The Effect of Induced Diabetes Mellitus on the Cerebellar Cortex of Adult Male Rat and the Possible Protective Role of Oxymatrine: A Histological, Immunohistochemical and Biochemical Study

ABSTRACT Diabetes mellitus (DM) represents a widespread metabolic disease with a well-known neurotoxicity in both central and peripheral nervous systems. Oxymatrine is a traditional Chinese herbal medicine that has various pharmacological activities including: anti-oxidant, anti-apoptotic and anti-inflammatory potentials. The present work aimed to study the impact of diabetes mellitus on the cerebellar cortex of adult male albino rat and to evaluate the potential protective role of oxymatrine. Fifty-five adult male rats were randomly divided into three groups: group I served as control, group II was given oxymatrine (80 mg/kg/day) orally for 8 weeks and group III was given a single dose of streptozotocin (50 mg/kg) intaperitoneally to induce diabetes. Then diabetic rats were subdivided into two subgroups: subgroup IIIa that received no additional treatment and subgroup IIIb that received oxymatrine similar to group II. The diabetic group revealed numerous changes in the Purkinje cell layer in the form of multilayer arrangement of Purkinje cells, shrunken cells with deeply stained nuclei as well as focal loss of the Purkinje cells. A significant increment in glial fibrillary acidic protein (GFAP) and synaptophysin expression were reported in immunohistochemistry compared with the control group. Transmission electron microscopy showed irregularity and splitting of myelin sheaths in the molecular layer, dark shrunken Purkinje cells with ill-defined nuclei, dilated Golgi saccules and dense granule cells with irregular nuclear outlines in the granular layer. In contrast, these changes were less evident in diabetic rats that received oxymatrine. In conclusion, Oxymatrine could protect the cerebellar cortex against changes induced by DM.


Introduction
Diabetes mellitus (DM) is a common metabolic disease caused by deficiency in insulin secretion and/or insulin resistance. 1 Insulin deficiency usually results in a condition of chronic hyperglycemia with alterations in carbohydrates, lipids and proteins metabolism. 2 Uncontrolled diabetes usually results in various complications such as heart disease, gastrointestinal disorders, eye disease, neuropathy and nephropathy together with increased mortality. 3 Neuropathies in the autonomic and peripheral nervous systems are considered the most common complication of DM. Besides these neuropathies, diabetes is also correlated with development of end-organ damage in the central nervous system. 4 This condition is termed 'diabetic encephalopathy', which is accompanied by memory dysfunction and electrophysiological alterations. 5 Structural, neurochemical and degenerative changes in the brain are also associated with these functional abnormalities. 6 In addition, it was reported that insulin signaling in the brain not only controls food consumption and glucose homeostasis but is also involved in both survival and maintenance of the cognitive activity of the general neural network, thus, the changes in serum insulin levels secondary to diabetes mellitus can directly cause brain injury and neurodegenerative diseases. 7 Cerebellum is one of the particular brain regions that has a high expression of insulin receptors in its cortex. 8 Besides its motor functions, which include balance maintenance and movement coordination, cerebellum is also concerned with learning and memory. 9 Image analysis of the brain white matter in diabetic patients showed a decrease in the connections within the cerebellum and cerebrocerebellar pathways. 10 The apoptotic activity has been reported to increase in the cerebellum of diabetic rats. 6 Moreover, Hami and coworkers revealed that maternal diabetes affected the synaptogenesis of the developing cerebellum in the offspring. 11 Streptozotocin (STZ)-induced diabetes mellitus in rats is a well-established model for induction of DM owing to its harmful impact on the pancreas through its direct toxicity on pancreatic β-cells as well as generation of cell-mediated autoimmune reaction directed toward them. 12 This model is employed in the current research for evaluating the histological changes in the cerebellum caused by DM.
Traditional Chinese remedies are commonly employed in both prophylaxis and management of refractory health problems due to their available resources, therapeutic outcomes on various targets, less adverse effects and reduced cost care. 13 Oxymatrine is the major quinolizidine alkaloid extracted from the root of a traditional Chinese herb; Sophora flvescens. 14 Oxymatrine has extensive pharmacological activities, including anti-viral, 15 anti-oxidant, 16 anti-tumor 17 and anti-inflammatory effects. 18 Also, it has been shown to suppress the neuroinflammatory reactions and to delay the development of neurodegenerative diseases. 19 Additionally, oxymatrine was reported to exert neuroprotective effects on cerebral ischemia/reperfusion injury in animal models. 20 Previous studies showed that administration of oxymatrine significantly reduced the level of blood glucose in mice fed with high fructose diet 21 and improved the insulin sensitivity in diabetic rats, thus, it was predicted that oxymatrine may have a beneficial impact on diabetes mellitus. 22 Hence, this work was performed to investigate the potential role of oxymatrine against cerebellar cortex injury caused by DM in rats using different histological methods.

Experimental animals
This research was performed on 55 adult male Wistar albino rats. Their average weight was 180 to 200 g. The animals were acclimatized for seven days before the experimental work began. The animal procedures were approved by the Faculty of Medicine's local Institutional Animal Ethics Committee, Tanta University, Egypt (Approval number: 34265/11/20).

Chemicals
Streptozotocin and oxymatrine (with a purity >98%) were bought from Sigma-Aldrich Saint Louis, Missouri, USA.

Experimental design
The rats were randomly divided into the following groups: Group I (Control group): included 15 rats that were divided into three equal subgroups of 5 animals each: Subgroup (Ia): received no treatment during the experiment.
Subgroup (Ib): received 1 ml of normal saline (0.9%NaCl), the diluting vehicle for oxymatrine. Subgroup (Ic): received a single intraperitoneal injection of 0.2 ml of 0.1 M sodium citrate, the diluting vehicle for streptozotocin.
Group II (Oxymatrine group): included 10 rats that received oxymatrine dissolved in normal saline at a dose 80 mg/kg/day by oral gavage for 8 weeks. 23 Group III (The streptozotocin-induced diabetic group): included 30 rats that were subjected to a single intraperitoneal injection of 50 mg/kg STZ freshly dissolved in cold 0.1 M sodium citrate buffer, pH 4.5. 24 Seventy-two hours after STZ injection, diabetes mellitus induction was confirmed by evaluation of the glucose levels in blood samples obtained from the tail vein using the ACCU-CHEK glucose meter (Roche Diagnostic Corporation, Indianapolis, IN). Only animals with blood glucose level of (≥250 mg/dl) were considered diabetic. 25 In this research, only 22 rats were proved to develop DM after STZ injection and were further subdivided into 2 experimental subgroups: Subgroup IIIa (Induced diabetic group): included 11 rats that received no additional treatment.
Subgroup IIIb (Induced diabetic-oxymatrine group): included 11 rats that received oxymatrine at a dose, route and duration similar to that of group II.
Twenty-four hours after the last dose of oxymatrine, the animals were weighed. The glucose levels in blood samples obtained from the tail vein were measured and the rats were anesthetized by ketamine (60 mg/kg i.p). 24 Animals were perfused with 2% paraformaldehyde prepared in 0.01 M phosphate buffer. The heads were then dissected to collect the cerebellar specimens. Parts of the cerebellum were used for the biochemical analysis of the oxidative stress and the other part was split into two pieces. One was processed for light microscopic study and the other for electron microscopic study.

Assessment of Oxidative Stress
Tissue malondialdehyde (MDA) levels, glutathione peroxidase (GP-x) activities, and superoxide dismutase (SOD) were determined according to the previous methods described by Uchiyama and Mihara, 26 Paglia and Valentine, 27 and Marklund and Marklund, 28 respectively.

Light microscopic studies
The cerebellar specimens from each animal were fixed in 10% buffered formalin solution for 24 hours, then dehydrated in graded alcohol and embedded in paraffin. Serial sections of 5 μm thickness were cut and the following techniques were applied: • Hematoxylin and eosin (H&E) stain: to illustrate the general histological features of the cerebellum. 29 • Immunohistochemical stains using streptavidin-biotin-peroxidase technique according to Shalaby et al. 30 The sections were incubated for 2 hours with Diluted primary antibody against:

Electron microscopic study
The processing of the cerebellar specimens for examination by transmission electron microscope was carried out according to Woods and Stirling. 31 Briefly, very small parts of the cerebellar tissue (1 mm 3 ) were fixed in 2.5% phosphate-buffered glutaraldehyde, post-fixed in 1% phosphate buffer osmium tetroxide, dehydrated in ascending grades of alcohol then immersed in propylene oxide and finally embedded in epoxy resin mixture. Ultrathin sections (80-100 nm thick) were cut, contrasted with uranyl acetate and lead citrate and examined by JEOL-JEM-100 transmission electron microscope (Japan) in Electron Microscopic unit, Faculty of Medicine, Tanta University, Egypt.

Morphometric study
Leica Qwin 500 image analyzer computer system was employed to measure: Ten non-overlapping random fields per slide from each rat in all studied groups were used in these measurements.

Statistical analysis
One-way analysis of variance and Tukey's procedure were employed to compare between the different groups. The values were expressed as mean ± SD. Differences were considered significant if the probability value (p value) was less than 0.05. 32

Results
Regarding rats' mortality during the experiment, two rats from the induced diabetic group (subgroup IIIa) and only one rat from the induced diabeticoxymatrine group (subgroup IIIb) died. Regarding the biochemical, histological and immunohistochemical findings, all subgroups (Ia, Ib, and Ic) of the control group, they exhibited no statistical differences, thus they were collectively referred to as the control group. Meanwhile, there were nonsignificant statistical differences between the control group (group I) and oxymatrine group (group II).

Body weight result
The induced diabetic group (subgroup IIIa) showed a significant decrement (P ≤ 0.001) in the body weight of rats (175.54 ± 12.88 gm) when compared with the control one (286.37 ± 32.78 gm). Administration of oxymatrine to the diabetic rats significantly minimized body weight loss induced by DM but it failed to normalize the weight of rats as it showed a significant decrement (P ≤ 0.001) (211.18 ± 9.12 gm) when compared with the control group (Table 1).

Biochemical results
The induced diabetic group (subgroup IIIa) exhibited a significant increment (P ≤ 0.001) in the blood glucose level (323.88 ± 20.68 mg/dl) when compared with the control one (104.43 ± 9.59 mg/dl). Administration of oxymatrine to the diabetic rats significantly minimized the blood glucose increase but it failed to normalize the level of blood glucose, thus it showed a significant increment (P ≤ 0.001) (265.98 ± 62.84 mg/dl) when compared with the control group (Table 1).

H&E-stained sections
The cerebellar sections of the control group displayed the normal histological architecture of the rat cerebellar cortex. It was formed of three layers; the molecular layer, the Purkinje cell layer that consisted of one row of large pearshaped cells with vesicular nuclei and prominent nucleoli and the granular cell layer that demonstrated densely packed rounded cells having darkly stained nuclei with cerebellar islands inbetween ( Figure 1A). Cerebellar sections from the induced diabetic group (subgroup IIIa) displayed numerous changes that were more obvious in the Purkinje cell layer. Purkinje cells were arranged in many layers with many neuroglial cells in-between ( Figure 1B). Most of them were irregular in shape with shrunken deeply stained nuclei ( Figure 1C). Vacuolation of the Purkinje cell layer ( Figure 1B & C), focal  Figure 1D). With regard to the molecular layer, it showed vacuolations in some sections ( Figure 1C). The granular cell layer showed presence of vacuoles ( Figure 1B). Some granule cells appeared darkly stained ( Figure  1D). Examination of cerebellar sections obtained from the induced diabetic-oxymatrine group (subgroup IIIb) revealed preservation of the normal cerebellar architecture as regards its three layers except for presence of few darkly stained shrunken Purkinje cells ( Figure 1E).
Statistical analysis of the mean number of Purkinje neurons/mm of the cerebellar lobules showed a significant decrement (P ≤ 0.001) (9.05 ± 1.53) in the induced diabetic group (subgroup IIIa) compared with the control group (21.12 ± 1.51). Whereas the induced diabeticoxymatrine group (subgroup IIIb) revealed a non-significant change (P ≥ 0.05) (19.50 ± 1.49) when compared to the control group (Table 2).

Immunohistochemical results
Examination of GFAP-immunostained cerebellar cortex sections of the control group displayed mild GFAP expression in the cytoplasm of the cell body and processes of the astrocytes of all cortical layers (Figure 2A). Whereas the GFAP expression in the astrocytes was intense in all cortical layers of the induced diabetic group (subgroup IIIa) ( Figure  2B). While the induced diabetic-oxymatrine group (subgroup IIIb) revealed moderate GFAP expression in the astrocytes of all cortical layers ( Figure  2C). Statistical analysis of the mean area percentage and the mean optical density of GFAP expression in the cerebellar cortex showed a significant increase (P ≤ 0.001) (30.02 ± 0.62, 27.20 ± 3.51 respectively) in the induced diabetic group (subgroup IIIa) when compared with the control one (9.70 ± 0.79, 11.44 ± 1.57 respectively). In regards to the induced diabetic-oxymatrine group (subgroup IIIb), there was a non-significant change (P ≥ 0.05) in the mean area percentage and the mean optical density of its GFAP expression (10.15 ± 0.43, 8.96 ± 1.27 respectively) when compared to the control group ( Table 2).
Examination of synaptophysin-immunostained sections of the cerebellar cortex of the control group showed weak expression of synaptophysin in the molecular and granular layers ( Figure 2D). While the cerebellar cortex sections of the induced diabetic group (subgroup IIIa) exhibited strong expression of synaptophysin in both molecular and granular layers ( Figure 2E). As regards the induced diabetic-  oxymatrine group (subgroup IIIb), it showed moderate synaptophysin expression in molecular and granular layers ( Figure 2F). Statistical analysis of the mean optical density of synaptophysin immunoexpression revealed a significant increase (P ≤ 0.001) (33.85 ± 0.73) in the induced diabetic group (subgroup IIIa) in comparison with the control group (12.53 ± 0.80). Meanwhile, the induced diabetic-oxymatrine group (subgroup IIIb) showed a significant increment (P ≤ 0.001) (15.43 ± 0.58) in comparison with the control group (Table 2).

Transmission electron microscopic results
Examination of the cerebellar cortex from the control group revealed regular arrangement of myelin sheaths in the molecular layer ( Figure 3A). The perikaryons of Purkinje cells in the Purkinje cell layer contained large euchromatic nuclei with prominent nucleoli, mitochondria and rER ( Figure 3B). The granular layer displayed two cell types: granule cells with heterochromatic nuclei surrounded by thin rim of cytoplasm and Golgi cells ( Figure 3C). The cerebellar cortex had many synapses in which multiple synaptic vesicles were present and each synaptic site showed dense presynaptic and postsynaptic membranes ( Figure 3D). The blood capillaries were enveloped by processes of astrocytes ( Figure  3E). On the other hand, sections from the induced diabetic group (subgroup IIIa) showed irregularity and splitting of the myelin sheaths in the molecular layer ( Figure 4A). Moreover, many Purkinje cells were shrunken and dark with ill-defined nuclei and dilated Golgi saccules. Vacuolation of the surrounding neuropil was also detected ( Figure 4B). Other Purkinje cells had irregular nuclei with dilated perinuclear cisternae, dilated rER and vacuolated cytoplasm ( Figure 4C). Cells in the granular layer appeared dark with irregular nuclear outlines and vacuolated cytoplasm. Some cells revealed shrunken nuclei ( Figure 4D). Multiple synapses containing numerous synaptic vesicles were demonstrated ( Figure 4E). In addition many processes of astrocytes related to blood capillaries were swollen and markedly enlarged ( Figure 4F). As regards the induced diabetic-oxymatrine group (subgroup IIIb), it displayed partial preservation of the cerebellar cortex structure. Most of the nerve fibers in the molecular layer showed regular arrangement of the myelin sheaths. However, few nerve fibers revealed irregularity of their myelin sheaths ( Figure 5A). Most of the Purkinje cells appeared normal ( Figure  5B). Others had dark cytoplasm ( Figure 5C). Also, most of the granule cells were normal, whereas some cells revealed rarefaction of their cytoplasm ( Figure  5D). Many synapses containing synaptic vesicles were present ( Figure 5E). The blood capillaries were surrounded by moderately enlarged processes of astrocytes ( Figure 5F).

Discussion
Diabetes mellitus represents a widespread metabolic disease with well-known side effects. Prolonged hyperglycemia in DM usually causes neurotoxicity in peripheral and central nervous systems. 24 In our research, we evaluated the role of oxymatrine in alleviating cerebellar cortex changes induced by DM in rats. In our study; STZinduced diabetic rats displayed the typical diabetic signs such as elevated blood glucose levels and body weight loss. This was in agreement with Choi et al. 33 who detected similar results. The current study revealed that DM induced structural changes in rats' cerebellar cortex as evidenced by light and electron microscopic results. These changes were more obvious in the Purkinje cell layer. The Purkinje cells were arranged in many layers. Most of them were shrunken with deeply stained nuclei. Others showed eosinophilic homogenization of their cytoplasm with fainting of the nuclei.
Moreover, focal disappearance of Purkinje cells was also detected. Results of the current work came in agreement with Sherif,34 and Hussein,35 in their studies on the cerebellum of diabetic rats. The hyperglycemia developed in DM induces oxidative stress which accounts for generation of reactive oxygen species (ROS) and lipid peroxidation. This usually leads to oxidation of proteins, damage to DNA together with peroxidation of the lipids present within the plasma membranes and causes an increase in neuronal cells death. 36 The condition of oxidative stress developed in diabetes mellitus also leads to glucose auto-oxidation, a decrease in tissues' glutathione concentration and depletion of the antioxidant enzyme activities with subsequent increase in the sensitivity of the brain tissue to oxidative damage. 37 This was in agreement with the significant increment in the MDA level and the significant decrement in the levels of GP-x and SOD that were reported in our research.
The Purkinje cells are considered the largest and the most obvious neuron in the cerebellum. 38 They are concerned with learning and motor function coordination, therefore, their damage causes motor disorders. 39 In the present research, a significant decrement in the Purkinje cells' number was detected in the induced diabetic group (group IIIa). This result came hand in hand with the results of previous research by Razi et al. 40 and Bak et al. 41 who reported a decrement in the number of Purkinje cells in their study on diabetic rats. Such decrease in number could be attributed to either programmed cell death or decreased neurogenesis in the CNS 6 or acidosis of the tissues 42 that all occur due to hyperglycemia. Meanwhile, exposure of Purkinje cells to the oxidative stress caused by DM causes their damage secondary to glycation and intracellular accumulation of α-synuclein; a neuronal protein associated with synaptic plasticity, neurotransmitters release, neurons differentiation and also neuronal viability regulation. 43,44 The multilayer arrangement of the Purkinje cells reported in this research could be attributed to focal crowding of the cells as a part of an adaptive mechanism of the Purkinje cells in order to reestablish connections with each other in order to perform their functions. 45 The vacuolations that were detected in the three layers of the cerebellar cortex in our study were described by other researchers as spongiform changes 46 which occurred due to loss of the cellular parts in the cerebellar cortex. 47 In this work, a significant increment in the mean area percentage of GFAP immunoreactivity in both molecular and granular layers was detected in the DM-induced group (group IIIa). GFAP is the main intermediate filament protein present in astrocytes and is considered a specific marker for them. The increase in the GFAP expression could be due to glial activation that occurred in association with the damage to the Purkinje cells secondary to DM. In this condition, the astrocytes enlarged in size and acquired thick tortuous processes and become strongly immuno-stained with the GFAP antibody. 