Abstract
The present study was conducted to investigate the effect of estradiol treatment and/or ovariectomy (OVX) on non-neoplastic lesions in the pancreatic islets of Sprague-Dawley rats. Males were divided into non-treatment (naïve) and β-estradiol 3-benzoate (EB) treatment groups and females into naïve, sham-operation, OVX, and OVX plus EB treatment groups. EB was subcutaneously administered once a week from seven to twenty-six weeks of age. The animals were euthanized at twelve, eighteen, and twenty-six weeks of age, and the serum estradiol concentrations were measured in conjunction with the pancreatic islet histopathology. The histological stages of pancreatic findings were classified into three groups, hemorrhagic, fibrotic, and inflammatory lesions, and the incidence of each type of lesion was enumerated. In males, both the total and individual incidence of pancreatic lesions increased age dependently in the naïve group. EB treatment significantly decreased the total incidence at twenty-six weeks. This alteration consisted of fibrotic and inflammatory lesions, but not hemorrhagic lesions. Additionally, the incidence of hemorrhagic lesions was at the same level between male naïve and male EB groups at twelve weeks, despite a markedly higher concentration of serum estradiol in the EB group. In females, a similar tendency was seen, and the total incidence was generally low in the naïve group, whereas it was increased by OVX. OVX plus EB treatment tended to decrease the incidence accompanied by a marked increase in estradiol concentrations. In conclusion, estrogen was shown to inhibit the development of pancreatic islet lesions toward inflammation and fibrosis but did not inhibit the occurrence of hemorrhagic lesions.
Introduction
Non-neoplastic histological changes in the pancreatic islets of rats have been reported: pigmentation and fibrosis in diabetic WBN/Kob rats (Saegusa et al. 1992) and spontaneous diabetic Torii (SDT) rats (Shinohara et al. 2005), and β-cell hyperplasia and fibrosis with pigmentation in naïve Sprague-Dawley (SD) rats (Dillberger 1994; Hajdu and Rona 1967, 1971). Recently, we have reported additional morphological changes, hemorrhage, inflammation, and the progress of islet lesions in SD rats aged from eight to twenty-six weeks (Imaoka et al. 2007). The slightest change, such as leakage of red blood cells from the capillaries in the islets, is considered to be followed by hemosiderin deposition and/or infiltration of pigment-laden macrophages. Furthermore, some islets were infiltrated by inflammatory cells, or dissected into small segments by fibrous septa. The incidence of animals having lesions increased age dependently, especially in males, and the morphological severity of the lesions was markedly higher in males than in females.
With respect to the gender difference in the incidence of islet lesions, there has been a focus on the contribution of sex hormones. Weekly treatment of estrogen for ten weeks in male SD rats aged twenty weeks reduced the incidence of animals having fibrosis of the islets to 40%as compared to the naïve group (Hajdu et al. 1969). The authors further reported the effect of three estrogenic compounds on the occurrence of lesions in the islets, such as fibrosis and enlargement (increased diameter of the islet in sections), by daily treatment for nineteen months with a dose-dependent decrease in the number of animals having lesions (Hajdu and Rona 1971). In ovariectomized SDT rats, the severity of pigmentation and fibrosis in the islet was increased compared to naïve females, and this was suppressed by weekly subcutaneous injections of β-estradiol 3-benzoate (EB) for twenty weeks, suggesting that estrogen played a pivotal role in the occurrence of islet lesions (Shinohara et al. 2005). However, these reports do not mention early (initiating) lesions, that is, hemorrhagic and inflammatory lesions, in the pancreatic islets. Detailed morphological observation would be necessary to clarify estrogen’s contribution to the onset and progress of lesions. Additionally, serum estrogen concentrations were not reported. Accordingly, in the present study we investigated the effect of estradiol treatment and/or ovariectomy (OVX) on the incidence of hemorrhagic, inflammatory, and fibrotic lesions in the pancreatic islets (number of islets having each individual lesion/total islet number in the left pancreatic lobe) of male and female rats from twelve to twenty-six weeks of age, with sequential measurement of the serum estradiol concentrations.
