Subchronic peroral exposure to 5 mg Na2SeO3/L in drinking water for 90 days resulted in a significant decrease in body weight and femoral length in adult male rats. Ip  similarly reported a reduction in growth in rats after 12 weeks of dietary exposure to 5 mg/kg Na2SeO3. Selenium is known to accumulate in the anterior part of the pituitary gland. Thorlacius-Ussing et al. mention decreased secretion of growth hormone (GH) and somatomedin C in rats after receiving 15 mg/L Na2SeO3 in their drinking water, suggesting that growth retardation could be mediated by reduced GH and somatomedin C production. Next Gronbaek et al. showed that a Se dose of 3.3 mg Na2SeO3/L in drinking water for 35 days induced a significant reduction in circulating insulin-like growth factor I (IGF-I). These rats also displayed a significantly shorter tibia.
The thickness of cortical bone is generally accepted as an important parameter in the evaluation of cortical bone quality and strength. We observed significantly decreased cortical bone thickness in Se-exposed rats. Studies of cortical bone thickness in Se-exposed rats has not been published previously. The value of cortical bone thickness in control group rats differed from values reported by Comelekoglu et al. and Chovancová et al., who analysed rats of different age.
The results of the qualitative histological analysis of the control rats corresponded to those of previous works [23–26]. We identified non-vascular and primary vascular radial tissues and irregular Haversian bone tissue. However, there was no evidence of true Haversian intracortical bone remodelling. It is generally known that aged rats and mice lack true Haversian cortical bone remodelling but not cancellous bone remodelling [25, 27]. Therefore, some secondary osteons can be observed in the long bones near the endosteal border. In our study, the newly formed remodelling units within compact bone originated from the endocortical surface and extended deep into the underlying compact bone. The same findings have also been documented in 13 month-old male rats .
Prolonged intake of a high dose of Se induced changes in the middle part of compact bone where primary vascular radial bone tissue was present. We propose that the formation of this type of bone, in the central area of the femur, could be explained as an adaptive response to Se toxicity to protect bone tissue against cell death. It is generally known that Se at high doses induces apoptosis in mature osteoclasts  and death of osteoblast-like cells . In a study by Turan et al. osteocyte loss was identified because of the destruction of the bone tissue and its replacement with a large uncalcified mass of new bone matrix in rabbits fed excess Se (10 mg Na2SeO3/kg of diet for a period of 12 weeks). These authors also reported decreased biomechanical strength of the femur in the Se-exposed animals.
Data obtained from the histomorphometric analysis showed a significant increase in area, perimeter, maximum and minimum diameters of the primary osteons’ vascular canals and on the other hand a significant decrease of the Haversian canals’ variables in the Se-exposed rats. In general, the vascular system is a critical target for toxic substances and their effects on the vascular system may play an important role in mediating the pathophysiological effects of these substances in specific target organs . Blood vessels readily adapt structurally in response to sustained functional changes, particularly those related to chronic pressure alterations or changes in the nutritional demands of the tissue [30, 31]. Results obtained by Ruseva et al. demonstrate that rats exposed to increased amount of dietary Se had higher glutathione peroxidase 1 (GPx-1) activity and a lower level of anti-elastin antibodies (AEABs) than those of control group, and the aortic wall histology showed degenerative changes associated with reduced thickness of the wall of the left coronary artery. Our results indicate that an excess of Se had a different impact on the primary osteons’ vascular canals and Haversian canals. The main difference between these structures is the presence of a cement line in Haversian canals (cement line delimits the canals) and its absence in primary osteons. We surmise that the cement line could be the main reason for the different results in the histomorpometry of both canals.
We found significantly lower values of all variables of secondary osteons in Se-exposed rats. According to Jowsey , the values of secondary osteons are higher in individuals with longer bones. Our results correspond with those found by Jowsey , as the femurs were longer in the control rats, which also displayed higher values for osteons. Moreover we propose that the observed differences in histomorphometry between the Se-exposed and control rats could be associated with changes in bone remodelling, which is mediated by osteoblasts and osteoclasts and subsequent calcification of bone tissue. The results of Boyar  showed that an excess of Se increased the amount of carbonate content in bones of Wistar rats injected intraperitoneally with 5 μmol Na2SeO3/kg for 4 weeks. The incorporation of carbonate ions into the crystal structure of hydroxyapatite (HA) results in changes in the physical and chemical properties of HA . HA crystals, as a major mineral component of bones, are aligned with their long axis parallel to the collagen fibre axis  creating concentric lamellae of secondary osteons. Another factor that could have contributed to the minimization of secondary osteons could be Se intoxication, which induces apoptosis in mature osteoclasts . Decreased osteoclast activity is associated with smaller osteon size . Therefore, significantly decreased secondary osteons may be present in Se-exposed rats.
The paper is the first report of femoral bone structure in rats after subchronic peroral administration of Se. Our results contribute to the insight into the complexity of Se toxicology in rats. Moreover, since the microstructural changes in the bone typically follow biochemical action of Se in the tissue, the results of the study can highlight the physiological implications and possible mechanisms of Se effects.
Our results demonstrate the effect of Se on bone microstructure of rats; however, possible extrapolation of the results to humans may be an interesting topic for discussion. Although Se poisoning is not reported frequently in humans and human toxicity from environmental exposure to Se is rare, incidents include industrial accidents, accidental ingestions, suicides, and attempted murder . In humans, the minimum lethal dose seems to be similar to that for animals. The chronic toxic dose for human beings is about 2.4 to 3 mg of Se per day . Although there are some differences in bone microstructure between humans and rats we suppose that observed changes in the variables of Haversian canals and secondary osteons in our rats will correspond to those of humans. Human femur in adult individuals is namely composed of many secondary osteons, which contain Haversian canals (dense Haversian bone tissue). Additional research dealing with the influence of Se on human bone structure is required to obtain more information on comparative aspects.