48,49 This finding is referred to as reactive gliosis and coincided with the findings of El-Akabawy and ElKholy, 50 who attributed it to oxidative stress and changes in the blood glucose level. Moreover, Cerrato 51 reported that reactive astrocytosis was accompanied with Purkinje cells degeneration in a mouse model of spinocerebellar ataxia. In addition, other researcher illustrated that the increase in GFAP expression could be due to the formation of myeloarchitectonics and new synaptic connections. 34 In the current research, the impact of DM on synaptogenesis in the cerebellar cortex was evaluated. Synaptophysin is a presynaptic protein associated with synaptogenesis and thus it is employed as a sensitive marker for the synaptic density. 52 Our findings revealed that DM caused a significant increment in the immune-expression of synaptophysin in granular and molecular layers of the cerebellar cortex. This was supported by our electron microscopic results that revealed multiple synapses containing numerous vesicles in the diabetic rats. These results came in accordance with Sherif, 34 who detected similar results in his study on diabetic rats. Another previous study by Grillo et al. 53 also reported that DM caused an increased expression of synaptophysin in the rat's hippocampus and attributed that increase to either the active synaptogenesis with formation of new synapses or the redistribution of proteins in the synapses that remained. These alterations were followed by changes in learning and memory indicating that these new synapses were usually abnormal.
Our electron microscopic results revealed further ultrastructural changes such as irregularity and splitting of myelin sheaths in the molecular layer. This result came in agreement with Ozdemir et al. 54 who demonstrated separations and breaks of myelin sheaths in cerebellar cortex of diabetic rats. These alterations of the myelin sheaths were demonstrated to occur as a part of brain encephalopathy that develops in different brain regions secondary to DM. Disarrangement of the myelin sheaths could be explained according to other researchers 6,35 to be due to inhibition of oligodendrocytes' function, decreased levels of myelin-associated glycoprotein and also formation of antibodies against myelin basic proteins.
Dilatation of Golgi, rER, and perinuclear cisternae together with the vacuolations that were detected in our work was attributed to the effect of oxidative stress and lipid peroxidation induced by hyperglycemia developed in DM. This caused membranes' damage with subsequent increase in intracellular sodium and water content resulting in edema. 55 Moreover, condensation and shrunken nuclei of the granule cells in the granular layer were considered as a form of injury in which DNA was compacted resulting in clumping of chromatin with stoppage of transcriptional activities. 56 Astrocytes are essential structural components of the blood-brain barrier (BBB), controlling the bidirectional passage of molecules and cells between blood and CNS through various pathways. Their end-feet wrap the entire vascular surface. Thus, their dysfunction is usually associated with BBB disturbance which occurs in many neurological disorders. 57 Previous studies on demyelinating disease reported that astrogliosis is associated with degeneration of junctional endothelial proteins as well as extensive disruption of the end-feet binding them to the vasculature with subsequent loss of the integrity of BBB structure. 58,59 Astrocytes are connected to the endothelial cells through gap junctions which are essential for maintaining BBB function. The expression of connexin 43 (Cx43), the key structural part of astrocytic gap junctions, has been found to be reduced in demyelinating disease. Cx43 deficiency causes astrocyte end-feet edema, which causes detachment from the basal lamina and degradation of the BBB, which has been attributed to BBB disruption. Moreover, the depletion of Cx43 was followed by a patchy loss of the water channel protein Aquaporin-4 (AQP4) at the astrocyte end-feet. Since this is also an important part of the BBB structure, it is possible that having both Cx43 and AQP4 astrocytopathy leads to further BBB injury. 