Materials and Methods
Animals
Seventy male and 140 female Crl:CD(SD) rats aged six weeks were purchased from Charles River Laboratories Japan, Inc. (Yokohama, Japan). Seventy and thirty-five females were subjected to OVX and a sham-operation, respectively, at four weeks of age by the supplier. The animals were housed two or three animals per wire mesh cage in an air-conditioned room (temperature: 23°C ± 2°C, relative humidity: 55%± 20%) with a twelve-hour light/dark cycle. They were allowed free access to a commercially standard diet (CRF-1, Charles River Laboratories Japan, Inc.) and chlorinated tap water during the experimental period, including quarantine and an acclimation period of one week, before being used at seven weeks of age. One sham-operated female rat was excluded from the study because of spontaneous myeloid leukemia at eleven weeks of age.
Animal Welfare
All the experimental procedures were performed in accordance with the Guideline for Animal Experimentation issued by the Japanese Association for Laboratory Animal Science (JALAS 1987).
Chemicals
EB was obtained from Wako Pure Chemical Industries, Ltd. (Tokyo, Japan) and dissolved in corn oil (Wako Pure Chemical Industries, Ltd.) at 100 μg/mL immediately before administration.
Experimental Design
Figure 1 shows the detail of the study design. Males were divided into two groups of thirty-five animals each, the non-treatment (naïve) and EB treatment groups; and females were divided into four groups, the naïve, sham-operation, OVX, and OVX plus EB treatment groups. EB was subcutaneously administered at 50 μg/0.5 mL corn oil/animal once weekly for up to twenty weeks (seven to twenty-six weeks of age), consistent with the dosage level reported by Shinohara et al. (2005). The animals of both sexes were further subdivided into three groups for euthanasia at twelve (ten animals), eighteen (ten animals), and twenty-six (fifteen animals) weeks of age. All the animals were observed for general conditions once daily on weekdays throughout the experimental period.
Laboratory Examinations
Blood was collected from the jugular veins of the rats under ether anesthesia. The serum glucose concentrations were measured with an automatic analyzer (Type 7350, Hitachi, Tokyo, Japan). The insulin levels were measured by an enzyme-linked immunosorbent assay with a commercial kit (AKRIN-010T, Shibayagi Co., Tokyo, Japan), and the estradiol levels were measured using a dissociation-enhanced lanthanide fluorescence immunoassay kit (PerkinElmer Life and Analytical Sciences, Wellesley, MA, USA).
Light Microscopy
Following blood collection, rats were euthanized by exsanguination under ether anesthesia. The whole pancreas was carefully removed and immediately fixed in 10%neutral buffered formalin. The tissues were trimmed into three parts, the right (duodenal segment), the body (parabiliary and gastric segments), and the left (splenic segment), embedded in paraffin wax, cut at 4 μm in thickness, stained with hematoxylin and eosin (H&E), and examined light microscopically. In addition, Prussian blue staining was applied to representative sections with pigmentation.
Incidence of Lesions
The histological findings in the pancreatic islets were classified into the following three stages: hemorrhagic lesion consisting of pigmentation with or without hemorrhage, fibrotic, and inflammatory lesions with or without pigmentation and/ or hemorrhage. The number of animals that had any of these lesions in any of the three pancreatic parts was counted. To obtain the exact incidence of each individual lesion, the incidence (percentage) of the pancreatic islets that had the lesion was calculated in the total number of islets examined in the left pancreatic lobe of an individual rat, because no significant difference in incidence had been observed between the three pancreatic parts (Imaoka et al. 2007). The individual incidence was used for calculation of the group mean incidence. Moreover, the total incidence of the islets that had any of the above three lesions was calculated and is described as “the total incidence” in this paper. The total numbers of the islets examined in the left pancreas ranged from 80.2 ± 23.9 to 91.3 ± 21.5 in the male groups, and from 46.3 ± 19.9 to 118.3 ± 46.3 in the female groups, respectively, with no significant differences between the same age groups in both sexes.
Immunohistochemistry
Immunohistochemical staining for estrogen receptors αand β1 (ERαand β1) was performed using immunoglobulin conjugated to a peroxidase-labeled dextran polymer (EnVision, DakoCytomation Co., Ltd., Kyoto, Japan). In brief, sections of the pancreas were deparaffinized, treated with mouse anti-ERαor mouse anti-ERβ1 antibody (1:20 each, DakoCytomation Co., Ltd.) as a primary antibody, and stained with the EnVision system (DakoCytomation Co., Ltd.). The sections were then counterstained with hematoxylin. As a positive control, sections of the uterus obtained from naïve females of the same age as those for the pancreas staining were processed in parallel, and the uterine sections untreated with primary antibody were used as the negative control.
Statistical Analyses
The group mean quantitative values of the laboratory examinations and the incidence of islets having each histological finding were statistically analyzed by a Student’s t test for the difference between the two male groups and by a Tukey’s multiple comparison test for the difference between each of the four female groups. The incidence of animals having pancreatic islet lesions was analyzed by a Fisher’s exact test for males and by χ2 test for females. Age-dependent differences in the serum estradiol concentrations were analyzed by a Dunnett’s multiple comparison test versus the twelve-week-old group in both sexes. A p value less than 5%(two-tailed) was considered to be significant.
Results
Laboratory Examinations
The serum estradiol concentrations in the male and female rats are shown in Tables 1 and 2, respectively.
In males, the estradiol concentrations were increased at eighteen and twenty-six weeks of age in the naïve group, compared to twelve weeks of age. The EB group showed much higher concentrations than those in the naïve group at all sampling points. Additionally, at twenty-six weeks of age, the level markedly increased to about twice that at twelve or eighteen weeks of age, with a significant difference from the twelve-week-old group.
In females, the estradiol levels in the naïve group were approximately twice as high as those in the naïve male group. In the OVX group, the levels were significantly decreased at eighteen and twenty-six weeks of age, compared to that at twelve weeks of age, with the values being similar to those of the male naïve group. On the other hand, in the OVX plus EB group, the estradiol levels were much higher than those in the other groups at all three sampling points, associated with a significant difference at twenty-six weeks of age from the twelve-week-old group.
Both males and females revealed no important changes in the serum glucose and insulin concentrations at any sampling ages (data not shown).
Light Microscopy
Histological changes in the pancreatic islets, such as hemorrhage, fibrosis, and inflammation, were observed in both the male and female rats and are presented in Figure 2. In the mild hemorrhagic lesions, a few erythrocytes considered to have leaked from the capillaries with hemosiderin (positive for Prussian blue staining), or hemosiderin deposits without erythrocytes, were evident in the pancreatic islets (Figure 2a). These changes were located in very limited areas in the islets in the mild cases; in severe cases, the changes were expanded to wider areas in the islets (Figure 2b). In other pancreatic islets, fibrous tissues were increased with or without the above hemorrhagic changes, and fibrous septa were formed, dividing the islet cells into small segments (Figure 2c). The septa became thicker with age. In females, the fibrotic tissue was generally less prominent than in males. As another lesion associated with hemorrhage, inflammation consisting of edema and inflammatory cell (mainly neutrophils and lymphocytes, and a few macrophages) infiltration was observed within and around the islets (Figure 2d).
Incidence of Lesions
The total number of rats having lesions in the pancreatic islets is presented in Table 3. The number increased with age in both males and females and was associated with no intergroup (treatment) differences. However, the female OVX group showed numbers higher than those of the other female groups and were similar to those in the male groups. Interestingly, the EB treatment and/or OVX groups already had lesions at twelve weeks of age.
The incidence of pancreatic islets having hemorrhage, fibrosis, or inflammation, and the total incidence in the left pancreases of male and female rats are presented in Figures 3 and 4, respectively. In males, the total incidence markedly increased up to 19.9%at twenty-six weeks of age in the naïve group, compared to a lower incidence at twelve and eighteen weeks of age. Hemorrhagic and fibrotic lesions were observed from twelve weeks of age, and inflammatory lesions were noted from eighteen weeks of age. In the EB group, the total incidence was significantly decreased to 10.5%at twenty-six weeks of age with no changes at other ages. The incidence of each lesion showed a similar increasing pattern to that of the total incidence in the naïve group, and both hemorrhagic and fibrotic lesions were observed from twelve weeks of age, whereas inflammatory lesions were not. EB treatment provoked significant decreases in the incidence of fibrosis and inflammation, but not of hemorrhage, at twenty-six weeks of age.
