well as determining any differences in islet cell hormones in the obese state. Further studies are required to explore the physiological implications of these finding on the development of diabetes.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

The authors would like to acknowledge the excellent technical assistance of Dr. Cheng-Han Chen. The authors would also like to acknowledge Dr. Michael Jakowec (Department of Neurology, Keck School of Medicine) for use of the Bioquant image analysis instrumentation.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED