Age- and Sex-Related Differences in Spontaneous Hemorrhage and Fibrosis of the Pancreatic Islets in Sprague–Dawley Rats

Abstract

To elucidate the spontaneous occurrence of hemorrhage in the pancreatic islet, naïve Crj:CD(SD)IGS rats were given a commercially available standard diet ad libitum over 20 weeks, and were sequentially examined. Islet hemorrhage was morphologically observed from 12 weeks of age, and its incidence was significantly higher in males than in females, with a wide distribution in all pancreatic lobes. The incidence (%) of affected islets to examined islets was increased with age. Hemorrhage was accompanied by brownish pigmentation, and reacted positively for iron by Prussian/Berlin blue staining. In 26 weeks old, most of the islets were dissected by dense fibrous tissue into small nests, and disarranged β cells were detected by insulin immunostaining. Ultrastructurally, no apparent morphological change was seen in any islet endothelial cell, although blood leakage with migrated macrophages and dense collagen fibers was observed around the capillaries. In serum biochemistry of rats aged 26 weeks, the estradiol level in males with hemorrhage was significantly lower than that in males with non-hemorrhage, presumably suggesting the lack of capillary protective ability. Next, when rats were given a high fat/protein diet over 20 weeks to clarify whether it accelerated the frequency or timing of hemorrhage, its occurrence was essentially identical to that of the animals fed the standard diet. In conclusion, the onset of spontaneous islet hemorrhage was observed predominantly in aged males, resulting from the low estradiol level in serum.

Introduction

In aged rats, there is much information on spontaneous alterations of the pancreatic islet; representative morphological changes are β-cell hyperplasia and neoplasia in the islet (Boorman and Hollander, 1973; Spencer et al., 1986; Majeed, 1997). For example, in Sprague–Dawley (SD) rats, β-cell hyperplasia and intra-insular fibrosis with accumulation of hemosiderin-containing macrophages have been recognized (Hajdu and Rona, 1967; Dillberger, 1994) as an initial event that occurs primarily in aged males. Additionally, in several diabetic and obese strains such as Zucker, eSS, and LA/N rats, hyperplasia, fibrosis, and pigment deposition in the islet with metabolic disorders are seen. These disorders, including hyperinsulinemia, hyperlipaemia, and hypercholesterolaemia, have been reported as age-related findings, which closely resembled those seen with SD rats (Larsson et al., 1977; Ahuja et al., 1987; Gomez Dumm et al., 1990). Interestingly, these changes have also been seen predominantly in males.

In contrast, little information is available on spontaneous islet hemorrhage in rats. In this study, we explored age-related degenerative lesions, including hemorrhage of the pancreatic islet in naïve CD(SD)IGS rats. First, to elucidate whether spontaneous islet hemorrhage was seen in rats fed a commercially available standard diet ad libitum over 20 weeks, the morphological alteration of the pancreas was examined sequentially in conjunction with related laboratory tests to ascertain the existence of functional abnormalities. Afterward, to compare the effect on frequency and timing of hemorrhage, the pancreas of rats fed a high fat/protein diet ad libitum for the same period was histopathologically investigated.

Materials and Methods

Animals

Experiment I

A total of 80 male and 80 female Crj:CD(SD)IGS rats, 5 weeks old, were purchased from Charles River Laboratories Japan Inc. (Yokohama, Japan). They were housed 2 to 3 animals per wire-mesh cage in an air-conditioned room (temperature: 23 ± 2°C, relative humidity: 50 ± 10%) with a 12-hour light/dark cycle. Rats were allowed free access to a commercially standard diet (CRF-1, Charles River Laboratories Japan Inc., protein 22.6%, fat 5.6%, sodium 0.32%, potassium 0.85%, chlorine 0.58%, magnesium 0.25%, and iron 16.6%) over 20 weeks (rat age: 5–26 weeks old), and chlorinated tap water was available ad libitum during the experimental period.

Experiment II

Sixty male and 60 female Crj:CD(SD)IGS rats, 5 weeks old, were provided from Charles River Laboratories Japan, Inc. The animals were kept in the same conditions as Experiment I. A high fat/protein diet (CMF-R, Oriental Yeast Co., Ltd., Tokyo, Japan, protein 29.5%, fat 8.5%, sodium 0.27%, potassium 1.10%, chlorine 0.41%, magnesium 0.24%, and iron 24.2%) and chlorinated tap water were given to rats ad libitum over 20 weeks.

Animal Welfare

The animal care and experimental procedures were conducted in accordance with the Guideline for Animal Experimentation issued by the Japanese Association for Laboratory Animal Science (JALAS, 1987), and the protocol was approved by the Institutional Animal Care and Use Committee of Daiichi Pharmaceutical Co., Ltd. (Tokyo, Japan).

Experimental Design

In Experiment I, the animals were divided into 4 groups of 20 rats each of either sex, and sequentially euthanized at ages 8, 12, 18, and 26 weeks. All animals were observed for general conditions once daily on weekdays throughout the experimental period. Body weight was determined each week from ages 5 to 20 weeks and every 2 weeks from ages 22 to 26 weeks. In Experiment II, rats were divided into 3 groups of 20 animals each of either sex and euthanized at ages 8, 12, and 26 weeks.

Laboratory Tests

In Experiment I, blood was collected once in the morning of 1 week before necropsy, by inserting a 22G needle into the tail vein of conscious animals (Furuhama and Onodera, 1983) under the feeding conditions. Using the serum obtained, insulin was measured with a commercial ELISA kit (AKRIN-010T, Shibayagi Co., Tokyo, Japan), and estradiol was assayed with a dissociation enhanced lanthanide fluorescence immunoassay kit (DELFIA, PerkinElmer Life and Analytical Sciences, Wellesley, Massachusetts, USA). Immediately before necropsy, with serum obtained from the jugular vein of rats under ether anesthesia, glucose (GLU) was measured with an automatic analyzer (Hitachi 7350, Hitachi, Tokyo, Japan).

Light Microscopy

In Experiment I, full internal and external macroscopic examinations were performed at necropsy. The pancreas was carefully removed and weighed, and immediately fixed in 10% neutral-buffered formalin. The tissues were trimmed into 3 regions, including right (duodenal segment), body (parabiliary and gastric segments), and left (splenic segment), embedded in paraffin wax, cut at 4 μm thickness, stained with hematoxylin and eosin (H&E), and examined with a light microscope (Table 1). The incidence and location of the lesions were recorded for both sexes, and then the distribution of the lesions in 3 regions and their incidence (numbers of affected islets/numbers of total examined islets on the slide) were counted. Further, special stainings (Prussian/Berlin blue stain or Masson’s trichrome stain) were applied to the representative sections. In Experiment II, to delineate the effect of a high fat/protein diet on the pancreatic lesions, the left region of the pancreas was removed and examined with a light microscope.

Immunohistochemistry

In Experiment I, immunohistochemical staining for 5-bromo-2′-deoxyuridine (BrdU) or insulin/proinsulin was performed. One hour before necropsy, 5 males at each age were intraperitoneally injected with 1 mg/mL/kg of 1% BrdU dissolved in phosphate-buffered saline. Sections were processed by a method with immunoglobulin conjugated to a peroxidase-labeled dextran polymer (EnVision, DakoCytomation Co., Ltd., Kyoto, Japan). In brief, the paraffin section was deparaffinized, treated sequentially with mouse anti-BrdU antibody (1:10, Beckton Dickinson Pharmingen, Franklin Lakes, NJ, USA) or mouse anti-insulin/proinsulin (1:100, Biogenesis Ltd., Poole, UK) as a primary antibody, followed by staining with the EnVision system (DakoCytomation Co., Ltd.). The sections were then counterstained with hematoxylin. For BrdU, as a positive control, the sections from the same animal’s intestine were processed in parallel. For insulin/proinsulin, as a negative control, the sections were processed in the absence of the primary antibody.

Electron Microscopy

In Experiment I, small blocks from the left region of the pancreas from groups of 5 males each were fixed in 2.5% glutaraldehyde, trimmed, and postfixed in 1% osmium tetroxide solution. Semithin sections were cut to be 1 μm, stained with toluidine blue, and examined with a light microscope to choose the target area. Ultrathin sections of these blocks were cut, stained with uranyl acetate and lead citrate, and examined by a transmission electron microscope.

Statistical Analyses

The incidence of hemorrhage, pigmentation, and fibrosis in the islets of each sex of either diet was analyzed by Fisher’s exact test. To focus on the development of the incidence with aging, linear regression analysis was applied. The incidence (%) of the affected islet in 3 regions was analyzed by Dunnett’s test (vs. the incidence at 12 weeks of age). In Experiment I, the animals at each age were divided into 2 groups consisting of non-hemorrhage and hemorrhage in the pancreatic islet on the basis of histopathological findings. The quantitative laboratory tests are expressed as the group mean ± standard deviation (S.D.), and were statistically analyzed by Student’s t-test. A p value less than 5% (2-tailed) was considered to be significant.

Results

Experiment I

Laboratory Tests

Islet hemorrhage was microscopically observed from 12 weeks of age mainly in males (Table 3). On the basis of these results, when male rats in 12, 18 and 26 weeks were divided into 2 groups with respect to non-hemorrhage and hemorrhage in the islets, no changes in clinical signs or body weights were observed throughout the experimental periods between these 2 groups (data not shown). Serum GLU in the hemorrhage group, however, was significantly increased without changes in serum insulin only in 12 and 18 weeks, as compared with that in the corresponding non-hemorrhaged group (Table 2). Serum estradiol in the hemorrhage group was evidently decreased in 26 weeks. On the other hand, there was no correlation between the microscopic finding (pigmentation or fibrosis) and serum insulin or estradiol (data not shown).

Light Microscopy

Although no macroscopic changes were observed in the pancreas, islet alterations were microscopically observed as mentioned previously (Table 3). For males, islet hemorrhage was seen in 5 animals (25%) at 12 weeks, 11 animals (55.5%) at 18 weeks, and 16 animals (80%) at 26 weeks. Pigment deposition in the center and/or periphery of the islet was observed in 6 animals (30%) at 12 weeks, 15 animals (75%) at 18 weeks, and 18 animals (90%) at 26 weeks. For females, hemorrhage was observed only in 1 animal (5%) each at 12 and 18 weeks, and 4 animals (20%) even at 26 weeks. Pigment deposition was seen in 5 animals (25%) at 12 weeks, 7 animals (35%) at 18 weeks, and 12 animals (60%) at 26 weeks.

When the relationship between the incidence (%) of hemorrhage or fibrosis and age (weeks) was examined, there was an age-dependent increase, with better correlation in males than in females. There was a similar increase in the incidence of pigmentation over time, although the degree of correlation was roughly equal for both sexes (Figure 1). With regard to lesion distribution (hemorrhage, pigmentation, and fibrosis) in 3 regions (right, body, and left) of the pancreas in males, there were no differences in their occurrence (Figure 2). The incidence (%) of the respective lesions (hemorrhage, pigmentation, and fibrosis) of affected islets, however, was increased in an age-dependent manner (Figure 3).

Morphologically, the hemorrhage in males was characterized as a leakage of red blood cells with pigment deposition and/or pigment-laden macrophages in the center or periphery of the islets. The pigment appeared blue in color by Prussian/Berlin blue staining, indicating that the pigments contained iron as hemosiderin of erythrocyte origin (Figure 4a). The pigmentation was also seen in the islet without hemorrhage. Some islets elicited active reactions with inflammatory cell infiltration and edema within and/or around the islet (Figure 4b).

This active hemorrhage was seen even in 26 weeks. In 12 weeks or more, islets were occasionally dissected into small nests by fibrous septa. The fibrosis was observed in aged animals with higher incidence and greater severity, and consequently the islets displayed the irregular margins. The fibrous tissue in 26 weeks was much thicker than that in 12 or 18 weeks (Figure 4c). Entrapment of exocrine acinar cells within the islet or fibrous tissue was sometimes seen. In females, the histological features were essentially similar to those of males but were much less severe with no visible active hemorrhage or dense collagen tissue.

Immunohistochemistry

Islets with fibrosis in males tended to be enlarged, extending into the surrounding exocrine parenchyma. Immunohistochemistry for insulin/proinsulin revealed disarranged β-cells in the affected islets (Figure 4d). Although BrdU-positive cells were sporadically detected in the islets, no increase in the frequency of positive β-cells was seen. Therefore, a positive labeling index was not determined.

Electron Microscopy

Blood leakage from the islet capillary was observed in males at 12, 18, and 26 weeks. In 8 weeks, capillaries are separated from the adjacent islet cells by 2 basement membranes, and a few collagen fibers were noted (Figure 5a). In 26 weeks, migrated and erythrophagocyted macrophages were seen around the capillary, but even then there was no apparent change in the endothelium. Mature and abundant collagen fibers were observed around the capillaries, showing thick basement membranes (Figures 5b and 5c).

Experiment II

Light Microscopy

When the incidence of microscopic changes in the pancreatic islets of the left lobe in males aged 8, 12, and 26 weeks was compared between standard and high fat/protein diets, there was no difference in the onset of hemorrhage at any age (Table 4).

Discussion

In the present investigation, we described in detail spontaneous non-neoplastic pancreatic islet lesions in SD rats. For this strain, there have been many reports dealing with age-related islet changes (Hajdu et al., 1967; Dillberger, 1994), in which their representative islet lesions have been recognized as fibrosis and hyperplasia.

Morphologically, the incidence and severity of hemorrhage were evidently higher in males than in females. Since there was no clear sex-difference in the incidence of pigmentation in the pancreatic islet, it is possible that micro-hemorrhage is common of SD rats and might even be a trigger for more overt hemorrhage. Generally, the incidence of spontaneous islet lesions including tumors has been reported to be more common in males than in females (Hajdu et al., 1968; Reaven et al., 1987; Ruhe et al., 1992, Dillberger; 1994; Majeed, 1997). Our results were also consistent with those in these previous investigations (Larsson et al., 1977; Nakama et al., 1985; Ahuja et al., 1987; Gomez Dumm et al., 1990; Mori et al., 1990; Dillberger 1994). According to data from Dillberger, β-cell hyperplasia was observed in males aged 3.5 months, and the incidence was increased to 90% in 14 months, whereas it was only 20% for the same-aged females.

Although the cause of the difference between sexes was unknown, it is likely that there is an influence of sex hormones on the etiopathogenesis of this finding. Regarding sex hormones, several reports were suggestive of a protective effect on the pancreatic islet changes in rats. For example, castration, followed by estrogen reagent treatment, has suppressed the incidence of fibrosis and enlargement of the pancreatic islet in male SD rats (Hajdu and Rona, 1971). Additionally, estrogen has been shown to have a protective action on the endothelial cells of veins (Arnal and Bayard, 2001; Schwertz and Penckofer, 2001).

In our work, serum estradiol level in males, age 26 weeks, with hemorrhage in the pancreatic islet was significantly lower than that in males with non-hemorrhage. The possibility is raised that the lower serum estradiol level implies the lack of vascular protective function in older males. This notion is, at least in part, supported by the fact that the incidence of pancreatic hemorrhage in females, which possess much higher serum estradiol level than males, was obviously lower than that in males. Unfortunately, in our further study with males, age 39 weeks, the correlation between the hemorrhage incidence and serum estradiol level was not determined, because the extremely high incidence of hemorrhage was seen in almost all males utilized (unpublished data).

The reason for the time lag between the appearance of islet hemorrhage (12 weeks) and a decrease in serum estradiol (26 weeks) may be explained by the hypothesis that local (target organ) and circulating estradiol levels may be different. Alternatively, additional studies are required to substantiate the relationship between the sex-difference and the pathogenesis of hemorrhage itself and to explain why only hemorrhage would be present in the islet but not in other organs or endocrine organs. With respect to increased serum glucose in males with islet hemorrhage, age 12 and 18 weeks, since these changes were within the basal range (186–211 mg/dL) and were not accompanied by changes in serum insulin throughout the experimental period, these were considered of doubtful biological significance.

It is noteworthy that spontaneously diabetic Torii (SDT) rats, derived from the SD strain, also have been reported to have hemorrhage and inflammation in the pancreatic islets of mainly male animals (Masuyama et al., 2004) and that the descriptions match the findings in our study. Moreover, in ovariectomized female SDT rats, more severe islet changes such as inflammatory cell infiltration, fibrosis and pigment deposition in/around the islet were observed, as compared to sham-operated or non-treatment females (Shinohara et al., 2005). This phenomenon was suppressed by injection of estrogen, demonstrating that estrogen played a critical role in the occurrence of the islet lesion and its progression.

In males 12 weeks or older, the islet areas were expanded into the surrounding parenchyma. This finding may be partly due to fibrous dissection as was the case reported by Dillberger (1994). However, it has been reported that the ratio of the fibrous tissue to islet cell mass increases with aging and that over time, some islets are reduced to scattered small islands of β-cells imbedded in a mass of scar tissue (Dillberger, 1994). Furthermore, it has been suggested that fibrosis of the islets may lead to the development of focal hyperplasia, adenoma, and carcinoma (Molon-Noblot et al., 2001). In the present study, however, neither morphological evidence of hyperplasia, nor were increased numbers of BrdU-positive cells detected in the islets of rats with hemorrhage, pigmentation, or fibrosis. It is possible that a longer period of observation may be necessary to observe such effects.

In rats, a close relation has been found between the onset of spontaneous islet lesions and the fat/protein levels in diet (Molon-Noblot et al., 2001). Briefly, the dietary restriction has delayed the incidence and timing of spontaneous changes in the islet. In general, ad libitum feeding has been considered to accelerate the occurrence of age-related prolifelative/degenerative changes in rats (Keenan et al., 1995a, 1995b). In our study, however, there was no difference in the occurrence of islet hemorrhage between rats fed the standard and high fat/protein diets. The reason for these findings still remains to be found.

In conclusion, spontaneous hemorrhage in the pancreatic islet was observed in CD(SD)IGS rats with a high frequency in older males. The pigmentation seen in the hemorrhagic region contained iron, presumably indicating the occurrence of micro-hemorrhage at an early stage. Hence, fibrosis and cell infiltration were considered to be secondary regenerative changes in response to this event. The hemorrhage in the pancreatic islet would be related to decreased estrogen levels in serum.