In naïve females, the total incidence reached a plateau at eighteen weeks of age, which was different from the males, and the highest incidence was only 3.2%at twenty-six weeks of age. The sham-operation group demonstrated similar incidence to that of the naïve group. The OVX group showed an age-dependent increase in the total incidence, with the highest incidence of 11.4%at twenty-six weeks of age, which was significantly higher than those in the naïve and sham-operation groups. In terms of each lesion, fibrotic lesions at twelve weeks of age showed significantly higher incidence than that in the naïve group. Inflammatory lesions were seen only at eighteen weeks of age. The age-dependent pattern of increase in these lesions was the same as in the naïve groups, and a significant increase in hemorrhagic lesions was seen at twenty-six weeks. Although there were no significant differences, the incidence of inflammatory and fibrotic lesions increased at eighteen and twenty-six weeks of age, respectively. In the OVX plus EB group, there were no significant differences in the incidence from the naïve and OVX groups at any sampling ages. However, the incidence of fibrosis and inflammation revealed a decrease, compared to that of the OVX group. Notably, although inflammatory lesions were observed with very low incidence in both the naïve male and female OVX groups, these lesions were completely blocked from occurring in the male EB and female OVX plus EB groups.
Immunohistochemistry
The immunohistochemistry for ERαand β1 revealed a positive reaction in the cytoplasm of almost all the islet cells in the presence and absence of histological changes in the islets (Figure 5). Furthermore, no difference was observed in the staining intensity between the groups, sexes, and ages. No positive reaction was observed in the capillary endothelial cells in the islets, unlike the uterus or aorta, in which the endothelial cells have been stained positively for the receptors (Andersson et al. 2001).
Discussion
The present study examined the effect of EB treatment and/ or OVX on the spontaneous non-neoplastic lesions in the pancreatic islets of SD rats. These lesions included hemorrhage, fibrosis, and inflammation in the islets and make us speculate that the initiating event, hemorrhage, led to inflammation that further induced fibrosis by the effect of cytokines. In naïve males, the incidence of islets having each lesion increased, particularly at twenty-six weeks of age. EB treatment decreased the incidence of fibrotic and inflammatory lesions at twenty-six weeks of age, but not the incidence of hemorrhagic lesions at any age. In females, the incidence of lesions was sustained at a low level in the naïve and sham-operated groups throughout the experimental period. OVX increased the total incidence of the lesions at twenty-six weeks of age, accompanied by an increase in the incidence of fibrosis at twenty-six weeks of age and inflammation at eighteen weeks of age. OVX plus EB treatment decreased the incidence of fibrotic and inflammatory lesions, but not hemorrhagic lesions. These results suggest that estradiol is involved in the pathogenesis of spontaneous lesions in the pancreatic islets of male and female rats and that it particularly inhibits marked increases in the incidence at twenty-six weeks of age, as well as inhibiting the progression toward fibrotic and inflammatory lesions, but not the onset of hemorrhagic lesion itself.
Estrogen has been reported to have antifibrotic and anti-inflammatory effects based on its antioxidant properties by scavenging oxygen free-radicals (Gómez-Zubeldia et al. 2000; Huh et al. 1994). For instance, estradiol suppresses hepatic fibrosis in rat models, with attenuation of hepatocyte apoptosis and production of collagen type I in hepatic stellate cells, by inhibiting the generation of reactive oxygen species (Lu et al. 2004; Shimizu et al. 1999; Yasuda et al. 1999). On the contrary, the use of neutralizing antibody against estradiol in male rats and ovariectomy in female rats leads to enhanced fibrogenesis in the liver (Yasuda et al. 1999). Estradiol also ameliorates neutrophil infiltration and oxidative tissue injury, which is associated with a decreased antioxidant level and enhanced lipid peroxidation, in the liver and ileum in a rat sepsis model produced by cecal ligation and puncture (Şener et al. 2005). Furthermore, estradiol has been shown to improve colitis caused by immune cell activation in an HLA-B27 transgenic rat model, as expressed by reduced inflammatory cell infiltration with decreased mRNA levels of mast cell protease, which promotes neutrophil infiltration (Harnish et al. 2004). Accordingly, the inhibition of the fibrotic and inflammatory lesions, which was observed in the present study, is considered to be a result of the anti-oxidative effect of EB. In particular, the inflammatory lesions were induced by OVX in females and were completely blocked by EB treatment in males and females. This finding may also reflect the fact that the decline in ovarian function with menopause is associated with spontaneous increases in proinflammatory cytokines, such as interleukin-1 (IL-1), IL-6, and tumor necrosis factor-α(Pfeilschifter et al. 2002; Shimizu 2003), and that estradiol inhibits these cytokines at physiological concentrations in whole blood cultures (Rogers et al. 2001).