60 This could account for the enlarged processes of astrocytes around blood capillaries that were detected in our study. Also, Hernandez-Fonseca et al. 6 demonstrated perivascular swelling of astrocytes in diabetic rats which accounted for disruption of the blood brain barrier functions. Moreover, a previous research on animals had depicted that diabetes had harmful effects on the blood brain barrier which could account for the neurological complications occurred in this condition. 61 In the current work, administration of oxymatrine in diabetic rats in group IIIb (diabeticoxymatrine group) greatly preserved the cerebellar cortex structure and protected against the changes induced by DM. Oxymatrine was proved to have antioxidant properties evidenced by enhanced production of antioxidant enzymes such as catalase, GP-x and SOD. 62 This was in agreement with the significant increment in the levels of GP-x and SOD that were detected in our study. The antioxidant effect of oxymatrine was also reported in various tissues, such as brain, 63 heart, 64 liver 65 and kidneys. 66 Moreover, in their study on rats' brain following intracerebral hemorrhage, Huang et al. 60 illustrated that oxymatrine prevented brain's oxidative injury through reducing production of 12/15-LOX proteins which play an essential role in developing oxidative damage in the brain tissues. In addition, Huang et al. 67 reported that oxymatrine improved hippocampal histological structure and cognitive brain function in diabetic rats through suppression of the oxidative stress through reduced expression and activity of NOX2 and NOX4 which are considered main sources of ROS in brain tissue that play a role in neuronal damage. Other studies also demonstrated that oxymatrine increased activity of SOD and decreased synthesis and release of lipid peroxides and malondialdehyde, contributing to its antioxidant activities. 68,69 Thus, the protective effect of oxymatrine that was detected in group IIIb of the current study could be attributed to its antioxidant properties that minimized the impacts of the oxidative stress induced by DM and so exerted a neuroprotective effect on the cerebellar cortex of diabetic rats.
In addition to its antioxidant effects, accumulated data demonstrated that oxymatrine had anti-apoptotic activity in different models of diseases. In most of these models, oxymatrine protected the tissues and organs from damage by minimizing apoptosis through enhancing expression of anti-apoptotic genes and reducing the apoptotic ones. 70,71 A previous study of Zhang et al. 72 revealed that oxymatrine prevented apoptosis of the hepatocytes and protected against acute liver failure in rats. Also, another study proved that oxymatrine protected the brain against hypoxic-ischemic brain injury by minimizing apoptosis in neonatal rats. 73 Moreover, oxymatrine prevented neuronal cells from apoptosis that occurs in the rat brain after cerebral hemorrhage 74 and also inhibited apoptosis of neuronal cells in diabetic rats through reduction of both expression and activity of cleaved caspase-3 protein. 67 In addition, it was documented that oxymatrine had a regulatory role in DM and succeeded in reducing glucose blood level in diabetic rats. 67 A previous work also showed that oxymatrine preserved the structure of the pancreatic islets, enhanced insulin production and sensitivity and reduced both hyperglycemia and hyperlipidemia in streptozotocin-induced diabetic rats. 22

Conclusion
Diabetes mellitus induced histopathological alterations in the cerebellar cortex of rats. Oxymatrine has good impacts and has some neuroprotective effect on DM-induced cerebellar cortex changes. Thus, it is recommended for diabetic patients to administer oxymatrine as an adjuvant therapy.

Disclosure statement
The authors declare no conflicts of interest.

Funding
This work was not funded by any grant sponsors or organizations.

Author Contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by [

Ethical statement
All animal work was conducted under the guidelines for the use of animals in research established by the local ethical committee of the Faculty of Medicine, Tanta University, Egypt (Approval number: 34265/11/20).

Data Availability
Datasets are available in a public repository that assigns persistent identifiers to the datasets.