Furthermore, estrogen has a vasoprotective effect: it offers protection against atherosclerosis by modulating low-density oxidation and binding free radicals (Barp et al. 2002) and by the attenuation of both the adventitial activation and potential translocation of adventitial fibroblasts into medial and neointimal compartments after balloon injury of rat carotid arteries (Chen et al. 1996; Oparil et al. 1997). Moreover, estrogen promotes vasodilation in part by stimulating prostacyclin and nitric oxide synthesis in the cardiovascular system of humans and experimental animals (Farhat et al. 1996). However, no literature exists on the effect of estrogen on hemorrhage from capillaries or venules in the pancreatic islets, as shown in the present study. In any case, our findings suggest that EB does not inhibit the occurrence of hemorrhage in the pancreatic islets of rats.
The total incidence of affected pancreatic islets in the naïve males and OVX females at twenty-six weeks of age was relatively high (19.9%and 11.4%, respectively), and the serum estradiol concentrations were relatively low (11.99, and 8.19 pg/mL, respectively). On the contrary, in the other groups of both sexes, the low incidence rates of 10.5%, 3.2%, 5.4%, and 7.1%were associated with high concentrations of 643.3, 20.80, 16.31, and 460.5 pg/mL in the male EB and female naïve groups and in the sham-operation and OVX plus EB groups, respectively. These negative correlations lead us to the speculation that a serum estradiol concentration of about 15 pg/mL or less may increase the incidence of pancreatic islet lesions to about 10%or more, and vice versa. Thus, an inverse relationship between the total incidence of spontaneous islet lesions and the serum estradiol concentration was roughly shown in the rats in the present study. Additionally, it is important to note that a significant increase in serum estradiol concentrations at twenty-six weeks of age from those at twelve weeks of age accompanied a decreased total incidence of lesions in the EB treated males and in the females with OVX. However, EB treatment could not completely block the occurrence of lesions, despite extremely high serum estradiol concentrations in both sexes. Therefore, unknown factors other than estradiol could also contribute to the development of pancreatic islet lesions and will be an issue to be examined in the future.
The biological actions of estrogen are mediated by binding to either ERαor ERβ, which have been shown in many tissues (Andersson et al. 2001; Pelletier et al. 2000; Saunders et al. 1997). Both ERαand ERβ have been histochemically demonstrated to be expressed in the pancreatic islets of baboons (Winborn et al. 1983), and their mRNA expression has also been reported in rats (Adachi et al. 2005). Therefore, we immunohistochemically investigated any changes in the ERs of the pancreatic islets. Positive staining for both ERαand ERβ1 was confirmed in the islets, but no difference in the staining intensity and distribution was observed between the islets with and without lesions or between groups. However, the present results are not sufficient to determine the reason for the lack of changes occurring in ERs and further studies, such as genomic analysis using laser microdissection, will be required to confirm the present results.
In conclusion, estradiol was demonstrated to be involved in the development of non-neoplastic lesions in the pancreatic islets of SD rats. Weekly subcutaneous injections of estradiol for twenty weeks were shown to inhibit fibrotic and inflammatory lesions, but not hemorrhagic lesions, in male rats at twenty-six weeks of age. Ovariectomized females showed a similar tendency, whereas the incidence was lower than that in naïve males. These phenomena were accompanied by an alteration of the serum estradiol levels. Detailed morphological analysis revealed estrogen’s effect on the series of the lesions, and our data compensates for the lack of information in the previous reports.